ELECTROCATALYTIC UPGRADING OF BIOMASS PYROLYSIS OILS TO CHEMICAL
AND FUEL

By

Chun Ho Lam

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirement
for the degree of
Chemistry – Doctor of Philosophy
2014

ABSTRACT
ELECTROCATALYTIC UPGRADING OF BIOMASS PYROLYSIS OILS TO CHEMICAL
AND FUEL
By
Chun Ho Lam
The present project’s aim is to liquefy biomass through fast pyrolysis and then upgrade
the resulting “bio-oil” to renewable fuels and chemicals by intensifying its energy content using
electricity. This choice reflects three points: (a) Liquid hydrocarbons are and will long be the
most practical fuels and chemical feedstocks because of their energy density (both mass and
volume basis), their stability and relative ease of handling, and the well-established infrastructure
for their processing, distribution and use; (b) In the U.S., the total carbon content of annually
harvestable, non-food biomass is significantly less than that in a year’s petroleum usage, so
retention of plant-captured carbon is a priority; and (c) Modern technologies for conversion of
sunlight into usable energy forms—specifically, electrical power—are already an order of
magnitude more efficient than plants are at storing solar energy in chemical form.
Biomass fast pyrolysis (BFP) generates flammable gases, char, and “bio-oil”, a viscous,
corrosive, and highly oxygenated liquid consisting of large amounts of acetic acid and water
together with hundreds of other organic compounds. With essentially the same energy density as
biomass and a tendency to polymerize, this material cannot practically be stored or transported
long distances. It must be upgraded by dehydration, deoxygenation, and hydrogenation to make
it both chemically and energetically compatible with modern vehicles and fuels. Thus, this
project seeks to develop low cost, general, scalable, robust electrocatalytic methods for reduction
of bio-oil into fuels and chemicals.

ACKNOWLEDGEMENTS

I would like to express my deepest appreciation to my graduate school advisor, Prof. Ned
Jackson, who has been a very patient scholar nurturing my academic growth. Ned taught me a lot
– from designing experiments for proofing hypothesis to the elegant use of the English language
for article composing (he even teaches me Latin and French occasionally). I often admire the fact
that Ned can use every word to its full potential to produce a flawless convincing article easily at
the tip of his fingers. With an in depth knowledge on almost every subjects, Ned cannot be
replaced with a better advisor. Aside from being a great writer, Ned is also an excellent teacher.
He would spent enough time for anyone to help them to understand (and think forward!) on
chemistry problems. This explains why undergraduate students from his classes always ask for
research opportunities in his research group. I am very fortunately that he entrusted me with his
carefully selected students, letting them augment my data throughputs and the quality of my
research. He taught me the patience required to mentor these students as well as the wisdom
needed to guide them forward. I am forever grateful to have him to be my research advisor for
the past 5 years – it has been a wonderful experience.
Thanks to Ned, I had the opportunity to work with over 30 undergraduate students over
the past five years. Of these many individuals, Miss Kelsey Longe, Miss Michaelyn Lux, Miss
Carly Houdek, Miss Christy Lowe, Miss Laura Schultz, Mr. Michael Caldwell, Mr. Jorden
Rayburn, Mr. Andrew Henika, Mr. Ryan Barlow and Mr. Jacob Kunan stood out the most. These
students took their own time and intellectual power to assist me in my graduate work at MSU.
Special mention to Kelsey, who had been working with me for her entire 4 years of college: 2

iii

years as a professorial assistant and 2 years as a volunteer, she continued to work dedicatedly
years after years on different projects.
I would also like to thank my co-advisor, Prof. Chris Saffron, and my colleagues in the
Jackson group. Dr. Saffron has been a very kind mentor to me in proofreading my work, giving
me insightful feedbacks, and introducing me to various connections during academic conferences.
His kindness and dedication are the two of his many positive qualities I looked up to.
Collaborations with my colleagues have been some valuable experiences to me; learning from
other members from different projects often inspires new ideas.
Finally, I would like to thank my family and friends. My parents never had the pleasure
to attend college, but they realize the importance of education, and thus providing an enormous
amount of support both emotionally and financially for me to complete this PhD degree in
Chemistry. I can never reply what they have done for me; it is the best gift I can ever receive
from anyone in my life. My friends outside of the program are also very supportive, my graduate
life would not be as smooth without them.
There are countless more individuals I would like to mention – perhaps another thesis
worth of acknowledgement can be written. I apologize if I missed out anyone on this page but I
promise you that your contributions have not been forgotten. So, thank you!

iv

TABLE OF CONTENTS

LIST OF TABLES…………………………………………………………………….…………vii
LIST OF FIGURES……………………………………………………………………………….x
LIST OF SCHEMES……………...………………………………………………………….…xvi
CHAPTER 1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.7.1
1.7.2
1.7.3
1.7.4
1.7.5
1.7.6
1.8
1.8.1
1.8.2

Literature Review on Common Renewable Mobile Fuel Candidate.................... 1
The Importance of Hydrocarbon ............................................................................. 1
Emergence of Hybrid Vehicle ................................................................................ 3
Hydrogen as a Fuel ................................................................................................. 5
The Biofuel: E85 ..................................................................................................... 9
Biofuel production via Fischer-Tropsch Process .................................................. 11
CO2 Recycling and Carbon Capture & Storage (CCS) ......................................... 13
Biomass as a Renewable Energy Feedstock ......................................................... 16
The Need of Renewable Energy ........................................................................... 16
Common Pathway for Biomass to Biofuels .......................................................... 18
Biomass as Feedstock ........................................................................................... 19
Biomass Fast Pyrolysis (BFP) .............................................................................. 20
Nature of Bio-oil ................................................................................................... 21
Bio-oil Upgrading via ECH .................................................................................. 24
Chapter Summary ................................................................................................. 26
Summary Table of the Discussed Alternative Fuel Candidates............................ 28
Goal of this Research Project ................................................................................ 31

CHAPTER 2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.10.1
2.11
2.11.1
2.11.2
2.11.3
2.12

Literature Review on Electrocatalytic Hydrogenation ..................................... 33
Introduction ........................................................................................................... 33
Catalytic Hydrogenation for Bio-oil Upgrading and its Concerns ....................... 33
From Catalytic Hydrogenation to Electrocatalytic Hydrogenation ...................... 35
Brief Introduction to Electrocatalysis ................................................................... 36
Current Efficiency................................................................................................. 38
Use of Cationic Surface Active Reagent in ECH ................................................. 39
Selection of Electrodes and the volcano plot ........................................................ 40
The use of Raney-Nickel and Other Skeletal Metals ............................................ 43
Preparation of Raney-Nickel Electrode ................................................................ 44
Introduction of the Cobalt Phosphate Water Oxidation Catalyst.......................... 45
Preparation of the Cobalt Phosphate Water Oxidation Catalyst ........................... 46
Overview of System Setup.................................................................................... 47
Constant Voltage vs. Constant Current Electrolysis ............................................. 47
Three Electrode System ........................................................................................ 50
Two Electrode System .......................................................................................... 51
Summary ............................................................................................................... 52
v

CHAPTER 3 Electrocatalytic Upgrading of Model Lignin Monomers with Earth Abundant
Metal ........................................................................................................ 55
3.1
Introduction ........................................................................................................... 55
3.2
Electrocatalytic Hydrogenation of Acetophenone in Undivided cell ................... 56
3.2.1
Experimental Procedures ...................................................................................... 57
3.2.2
Analysis................................................................................................................. 58
3.2.3
Results and Discussion ......................................................................................... 58
3.3
Reduction of Phenol in Undivided Cell ................................................................ 60
3.3.1
Experimental Procedures ...................................................................................... 60
3.3.2
ECH of Phenol ...................................................................................................... 62
3.3.3
Results and discussion .......................................................................................... 62
3.4
Reduction of Guaiacol in Divided Cell:................................................................ 63
3.5
Experimental Procedures ...................................................................................... 64
3.6
Results and discussion .......................................................................................... 65
3.6.1
Mechanistic Investigation of Guaiacol Demethoxylation..................................... 68
3.6.2
Material Balance ................................................................................................... 74
3.6.3
Current Efficiency................................................................................................. 77
3.6.4
Mechanistic Insight ............................................................................................... 79
3.7
Summary ............................................................................................................... 83

CHAPTER 4 Electrocatalytic Hydrogenation of Aliphatic Model Compounds in a Solid
Polymer Electrolyte Electrolyzer (SPEE) ...................................................... 85
4.1
Introduction ........................................................................................................... 85
4.2
Preliminary Experiment I – SPEE Designs and Construction .............................. 88
4.2.1
Preparation of the Membrane Electrode Assembly (MEA) .................................. 89
4.2.2
Construction of the 1st Generation SPEE – The Prototype ................................... 90
4.2.3
Construction of the 2nd Generation SPEE ............................................................. 92
4.2.4
Construction of the 3rd Generation SPEE ............................................................. 93
4.3
Preliminary Experiment II – Exploring the 4thSPEE ............................................ 95
4.3.1
Construction of the 4th Generation SPEE – The Current Version ........................ 95
4.3.2
Direct Metal deposition on Nafion Membrane ..................................................... 96
4.3.3
MEA Preparation with Different Reductants ........................................................ 97
4.3.4
Experimental Procedures ...................................................................................... 98
4.3.5
Results and Discussion ......................................................................................... 99
4.3.6
Formation of Ag-Cu Bimetallic Layer Catalyst.................................................. 101
4.4
New Perspective on the SPEE and C.E. % Improvement Effort ........................ 105
4.4.1
Experimental Procedures .................................................................................... 105
4.4.2
ECH of acetol in the presence of acetic acid ...................................................... 110
4.4.3
Effect of Anolyte on Current Efficiency Improvement ...................................... 113
4.4.4
Influence of Metal Ion Re-deposition ................................................................. 115
4.4.5
Mesh Size Effect on Current Density ................................................................. 119
4.4.6
The Effect of Current Density on Current Efficiency ......................................... 120
4.4.7
Isopropyl Alcohol Study ..................................................................................... 122
4.5
Summary ............................................................................................................. 123
vi

4.5.1

Miscellaneous Ideas and Future Directions ........................................................ 126

CHAPTER 5
5.1
5.2
5.3
5.3.1
5.3.2
5.4
5.5
5.6

Pedagogy: Electrocatalytic Hydrogenation of Acetol with US. Coins ............. 128
Introduction ......................................................................................................... 128
Hazards and Disposals ........................................................................................ 131
Experimental Procedures .................................................................................... 131
Materials and Preparations .................................................................................. 131
Experiment .......................................................................................................... 132
Current Efficiency Analysis ................................................................................ 134
Results and Discussion ....................................................................................... 136
Summary ............................................................................................................. 140

CHAPTER 6
6.1
6.1.1
6.1.2
6.2
6.2.1
6.2.2
6.3
6.4
6.5

Conclusions ............................................................................................. 141
Project I - ECH of Bio-oil Model Compound, Guaiacol to Chemicals and Fuel 141
Improvement I – Electrode Material Synthesis .................................................. 142
Improvement II – Substrate Scope and Mechanistic Studies.............................. 144
Project II – Development of SPEE system for Bio-oil Hydrogenation .............. 144
Improvement I – Deposition of Ruthenium on Stainless Steel Mesh ................. 145
ECH of Aqueous Bio-oil using Ru-SPEE ........................................................... 147
New Cell Design – The High Pressure Tube Reactor......................................... 149
Overall Progress Summary ................................................................................. 153
Final Remarks – The Carbon Balance ................................................................ 154

REFERENCES .............................................................................................................. 156

vii

LIST OF TABLES

Table 1-1: Composition comparison between bio-oil and crude oil are shown. Note: actual
composition of bio-oil varies depending on the biomass feedstock. ......................... 22
Table 1-2: Summary of each discussed alternative energy candidate. 5 major factors: adaption to
the current fueling structure, relationship to fossil energy, Cost to consumers,
sustainability, and the environmental impacts are explored. ..................................... 29
Table 2-1: Terms comparison between chromatography and electronic system .......................... 48
Table 2-2: An example to illustrate how the absolute potential on individual electrode changes
during a constant voltage electrochemical reaction. Note: Ta and Tb describes
different stage during the ECH .................................................................................. 49
Table 3-1: Condition for sacrificial and non-sacrificial of Acetophenone (AP) to 1-phenylethanol
(PE) at constant current (50mA) ............................................................................... 57
Table 3-2: Deposition rate and deviations on Ni-Al alloy deposition. Mono side physical area
(2.5 x 2.5 cm) of the used for the J value calcuation ................................................ 61
Table 3-3: Reduction of 20 mM of Guaiacol in 0.1 M pH 8.2 borate buffer at 82 ± 2 oC, and j= 8
mA cm-2 (current density was calculated by applied current (50 mA) divided by the
electrode area that is facing the anode 2.5 x 2.5 cm). G: Guaiacol, P: Phenol, 2M, 2Methoxycyclohexanol and C: Cyclohexanol ............................................................ 67
Table 3-4: Values are percentages relative to starting material buffer concentration, 12.1 ± 1.5
mM as determined by gas chromatography. Reductions employed a Raney-Nickel
cathode in 30 ml of 0.1 M pH 8.0 potassium borate buffer, and a stainless steel
anode coated with cobalt phosphate catalyst in 30 ml of 0.1 M pH 7 potassium
phosphate buffer. Reaction started with 60 minutes of pre-electrolysis and followed
immediately by compound addition and then run for 6 hours at 50 mA (8 mA cm-2)*
at 75 ˚C. Total current passed in this time represents 4.1 x the amount required for
substrate reduction to ROH and cyclohexanol. b) Cis and trans product are in equal
amounts. c) Only a single peak was observed by GC. *Note: current density is
calculated based on the single side electrode facing the membrane and anode
compartment. ............................................................................................................. 70
Table 4-1: 10 ml of 5% acetol (924.5 mM) dissolved in deionized water undergo ECH to
propylene glycol after 6 hours at a constant current 98 ± 2 mA, at room temperature
and ambient pressure. All current collectors are mesh 200 in size, and had been
cleaned with hexane, acetone and water prior to use. Stainless steel 304 (Mesh 200)
was used in a 0.1M pH 7.0 phosphate buffer as an anode for all trials. Material
viii

balance was calculated by summing acetol and propylene glycol alone. *Initial
acetol concentration for the “copper with 5% acetic acid” trial was 716 mM ........ 107
Table 4-2: ECH of 10 ml of 5% acetol with 5% (v/v) glacial acetic acid added. Condition: copper
mesh with various mesh size at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 oC)
for 6 hours. Note: mesh size 2.3 x 2.3 cm = 5.29 cm2............................................ 119
Table 4-3: ECH of 10 ml of 5% acetol with 5% (v/v) glacial acetic acid added. Condition: Monel
400 mesh 200 at different current density (mA cm-2) 1 atm at room temperature (22
± 1 oC). Note: charge passed was calculated by the summation of multiplying current
level (recorded hourly) with the time interval between sampling time. Note: current
density was calculated based on the cut dimension of 2.3 x 2.3 cm = 5.29 cm2 ..... 120
Table 4-4: Factors that enhance current efficiency towards organic substrate hydrogenation ... 125
Table 5-1: Results of ECH of acetol in various conditions. All trials were run using a penny as
the anode at room temperature. Note: results were determined by 1H-NMR peak
integration to verify the formation of propylene glycol (See Figure 5-6). 1) 3 V d.c.
power supply; 2) 3 V (two AA cells in series) batteries; 3) 3 V d.c. power supply
with bag/coins in an ice bath. .................................................................................. 135
Table 6-1: atomic % from EDX analysis on the surface of the electrodes, SS-Ni-Pd and SS-NiPd-Ru, showing palladium and ruthenium were deposited onto the surface. Note: the
remaining percentage is made of nickel and chloride. ............................................ 147

ix

LIST OF FIGURES

Figure 1-1: Comparison of the specific energy (MJ kg-1) of various common energy carrying
chemicals. Hydrogen has a specific energy of 124 MJ kg-1 and its scale is associated
with the secondary axis on the right. ........................................................................... 2
Figure 1-2: General operating principle of hydrogen fuel cell ....................................................... 6
Figure 1-3: Comparison of the overall carbon and energy balance in the US transportation sector
and harvestable biomass. Red bars compare the weight of the petroleum used, and
the weight of the harvestable biomass. Blue bars compare the actual carbon content.
Green bars compare the energy spent annual, and the harvestable amount of energy
from biomass. Projections are made based on the available biomass by 2030. ........ 17
Figure 1-4: General chemical distribution of organic substrates found in bio-oil. Percentage is
given percent of mass. ............................................................................................... 23
Figure 1-5: Phenol is used as a model compound to illustrate the increases in specific energy (MJ
kg-1) upon hydrogenation and deoxygenation. Note: higher heating values mean the
heat of combustion is calculated with the heat of formation of water in liquid state,
which is lower in energy compared to gaseous state, and thus gives a greater value of
heat of combustion (heating values). ......................................................................... 25
Figure 2-1: Cetyltrimethylammonium bromide (CTAB) is used as an example to illustrate the
positively charged hydrophilic head and hydrophobic end (the hydrocarbon tail)
nature of surfactant .................................................................................................... 40
Figure 2-2: The volcano plot of the common metals used in electrocatalysis. ............................. 42
Figure 2-3: Co-deposition of Ni-Al alloy onto the electrode surface. Condition: Preparation of
the Ra-Ni cathode uses the Lessard method of trapping nickel-aluminum alloy
particles in an electrodeposited nickel matrix. .......................................................... 45
Figure 2-4: Three electrode system where the reference electrode is referencing the cathode
(working), substrate A is oxidized on the anode surface and substrate B is reduced
on the cathode surface. The separator, typically a glass frit or Nafion membrane,
may or may not be necessary to the system. Note: V = Voltmeter, to measure the
voltage between the reference electrode and the working electrode. A = Ammeter, to
measure the current at any time during the reaction ................................................. 50
Figure 2-5: Two electrodes with the cathode (working) and the anode (counter); substrate A is
oxidized on the anode surface and substrate B is reduced on the cathode surface.
Separator is optional, and is typically a proton exchange membrane or glass frit.
Note: V = Voltmeter, to measure the voltage between the reference electrode and the
x

working electrode. A = Ammeter, to measure the current at any time during the
reaction ...................................................................................................................... 51
Figure 3-1: Electrocatalytic Hydrogenation of acetophenone in an undivided cell with different
electrodes ................................................................................................................... 56
Figure 3-2: Reduction of a0;cetophenone with different electrode at room temperature ............. 58
Figure 3-3: Raney Nickel cathode before (left) and after (right) leaching of Al with NaOH at 70
o
C for 7 hours. ........................................................................................................... 62
Figure 3-4: Electrocatalytic hydrogenation of Guaiacol and the possible product and their
annotations................................................................................................................. 65
Figure 3-5: Electrocatalytic Hydrogenation of Guaiacol in high pressure setting (Left) the
physical setup of the overall system (Right) Schematic depiction of the reaction
chamber ..................................................................................................................... 65
Figure 3-6: Electrocatalytic hydrogenation (ECH) of organic molecule in a divided cell separated
by a Nafion 117 proton exchange membrane. ........................................................... 66
Figure 3-7: Depiction of CTAB at different concentration in solution. Low and optimal will
cause CTAB to form on adsorption on surface while medium high to high will
encourage formation of michelle and may encourage diffusion of organics through
the cationic exchange membrane (Nafion) ................................................................ 68
Figure 3-8a. (upper) Trial a: 1:1:1 mixture of guaiacol (---), 2-ethoxyphenol (---), and 2isopropoxyphenol (---) undergoes electrocatalytic hydrogenation (ECH) under the
same condition described in table 1. All three reactants form a common intermediate,
phenol (---), which is further hydrogenated to cyclohexanol (---). Only traces of the
alkoxycyclohexanol products of direct hydrogenation were detected. Figure 3-8b
(lower) Trial b: 1:1:1 mixture of guaiacol (---), 3-methoxyphenol (---), and 4methoxyphenol (---) undergoing ECH as above.
Formation of 3methoxycyclohexanol (---) and 4-methoxycyclohexanol (---) is observable. Phenol (--) and cyclohexanol (---) were also formed as the major products. Curves are
polynomial fits included to guide the eye. ................................................................ 72
Figure 3-9: ECH of Syringol (IUPAC: 2,6-dimethoxyphenol) in 30 ml of 0.1M pH 8.0 Borate
Buffer with 0.5mM CTAB at 75 oC. Syringol(---), cyclohexanol(---), guaiacol(---),
phenol(---) and current efficiency(---), vertical axis on the right .............................. 73
Figure 3-10: (a) Reactants, products and C.E. (%) data from ECH at 50 mA (8 mA cm-2)* and 75
˚C of 30 ml 18 mM guaiacol in 0.1 M pH 8.0 potassium borate buffer: guaiacol(---),
phenol(---), methanol(---), cyclohexanol(---), and material balance (---). ................ 74
Figure 3-11: Adsorption study to investigate the cause of material loss. ..................................... 76

xi

Figure 3-12: Diffusion test on guaiacol and phenol under experimental condition. Raney nickel
electrode is substituted with a plain nickel bar to prevent the hydrogenation of the
substrates. .................................................................................................................. 77
Figure 3-13: Total charged passed of the reaction vs. amount of charge spent on guaiacol
reduction for the reaction shown in Figure 3-10. ...................................................... 78
Figure 3-14:Proposed mechanism for the alkoxy-demethoxylation. Scheme 1 has been
eliminated with 1H-NMR study. Scheme 2 and 3 are the probable route for the
reaction. ..................................................................................................................... 79
Figure 3-15: Competition study of 10 mM guaiacol in the presence of various amount of p-cresol.
Upper left (a) guaiacol analysis, upper right (b) phenol analysis, lower left (c)
cyclohexanol analysis and lower right (d) p-cresol and 4-methylcyclohexanol
analysis: With 20 mM of p-cresol added (----), with 10mM of p-cresol added (----),
and no p-cresol added (----). Legends are same across all figures, each figure focuses
on the specific compound of interests specified on the vertical axis. ....................... 81
Figure 3-16: Possible mechanism for the de-alkoxylation of 2-alkoxyphenol ............................. 83
Figure 4-1: Schematic diagram to show the (Left) operating principle of a chemical (H2) fuel cell
and (Right) how the SPEE operates. Note: App. is short for electronic applications 86
Figure 4-2: Solid Polymer Electrolyte Electrolyzer processing unit that connects to the fast
pyrolysis unit. Bio-oil can be upgraded immediately upon exit................................ 88
Figure 4-3: Schematic drawings of the 1st SPEE cell made of two identical pieces as shown in the
figure on the right, these two halves are held together with a cell metal clip. .......... 90
Figure 4-4: Schematic drawings of the 2nd SPEE cell which combines conventional dual
compartment electrochemical cells with SPEE concepts together. ........................... 92
Figure 4-5: Schematic drawings of the 3rd SPEE cell – 2 pieces of glass plate that host the MEA
supported by a pair of Parafilm gasket and Carpenter clamp. ................................... 94
Figure 4-6: Production of furfuryl alcohol from furfural ECH in SPEE. (Left) MEA was prepared
with sodium borohydride (Right) MEA was prepared with hydrazine. .................... 99
Figure 4-7: Used Ag-MEA after 3 uses of electrolysis. (Left) a used Ag-MEA prepared with
sodium borohydride, and (right) a used Ag-MEA prepared with hydrazine.
Conductivity measure was taken using multimeter for both membranes: .............. 100
Figure 4-8: Visual comparison between (left) before, and (right) after copper plating onto the
Ag-MEA .................................................................................................................. 101
Figure 4-9: (Right) ECH of 10 ml of 10 mM furfural. Condition: Ag is deposited on Nafion using
1% Hydrazine solution. Reactions were run at 98 ± 2 mA at 1 atm at room
temperature (22 ± 1 oC) for 6 hours. Note: Cu-Ag is prepared by plating copper onto
xii

the Ag-MEA electrochemically using 50 mM CuSO4 at 98 ± 2 mA for 5 minutes.
(Left) the voltage log for the reactions displayed on the right ................................ 102
Figure 4-10: Reduction of acetol in examining the necessity of the silver foundation and the
importance of the current collector. ........................................................................ 104
Figure 4-11: Solid Polymer Electrolyte Electrolyzer (SPEE) parts, and the setup ..................... 106
Figure 4-12: (Left) typical experimental setup of a SPEE. (Right) close-up to a SPEE with
alligator clips and silicon tubing connected ............................................................ 106
Figure 4-13: Proposed reaction pathways of acetol. Candidates are proposed based on the
observed molecular weight from aqueous phase MS analysis. Percentages of the
compounds are given based on the relative intensity of the peak. .......................... 109
Figure 4-14: ECH of 10 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added. Condition:
copper mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 oC) for 6 hours.
Copper mesh is being reused multiple times and compared with the Monel 400 mesh
200 Note: reaction is evaluated using the cumulative current efficiency, each point
accounts for the product accumulated previously. For equation, see Scheme 2-2 .. 111
Figure 4-15: SEM examination on copper mesh before (Top, Left) and after (Top, Right) 6 hours
of electrolysis. New Stainless Steel (Bottom, Left) and used stainless steel (Bottom,
right). Samples were washed with water, acetone and chloroform to remove organic
residues. Then they were dried in vacuum desiccator for 2 hours before SEM
examination at 12kV and 10 mm working distance under ultrahigh vacuum. ........ 112
Figure 4-16: ECH of 10 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added. Condition:
Monel 400 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 oC) for 6 hours. The
anode compartment conditions are listed in the legend .......................................... 113
Figure 4-17: ECH of 100 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added. Condition:
Monel 400 mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 oC) with a
platinum mesh anode in deionized water for 6 – 12 hours. Metal salts were added at
the beginning of the experiment. ............................................................................. 116
Figure 4-18: ECH of 100 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added. Condition:
Monel 400 mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 oC) with a
platinum mesh anode in different anolytes for 12 hours. ........................................ 117
Figure 4-19: ICP-OES analysis of nickel and potassium in the cathode solution for trials
described in Figure 4-18. Samples were diluted 125 times with 2% nitric prior
analysis (0.2 ml of sample in 25 ml of 2% nitric acid). .......................................... 118
Figure 4-20: graphical interpretation of A) at low current, hydrogen adsorption is low, resulting
in low hydrogenation efficiency. when the applied potential is too low (or too high),
energy of the electron and the LUMO is mismatched, causing an inefficient transfer
and lower the C.E. B) at moderate level of current, surface is filled with enough
xiii

adsorbed hydrogen and can fit in organics to achieve hydrogenation. C) at high
current, surface is saturated with adsorbed hydrogen and disfavors organic
adsorption. H2 formation becomes dominated result. energy of the transfer electron
matches the LUMO of the pi bond to give an efficient electron(s) transfer ........... 121
Figure 4-21: ECH of 10 ml of 5% acetol in water or isopropanol, with and without acetic acid.
Condition: Monel 400 mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1
o
C) for 6 hours. ........................................................................................................ 122
Figure 5-1: Schematic diagram of ECH of unsaturated organic molecules (a-d) and hydrogen
evolution (e -f) ......................................................................................................... 130
Figure 5-2: (left) Experimental setup of ECH of acetol with coins. A copper coin anode is
clipped with red wire, and the nickel coin cathode is clipped with the black wire. An
extra coin is added on each side to act as a current collector and provide enough
pressure for the clamp (right). Setup is in a small sample bag to prevent filter paper
drying....................................................................................................................... 132
Figure 5-3: Electronic setup of the experiment. Note: Drawing is not to scale .......................... 133
Figure 5-4: Color changes of Q.B. solution in response to concentration of acetol. From left to
right: 100:0, 80:20, 60:40…..0:100 acetol (5% v/v): propylene glycol (5% v/v) ... 134
Figure 5-5: Change in absorbance of QB solution in response to % of acetol in standard solution.
Note: students can take readings in transmittance and calculate the absorbance using
the equation: A = 2 – log (%T)................................................................................ 134
Figure 5-6: 1H -NMR spectrum (Red) propylene glycol standard (Orange) ECH of acetol with
penny for 10 minutes; low resolution of the orange spectrum was due to Cu2+ ion
contamination. (Green) acetol standard (Blue) experiment 10 minutes with no
electricity passed ..................................................................................................... 135
Figure 5-7: An example of current-time profile. Total charge passed (Ctotal) can be estimated by
calculating area of the curve. Area under the curve can be estimated by simple
summing of the trapezoidal areas. ........................................................................... 136
Figure 5-8: (left) SEM spectroscopic analysis of plain penny surface, and (right) the black copper
layer after reaction ................................................................................................... 136
Figure 5-9: Change in absorbance of QB solution in response to % of acetol in standard solution.
................................................................................................................................. 138
Figure 6-1: ECH of guaiacol using Raney-Nickel cathode. Reaction proceeds with excellent
selectivity to form cyclohexanol as the sole product, and does not form the planar 2methoxycyclohexanol .............................................................................................. 142
Figure 6-2: Multiple layers of metals plated onto stainless steel to support the deposition of
ruthenium................................................................................................................. 146
xiv

Figure 6-3: GC-MS results on the aqueous fraction of bio-oil before hydrogenation in SPEE cell.
Temperature program: 50 oC hold for 3 minutes, then 15 oC/minute until 270 oC and
hold for 5 minutes.................................................................................................... 148
Figure 6-4: GC-MS results on the aqueous fraction
of bio-oil after hydrogenation in SPEE
cell. Temperature program: 50 oC holds for 3 minutes, then 15 oC/minutes until 270
o
C and hold for 5 minutes
............................................................ 148
Figure 6-5: HPLC-Flow style cell anatomy ................................................................................ 150
Figure 6-6: the general overall setup of the Flow cell reactor .................................................... 150
Figure 6-7: Prototype testing of the flow cell reactor. Reaction condition: 5% of acetol dissolved
in 0.25 M of ammonium acetate, 50 mA (117.9 mA cm-2), room temperature, 1 ml
min-1. Anolyte was the platinum wire inside of 0.1 M H2SO4 solution. 0.01 ml of
sample was taken hourly and was mixed with 0.75 ml of D2O for H-NMR analysis.
The absolute concentration of product was calculated by the internal concentration
reference, 1,4-dioxane, that was pre-dissolved in the D2O solvent. Note: the current
efficiency is calculated based on the charge passed and product formed hourly, NOT
accumulatively......................................................................................................... 152
Figure 6-8: Overall time summary of the progress of this research project. .............................. 153

xv

LIST OF SCHEMES

Scheme 1-1: General descriptive chemical equation describes common biomass to biofuel
processing pathways. Fermentation and Gasification both disposes 33% of the
carbon content as CO2 while fast-pyrolysis retains most carbon in the form of biooil: 70% bio-oil, 15% bio-char, and 15% gas. ...................................................... 18
Scheme 1-2: Overall scheme of the research project: a maximum carbon retention strategy to
produce renewable carbon fuel through biomass pyrolysis, followed by
electrocatalytic upgrading that is powered by renewable electricity. ................... 24
Scheme 2-1: general chemical equations for electrocatalytic hydrogenation of organic substrate
............................................................................................................................... 37
Scheme 2-2: chemical equations illustration on how H2 can be generated on the electrode surface
with two different mechanisms ............................................................................. 38
Scheme 4-1: Possible cathodic reactions: a) Hydrogen evolution b) acetol undergoes ECH to
propylene glycol using electrons and protons c) acetol undergoes ECH to
propylene glycol using water and electron as a reductant. d) cathodic water
reduction forming hydroxide anion. ................................................................... 109
Scheme 4-2: Revised mechanism for the SPEE cell that accounts for all observed reactions.
Reaction mechanism is proposed based on the conditions described in the diagram.
............................................................................................................................. 124
Scheme 5-1: Acetol hydrogenation to propylene glycol electrocatalytically. and hydrogenation
evolution ............................................................................................................. 134

xvi

Chapter 1 Literature Review on Common Renewable Mobile Fuel Candidate
1.1

The Importance of Hydrocarbon
There is no replacement for hydrocarbon’s low toxicity, well-developed and ease of

handling as storable stable liquid. These excellent qualities make liquid hydrocarbon fuels
irreplaceable by any other energy carriers. It is the ideal source of energy for transportation
purposes owning to its high energy: mass ratio, formally known as specific energy (MJ kg-1).
Today, hydrocarbon fuels are still playing a key role in the transportation sector. In 2011, over
250 million cars were thought to be on the roads in the US.1, but only 1.19 million of these were
powered by non-traditional sources such as batteries, hydrogen, propane, or ethanol2. Although
the trend towards alternative fuel vehicle usage is rising, traditional fuels will still play a key role
far in the foreseeable future. At a glance, gasoline carries approximately 25 times more energy
per kilogram (45 vs. 1.8 MJ kg-1) than the most advanced commercially available lithium battery,
as shown in Figure 1-1.3 The difference in energy immediately explains why gasoline is still the
predominant choice of energy carrier for transportation; it has a high energy to mass ratio
compared to other carriers available. Specific energy is a particularly important factor in
transportation fuel consideration because spending energy on moving the fuel itself is an energy
wasting process. Even if energy supply were easily available and come in unlimited quantity, the
mass of the energy carrier would still matter. Vehicle that is powered by low specific fuel would
suffer from frequent fuelling stops and low transportation capacity as the majority of the vehicle
space would be occupied by the energy carrier.
Therefore, this is why crude oil is being to refined to hydrocarbon fuel to power internal
combustion engine (ICE) rather than being used to generate electricity in a power plant to power
electric motors. This is because hydrocarbon stores a lot more energy than battery per given
weight, so hydrocarbon can cover a lot more mileage compared to battery. It is worth noting,
1

however, using a hydrocarbon fuel cell to generate electricity in a hydrocarbon fuel cell would be
a different story because it would still be best to have an efficient way of converting hydrocarbon
energy to electricity without the ICE. The vast value difference in specific energy between
hydrocarbons and batteries makes the ICE still the preferred converter of chemical to mechanical
energy, even though it has a lower energy efficiency than the electric motor. For the foreseeable
future, hydrocarbons will continue to be to the predominant form of energy carriers over any

Specific Energy (MJ kg-1)

other sources.

50
45
40
35
30
25
20
15
10
5
0

46.4

46.2

124
120

42.8

100

36.6
30.0

80
60

19.7

40
20
0.95

0.29

0.17
0

Figure 1-1: Comparison of the specific energy (MJ kg-1) of various common energy carrying
chemicals. Hydrogen has a specific energy of 124 MJ kg-1 and its scale is associated with the
secondary axis on the right.
Nature has affirmed this assessment: hydrocarbon-like fats and oils have been powerfully
selected by evolution as the ideal lightweight energy carriers for essentially all mobile organisms.
The use of hydrocarbons in engines is thus not a coincidence, but a logical practice that mimics
nature’s highly optimized energy transport materials. However, this practice and the combustion

2

of the fuel, does not only depletes the earth’s finite reserves of fossil resources, but also increases
the CO2 concentration in the atmosphere causing global warming and various unpredictable
climate changes.
The world is making great efforts to develop CO2 emission free technology while
accommodating the energy needs of the ever growing population. However, as of today, NO
technology is entirely free from CO2 emission because these technologies, e.g. hydrogen fuel,
E85, Fischer-Tropsch diesel synthesis and electric vehicles all depend on the current energy
framework. As long as fossil fuel continues to serve as the fundamental energy source, these
“renewable fuel” options can only be considered mitigation plans that minimize localized CO2
emission.
In this chapter, these renewable fuel options will be explored, their pros and cons will be
evaluated from various viewpoints such as cost, safety and ease of implementation within the
current fueling infrastructure, and in some cases the overall environmental impact. Bio-oil will
be presented as the renewable fuel candidate with the most immediate promise to potentially
displace the current fossil fuel demand in the transportation sector.
1.2

Emergence of Hybrid Vehicle
Hybrid vehicles powered by gasoline and batteries together have begun to emerge as the

world is slowly moving towards low carbon-emission technologies. Such hybrid vehicles operate
by converting the spare mechanical energy produced from internal combustion in the engines, or
by braking, into the form of electrical energy, which is then reversibly stored in the rechargeable
batteries installed in the vehicle. The stored energy is delivered when driving conditions call for
acceleration or other heavy duty work that would otherwise require high gasoline input. This
electrical energy compensation mechanism reduces gasoline consumption and thereby improves

3

the vehicle’s gas mileage. Today, some of the hybrid models even come in “plug-in” versions
with zero vehicle emission modes in which the vehicle can be recharged from the electrical grid
and be driven modest distances without combusting gasoline. Although the zero emission mode
typically only lasts several tens of miles, it is certainly a great step in the right direction towards
reducing greenhouse gas emission. However, as mentioned before, since the majority of
electrical power on the grid still comes from combustion of fossil fuels, mainly coal and natural
gas, this mode of powering transport is not actually “emission free” when the source is
considered.
Along the same line, some companies, such Nissan, Honda, and Tesla have begun to
market total zero emission vehicles – cars that run completely on electricity, and release zero
carbon dioxide. This seems excellent at a first glance, but it does come with drawbacks when
viewd in the bigger picture. First, electricity from the grid is still generated from coal or natural
gas burning, so zero emission vehicles are not entirely emission free. But localized, fixed source
CO2 emission could potentially enable effective CO2 recapturing, a process that will be discussed
later. Second, even if these vehicles are powered by renewable electricity, their manufacture
requires intensive energy inputs from the grid which as noted comes mainly from fossil sources.
Lastly, these vehicles typically suffer from high production costs and frequent fueling stops
because the energy storage module is manufactured with costly advanced materials and limited
by weight in the vehicle. Tesla Motors, a leading company in production of electric cars, has
been marketing its latest model, capable of travelling more than 300 miles (driving at 40 mi/hr)4,
a typical driving distance of a passenger car with a 14 gallon gasoline tank. Unfortunately, such
performance does not come without cost. A car from Tesla costs anywhere between $58,570 -

4

$106,570 depending on the model and accessories purchased, while most cars today on the
market cost between $25,000 – $45,000.
Although the US government has introduced various tax subsidies to promote the growth
of the alternative fuel vehicle business, the current cost is simply beyond what an average
American household can afford. These overwhelming prices are mainly due to the advanced
patented battery technology, and high unit production costs due to low present demand. After all,
battery charging stations are not yet widely established in the country compared to gasoline
stations. As of March 2014, only 7700 fueling sites in the US come with charging option. 5
Recharging the battery to cover 300 miles in the Tesla takes approximately 53 hours on a 120
volt outlet6 whereas filling a 14 gallon gasoline tank takes less than 5 minutes. So even though
the fueling option of electric car seems more convenient – plug the vehicle at home vs. driving to
gas station, the long refueling time without high cost equipment is a major setback to its sales.
Though a world operating with 100% zero emission vehicles would be wonderful, the high
production and long refueling time will continue to let conventional gasoline cars dominate the
transportation system for years to come.
1.3

Hydrogen as a Fuel
One highly touted carbon free transportation energy carrier has also been the focus of great

effort – Hydrogen. In 1970, John Bockris coined the term “hydrogen economy” in which
hydrogen would serve as the fundamental energy source to power various applications run in a
society.7, 8 In his theory, hydrogen could be generated from hydrocarbon steam reforming (CH4 +
H2O  CO + 3 H2) or water electrolysis (2H2O  O2 + 2H2. However, steam reforming
makes use of methane which is a fossil fuel, and electrolysis requires electricity that usually
comes from burning coal or natural gas. In other words, the hydrogen economy can merely serve

5

as a strategy to alleviate the global dependence on fossil fuel but it not a long term solution
unless there is affordable means to practice water electrolysis through renewable electricity.
Unlike hydrocarbons, which generate carbon dioxide and water upon combustion, hydrogen only
generates water in the form of steam, which is bio-friendly and hazard free. To maximize the
energy extraction from hydrogen, the hydrogen fuel cell (HFC) is currently one of the most
studied topics of research. In an HFC, electrical energy is directly generated through oxidation of
hydrogen and reduction of oxygen. Electrons are driven in the circuit by the formation of water,
which is a more thermodynamically stable product. This direct conversion of chemical to
electrical energy is substantially more efficient than the conventional internal combustion engine.
(ICE). The ideal efficiency of a hydrogen fuel cell can be as high as 83% under standard testing
conditions (1 atm and 25 oC) while an ICE only gets up to 34%.9 The general fuel cell operation
principle is depicted in Figure 1-2.
App.
e-

eH+

nH+
nH2

H+

O2

H+

2H2O

H+
H+
Electrode
loaded with H2
oxidizing
catalyst

Proton
Exchange
Membrane

Electrode
loaded with
O2 reduction
catalyst

Figure 1-2: General operating principle of hydrogen fuel cell
H2 fuel cell powered mass transport vehicles (buses) have been produced by various
manufacturers such as Toyota and Mercedes-Benz since early 2000 and are becoming

6

increasingly popular in several European countries.10, 11 The H2 tank and the fuel cell system are
typically located on the roof of the bus. Electricity generated is stored or dissipated through a
super capacitor or battery to fuel the motion of the vehicle. H2 refilling typically takes place at a
central H2 storage and fueling station. Prototype of passenger fuel cell vehicles have also begun
to emerge from many different vendors such as Audi, BMW, Nissan, Toyota, and Volkswagen.
They are often showcased in automobile expositions. However, H2 powered small cars have
generated less interest than gasoline powered cars for several reasons: H2 is a highly flammable
and, when in a mixture with air, explosive gas upon ignition. The 1937 Hindenburg disaster in
which a German passenger airship with a flawless safety record caught fire and burned in
seconds, resulted in the death of 35 of the 97 on board. While risk assessment reports indicate H 2
can be handled safely with proper public education on its properties, the general public would
still fear driving with such a highly explosive gas that reminds them of the airship tragedy. 12, 13
Furthermore, the H2 economy has failed to take off because of the need for a completely new
infrastructure of H2 filling stations, a distribution system, and low cost H2 storage methods and
equipment.14, 15
While large vehicles have room for large H2 tanks sufficient to cover moderate distances,
H2 tanks for passenger cars significantly increase the engineering challenge, weight, and
production cost for high pressure vessels or cryogenic systems. Moreover, fuel cell systems lack
durability compared to internal combustion engines; they are sensitive to real world fluctuations
in environmental variables such as humidity and temperatures. Although service life
improvements from 29,000 to 75,000 miles have been made in the last few years, consumers are
still better off with the internal combustion engines which typically last more than 150,000
miles.16 The cost per unit power of the H2 fuel cell technology has improved greatly over past

7

decade. The delivery and dispensed price of hydrogen has lowered to $4.10/gge (gallon of
gasoline equivalent) in 2012, but there remains a substantial gap to the ultimate target of $12/gge,17 a value that would give H2 fueled vehicles an economical advantage over gasoline
fueled vehicle.
The needed infrastructure of H2 fueling/delivering systems is still unavailable for
commercial use. H2 fuel cell powered public transportation has the advantage of centralized,
company-built H2 refilling facilities, standard fleets of vehicles, and regular travel patterns, but
there are only a handful of H2 refilling station in the US for H2 fuel cell car owners. This lack is
mainly due to the extreme capital investments required, ranging from $500,000 to $5,000,000
per station.18 Even if H2 refilling stations were subsidized by the government (i.e. taxpayers), and
were built across the nation, the current high cost of H2 ($4.10/gge) would still put H2 powered
vehicles at an economic disadvantage compared to gasoline (~$3/gal) for the general public due
to lower durability, higher vehicle costs, and safety concerns.
Although production of H2 through biomass gasification has been shown to cost as little
as $2.00/gge,17 the gasification process requires both intensive energy input and heavy capital.
As a path to valuable chemical product, gasification would be a reasonable strategy because its
large energy demand does not contradict with the goal. However, production of fuel in an energy
intensive environment would be counterintuitive (with the exception of making rocket fuel or
military jet fuel). Therefore, even if the government would subsidize construction of H2 fueling
stations and advanced H2 storing and fuel cell technology, the H2 fuelling option cannot be
considered as carbon free technology, unless H2 comes from water electrolysis and every stage
of the process, from production to distribution, is powered by renewable electricity.

8

Hydrogen has the highest specific energy of any chemical energy storage medium,19 though that
is still several orders of magnitude lower than uranium-235 nuclear fission reaction, which has a
specific energy of 6.73 x 107 MJ kg-1.20 It must be noted, however, that the number does not
include the weight of the nuclear plant, and various safety structures associated with it. While
nuclear power plant construction is becoming more of a common practice, nuclear power as a
transportation fuel is impossible because of the mass of the energy extraction system (with the
exception of large scale military vessels). Along the same time, hydrogen fuel also requires
heavy gas tanks, compression pumps, cryogenic system for storage. When these weights were
included in the specific energy calculation, the effective specific energy value is lower.
Hydrocarbon fuel (gasoline), on the other hand, bypasses the necessity of heavy tanks, because it
exists naturally in liquid form under ambient conditions; gasoline storage requires much less
attention than that of hydrogen. Their safety, established practice, chemical stability and high
specific energy make hydrocarbons still the ideal choice of energy carrier in today’s and
tomorrow’s markets.
1.4

The Biofuel: E85
The most popular and developed alternative fuel today is 85% ethanol (E85), which is a

blend of 85% denatured ethanol and 15% gasoline by volume. To date, E85 is viewed as a
promising renewable energy candidate. Like gasoline, E85 is a stable liquid and as such requires
no major modification to the current fueling infrastructure. It is a blend of gasoline with ethanol
produced from corn starch. In other words, this strategy opens a route to turn CO2 and sunlight
into mobile carbon fuel. Production of ethanol starts with mixing ground corn kernel or other
starch rich component with water and enzymes. The starch is converted enzymatically into
glucose for fermentation, and the fermenting processing typically takes 40 – 50 hours. The

9

resulting ethanol is dehydrated, by distillation (an energy intensive process in its own right)
denatured, and then blended with gasoline.21
E85 was proposed as a means to alleviate the US dependence on foreign crude oil
imports and to reduce greenhouse gas emissions. E85 compatible vehicles, aka flexible fuel
vehicles (FFV), require only little modifications to their engine and fuel systems, and hence do
not pose significant production costs to consumers.
In sum, E85 establishes a connection between biomass and carbon fuel, avoids the safety
concerns posed by H2 fuel, and most importantly requires only modest effort to implement
within the current transportation infrastructure. All these positive qualities explain why E85
quickly became the first national renewable fuel option, making its appearance in many local
gasoline stations. Of the 121,446 gas stations in the nation, 2400 of them offer E85 refueling. 5
Although that is merely a 2% coverage, it is nonetheless a good start toward widely available
renewable carbon-based biofuel.
As wonderful as the E85 scheme sounds, there are several problems associated with the
E85 productions. First of all, Atmospheric simulation studies have shown that use of E85
actually makes no improvement to emissions of CO2 or to tropospheric formation of NOx and
ozone. In fact, NOx increased in the Midwest and Southeastern region as a result of rising
biogenic emissions from increased corn acreage.22 Moreover, detailed environmental studies
reveal that increasing production of E85 may cause increases in CO2 emission over conventional
gasoline when other factors were considered. Second, production of fertilizer and pesticides, and
fueling of tractors actually boosted the energy consumption in the big picture. Furthermore,
boosting corn farming to match fuel demand has led to undesired environmental effects.
Increased water usage, smog formation and land use are all negative impacts due to expanded

10

corn cultivation, and hence disfavor corn ethanol production.23 Third, nitrogen rich fertilizers,
and pesticides themselves, require energy to synthesize and are potential pollutants to the
ecosystem upon disposal.24, 25 Fourth, Although the US currently has no shortage in food supply,
increased use of food materials for fuel production may eventually lead to an increase in food
prices due to land displacement.26, 27 Most importantly, E85 only slightly reduces the usage of
fossil petroleum and is not a permanent solution to the energy problem. Efforts today should be
devoted to seeking long term solutions that are environmentally friendly, sustainable, and can
address the scale of the US’s liquid fuel economies.
1.5

Biofuel production via Fischer-Tropsch Process
Aside from glucose fermentation, a long studied catalytic process known as the Fischer-

Tropsch process (FT) is another approach to fuel production. It was invented by Franz Fischer
and Hans Tropsch in Germany 1925 as a strategy to convert other forms of carbon resources
such as natural gas, coal, or biomass into liquid hydrocarbons for production of long chain oil
lubricants and synthetic fuels. Depending on the target products, FT processes and conditions
can vary vastly, but the overall process can be broken down into three stages: synthesis gas
(syngas) generation, FT synthesis of hydrocarbons from syngas, and FT product refining.
Biomass gasification via FT process has been considered as one of the most attractive routes
towards fuels and chemicals production.
Syngas is essentially a 1:2 mixture of carbon monoxide and hydrogen, which are
chemicals of high reducing power. The production of syngas starts with gasification of biomass
or other carbon feedstock in the temperature range of 900 – 1600 oC. Gasified carbon feedstocks
become syngas, which occasionally includes some fully oxidized species such as carbon dioxide
(CO2) and water (H2O). This mixture can be refined downstream via the water gas shift

11

equilibrium (CO + H2O  CO2 + H2). The resulting mixture of gases is then purified of
contaminants such as H2S and NH3 which may be present if the feedstock contained nitrogen or
sulfur.
The purified syngas can then be converted to a variety of hydrocarbon products: nCO +
(2n+1)H2  CnH2n+2 + nH2O, where CnH2n+2 represents any hydrocarbon product desired.
Among many conditions criteria, temperature plays the most important role in dictating the
growth of the hydrocarbon product. The high temperature Fischer-Tropsch (HTFT) reaction,
typically operating at 340 oC, tends to favor production of relatively short chain hydrocarbons
C1-C30 that consist mostly of alkenes along with some alkanes and aromatics28. On the other
hand, the low temperature Fischer-Tropsch (LTFT) reaction, typically around 220 – 240 oC,
favors long chain hydrocarbon formation, C1-C120, producing waxes and saturated alkanes along
with some alkenes.29, 30 The flexibility of FT synthesis enables production of various organic
chemicals from virtually any common carbon feedstock. In the context of renewable fuels as the
target, ideally non-food-grade biomass would serve as feed.
Hydrocarbon synthesis, particularly diesel fuel production, using biomass gasification has
been an attractive strategy in the recent years. The sulfur free and low nitrogen nature of biomass
makes it a better candidate than coal in the gasification process. Syngas (CO: H2 in ratio of 1:2)
conversion yields CO + 2H2 ⇌ (-CH2-) + H2O. Production of (-CH2-) creates building block for
hydrocarbon production.31, 32 However, gasification of biomass, which has a general formula of
“-CH2O-”, yields a syngas ratio of CO: H2 of 1:1. As a result, CO2 will be generated as a result
of hydrogen production. via the water gas shift 6CO + 6H2 + 2H2O  4CO + 8H2 + 2CO2
 4H2O + 2CO2 + 4(-CH2-) To avoid wasting carbon as CO2 in the product, addition of H2 is
required. As discussed before, today’s H2 supply principally comes from reforming methane in

12

steam, or high cost water electrolysis. The addition of H2 towards fuel production limits the
sustainability of FT synthesis using biomass.
Moreover, FT plants come with high capital and operating energy costs. This is because
the actual chemical process is rather complex and energy demanding. These difficulties and costs
explain why common commodities today are still produced from refining petroleum.
Nonetheless, there are circumstances when a FT plant is favored over a petroleum refinery. For
example, when crude oil supplies are distant or unstable, or transportation of petroleum
feedstocks is economically impractical, the FT process can be useful in the sense that it can
condense low energy density feedstocks into higher energy value products, which lowers the cost
for the chemical transportation. Of the several stages of the mentioned FT process, the
gasification utility poses the most significant cost and has the greatest energy demand. While FT
could serve as a flexible hydrocarbon production platform, the energy input and inefficient
carbon utilization (water gas shift equilibrium produces CO2 when the syngas ratio is CO: H2 1:1)
may exceed the value of products formed, and hence it is less attractive as a strategy for the
renewable fuel chemist. Nonetheless, addition of H2 from water splitting could represent a
strategy to maximize carbon utilization and raise its energy content. However, that would be a
different subject matter.
1.6

CO2 Recycling and Carbon Capture & Storage (CCS)
Replacing fossil fuel with bio-fuel can reduce the atmospheric CO2 because biomass is

considered as a carbon neutral fuel, a source that does not put fossil carbon into the current
atmosphere. In principle, the emitted CO2 from coal plant can be collected and stored in
pressurized vessels underground. The stored CO2 could potentially be recycled to chemicals or
fuels using electrochemistry.

13

Electrocatalytic hydrogenation of carbon dioxide (CO2) to chemicals and fuels is one of the
current biggest areas of research. It aims to fix CO2 in the form of methanol or related mono
carbon substrates. Electricity captured from renewable sources such as solar, wind, hydroelectric
and geothermal could be used for the formation of carbon hydrogen bonds. In other words, CO2,
a greenhouse gas, could be recycled into an energy rich chemically valuable resource through
storing captured renewable energy. These efforts could be fruitful in reducing global warming if
CO2 could be captured and effectively recycled.
Electrocatalytic Hydrogenation of CO2 was first reported in the beginning of the 19th
century, when a zinc cathode was used to reduce (hydrogenate) CO2 to formate (HCOO-) in an
aqueous environment. Today, hydrogenation of CO2 has been explored to a great extent both
homogeneously and heterogeneously.33 In general, the reaction is done in a neutral to basic
environment in which CO2 is solubilized as carbonate (CO32-).34 Mechanistic insights into CO2
reduction have been reported in many review articles or book chapters. 35 Product selectivity
seems to be dependent on choice of electrolyte and electrode. In general, inorganic electrolytes
seem to afford formate (HCO2-) as the sole product while organic electrolytes, e.g.
tetraalkylammonium salts, favor formation of a series of organic products, such as formate
(HCO2-), glycolate (HOCH2COO-), malate (C(O)CH2CHOHC(O)O)2-, oxalic ((OC(O)C(O)O)2-),
and other carboxylic acids.36-38 Other reviews noted that the coinage metals, gold, silver, and zinc
yield mostly CO as the major product while indium, lead, tin and cadmium yield mostly formate
as the major products with the condition unspecified. Recently, interest has begun to shift
towards the use of copper, an abundant and environmentally friendly metal, showing formation
of a variety of hydrocarbon products depending on applied potential.39-41 Although research on

14

CO2 hydrogenation has taken off and began to demonstrate promising results (CO is a valuable
chemical), there are yet any effective scheme in recapturing atmospheric CO2.
Some newly built coal plants are designed with carbon capture and storage (CCS)
features. CCS concentrates the CO2 released from fuel combustion, and stores it deep
underground.42 A CCS power plant is estimated to release 88% less CO2 into the atmosphere.43
However, CCS is not a sustainable option to address the energy problem. First of all, CCS power
plants are still burning fossil coal to generate electricity; CO2 recapture may be carbon neutral in
terms of the atmosphere, but it is no solution to the greenhouse effect. Moreover, gas collection
and compression takes 10 – 40% extra energy from the power plants.44 Manufacture of CO2
capturing chemicals such amine, ammonia, or amino salt for CO2 recapture is also an energy
investment.45
In a CCS scheme, CO2 generated from fossil fuel combustion is pumped and sealed deep
underground permanently. Even if, hypothetically, the stored CO2 can be retrieved and recycled
to CO electrocatalytically, hydrogen is still required to mix with CO for fuel production. In
summary, CO2 hydrogenation is a promising technology that is crucial for greenhouse gas
reduction in the future. However, it is not a practical solution for fuel production today because a)
mono-carbon species carry too little energy as fuel, and b) fuel production using CO generated
from biomass gasification requires addition of H2 to became effective synthesis gas. In theory,
CO2 could be reduced to methanol (CH3OH) to power methanol fuel cell device. However, the
specific energy of methanol is much smaller than that of hydrogen or hydrocarbons. Methanol is
also much more chemically aggressive towards polymer based container and is lesser likely fit
the current transportation infrastructure.

15

In summary, CO2 recycling to fuel and chemicals is a potentially wonderful technology. But
CCS is flawed; even if the underground CO2 can be retrieved and reduced to CO, it will require
addition of H2 for hydrocarbon production. Most important, CCS is not a sustainable option, it
merely reduces the net CO2 output to the atmosphere while consuming an ever depleting fossil.
A low cost renewable carbon based strategy that can be implemented easily within the current
infrastructure. If the cost is too high, it will lack the economic driving force to displace fossil fuel.
This fuel must be carbon based so that it can be implanted to the current system without major
modification, similar to E85. The fuel must be free from using any fossil related products for its
processing. For this, the project seeks biomass to be the optimized candidate towards renewable
fuel production.
1.7

Biomass as a Renewable Energy Feedstock

1.7.1 The Need of Renewable Energy
As of 2012, the US continue to be the world greatest oil consumer, consuming 18.49
million barrels of oils daily.46 Although the consumption has declined slightly from 19.18
million barrel/day since 2010 owing partly to improved gas mileage in modern engine design 47
and partly to less use of liquid fuels in fixed settings (natural gas boom), the US is still leading
by far compared to the 2nd runner, China, where 10.3 million barrels of oil are consumed daily.
Moreover, over half of the oil supply is dependent on foreign regions; as such the US should find
a renewable domestic source of fuels to secure its energy supply and independence. 48 Even more
important, ongoing combustion of fossil carbon-based fuels continues to raise atmospheric CO2
concentrations, causing ever-increasing greenhouse warming and ocean acidification with the
attendant drastic shifts in the earth’s global climate, sea levels, and biosphere.49 In January 2014,
the atmospheric CO2 concentration, corrected for seasonal fluctuation, at Mauna Loa was 397.80

16

ppm.50 a 24.4% increase compared to 50 years ago (Jan. 1964. 319.67ppm); transportation fuels
contribute roughly 1/3 of this excess CO2 load.51 Meanwhile, individual vehicle use is ever
growing in the United States.52 Ultimately, if humans are to continue enjoying the quality of life
we have built on a hydrocarbon foundation, future supplies of these critical substances must
come from renewable, carbon-neutral (or preferably carbon-negative) sources.
As of 2011, US petroleum use is roughly a 9.6 x 1011 kg of petroleum per year, almost all
for fuels. Estimates for annual potential production by 2030 of non-food biomass are very
optimistically 1.4 x 1012 kg.53 But the carbon and energy contents of these starting materials are
vastly different, as shown in data from the US Energy Information Agency (EIA) and the US
Department of Energy’s (DOE) office of Energy Efficiency and Renewable Energy, presented in
Figure 1-3 below:
Annual Usage
(Weight)
2011 Petroleum Usage
Harvestable Biomass

9.6 x 1011 kg
1.4 x 1012 kg

Based on general formula
of “CH2”
(12/14) x 9.6 x 1011

Annual Usage
(Carbon Content)
2011 Petroleum Usage

8.2 x 1011 kg

Harvestable Biomass

5.6 x 1011 kg

Annual Usage
(Energy Content)
2011 Petroleum Usage

4.3 x 1013 MJ

Harvestable Biomass

2.1 x 1012 MJ

Based on general formula
of “CHOH”
(12/30) x 1.4 x 1012

*Diagrams are drawn to approximated scale

Based on general gasoline
specific energy – 45 MJ kg-1
9.6 x 1011 x 45

Based on general biomass
specific energy – 15 MJ kg-1
1.4 x 1012 x 15

Figure 1-3: Comparison of the overall carbon and energy balance in the US transportation
sector and harvestable biomass. Red bars compare the weight of the petroleum used, and the
weight of the harvestable biomass. Blue bars compare the actual carbon content. Green bars
compare the energy spent annual, and the harvestable amount of energy from biomass.
Projections are made based on the available biomass by 2030.

17

In the US, there is neither enough carbon nor enough energy in the most optimistically
projected biomass supply to match the amounts in the petroleum used. The scenario is even less
optimistic when coal and natural gas are considered. Both biological and thermochemical
processes to convert biomass to desirable fuels have seen impressive advances in recent years,
yet most of these biofuel strategies simply extract the small energy content of biomass as fuel,
and discard the remaining carbon, usually as CO2.
1.7.2 Common Pathway for Biomass to Biofuels
Method

General Descriptive Chemical Equations

Carbon Loss

Fermentation

33%

Gasification
(F-T Process)

33%

Fast-Pyrolysis

0%

Scheme 1-1: General descriptive chemical equation describes common biomass to biofuel
processing pathways. Fermentation and Gasification both disposes 33% of the carbon content
as CO2 while fast-pyrolysis retains most carbon in the form of bio-oil: 70% bio-oil, 15% biochar, and 15% gas.

Both gasification and pyrolysis could in principle capture all the input carbon as fuel, if
H2 were freely available. But current production of H2 comes from reforming natural gas
(methane) with CO2 as a byproduct. So despite its lack of carbon, H2 is a fossil fuel product in
today’s market; its use as primary reductant in the field would bring extra costs in complexity,
energy, and fossil carbon. Of course, hydrogen can be produced by electrolysis of water, and a
large amount of effort has been put into development of water splitting catalysts. An optimistic
estimation, based on today’s electricity cost from burning cheap coal, of H2 production will be
still at the high end of the overall candidates ($6.75/kg) – 4.5 times more expensive than methane
18

reforming ($1.47/kg).54 The associated high cost of H2 production from water electrolysis would
substantially increase the cost of fuel production. If these productions were carried out with
renewable electricity, the fuel prices will increase dramatically.
In summary, any renewable fuel scheme that uses gaseous hydrogen as a reductant or that
releases CO2 during its process is flawed as a long term solution towards creating a sustainable
carbon fuel cycle. A new approach is necessary to address all the mentioned challenges. This
renewable fuel must be able to fit into the current fueling structure, have must a high specific
energy, and must be independent from fossil carbon. Hence, this project seeks an innovative
strategy to produce a carbon-based fuel using the renewable carbon from biomass, via the
unprecedented combination of fast-pyrolysis and electrocatalytic upgrading.
1.7.3 Biomass as Feedstock
Any strategy for direct biomass-to-fuels conversion is fundamentally a solar energy
capture scheme. Plants, natural CO2 collectors designed by evolution, are the cheapest carbon
collectors available. However, while excellent at fixing carbon, plants are inefficient at capturing
the sun’s energy and storing it in chemical form. With the sun rising every day, they have no
evolutionary driving force to store more energy than they need for growth. Most plants in nature
has a photosynthetic efficiency ranged 1 – 2%; the other 98% energy is lost to optimal
wavelength mismatch, respiration, reflection and transmission.55, 56 Furthermore, biomass has
only about one third of the specific energy of gasoline as mentioned above; therefore,
unmodified plant matter is energetically unsuitable as transportation fuel.
Human technologies can capture solar energy much more efficiently than plants, but they
do so in the form of electricity. Commercially available lower-end solar voltaic cells, mostly
made of single crystalline silicon, have an energy efficiency of 12 – 15%.57 In other words, even

19

the oldest form of photovoltaic device on the market still captures solar energy at least ten times
more efficiently than plants. It is a national goal to replace as much fossil (petroleum oil) with
biological carbon in liquid fuels as possible. If solar electricity could be used to raise the energy
content of biomass, the limiting quantity would become the content of carbon rather than energy,
enabling higher yields of biomass-derived fuels. Furthermore, due to a mismatch between the
times of human power demand and the cycles of light and wind availability, electric power
generating capacity may go unused at some times, so an alternative, time-insensitive method of
using and storing alternative energy is sorely needed.
1.7.4 Biomass Fast Pyrolysis (BFP)
Biomass Fast Pyrolysis is one of the most popular strategies in biomass liquefaction. It
heats up biomass rapidly to 400 – 600 oC for a few seconds in the absence of oxygen and causes
it to “melt”.58 The resulting products can be categorized based on their physical states: gas,
which can be used as a fuel for the pyrolyzer; liquid, also known as “bio-oil” that consists of
hundreds of different small organic compounds to be discussed in detail in the next section; and
solid “bio-char” which is used primarily as a soil amendment for plant cultivation.
BFP may seem simple – it is not. It is a complex physical and chemical process that
transforms solid matter under rapidly changing temperature. Various factors such as feedstock
powder size, feed rate, temperature program and heat transfer mechanisms, vessel pressure, and
residence time can control the quality and quantity of bio-oil58-66 which is the key ingredient in
biofuel production. This research focuses on upgrading of bio-oil to bio-fuel and hence BFP will
only be discussed at an introductory level in this chapter.
Biomass must first be dried and ground prior to feeding to the pyrolyzer. Drying
minimizes water input to the system, which reduces the amount of water in the product bio-oil.

20

Small particle size of the ground biomass ensures the uniformity of heat transfer to maximize
bio-oil production. In general, the particle size can range from less than 1 – 6 mm.61 The optimal
biomass particle size depends on the type of reactor, feedstock, and feed rate. Different
feedstocks require different temperature programs to decompose properly. For instances,
hemicellulose and cellulose decompose in a fairly narrow temperature range compared to lignin,
which degrades slowly over a wider temperature range. The ratio of cellulose to lignin varies
among plants, and hence there is no one single best universal setting for bio-oil production. In
general, maximum bio-oil product yield is obtained in the range 500 – 520 oC with less than a 2
second residence times.58,

66

Prolonged residence times trigger secondary reactions, in which

dehydration of oxygenated organics occurs and leads to char formation, thus decreasing the biooil yield.
1.7.5 Nature of Bio-oil
Compared to crude oil, bio-oil is a dark syrupy emulsion comprising many low carbon
number organics and water. It cannot be used directly as transportation fuel without pretreatment for several reasons, most important of which is its low energy density. Like biomass,
bio-oil has ca. 1/3 the energy content of hydrocarbons (ca. 15 vs. ca. 45 MJ/kg). In addition, biooil is highly acidic and contains many oxygenated organics. In fact, almost 20% of bio-oil is
comprised of compounds like acetic acid, hydroxyacetaldehyde, hydroxyacetone, furfurals,
levoglucosan (derived from sugars), and a variety of lignin-derived aromatics.

66

A general

chemical distribution of bio-oil67-70 is depicted in Figure 1-4. This mixture corrodes steel storage
and pipes aggressively, especially at elevated temperature.71 Furthermore, the reactivity and
viscosity of bio-oil are both high; bio-oil tends to polymerize via acid-catalyzed condensation of
its high content of reactive oxygenates such as aldehydes and ketones and phenols. Lastly, bio-

21

oil has a high water contents.

Bio-oil is an emulsion, consisting of liquid phase organic

compounds and water. All of the above properties make it incompatible with storage and
transport over extended periods. Table 1-1 shows a composition comparison between bio-oil and
crude oil.63
Table 1-1: Composition comparison between bio-oil and crude oil are shown. Note: actual
composition of bio-oil varies depending on the biomass feedstock.
Characteristics?
Bio-oil
Crude Oil
Moisture wt%

15-30

0.1

pH

2.5

-

Specific gravity

1.2

0.82 – 0.875

Element composition wt%
Carbon

39.5-55.8

85.2

Hydrogen

6.1-7.5

11

Oxygen

37.9-52.6

1.0

Nitrogen

<1

0.3

Specific Energy MJ/Kg

18 - 22

43 – 46

22

Guaiacols

Figure 1-4: General chemical distribution of organic substrates found in bio-oil. Percentage is
given percent of mass.

Bio-oil must be pretreated for stabilization before it can even be considered as a crude
fuel because of its reactivity and aggressive character. The primary key to effective stabilization
is the reduction of the carbonyl moieties of the ketone and aldehyde functional groups, and acid
neutralization. Ideally, such reductions would occur immediately downstream of the pyrolysis
process. The stabilized material could then be further upgraded and refined, or transported to a
central facility such as a petroleum refinery for classical upgrading by hydrotreating, a process
that upgrades the energy value and stability of bio-oil using hydrogen. However, as noted earlier,
the current supply of hydrogen comes from steam reforming of methane, and thus it is a nonrenewable fossil-based product. Moreover, even if water electrolysis becomes cost feasible for
H2 production, the need for a storage system for H2 would pose an economic barrier towards biooil upgrading. However, it is worth noting that substantial work has gone into direct petroleum
style upgrading of bio-oil with H2, and more detail will be discussed in detail in Chapter 2.

23

Fortunately, electrocatalytic hydrogenation (ECH) upgrading bypasses both issues. It
achieves hydro-upgrading of organic compound without an explicit H2 gas supply. H2 (or more
precisely a supply of H-atom) is generated on the surface a catalytic electrode that can be used to
hydrogenate organic substrates adsorbed from solution. The details of ECH will be discussed in
the next chapter. In short, ECH can serve as a relatively energy feasible alternative to chemical
H2 hydrogenation. With this in mind, the overall scheme of this research project, biomass to
biofuel production through BFP and ECH, is depicted in Scheme 1-2.
Good at Capturing CO2
Bad at capturing light energy
(1 – 2%)

Cannot Capture CO2
Great at capturing light
(>10%)

Gas (~15%)
Electrocatalytic
Upgrading (My Focus)

Fast
Pyrolysis

Biomass
(E.g. Poplar, Switch
grass and Corn
Stover)

Bio-oil (~70%)
(Biomass Pyrolysis Oil)

Gasoline

“Bio-char” (~15%)

Scheme 1-2: Overall scheme of the research project: a maximum carbon retention strategy to
produce renewable carbon fuel through biomass pyrolysis, followed by electrocatalytic
upgrading that is powered by renewable electricity.
1.7.6 Bio-oil Upgrading via ECH
For fuel upgrading purposes, ECH could be the ideal route because it operates at mild
conditions below the electrolyte’s boiling temperature (typically water), while at the same time it
avoids the storage and use of H2 gas. By reducing the carbon-oxygen and carbon-carbon double
bonds of the organic compounds in bio-oil, it is possible to eliminate carbonyl condensation
24

processes that lead to a polymerization to form tars. Also, hydrogenation incorporates the
additional heat of combustion of H2 into the reduced species as shown in Figure 1-5.
Bio-oil also contains many oxygenated aromatic compounds. Hydrogenation of such electronrich aromatic compounds in general is difficult due to their aromatic stability. However, under
favorable pH and surfactant conditions, efficient ECH results can be achieved.72 The design and
investigation of an efficient ECH system is one of the major goals of this research project.

Specific Energy MJ/kg

48
46
44
42

46.61 MJ/kg
45.64 MJ/kg

40

38
36

34
32

37.07 MJ/kg
32.49 MJ/kg

• Data taken from NIST Webbook
• Higher Heating Values are used

Figure 1-5: Phenol is used as a model compound to illustrate the increases in specific energy
(MJ kg-1) upon hydrogenation and deoxygenation. Note: higher heating values mean the heat
of combustion is calculated with the heat of formation of water in liquid state, which is lower
in energy compared to gaseous state, and thus gives a greater value of heat of combustion
(heating values).

It is noticeable in Figure 1-5 that the specific energy increases substantially between
cyclohexanol (37.07 MJ/kg) and cyclohexane (46.61 MJ/kg) by replacing an –OH with –H. This
is because the weight of oxygen (15.999 g/mol) is 16 times heavier than hydrogen (1.008 g/mol),
and the presence of oxygen decreases the value of the heat of combustion, and thus decreases the
energy per unit weight. In terms of fuel mass, oxygen is considered as a free supply from the air
25

in the heat of combustion equation, so its mass is not counted in the specific energy calculation.
This explains the importance of bio-oil hydrogenation and hydrogenolysis; not only does it
incorporate the heating value of H2 into the liquid, but it can also potentially achieve
deoxygenation to increase the specific energy of the compound.
Conventional hydrogen upgrading is done in an industrial style where a reactant is subjected to
high pressure and high heat environment. These conditions require heavy energy input, counter
to the purposes of renewable fuel production. ECH avoids the use of H2 gas, operates at
relatively mild conditions, and most important of all, it incorporates the renewable energy
captured from solar, wind, and geothermal electricity, into the bio-oil to raise its energy content.
In other words, bio-oil is acting like a “battery” to store energy captured from renewable sources.
1.8

Chapter Summary
Several alternative energy schemes such as electric/hybrid vehicles, hydrogen fuel, E85

fuel, FT biomass gasification, CO2 capture and storage, and bio-oil have been surveyed in this
chapter. Their pros and cons have been evaluated from the stand point of sustainability, cost,
environmental impact, and infrastructure compatibility.
Hybrid vehicles are becoming more popular for their high gas mileage and successful
integration into the current fuelling system, but they suffer from high production costs and they
still use petroleum fuel as their major fuel source. Electric cars and hydrogen fueled vehicles, on
the other hand, have the potential to run on fossil free energy, but they suffer from both high
production costs and a lack of fueling stops. Also, current electricity supplies come from coal or
natural gas combustion, and hydrogen is generated from methane steam reforming. So even
though these vehicles provide a means to alleviate local gasoline consumption and CO2 emission,

26

they remain dependent on burning fossil carbon and so are not the ultimate solution to the fuel
problem. Also, H2 gas raises significant safety concerns for drivers.
Current strategies for biomass-derived fuel also have limitations; although E85 is making
its way to be the first available renewable fuel nation-wide, in-depth analyses have revealed that
an increase in ethanol production from corn starch poses a greater environmental burden to local
farming than petroleum refining. The FT process enables production of a great variety of
hydrocarbon products from virtually any carbon feedstock including biomass. However, FT
plants are both energy demanding capital cost intensive. Also, like E85, FT throws away about
33% of the carbon as CO2 during the water gas shift equilibrium stage because of the wrong CO:
H2 ratio. Although CO2 recycling has been explored as an option towards renewable fuel
production, it is still far from practical to be able to create a sustainable fuel cycle to support the
U.S. transportation sector.
Continued use of fossil coal via CCS is by no means a renewable energy strategy; it
merely puts the released CO2 back underground at the expense of greater energy consumption. In
a way, CCS may speed up fossil reserve depletion while causing just as much negative impact on
the environment because of chemical production for CO2 recapture. Land drilling and sites
construction requires careful selection around the shore area and close enough to the CO2
emission locations to minimize transportation costs. Ultimately, the multiple factors adds up
making it a challenge option to execute.
Hydrogen fueled vehicle is relatively more promising in creating a sustainable cycle
compared to other mentioned candidate. However, the high cost associated with water splitting
and gas storage makes it less favorable to the transportation system. Steam methane reforming
for hydrogen production is not a sustainable strategy. In terms of safety, hydrogen is a highly

27

explosive gas combing with oxygen. This put a greater safety and psychological burden to the
drivers . Hydrogen fuel cell has a shorter life time compared to gasoline engine, and the
operating efficiency fluctuates with the real world environment – carbon fuel is still a better
option even with its low specific energy that is only 1/3 of hydrogen.
Making hydrocarbon from FT synthesis using biomass as a feedstock seemed like an attractive
option, but the energy investment and the capital cost of a FT plant is too high to be
economically feasible for fuel production (chemicals would be fine). Gasification of biomass
requires high energy input, and worse, the produced bio-syngas ratio will require addition of H2
to prevent CO2 formation during the water-gas shift equilibrium. Although the CO2 release from
bio-syngas WGSE does not consider as carbon positive to the atmosphere, throwing away value
biomass carbon that was insufficient in the first place is problematic in the bigger picture.
BFP can produce liquid fuel through liquefaction of biomass. Biomass is heated rapidly
in an oxygen-free environment to produce an emulsive liquid called bio-oil. The highly
oxygenated bio-oil can be upgraded to gasoline through catalytic hydrogenation and
hydrogenolysis (discussed in Chapter 2) and be directly substituted into the current fueling
infrastructure. Although only 70% of the carbon of biomass will end up in bio-oil, the other 30%
(15% gas + 15% bio-char) can be used for process heat to the reactor or for soil amendment,
which will be recycled into biomass carbon again. This is the only option that can effectively
produce a carbon fuel without heavy involvement of fossil energy.
1.8.1 Summary Table of the Discussed Alternative Fuel Candidates
The overall argument of how biomass can be favored over other alternative energy is
summarized in Table 1-2 below:

28

Table 1-2: Summary of each discussed alternative energy candidate. 5 major factors: adaption to
the current fueling structure, relationship to fossil energy, Cost to consumers, sustainability, and
the environmental impacts are explored.
Hybrid Vehicle
Adaptation to the current
fueling structure
Relationship to fossil
energy
Cost to consumer
Sustainability
Environmental Impact

Highly adaptable to the current fueling system as they are run on gasoline. (121,446
gasoline stations in the US as of March 2014)
Gasoline production comes from refining petroleum, a form of fossil fuel
Hybrid vehicles are more expensive than non-hybrid vehicles, although tax credits are
given towards hybrid vehicle purchase, they are still cost typically $10K - 20K more.
Although hybrid vehicles have better gas mileage, it is not a sustainable option
because fossil fuel is a finite resource.
Gasoline from fossil sources will continue to release fossil carbon into the atmosphere,
though the rate of CO2 will reduced as a result of the hybrid technologies

E85 Fuel
Adaptation to the current
fueling structure
Relationship to fossil
energy
Cost to consumer

Sustainability
Environmental Impact

Electric Vehicle
Adaptation to the current
fueling structure
Relationship to fossil
energy
Cost to consumer

Sustainability
Environmental Impact

Around 2400 gasoline stations offer E85 fueling option5, a 2% coverage of the current
gasoline stations
Only 85% of the fuel is ethanol from corn starch fermentation. The other 15% is
gasoline from fossil fuel
E85 in general costs $0.6 – 0.7 less per gallon of gasoline. E85 fuel only requires
minor engine modification (flex fuel engine) and so poses no significant financial
burden to consumers. However, FFVs are 30% less efficient on ethanol and hence,
lower fuel price still does not compensated for that.
Not a sustainable option because 15% of which is still gasoline that comes from
petroleum refinery
Combustion of E85 release lesser greenhouse gas to the atmosphere. However,
increases in agriculture land mining, production and use of fertilizer, fermentation
processing have negative impact towards the environment in a overall picture
Around 7700 charging stations available nationwide5 - 6% of the current gasoline
stations, however, home charging is also available for most models in the market.
Current electricity comes from burning natural gas and coal, with minor supply from
nuclear reactors and tiny fraction from renewable sources like solar and wind
Vehicles powered by electric motors in general are only slightly more expensive than
those of gasoline engines. However, high end electric vehicles can cost more than
100k.
Potentially sustainable is in a distant future when electricity is generated from
completely renewable sources.
Electric Vehicles themselves are emission free but the increase of electricity
consumption will lead to greater coal or natural gas consumption meaning more CO 2
release to the atmosphere. Greater demands on battery (E.g. lithium ions or Lead acid)
and power plants have impacts on the environment

29

Table 1-2: (cont’d)
CO2 Capture and Storage
Adaptation to the current
fueling structure
Relationship to fossil
energy
Cost to consumer

Sustainability
Environmental Impact

Not applicable - CCS does not alter the current fueling infrastructure
CCS requires more 10 - 40% more energy to transport and store the CO2 generated.73
Thus increases the consumption of coal.44
Although no literature found on how CCS plant would change the electricity price, it is
reasonable to assume electricity price would go up as a result of the CCS recapture
system construction.
CCS is not a sustainable option; it only reduces fossil carbon release into the
atmosphere and makes fossil fuel more “environmentally friendly”.
CCS should have positive impact on the environment, but increase in chemical
production for CO2 capturing, extra land use for CO2 storage site construction with
deep drilling would all have a great negative impact on the environment.

Hydrogen Fuel Cell
Adaptation to the current
fueling structure
Relationship to fossil
energy
Cost to consumer

Sustainability
Environmental Impact

Only 10 H2 fueling stations are available nationally5, 9 of them are in California, a 2%
coverage of the current gasoline stations in that state.
Current H2 supply comes from methane reforming, making H2 a non-renewable energy
source.
$3 - 4 /gge but projected to drop in the course of next decade. Hydrogen vehicle,
however, in general costs around $50K and the fuel cell unit has a lower lifetime (most
recent reported ca. 75,000 miles) compared to internal combustion engine (typically
around 150,000 miles)
Although current hydrogen supply comes from methane reformation, hydrogen in the
future may come from water electrolysis powered by renewable electricity.
Hydrogen fuel probably has the least environmental impact in the pool of candidates –
it releases water upon combustion. However, manufacture of fuel cell requires
precious metals that would have negative environmental impact to the land mining.

Bio-Fuel via FT Process
Adaptation to the current
fueling structure
Relationship to fossil
energy

Cost to consumer

Sustainability

Environmental Impact

FT produces conventional hydrocarbon fuel through biomass, and so it can benefit
from the current fueling infrastructure
This uses biomass as feedstock; however, the bio-syngas formed has a 1:1 ratio of CO:
H2 which leads to formation of hydrocarbon and CO2. Addition of another equivalent
of H2, produced from methane reforming or water electrolysis, is required to avoid
CO2 formation
F-T produces hydrocarbon fuel that requires no modification on the vehicle. However,
FT plants comes with a high capital cost, and hence may boost the fuel prices due to
increase in cost production
As mentioned above, H2 is required to set the CO: H2 ratio right. Even if H2 can be
made from water electrolysis via renewable electricity, the energy demand of
gasification will most likely supplied by coal or natural gas burning
The increase in H2 demand will increase steam reforming thereby putting more CO2
into the atmosphere. Higher energy demand from the power plant may increase coal
and natural gas consumption. Biomass are selected to no interfere with the current
food supply.

30

Table 1-2: (cont’d)
Biomass Fast Pyrolysis (BFP)
Adaptation to the current
fueling structure
Relationship to fossil
energy

Cost to consumer
Sustainability
Environmental Impact

BFP produces bio-oil that can be upgraded to bio-oil fuel for gasoline engine use, and
so it benefits from the current fueling infrastructure.
BFP liquefies biomass to bio-oil. However, energy for the fast pyrolysis is likely to
come from power plants fueled by non-renewable sources. Depending on the
upgrading scheme, the H2 needed for upgrading can either come from methane for
catalytic hydrogenation or in-situ formation for electrocatalytic hydrogenation.
Product of upgraded bio-oil should function identically to gasoline fuel, and so does
not requires more financial investment.
Almost - projection of harvestable non-food grade biomass by 2030 cannot cover the
fuel usage in the transportation sector for today.
Unlike E85, where corn is the sole crop plant for fuel production, BFP includes
different biomass feedstocks. The multi-crop cultivation may cause lesser impact on
the environment due to mono cultivation. Moreover, BFP doesn't throw away 33% of
its carbon unlike ethanol production. 15% of the biomass will become bio-gases, and
can be fed back to the pyrolyzer for further heating.

1.8.2 Goal of this Research Project
Of the options explored, bio-oil is the renewable fuel candidate that can maximize carbon
utilization and potentially fit into the current fueling infrastructure after it is upgraded. Without
needing to change the fueling infrastructure, bio-oil could be marketed nationally similar to the
case of E85. The hydrocarbon nature of bio-oil allows consumers to continue to drive with a
peace of mind just like driving with gasoline. Utilization of non-food grade biomass does not
compete with the land for edible agriculture production, thus making bio-oil an economically
favorable route on a grand scale.
However, as discussed before bio-oil only carries 1/3 of the energy content compared to
gasoline. Bio-oil contains large amounts of water and polymerizable oxygenated organic
compounds, making bio-oil incompatible with modern vehicle engines. Hence, bio-oil must be
upgraded with a maximum carbon retention strategy to an energetically compatible and
chemically stable fuel. Such upgrading ideally does not require energy intensive processes such
as high heat or pressure or the use of H2. Hence, this project seeks to develop an electrochemical
hydrogenation method that would be an effective bio-oil upgrading strategy for renewable
31

carbon fuel production. The goal is a chemically and economically scalable electrocatalytic
reactor system in which bio-oil can be hydrogenated in a continuous process with good current
and energy efficiency. In the past, Lercher reported the use of Raney nickel catalyst to achieve
Hydrodeoxygenation on common bio-oil aromatic compounds such as eugenol, guaiacol,
syringol in a high H2 pressure and high heat setting.74, 75 The ideal short term goal would be
translate Lercher’s chemistry from catalytic to electrocatalytic cost efficiently. Traditional ECH
makes use of expensive catalysts such as platinum and ruthenium, which are fine for small scale
chemical productions, but they are insufficient to address the scale of transportation fuel
production. Hence, investigation on low cost catalytically active metals is the first stage of the
project, and will be discussed in chapter 2 in detail. An ultimate ideal scenario of would be for a
farmer to make a bucket of bio-oil through the fast pyrolysis reactor, use the electrodes or reactor
developed from this project to upgrade it to a fuel, an organic mixture that has the comparable
specific heat to fossil transportation fuel, and finally, use it for her tractor using the electrodes or
reactor developed from this project.
Hydrocarbon fuel is deeply rooted in today’s society but it comes from a finite resource –
fossil carbon compounds. An innovative strategy that does not disturb the current fueling
infrastructure will be needed to replace fossil fuel with biomass for a sustainable future.
Liquefaction of biomass to bio-oil has been shown to be a promising way to replace fossil
hydrocarbon fuel. Unfortunately, bio-oil itself comes with low specific energy, and is not a good
fuel. Moreover, bio-oil contains a large amount of acetic acid and oxygenated compounds which
makes it highly unstable to store, and energetically incompatible as fuel. Hence, this project
seeks an energetically feasible strategy that can utilize renewable electricity to stabilize and

32

upgrade bio-oil to usable fuel. The details of the upgrading strategy will be discussed in Chapter
2.
Chapter 2
2.1

Literature Review on Electrocatalytic Hydrogenation

Introduction
Due to Bio-oil’s low stability, low energy content and highly oxygenated nature, it must

be upgraded before it can be used as a transportation fuel. These negative properties can be
circumvented by hydrodeoxygenation (HDO), in which bio-oil undergoes catalytic
hydrogenation at high temperature (150 – 400 oC) and high pressure (~150 – 200 psi H2). HDO
can achieve deoxygenation by hydrogenating all possible carbonyl, alkene, and aromatic sites.
The resulting increase in the hydrogen to carbon ratio effectively raises the specific energy of
bio-oil (See Figure 1-5). The current HDO strategy is performed in the current petroleum
refinery setting, so it does not require major infrastructure modifications and is thus is
economically appealing. Many bio-oil upgrading studies have shown promising results toward
fuel production using petroleum upgrading catalysts such as cobalt or nickel molybdenum sulfur.
2.2

Catalytic Hydrogenation for Bio-oil Upgrading and its Concerns
In the petroleum refinery industry, many upgrading catalysts are designed to remove

sulfur, a process known as hydrodesulfurization (HDS). Without HDS, fuel combustion emits
sulfur dioxide (SO2) leading unpleasant smell and to acid rain in the environment. Also, sulfur
poisons noble metal catalysts such as platinum, and therefore reduces the life time of the
catalytic convertors at the exhaust. The most commonly used catalysts for HDS are a cobalt
molybdenum (Co-Mo) and a nickel molybdenum (Ni-Mo) alloy, both of which are available on
different supporting materials such as aluminum oxide (Al2O3) and silicon dioxide (SiO2).

33

Lifetimes of these catalysts and their catalytic functions have been studied extensively and
reported in these selected reviews.76-79
Today, HDS is a well-established technology in every petroleum refinery site. Knowing
oxygen and sulfur are both group VI elements and the promising future that bio-oil can offer in
renewable fuel production, one might expect that petroleum HDS chemistry would be applicable
for hydrodeoxygenation (HDO) of bio-oil. Hence, HDO on reactive bio-oil aromatic model
compounds such as phenol, anisole, or guaiacol, using petroleum refinery catalysts has become a
popular research area in recent years.74, 75, 80-88 The results of these studies support the possible
transition from crude oil to bio-oil inside the current refinery infrastructure.
As discussed in the section on the hydrogen economy in chapter 1, the use of hydrogen in
renewable fuel production is questionable because H2 comes from methane, a fossil resource.
Moreover, the intensive energy requirement for HDO would most likely be supplied from
burning coal, which is the cheapest form of energy currently in the US. In other words,
upgrading bio-oil by HDO in a petroleum refinery would still be a fossil fuel dependent process;
it could partially mitigate US dependence on crude oil but it is not a completely sustainable
solution.
Ideally, bio-oil upgrading would be done without using fossil energy. While all
commercial forms of energy today are built on a fossil foundation, renewable sources (e.g. solar
electricity) are being rapidly implemented. This project seeks to develop a new chemical tool-Electrocatalytic Hydrogenation (ECH)—as an alternative HDO method for the fuel production
cycle based on these alternative energy inputs.

34

2.3

From Catalytic Hydrogenation to Electrocatalytic Hydrogenation
Compared to catalytic hydrogenation, ECH achieves hydrogenation using electrical

energy at atmospheric pressure and below the boiling temperature of its electrolyte, typically
water. ECH allows renewable electricity, captured from solar, wind, geothermal and other
alternative schemes, to be incorporated into bio-oil, raising its chemical energy content.
Although CH and ECH are run under vastly different conditions, both methods achieve effective
hydrogenations, suggesting that the harsh requirements of CH can be improved by using
electricity in ECH. In the following discussion, some reactions from the literature have been
selected to illustrate the similarities between CH and ECH. Benzene can be hydrogenated with a
Raney-Nickel catalyst at 300 oC and 20 atm pressure, but a reasonable conversion can also be
achieved at 45 oC with ECH.89 In the past, the Jackson group uncovered a similar reaction
improvement. Lactic acid had been chemically hydrogenated to propylene glycol in H2O with H2
over 5% Ru/C at 150 oC and 14.5 MPa H2.90 Using the same Ru/C catalyst as had been used for
chemical hydrogenation, lactic acid was hydrogenated electrocatalytically at 70 oC with traces of
perchloric acid as the electrolyte.91
For fuel upgrading purposes, ECH could be an ideal route because a) it operates at mild
conditions; b) it avoids the storage and use of H2 gas; c) the chemisorbed hydrogen density and
reactivity on the cathode surface can be controlled by current density. Protons can be generated
in-situ by anodic water oxidation and then recombine with electrons at the catalytic cathode to
reduce the organic components in bio-oil. By reducing C=O and C=C double bonds, it is possible
to eliminate carbonyl condensation processes which polymerize bio-oil, while increasing the
energy density of the bio-oil.

35

Bio-oil also contains many oxygenated aromatic compounds. Hydrogenation of such
electron-rich aromatic compounds is generally difficult due to aromatic stability. Under
favorable pH and surfactant conditions, useful ECH results can be achieved.92,

93

Because it

avoids use of H2 gas, operates at relatively mild conditions, and allows the potential
incorporation of renewable energy into the bio-oil to raise substrate energy content, ECH will be
the focus of this research project.
2.4

Brief Introduction to Electrocatalysis
Electrocatalysis is a familiar reaction in the field of organic electrochemistry, which itself

has been known since the late 18th century when Faraday first electrolyzed an acetate solution
and obtained a gaseous product, ethane.94 In that experiment, potassium acetate was electrolyzed
with platinum at high anodic potential to generate CO2, CH4, C2H6. This oxidative
electrochemical reaction is known today as the Kolbe oxidation (oxidation of fatty acids to
hydrocarbons with the loss of CO2). Another famous oxidative process is the Simons
electrochemical fluorination (replacement of all hydrogens and double bonds with fluorine using
a nickel anode). These reactions play vital roles in many industrial processes and natural product
syntheses. Fewer electrochemical organic reductions are familiar; perhaps the best known is the
reduction and dimerization of acrylonitrile to form adiponitrile.95
In general, reduction rates and selectivities depend on applied conditions such as catalytic
electrode material, supporting electrolyte and solvent, applied voltage, current density,
temperature, and pressure. Of the mentioned conditions, the supporting electrolyte and cathode
materials have the greatest influence on ECH.96 These two factors dictate how well the reactants
interact (adsorb) on the surface, which in turns affects the rate of hydrogenation reaction. When
the compound and hydrogen are adsorbed on the metal surface in close proximity, hydrogen can

36

bond with the substrate, reducing it. This set of cathode processes can be described in the form of
chemical equations:
Y=Z  Y=Z(ads)
H+(aq)

Adsorption of substrate Y=Z on electrode

-

Adsorption of protons to form hydrogen on electrode

2H(ads) + Y=Z(ads)  YH-ZH(ads)

Hydrogen reacts with substrate nearby on electrode surface

YH-ZH(ads)  YH-ZH(aq)

Desorption of reduced substrate YH-ZH from electrode

+ e  H(ads)

Scheme 2-1: general chemical equations for electrocatalytic hydrogenation of organic
substrate
Although no one single material is universally better, in general, the platinum group tends
to have a better catalytic effect. The reason was summarized by Navarro in his recent
publication:97, 98
“In the Co family ([Ar] 3d7 4s2), the differences in energy between s and d orbitals
hinder the re-hybridization of s valence electrons. These orbitals cause Pauli repulsion….Metals
such as Ni, Pt, and Pd have closer d and s orbitals, thereby allowing the re-hybridization of s
electrons and leading to chemisorptions of both hydrogen and unsaturated organic compounds”
-Navarro, M.
The cost of platinum is undesirably high so this metal is best avoided if other options are
available. Thus, the primary focus of this research project is to find a low cost, robust, and
simple catalyst that can achieve ECH of bio-oil at ambient conditions.
As mentioned above, substrate reduction takes place when there is enough adsorbed
hydrogen in close proximity with the substrate. However, this situation also would place two
adsorbed hydrogens close to each other, favoring formation of hydrogen gas, a process known as
the Tafel reaction. The closely related Heyrovsky reaction involves a proton from the electrolyte
solution reacting with a surface hydrogen and an additional electron from the cathode
(effectively, solution proton reacting with surface metal hydride). In either case, H 2 is formed
and current is wasted by being diverted from substrate reduction. Both mechanisms are
illustrated in Scheme 2-2.

37

The Tafel mechanism
2(H+(aq) + e-)  2H(ads)

2 protons are adsorbed on the electrode surface

H(ads) + H(ads)  H2(g)

Both adsorbed protons combine to form H2

The Heyrovsky mechanism
H+(aq) + e-  H(ads)

1 proton is adsorbed on the electrode surface.

H(ads) + H+ +e-  H2(g)

The adsorbed proton combines with a proton and electrons to
form H2

Scheme 2-2: chemical equations illustration on how H2 can be generated on the electrode
surface with two different mechanisms
If there is too much adsorbed hydrogen on the cathode, the reaction will favor hydrogen
evolution. In order to quantitatively evaluate the fraction of electrons spent on substrate
hydrogenation verses those spent on H2 formation, the Columbic efficiency fot the desired
reaction is calculated. If the efficiency drops during the reaction, either the electrode surface is
being deactivated for catalytic hydrogenation or the surface concentration of the reactant is
declining which diverts the reaction to form H2 instead.
2.5

Current Efficiency
Current efficiency is the percentage of total charge passed that effects the desired

reaction. An example of a Columbic efficiency calculation is as follows. In the case of
acetophenone hydrogenation to 1-phenylethanol, a 2e- transfer reaction, if 6 moles of electrons
were passed to form 1 mole of 1-phenylethanol, the Columbic efficiency would be 2/6 = 33.3%.
That means 33.3% of the electrons were used to perform the hydrogenation and the other 66.6%
were used for side reactions such as hydrogen evolution. The Columbic efficiency by the end of
the reaction can be calculated by:

38

( )

(

)

where Mole = moles of product formed. F = faraday's constant, 96485 C mol-1, n = number of
electrons per reaction, and Ctotal = the accumulative charge passed
For bio-oil hydro-upgrading, it is preferred to have an electrode that has a high Columbic
efficiency; an electrode that prefers hydrogenating the organic substrate rather than forming H2.
In general, Columbic efficiency of ECH in aqueous phase, which sometimes contains small
portion of organic co-solvent (e.g. methanol) to improve organic substrate solubility, has
Columbic efficiency ranges vastly. ECH of small aliphatic molecules (6 carbon atoms or fewer)
typically has a greater Columbic efficiency (~50 – 60%) than the larger ones (10 – 30 %); most
likely owe to how the organic molecules adsorb on the electrode surface. Given the majority
portion of bio-oil consists mostly small organic molecules, such as acetols, butandiones, and
furfurals, ECH is a promising strategy for bio-oil upgrading. However, bio-oil is a complex
mixture that also contains a large quantity of larger size aromatic substrates, like phenol,
guaiacol, and eugenol, which are harder to hydrogenate due to their aromaticity. ECH of these
molecules will require more catalytically active electrodes to break these stable π bonds. The
selection of electrode will be discussed later in the chapter. Aside from electrode surface, there
are chemical means to improve Columbic efficiency to favor hydrogenation of organic substrate
over side reactions such as hydrogen evolution.
2.6

Use of Cationic Surface Active Reagent in ECH
Surfactants are typically long molecules that have two ends with very different chemical

character – specifically one has hygroscopic, polar character and other is nonpolar and
hydrophobic.

39

Hydrophilic
Head

Hydrophobic
End

Figure 2-1: Cetyltrimethylammonium bromide (CTAB) is used as an example to illustrate the
positively charged hydrophilic head and hydrophobic end (the hydrocarbon tail) nature of
surfactant
As a surface active reagent, the hydrophilic end of the reagent adheres on the charged
surface of the electrode while the hydrocarbon end makes increases the hydrophobicity of the
surface. At an optimal concentration, the surface becomes more hydrophobic to favor organic
adsorption. As emulsifying reagents, surfactants can increase the solubility of the organic solute.
This in turn speeds up the exchange rate of organic between the surface and the bulk solution.
With these attractive properties, cationic surfactants have been employed in many ECH
reactions to increase CE.99-104 A common example would be cetyltrimethylammonium bromide
C19H42NBr (CTAB). However, if the concentration exceeds a limit, known as the Critical
Micelle Concentration (CMC), CTAB (or almost any surfactant) will form micelles and no
longer behave as a surfactant. For CTAB, the concentration necessary for the CMC varies a lot
depending on experimental conditions. In neutral water systems, it typically ranges from 0.9 –
0.98 mM at room temperature.105 For high Columbic efficiency, it is important to adjust the
amount of CTAB such that it is just below the CMC.
2.7

Selection of Electrodes and the volcano plot
The choice of electrode is critical to product selectivity and Columbic efficiency in ECH.

For instance, ECH of acetophenone on different surfaces changes the selectivity from simple
hydrogenation (forming 1-phenylethanol) to dimerization (forming pinacol).106 Other factors
40

such as current density, cathodic potential, temperature, and electrolyte can also change the
product outcome,98 but these factors typically have lesser influence on product selectivity
compared to cathode properties. In ECH, electrode metals can be categorized based on their H2
evolution overpotential. To understand this subject further, it is necessary to introduce the use of
the volcano plot.
The volcano plot shows the relationship between the metal-substrate binding energy
versus activity (typically measured in current density). The concept of a volcano plot is derived
from the Sabatier principle, which states that high activity happens only when the substrate and
metal binding energy is just right.107 If the metal-substrate binding energy is too low, the
substrate will not bind with the surface in the first place or binds too loosely to carry out the
reaction. On the other hand, if the metal-substrate binding energy is too high, the substrate will
be held tightly on the metal surface, and it will not be released for the reaction. The metals with
too low and too high binding energies are shown in the bottom left and bottom right corner,
respectively, of the volcano plot in Figure 2-2.108, 109 During the H2 evolution reaction, these
metals generate low current density, a measure of hydrogen evolution activity due to the
formation and release of every H2 which consumes 2 electrons, because hydrogen either adsorbs
too weakly or too strongly on their surfaces. Metals on the apex are the ones with the optimal
binding energy, which thus give the highest current density. However, these metals are typically
the precious metals that should be avoided if possible. Hence, the 2nd highest activity group,
which consists of copper, nickel, cobalt and iron, are of great interest to the ECH project.

41

Figure 2-2: The volcano plot of the common metals used in electrocatalysis.

In the past, the ECH literature has surveyed electrodes made of the metals
in the volcano plot. In general, high H2 activity metals tend to hydrogenate organic substrates
giving the expected hydrogenation product, whereas the low H2 activity metals tend to favor a
radical transfer mechanism, in which the surface prefers to transfer an electron to the substrate
and causes organic dimerization. In the case of acetophenone, 1-phenylethanol is the major
product when high H2 activity metals (Ni, Cu, Fe, Pt) are used,110-113 but the product selectivity
shifts to acetophenone pinacol when the reaction is done with low H2 activities metals (Hg, Sn,
Pb, Cd).

106, 114-116

The difference in product distribution is owed to the nature of hydrogen

adsorption on the electrode surface. Lead lies in the region of the “mismatched area” for
hydrogen evolution compared to nickel. In this case, under the same voltage, the hydrogen
coverage on lead’s surface is less than that of nickel due to the low M-H energy. The adsorbed
organic species on the lead electrode is therefore less likely to encounter an adsorbed hydrogen
atom. Therefore, hydrogenation of organics is less preferred on a lead surface. By comparison,
dimerization is more favorable compared to hydrogenation. In the case of nickel, the surface is
42

more frequently covered with adsorbed hydrogen compared to that of lead. So when an organic
species is adsorbed on the nickel surface, it will likely to meet an adsorbed hydrogen leading to
hydrogenation.94, 117
It is important to remember the mechanism of organic ECH consists of adsorption of both
hydrogen and the organic substrate shown in Scheme 2-1. Therefore, the hydrogen volcano plot
only covers half of the scenario.118 Unfortunately, the complexity of bio-oil would require the
survey or generation of many more volcano plots which is both time consuming and cost
ineffective (calculating organic adsorption energy costs computer time, especially with
complicated organic substrate). Hence, only the hydrogen volcano plot will serve as a model for
the electrode selection.
2.8

The use of Raney-Nickel and Other Skeletal Metals
Within the 2nd highest group in the volcano plot shown in Figure 2-2, nickel shows up as

the highest activity metal after gold, which is an expensive choice. This suggests the hydrogen
will be adsorbed on the nickel surface with the optimal amount of energy. For this reason, nickel
has been a popular catalyst for many organic ECH reactions.98, 119, 120 Various organic substrates
have been hydrogenated electrocatalytically under ambient conditions using a nickel cathode.
These ECH reactions occur at moderate voltages with reasonable efficiency in an aqueous
environment. However, plain nickel cannot hydrogenate aromatic rings. Fortunately, a special
nickel catalyst was discovered in the early 1900s that can address this problem – Raney Nickel.
Raney-Nickel, more formally known as skeletal nickel, was developed by the American
engineer, Murray Raney in 1924.121 In his original invention, a mixture of fine silicon powder
and nickel were mixed together in a 1:1 ratio before melting them together in coal fire
temperature. The resulting alloy was then treated in a basic solution (sodium hydroxide) to

43

dissolve (leach) out the silicon, leaving behind a porous, reduced nickel surface, which was
found to be surprisingly active in cottonseed oil hydrogenation. Two years after his discovery, he
modified the design by substituting silicon with aluminum forming the nickel-aluminum alloy,
which improved catalytic activity dramatically.122 Because of the porous surface, Raney-nickel is
also known as skeletal nickel for its non-uniform, high surface area, reticulated framework. Since
Murray’s discovery, other types of skeletal metals have been investigated with the same
methodology: melting aluminum powder with the metal of interest followed by leaching the
aluminum. Non-electrocatalytic hydrogenation using skeletal cobalt, copper, iron, platinum,
ruthenium have been examined and documented in great detail.123, 124
Today, Raney-Nickel is one of the most widely used catalysts in many areas. Although it
is made of low cost abundant metals, its catalytic effect is comparable to precious metal catalysis
in terms of aromatic hydrogenation.125, 126 It has been studied extensively in terms of leaching
rate,

dopants,127,

128

temperature

treatment,129

chemical

adsorption,130-139

organic

hydrogenation,126, 140-145 catalysis surface,146 kinetics147, hydrogen evolution148, 149 and more.
2.9

Preparation of Raney-Nickel Electrode
With its attractive properties, Raney nickel has been explored as an electrode in many

ECH reactions for organic substrates.103, 104, 129, 150-154 In these studies, preparation of RaneyNickel is either done by the hard pressing method151 or an electro-codeposition method155. In the
hard pressing method, the Ni-Al alloy is mixed with small portion of binding reagent and is hard
pressed into a cylinder shape before it is treated with in a hot base bath to leach out the
aluminum. The electro-codeposition works by trapping the Ni-Al alloy powder into the forming
nickel matrix during the nickel plating process. The trapped Ni-Al alloy powder is thus in good

44

electrical contact with the electrode and functions well electrochemically. A schematic depiction
is given in Figure 2-3 to illustrate to the plating process.
The deposition began with a 2.5 x 2.5 cm stainless steel 314 mesh 60 screen submerged
in 50 ml nickel-ammonia plating solution129 with 1.5 g of nickel-aluminum powder stirred in
suspension. A nickel plate facing parallel to the mesh screen serves as a sacrificial anode. Plating
current density, calculated in terms of the mesh side facing the anode, is maintained at 60 mA
cm-2 for 2 hours. The electrode is turned 180˚ every 30 minutes to ensure even deposition on
both sides.

Power supply

Sacrificial nickel bar
(provide nickel ion for the plating)

Stainless steel mesh (plate with
nickel and trap Ni-Al particles)

Stir bar
Ni-Al particles

Figure 2-3: Co-deposition of Ni-Al alloy onto the electrode surface. Condition: Preparation of
the Ra-Ni cathode uses the Lessard method of trapping nickel-aluminum alloy particles in an
electrodeposited nickel matrix.
The resulting Ni-Al deposited electrode is treated with 30% NaOH (6M) at 75 oC for 6
hours, and the electrode is stored in 4% NaOH solution until use.
2.10

Introduction of the Cobalt Phosphate Water Oxidation Catalyst
Platinum has always been a popular choice for the anode because of its high oxidative

potential and good catalytic activity compared to many other metals. When it is used in an
aqueous environment, it oxidizes water on its surface instead of being oxidized itself like other

45

more abundant metals such as nickel, copper, iron oxidize. However, platinum is not suitable for
large scale synthesis because of its preciousness and scarcity. For this reason, the use of a low
cost earth abundant metal water oxidation catalyst was proposed to satisfy the need of platinum.
Water oxidation catalyst development has been highly active for the last two decades for
renewable energy production purposes. In general, these studies explore and mimic
photosynthesis oxidation systems using different transition metals such as ruthenium, platinum,
cobalt, manganese, in combination with different ligand. These catalysts are designed to split
water to produce hydrogen and oxygen: 2H2O --> O2 + H2, effectively using electricity
presumably generated from renewable sources. When the O2 and H2 combine in a fuel cell to
form H2O, the reaction is exothermic releasing the stored energy to generate electricity.
In recent literature, the cobalt phosphate (Co-P) water splitting catalyst from 2008 has
received the most attention in the community for several of its outstanding merits.156 First of all,
compared to platinum, Co-P has a lower water oxidation overpotential, which means less energy
is required to split water. Second, the simplicity and low cost of the catalyst makes it easily
scalable so that it has great potential to address the global energy demand.157-160 Most reported
water splitting catalysts before 2008 either make use of precious metal or sophisticated ligands to
tune the catalyst reactivity.161, 162 Also, these catalysts typically operate in high pH environments
whereas Co-P can function in the neutral pH range (pH 6 – 8).
2.10.1 Preparation of the Cobalt Phosphate Water Oxidation Catalyst
Preparation of Co-P is very simple: dissolve 0.5 mM Co2+ in 0.1 M pH 7 phosphate
buffer followed a constant potential electrolysis reference to anode at (+1.3 vs. Ag/AgCl) for a
few hours. The formation of a black film with gas (oxygen) evolved marks the successful
deposition of the Co-P. The formed catalyst can be removed from its preparative solution and be

46

used in a new aqueous environment. Professor Daniel Nocera, whose group developed the
catalyst, demonstrated that the catalyst can operate in the water collected from the famously
polluted Charles River. Because of its inexpensive nature and ease of preparation, cobalt
phosphate will be explored as a low cost alternative to replace the use of platinum. The results
may dramatically increase the chances of applying ECH in large scale Bio-oil production.
2.11

Overview of System Setup

2.11.1 Constant Voltage vs. Constant Current Electrolysis
Before the discussion on cell setup, it is important to identify the difference between
constant current and constant voltage electrolysis because the reaction cell setup or apparatus
requirements can differ depending on setting. This concept requires an entry level background on
electricity that is usually taught in an introductory college level physics class. In here, the level
of knowledge will be covered just enough for the reader to understand how an electrochemical
system functions – however readers are still encouraged to learn how the basics of electronics
though.
In any electrical circuit, assuming the electrical resistance is constant, higher voltage will
result in higher current. This is because voltage is the driving force that pushes the electrons
through the circuit, and the current is the speed of the electrons travel. To make this more
understandable for organic chemists, imagine running a chromatography column; the mobile
phase travels slower if the gel is packed too tight. For the solvent to elute at a higher speed in the
same gel, pressure is needed to “push” the solvent harder in addition to the gravitational force. In
terms of electricity, the “push” from the chromatography example represents the voltage; the
speed of the mobile phase travel represents the current; and the tightness of the gel packing

47

represents the resistance of the electrical load. The overall analogical terms are summarized in
Table 2-1.
Table 2-1: Terms comparison between chromatography and electronic system
Chromatography
Electronics terms
Gel packing
Resistance
Flow of solvent
Current
Solvent driving force (gravity)
Voltage

If the gel gets tighter during the chromatography process, then a higher driving force is
need to push the solvent to maintain the same flow rate. The idea is the same for constant current
electrolysis: if the flow of electrons is required to continue at a fixed rate, the voltage needs to be
adjusted to compensate for resistance changes. On the other hand, for constant voltage
electrolysis, a change in the resistance will result in a change in current. This is because when the
driving force is fixed, changing the resistance will result in a change in the electron flow rate
(current). The relationship between resistance, voltage and current is summarized with V/R = I,
where V means voltage, R means resistance, and I means current.
For constant voltage setting, the subject can get more complicated when “absolute
potential” is required for some experiments. Although this project does not use constant voltage
(with or without absolute potential applied), knowing the topic will be beneficial for future
development.
A conventional direct current (DC) power supply can only regulate the voltage difference
between the cathode and anode in a constant voltage setting. However, the actual potential for
the individual electrode can fluctuate. In other words, the voltage potential on the cathode may
become increasingly negative while the anode shift more and more positive to compensate for
the change in cathode. The problem is depicted in Table 2-2.

48

Table 2-2: An example to illustrate how the absolute potential on individual electrode
changes during a constant voltage electrochemical reaction. Note: Ta and Tb describes
different stage during the ECH
Cathode
Anode
Potential
Potential
Potential
Difference
Initial
-2
+3
-5
Ta
-3
+2
-5
Tb
-1
+4
-5

In all 3 stages of the electrolysis, initial, Ta and Tb, the potential difference are all +1
while the absolute potentials of both electrodes are shifting. Such effect can dramatically affect
the results because the voltage dictates the energy level of the passing electrons (this subject will
be discussed in Chapter 5 with experimental data). To accurately maintain the absolute potential
on the electrode of interest, which is often referred as the working electrode, a reference
electrode is needed.
Reference electrode is an electrode that is electrochemically stable over a range of
potentials. The universal standard reference electrode is the Standard Hydrogen Electrode (SHE),
which generates H2 electrochemically and has a 0 v potential by definition. However, the
construction and maintenance of a SHE is rather inconvenient and time consuming. So, silver
chloride on silver wire electrode (Ag/AgCl) is the preferred choice for small scale short duration
reactions, such as cyclic voltammetry. (For preparation of a Ag/AgCl reference electrode, please
refer to the cited book chapter163). By controlling and monitoring the potential between the
reference electrode and the working electrode, an absolute potential electrolysis can be achieved.
Note that absolute potential electrolysis requires a sophisticated digital potentiostat that supports
three electrode inputs (working, counter, and reference electrodes) to operate.

49

2.11.2 Three Electrode System
Power is adjusted digitally to
ensure working electrodes
potential stays constant
e-

e-

A

e-

+

-e

-

e-

e-

B+e-

-eA+

B

Working Electrode

Counter Electrode

A

Reference Electrode

e-

e-

Counter electrode
potential is not
reference and is
observed not
monitored

Voltage difference
between Working and
Reference is monitored
digitally

V

Reference electrode
is placed closely to
the working
electrode.

Figure 2-4: Three electrode system where the reference electrode is referencing the cathode
(working), substrate A is oxidized on the anode surface and substrate B is reduced on the
cathode surface. The separator, typically a glass frit or Nafion membrane, may or may not be
necessary to the system. Note: V = Voltmeter, to measure the voltage between the reference
electrode and the working electrode. A = Ammeter, to measure the current at any time during
the reaction
The three-electrode system shown in Figure 2-4 is a typical electrochemical setup
employed in various electrochemical techniques. Among them, cyclic voltammetry is the most
frequently used technique to study redox mechanisms, substrate diffusion, and various
electrochemical aspects on the working electrode surface. (Note: working electrodes can be
either cathode or anode, depending on the reaction of interest) The potential of the counter
electrode in this case is not regulated, and is considered as a part to complete the circuit. The
separator, typically a glass frit or proton exchange membrane, is optional and is used to separate
the two compartments in order to isolate the reaction product.

50

2.11.3 Two Electrode System
Voltmeter is employed optionally to
monitor the potential difference between
the 2 electrodes
V
e-

e-

A

e-

+

-e

-

e-

e-

e-

Counter Electrode

e-

A

B+e-

-eA+

B

Working Electrode

Ammeter is
employed the
current flow.

Working and Counter electrodes are
regulated based on current passed; the
current density on both electrodes are
kept constant throughout the reaction

Figure 2-5: Two electrodes with the cathode (working) and the anode (counter); substrate A
is oxidized on the anode surface and substrate B is reduced on the cathode surface. Separator
is optional, and is typically a proton exchange membrane or glass frit. Note: V = Voltmeter,
to measure the voltage between the reference electrode and the working electrode. A =
Ammeter, to measure the current at any time during the reaction
Without the reference electrode, the electrolysis is simplified to two electrodes with a
power source shown in Figure 2-5. Since there will be no electrode potential regulation, the only
sensible setting would be a constant current electrolysis, in which the flow of electrons is
maintained at a fixed setting by an automated control potential. For electrocatalytic
hydrogenation, constant current is preferred because the surface hydrogen adsorption can be
controlled via applied current regulation. Also, in this research project, the use of Co-P catalyst
requires either a constant anodic potential or a fixed current density (current per given area) to
maintain the catalyst functionality. If a reference electrode (3 electrode system) is employed to
regulate the anode potential, the cathode current density will fluctuate making the hydrogenation
process difficult to monitor. However, if the cathode potential is regulated, then the anode
potential will fluctuate and disfavor the Co-P catalyst formation. Therefore, this project has
51

chosen the constant current electrolysis setting – fixing the current density to keep the flow of
electrons on both electrodes regulated. Also, Columbic efficiency calculations are simplest with
constant current electrolysis as opposed to constant potential electrolysis. This is because the
electrons are passed at a fixed rate and so the amount of charge is constant over every hourly
sampling block. Thus, no additional equipment, such as a coulometer, is needed to measure the
charge passed.
The undivided cell is essentially the two compartment cell without the separator.
Undivided cells can minimize voltage consumption because the contact between the two
electrodes is not “obstructed”. Moreover, the cost of large scale electrolysis can be reduced
because proton exchange membranes or glass frits are costly items and require regular
replacement. However, the lack of separator can result in undesired reactions. For example, the
reduced product released from the cathode can get oxidized when it contacts the anode or vice
versa. For bio-oil upgrading, a divided cell is preferable due to reactivity reasons. The large
amounts of oxidizable organics in bio-oil can undergo oxidation on the anode surface and then
potentially undergo polymerization. However, the lower voltage demand aspect of an undivided
cell is very attractive for large scale processing especially in the context of energy upgrading, so
it is worth trying to develop a system, or an anode that does not oxidize the organic substrate
during the bio-oil upgrading process. For this reason, this research project began by mimicking
Navarro’s work in organic hydrogenation in an undivided cell.97, 119
2.12

Summary
With low energy content, high acidity, and polymerizable nature, bio-oil has to be

upgraded via hydrogenation to increase its energy content and stability before it can be used as
fuel. Although the petroleum hydrodesulfurization catalysts studied are extensive and well

52

implemented in the current industrial upgrading infrastructure, the energy intensive environment
will decrease the sustainability of bio-oil as a fuel. Therefore, an alternative hydrogenation
scheme is needed – electrocatalytic hydrogenation (ECH).
ECH can serve as an alternative low cost strategy compared to catalytic hydrogenation. It
achieves hydrogenation using electrical energy at ambient conditions below the electrolyte
boiling temperature at atmospheric pressure. The electrical input of ECH allows incorporation of
renewable electricity capture from solar, wind, geothermal and various schemes and stores it as
chemical energy in the bio-oil during the upgrading process. In other words, ECH can be viewed
as an energy-upgrading bridge between biomass and fuel – allowing a vehicle to be powered
using renewable energy.
Two electrodes, skeletal nickel cathode, and cobalt phosphate coated anode, are proposed
as the key electrodes towards bio-oil hydrogenation. Skeletal nickel is effective in aliphatic and
aromatic hydrogenation, while the cobalt phosphate catalyst serves as a low cost alternative to
platinum to provide H+ (for hydrogenation) via water oxidation. The process would ideally use
electricity generated from renewable sources, and store the energy as chemical energy during the
upgrading process.
The goal of this project is to develop a scalable and low cost electrocatalytic system in
which bio-oil can be hydrogenated to fuel, thereby addressing the national demand for renewable
fuel. With this in mind, the first part of the project is to investigate the viability of using the
cobalt phosphate catalyst to bypass the use of platinum. Since the use of cobalt phosphate coated
anode in conjunction to cathodic organic hydrogenation is unprecedented, the first part of the
project will be devoted to understand the function of cobalt phosphates catalyst and its limits. In

53

addition to the cobalt phosphate study, other factors such as surfactants and current density will
be explored to build a solid foundation for the actual bio-oil hydrogenation.

54

Chapter 3

3.1

Electrocatalytic Upgrading of Model Lignin Monomers with Earth
Abundant Metal

Introduction
Electrocatalytic Hydrogenation (ECH) has been shown as the most promising upgrading

strategy in bio-oil. Compared to catalytic hydrogenation (CH), which operates at 150 – 200 psi
H2 and 150 – 400 oC, ECH can achieve hydrogenate pi bonds at atmospheric pressure and
relatively low temperature, typically below water boiling temperature. For fuel production
purposes, it will be the more preferred strategy because the upgrading stage will be consuming
much less energy. The energy required for the upgrading process will presumably come from
renewable sources generators.
This research project collaborates with Bio-system Engineering, where biomass will be
liquefied into bio-oil via fast pyrolysis (see Chapter 1 for details). The project aims to achieve
bio-oil upgrading electrocatalytically and investigation of the optimal conditions in which
maximum Columbic efficiency can be obtained. Several goals were set out during the project
initiation stage.
Achieve ECH of organic substrate in an undivided cell: traditional ECH experiments are
conducted in a divided cell where cathode and anode compartments are separated by an ion
exchange membrane, e.g. Nafion®. However, the membrane itself imposed a significant amount
of electrical resistance thereby increases the potential difference and thus the energy cost for
same amount of current delivered. To minimized resistance (or lower potential difference) it is
preferable to conduct ECH in an undivided cell.
Design of long life, low cost reaction system: for long duration system, the anode
material must have a high overpotential to prevent self-oxidation. This is why platinum is usually

55

employed for research purposes. However, the cost and rare nature of platinum would decrease
the cost effectiveness of the project. As mentioned above the Co-P Nocera catalyst has a great
current density (for oxygen evolution) compared to platinum. Hence, it is a goal to incorporate
the Nocera catalyst in the undivided cell.
Optimization of condition for multi substrate hydrogenation at minimal applied voltage:
Bio-oil consists of various different organic components. It is necessary to characterize the
selectivity and reduction order when different functional groups are present in the same reaction,
and to optimize potential to achieve efficient reduction of the multi-component system.
3.2

Electrocatalytic Hydrogenation of Acetophenone in Undivided cell
Navarro reported a sacrificial nickel anode system where simple non-aromatic olefins can

be reduced with high efficiency.96, 164, 165 Being able to reproduce literature results will be a key
step to design a non-sacrificial system, and gaining understanding of the reduction is crucial to
the project. Setup of the reaction is rather straight forward; it consists of two electrodes
connected to a power supply shown in Figure 3-1.

A

Anode

Sacrificial System:
Nickel anode was pretreated with
1M HCl followed by deionized
water wash. Soak in 1” into the
electrolyte

Cathode

both Sac. and Non sacrificial system
• Nickel
• Iron
Approximately 1” in the electrolyte

Non-Sacrificial System:
Stainless Steel was pre-deposited
with the cobalt phosphate
catalyst (See procedures). Soak in
1” into the electrolyte

Figure 3-1: Electrocatalytic Hydrogenation of acetophenone in an undivided cell with different
electrodes
56

3.2.1 Experimental Procedures
Sacrificial anode system, both electrodes were brushed with sandpaper. Then iron cathode was
pretreated with 1M CH3COOH and the nickel anode was pretreated with 1 M HCl for 1 hour
followed by a de-ionized water wash. Both electrodes were then immersed into 75 ml of
electrolyte (see Table 1) containing the organic substrate. Samples (0.5 ml) were taken initially
and at 30 minute intervals over the 150 minute reaction time.
Non-sacrificial system, the water oxidation catalyst was prepared separately in 0.1 M pH 7
potassium phosphate buffer with 0.5mM CoCl2•6H2O dissolved. Traces of white precipitate were
observed upon mixing of this solution due to the insolubility of Co3(PO4)2. Two pieces of 2.5 x 3
cm 304 stainless steel (60 mesh) screens were immersed as cathode and anode. Constant voltage
(Anode +1.30 vs. Ag/AgCl) was applied for 1 hour and the anode screen turned black indicating
the formation of the Co-P film. Analysis of the Co-P catalyst with cyclic voltammetry and SEM
imaging with EDS analysis both showed that the described procedure yields results in agreement
with the literature. Applications of such prepared anodes were explored with various cathode
materials. Results are summarized in Figure 3-2.

Supporting Electrolyte

Cathodic Reaction

Sacrificial Anode

Non-Sacrificial System

4:1 (Water:MeOH), 0.25M
NH4Cl

4:1 (KPi : MeOH)

AP + 2H+ + 2e----> PE

AP + 2H+ + 2e----> PE

2H+ + 2e- ---> H2(g)

2H+ + 2e- ---> H2(g)

2Ni2+(aq) + 2e- ---> Ni(s)
Anodic Reaction

Ni(s) ---> Ni2+(aq) + 2e-

2H2O ---> O2(g) + 4H+ + 4e-

Table 3-1: Condition for sacrificial and non-sacrificial of Acetophenone (AP) to 1-phenylethanol
(PE) at constant current (50mA)

57

3.2.2 Analysis
Samples were extracted with 1 ml CDCl3 and filtered with dry MgSO4 in disposable glass
pipette to filter out paramagnetic nickel ions (as their insoluble sulfate salts). Control
experiments show that extraction efficiency is 99.2% and extraction selectivity of both reactant
and product is in a good 50:50 ratio. Reaction conversions were then determined by NMR based
on the aromatic proton integration. For calculations for precautions of using H-NMR
quantitatively, please refer to Chapter 2.
3.2.3 Results and Discussion
60

Fe-Ni
Ni-Ni
Fe-CoP with 0.2mM CTAB
Ni-CoP with 0.2mM CTAB
Fe-CoP
Ni-CoP

Coulombic Efficiency (%)

50
40
30
20

10
0
30

50

70

90

110

130

150

Time (minutes)
Figure 3-2: Reduction of a0;cetophenone with different electrode at room temperature
The reduction of non-aromatic double bonds occurring at room temperature and high
electronic efficiency marks the beauty of the sacrificial anode system. However, the drawback is
that the metallic nickel of the anode must lose electrons, undergoing oxidation in order to send
electrons as current to the cathode. Thus the mass of the nickel anode limits the lifespan of the
system. Though not all were tried, experiments from the Navarro paper were easily and reliably
58

reproduced. For instance, results with an iron and nickel cathode were in good agreement with
the paper. The re-deposition (plating) of nickel onto the cathode constantly renews the surface
with Ni, which allows reduction to take place. Interestingly, the efficiency is slightly diminished
when the iron cathode is replaced with nickel or, even worse, with a Raney-Nickel cathode. This
was suggested by Navarro that the Fe-Ni combination may give special catalytic effect. As for
Raney-nickel, the high surface area and morphology was known to favor hydrogen evolution,
which decreases the columbic efficiency. As for the columbic efficiency, both Iron and Nickel
sacrificial system has agreeable and good C.E. However, even of the high surface area, RaneyNickel has a rather low C.E. which suggests it favors hydrogen evolution by far.
By replacing the sacrificial nickel with a stainless steel anode coated with the Co-P
catalyst, the duration of the system’s effective run-time increases significantly. No sign of anode
oxidation was observed even after 48 hours. However, the tradeoff of the long duration is a low
reduction rate. For the same amount of time, the overall C.E. dropped dramatically when
switching from sacrificial anode to non-sacrificial anode. To address this problem, cationic
surfactant was added in attempt to improve the C.E. With 0.2 mM of CTAB, the overall C.E. of
Fe-CoP increased from 7% to 15%.
The optimal concentration of CTAB was not investigated further because this preliminary
experimental was designed to test out the compatibility of the Co-P catalyst in conjugation with
organic hydrogenation. The effect of CTAB concentration verses efficiency will be shown later
in the chapter.
In summary, an electrochemically inactive surface can be modified with traces of
surfactant. More importantly, undivided cell with abundant material is demonstrated despite the

59

low C.E. This finding sets the path for ECH of other molecules that would be difficult to reduce
otherwise.
3.3

Reduction of Phenol in Undivided Cell
Seeing this promising system with the Co-P catalyst, reduction of phenol was

investigated with Raney-Nickel Cathode. Acidic, easily oxidized and polymerized phenolic
compounds are the main lignin-derived components in bio-oil, so successful reduction and
deoxygenation of these aromatics to simpler cycloalkanes represents an important mode of
upgrading and stabilization. In powder form, the high surface area Raney-Nickel catalysts are
effective in reduction of aromatic systems via classical hydrogenation and electrocatalytic
hydrogenation.166-168 Raney-Nickel particles can be electrochemically deposited in the matrix of
Nickel metal169 and the procedure is described below. Previously, Menard have shown that
reduction of phenol to cyclohexanol is possible with Raney-Nickel cathode in a divided cell
compartment even at 35 oC. With traces of cationic surfactant, CTAB, the Columbic efficiency
further improved. These experiments were repeated and the results are agreeable. Hence, the
second phase of the project was devoted to the development of an undivided cell for aromatic
reduction using Raney-Nickel as a cathode.
3.3.1 Experimental Procedures
The Raney-Nickel electrode was prepared by co-depositing Ni-Al alloy powder onto
stainless steel screen along with the electroplated nickel The nickel plating bath consists of
(213g/L NiCl2•6H2O, 30g/L NH4Cl, ~200ml/L NH4OH) and pH is maintained at 9-10 with
NH4OH. 3g of Ni-Al powder is added to an 80ml plating bath. Nickel rod is used as an anode
and a 2.5 x 3.0 cm stainless steel plate is used base support for the Raney-nickel cathode. The
aluminum is then leached out by 30% NaOH at 70 oC for 7 hours. The leached cathodes were

60

stored in 4% NaOH. The storage period was good for at least 3 months. The leaching condition
was studied carefully by Lessard based on the efficiency of ECH of phenanathrene. 170 On
average, 10 mg of Ni-Al alloy were deposited and deposition is every uniform with the constant
current density, 60mA cm-2. The deposition rate deviates on average 10 – 15 from average
implying the position and geometry of the plating plate may have influence to the results. The
results are summarized in Table 3-2 below. Dried samples of the cathode before and after
leaching were examined by SEM and EDX analysis to check the presence of nickel, aluminum
and changes on surface morphology.
Table 3-2: Deposition rate and deviations on Ni-Al alloy deposition. Mono side physical area
(2.5 x 2.5 cm) of the used for the J value calculation
Current
(A)
0.45
0.45
0.45
0.45
0.45
0.45
0.45
0.45

J
Rxn Time
-2
(mA cm )
(s)
60.0
7680
60.0
7200
60.0
7260
60.0
7380
60.0
7260
60.0
7260
60.0
7380
60.0
9000

Before
(g)
0.8911
0.8569
0.8739
0.8970
0.8970
0.8720
0.8975
0.9033

After
(g)
1.9978
1.8772
1.8951
1.9348
1.9366
1.9192
1.9330
2.1714

61

Dif.
(g)
1.107
1.020
1.021
1.038
1.040
1.047
1.036
1.268

Ni-Al Deposit Rate
(g)
(µg s-1)
% Deviation
0.107
13.99
12.4
0.084
11.60
15.9
0.077
10.55
17.4
0.078
10.52
17.1
0.095
13.09
14.0
0.103
14.13
13.0
0.075
10.20
17.7
0.097
10.79
13.7

Figure 3-3: Raney Nickel cathode before (left) and after (right) leaching of Al with NaOH at
70 ⁰C for 7 hours.
3.3.2 ECH of Phenol
A solution of 0.022g of phenol (without drying) was dissolved in 20ml of 0.1M KPi
buffer. Using Raney-Nickel cathode and the Co-P catalyst on stainless steel as anode, constant
current electrolysis was applied until the number of coulomb passed reaches 100% of theoretical
(assuming perfect Columbic efficiency). A 0.5 ml sample was taken, saturated with NaCl, and
extracted with CDCl3. NMR was used to determine the relative ratio of Phenol and cyclohexanol.
3.3.3 Results and discussion
Unfortunately, having tried almost all combination with pH 3, 5, 7, 0.25 mM, 0.5 mM
CTAB, 25, and 74 oC, none of these combinations showed reasonable amount of cyclohexanol.
At pH 7 and pH 9, with 0.25 mM, traces of cyclohexanol can be observed with NMR in the
baseline noise level. During the reaction, electrolyte turned to a brown color and the anode
surface turned brown. The color cannot be removed even with concentrated THF, HCl or HNO3,
which implies they were not attributed by the water oxidation catalyst. Literature revealed that
phenolic compounds can be polymerized even at low oxidation potential (+1.44 vs. Ag/AgCl) in
both acidic and basic environment on 304 stainless steel surface.171 The phenolic film adheres
62

strongly on the anode surface, blocking Co-P catalyst formation and water oxidation reaction and
the resistance of the system. Hence, reduction of phenol in undivided cell with the Nocera anode
was an unexpected challenge. Other phenolic derivatives, Guaiacol (2-methoxyphenol) and
Syringol (2,6-dimethoxyphenol) showed similar behavior. Hence, reduction of phenolic
compounds at this stage can only be done with a sacrificial nickel or in a divided cell where
cathodic material is separated from the anode compartment to prevent polymerization.
3.4

Reduction of Guaiacol in Divided Cell:
Having seen phenolic derivatives oxidized and polymerized, a divided cell is necessary to

achieve ECH of bio-oil. Further investigation on phenol reduction in divided cell will not be
conducted because a) it has been investigated in the past with platinum group metal and Raneynickel cathode.166, 167 b) the presence of phenolic derivatives is relatively less than Guaiacol
derivatives in bio-oil. Guaiacol derivatives are not costly, and they are good models compounds
for preliminary data purposes. To the best of our knowledge, ECH of guaiacol has never been
reported. Although Lessard has reported lignin treatment with Raney-nickel catalyst,172 in which
guaiacol was formed and maybe hydrogenated but the ECH of guaiacol was not studied
executively. Furthermore, Thomalla has reported ECH of phenol with traces of cationic
surfactant at different temperature and obtained reasonable yield.92 Hence, it is highly possible
that guaiacol can be hydrogenated into 2-methoxycyclohexanol electrochemically with the help
of cationic surfactant. However, the electron donating methoxy group may decrease the
reduction rate because it can stabilize the aromatic system through resonance with the oxygen
lone pair.

63

3.5

Experimental Procedures
Nafion 117 was boiled in 5mM H2SO4 for 1 hour, and boiled in 1% H2O2 solution for 1

hour. Cleaned Nafion membranes were stored in 1 mM H2SO4 solutions. All solutions were
prepared with HPLC grade water. Raney-nickel cathode preparation was described in part II and
a Co-P on stainless steel mesh was used as anode. 30mL of electrolyte was heated to the desire
temperature in water bath. Electrodes were connected through a rubber stopper by stainless steel
wire to slow down evaporation of electrolyte. Both cathodes and anode were placed closely to
the membrane to minimized internal resistance. After 20 minutes of pre-electrolysis, 20 mM of
guaiacol was added. Reaction time was based on charged passed for 100% theoretical yield
assuming perfect Columbic efficiency. 1 ml of sample will be taken at the end of the reaction
from cathode and anode compartment, saturated with NaCl, and extract with CDCl3 for NMR
analysis. Another 1 ml from both compartments will be analyzed by HPLC using mixed mode
C18 cation column using 13% acetonitrile in water as mobile phase and 0.8 ml min-1 flow rate.
Phenol and guaiacol are analyzed by the UV detector while 2-methoxycyclohexanol and
cyclohexanol are analyzed by water reflective index detector. Response curve of both detectors
and retention times were determined by standards injection. Note: Phenol was dried by
sublimation and checked with NMR for standard preparation. GC-MS also was employed to
analysis some of the samples due to unexpected technical difficulties with the HPLC system.
Standards with known concentrations were prepared in dichloromethane for standard curves. 0.1
ml of samples from experiment were taken from experiment and extracted with 1.0 ml of
dichloromethane and products were identified based on retention time and M.S.

64

Figure 3-4: Electrocatalytic hydrogenation of Guaiacol and the possible product and their
annotations.
3.6

Results and discussion
As expected, reduction of guaiacol is more challenging than phenol. In absence of CTAB,

guaiacol does not get reduced at 80 oC at 3 different pH buffer solutions: Ammonium Acetate,
Phosphate and Borate Buffer. Strong acid, such as HI, HCl, even SAC-13 (Nafion-H solid
superacid), have no effect on reduction of guaiacol. To achieve reduction without CTAB, a pyroreactor was modified with two stainless steel wire as electrical couple to achieve temperature
above boiling point.

Ar

Barometer

Power Supply

Ar Gas Inlet
Reaction Chamber
(See Left)

Power
Supply

Raney
Nickel
Electrode

Platinum
Mesh
Glass
Container
Glass Frit

Heating
Module

Figure 3-5: Electrocatalytic Hydrogenation of Guaiacol in high pressure setting (Left) the
physical setup of the overall system (Right) Schematic depiction of the reaction chamber
Trace of transformation (<4%) was observed at 120 oC at 100 psi of argon gas at both pH
7 and 8.2, but not with pH 3 ammonium acetate. However, with 0.50 mM of CTAB, guaiacol
began to undergo reduction at 80 oC in the normal atmosphere as shown in Table 3-3. Reaction
65

was carried out in a duel compartment cell without an elaborated pressurized setup shown in
Figure 3-6.

Figure 3-6: Electrocatalytic hydrogenation (ECH) of organic molecule in a divided cell
separated by a Nafion 117 proton exchange membrane.
At pH 8.2, reduction of guaiacol becomes more preferable as the relative ratio of guaiacol
to hydrogen on cathode surfaces decreases as the electrolyte becomes more basic. In the presence
of CTAB, the Columbic efficiency reaches as high as 35.9% at 0.50 mM, showing hydrogen
evolution is still the dominant reaction. At higher concentration of CTAB, the current efficient
drops again due to the formation of micelle and CTAB no long functions as a surface modifying
reagent.

66

Table 3-3: Reduction of 20 mM of Guaiacol in 0.1 M pH 8.2 borate buffer at 82 ± 2⁰C, and j=
8 mA cm-2 (current density was calculated by applied current (50 mA) divided by the electrode
area that is facing the anode 2.5 x 2.5 cm). G: Guaiacol, P: Phenol, 2M, 2Methoxycyclohexanol and C: Cyclohexanol
CTAB
Coulomb
Use
G
P
2M
C
C.E. (%) M. Bal (%)
mM
(C)
1
0.25
12.36
0.63
0.39
3.21
363.80
23.31
82.97
1
0.50
10.63
0.66
0.52
5.25
374.59
35.89
85.29
1
0.75
15.06
0.59
0.37
3.58
376.50
24.65
98.02
1
1.50
14.16
0.05
0.19
0.66
364.54
5.22
75.36
2
0.50
14.84
0.28
0.39
2.11
380.97
15.04
88.09
2
0.75
16.68
0.38
0.19
1.51
382.38
10.60
93.83
2
1.50
14.21
0.00
0.23
0.11
390.52
1.67
72.73
The influence of CTAB in mass balance was also examined. Interestingly, the mass
balance increased from 0.25 mM reaches highest at 0.75 mM CTAB and dropped after. This may
due to the positively charged micelle assisting the guaiacol diffuse across the membrane as
below in figure 8. A similar phenomenon is also observed in fuel cell application such as the
PEM-fuel cell and Methanol-fuel cell and is known as electro-osmosis (EOS).173 Despite the
membrane separation, neutral substrates are assisted by charged species to diffuse from one
compartment to another. In this case, it is possible that the micelle which has a positively
charged surface binds with guaiacol and the complex diffuses across the membrane. Interestingly,
EOS also occurs in the absences of CTAB as NMR shown traces of guaiacol was present in the
sample from the anode compartment. In this case, it may be the formation of water-guaiacol
complex to facilitate the diffusion across the membrane as shown in Figure 3-7.

67

Figure 3-7: Depiction of CTAB at different concentration in solution. Low and optimal will
cause CTAB to form on adsorption on surface while medium high to high will encourage
formation of michelle and may encourage diffusion of organics through the cationic exchange
membrane (Nafion)
After the 1st trial of experiment, the Raney-Nickel electrodes were washed with distilled
water, rinsed with acetone, and stored back in the 4% NaOH for 24 hours before the 2 nd use. It
was expected that Raney-nickel electrode prepared by the described method would have a good
life time and the C.E. would only decrease by a small amount.117 Surprisingly, in a second
identical application condition the Columbic efficiency dropped almost 50%. This unexpected
result may due to the diffusion of Co2+ ions, which was part of the oxidation catalyst, migrates
from anode to cathode compartment and be reduced on the Raney-Nickel electrode surface.
Further investigation with be done with ICP-MS to determine the concentration of Co2+ in the
cathode compartment. The result will be compared with control experiment where Pt wire
substitute as anode. If such hypothesis is valid, it is important to minimize the Co2+ deposition by
either employing ion participation method or consider other catalysts to prolong the life time of
Raney-Nickel electrode.
3.6.1 Mechanistic Investigation of Guaiacol Demethoxylation
Production of guaiacol to cyclohexanol was serendipity. The elimination of an oxygen
atom from guaiacol to phenol effective increased the energy content from 27.5 MJ kg-1 to 31.0

68

MJ kg-1 an increase of 12.7%. When phenol is hydrogenated to cyclohexanol, the specific energy
increased to 34.5 MJ kg-1, an additional boost of 11.3% of the energy.
The alkoxydemethoxylation occurred at an ambient condition was highly encouraging. In the
past, similar alkoxy cleavage was achieved by a purpose-built homogeneous nickel catalysts with
various reductants have also been reported.174-176 However, the conventional reductant,
molecular H2, is derived from natural gas or petroleum refining in today’s markets, so despite its
lack of carbon, it must be viewed as a fossil resource. An interesting transition metal-free alkoxy
cleavage system makes use of organosilane but requires stoichiometric amounts of reagent and
strong base which limits its potential for scale-up.177 The formation of phenol from guaiacol
under an ambient condition, atmospheric pressure, and in the absence of external H2 supply is
remarkable. Furthermore, the heterogeneous electrocatalytic nature of the reaction makes it more
favorable to industrialization compared to homogenous reaction.
Driven by the serendipities, a series of mechanistic investigation experiments were
conducted to see how the demethoxylation occurred, and see if the same transformation could be
performed other compounds with similar structure.

69

Table 3-4: Values are percentages relative to starting material buffer concentration, 12.1 ± 1.5
mM as determined by gas chromatography. Reductions employed a Raney-Nickel cathode in
30 ml of 0.1 M pH 8.0 potassium borate buffer, and a stainless steel anode coated with cobalt
phosphate catalyst in 30 ml of 0.1 M pH 7 potassium phosphate buffer. Reaction started with
60 minutes of pre-electrolysis and followed immediately by compound addition and then run
for 6 hours at 50 mA (8 mA cm-2)* at 75 ˚C. Total current passed in this time represents 4.1 x
the amount required for substrate reduction to ROH and cyclohexanol. b) Cis and trans
product are in equal amounts. c) Only a single peak was observed by GC. *Note: current
density is calculated based on the single side electrode facing the membrane and anode
compartment.
Material
R Group
Balance
b
2-MeO
1.1
Traces
78.9
80.0
2-EtO
1.8
90.7
92.5
2-iPrO
4.6
1.3
82.8
88.7
3-MeO
9.9
0.6
32.7c
41.1
84.3
b
4-MeO
Traces
Traces
48.1
44.6
92.7
As shown in Table 3-4, when guaiacol is subjected to exhaustive galvanostatic ECH,
cyclohexanol is the main product observed; only traces of 2-methoxycyclohexanols, direct
aromatic ring hydrogenation products, are seen. Thus, ECH effects both methoxy ether cleavage
and phenol hydrogenation. Importantly, under these same conditions, phenol itself readily and
rapidly undergoes hydrogenation to cyclohexanol. This excellent selectivity for deoxygenation
and hydrogenation appears promising for bio-oil energy upgrading. Though this aryl ether
cleavage had been seen in earlier work by Lessard et al., it was not particularly noted as
unusual.151
One possible model for interaction between the catalyst and phenolic substrates involves
the aryl ring lying flat on the catalyst surface.178, 179 Inhibition of such binding might be expected
as side chain bulk increases from methyl to isopropyl. However, with guaiacol analogues 2ethoxy- and 2-isopropoxy phenol, the results are essentially similar to those from guaiacol. As
shown in the time sequence in Figure 3-8a, the ether alkyl group size seems to have little effect,
at best slightly increasing reaction rate.
70

The guaiacol isomers 3- and 4-methoxyphenol were also subjected to the above reduction
conditions and produced cyclohexanol as major products, Figure 2b. Observable quantities of 3and 4-methoxycyclohexanol, the direct aromatic hydrogenation products, were also observed.
Individual trial results, shown in table 1, found that alkoxy group cleavage appears to be favored
by closer proximity to the phenolic hydroxyl group.

71

25

Concentration (mM)

20

15

10

5

0
0

1

2

3
Time (hr)

4

5

6

12

Concentration (mM)

10

8

6

4

2

0
0

1

2

3
Time (hr)

4

5

6

Figure 3-8a. (upper) Trial a: 1:1:1 mixture of guaiacol (---), 2-ethoxyphenol (---), and 2isopropoxyphenol (---) undergoes electrocatalytic hydrogenation (ECH) under the same
condition described in table 1. All three reactants form a common intermediate, phenol (---),
which is further hydrogenated to cyclohexanol (---). Only traces of the alkoxycyclohexanol
products of direct hydrogenation were detected. Figure 3-8b (lower) Trial b: 1:1:1 mixture of
guaiacol (---), 3-methoxyphenol (---), and 4-methoxyphenol (---) undergoing ECH as above.
Formation of 3-methoxycyclohexanol (---) and 4-methoxycyclohexanol (---) is observable.
Phenol (---) and cyclohexanol (---) were also formed as the major products. Curves are
polynomial fits included to guide the eye.
In further exploratory studies, Raney Cobalt (Raney Co) and Devarda copper, skeletal
metals other than Raney nickel, were examined as cathodic catalysts. Cobalt-aluminum and
Devarda copper150 (copper zinc aluminum) electrodes were prepared using the same method as
72

for the Raney nickel. Only Raney Co formed observable traces of phenol and cyclohexanol after
prolonged reaction time, as expected based on Raney Co's previously noted lower reactivity
relative to Raney nickel.124, 180 Importantly, control experiments showed that a plain nickel bar
electrode completely failed to reduce guaiacol or phenol. Together, these results indicate that
both demethoxylation and hydrogenations require a highly active skeletal nickel surface.

Figure 3-9: ECH of Syringol (IUPAC: 2,6-dimethoxyphenol) in 30 ml of 0.1M pH 8.0 Borate
Buffer with 0.5mM CTAB at 75 ⁰C. Syringol(---), cyclohexanol(---), guaiacol(---), phenol(--) and current efficiency(---), vertical axis on the right

Syringol (2,6-dimethoxyphenol) was also studied under the same condition, but the
efficiency of the process is a lot lower compare to the 20 mM trials (Figure 3-9) . The dimethoxy character of syringol further increases the electron density of the ring and decreases of
the rate of demethoxylation. This supports demethoxylation is indeed the rds. Throughout the
reaction, concentration of guaiacol and phenol stayed at low level while cyclohexanol continue
to build up. To our delight, we observe no formation of the 2,6-dimethoxycyclohexanol,
implying the system is exclusively de-alkylation process occurs regardless the nature of the
73

aromatic ring. Such results show this strategy is ideal for bio-mass pyrolysis liquid processing
and energy upgrading.
3.6.2 Material Balance
To probe the sequence of events in the dealkoxylation and phenol hydrogenation
processes, Columbic efficiency (C.E.), the percentage of current spent on the de-alkoxylation and
reduction to form cyclohexanol, was studied as a function of reaction time.

Figure 3-10: (a) Reactants, products and C.E. (%) data from ECH at 50 mA (8 mA cm-2)*
and 75 ˚C of 30 ml 18 mM guaiacol in 0.1 M pH 8.0 potassium borate buffer: guaiacol(---),
phenol(---), methanol(---), cyclohexanol(---), and material balance (---).

As seen in Figure 3-10, cyclohexanol is produced at the expense of guaiacol. Once
reaction began, phenol was observable in small amounts throughout, suggesting that it is formed
and liberated but undergoes hydrogenation much faster than guaiacol conversion to phenol. In
other words, the demethoxylation of guaiacol appears to be the rate limiting step.
The material balance of the reaction is calculated by the sum of substrate, and is dropping
slowly during the course of reaction. A small amount of cyclohexanol was detected in the anodic
chamber and suggests that diffusion of organics across the membrane may account for the
74

observed material losses. Compare to cyclohexanol, the alkoxyphenols might be more vulnerable
to membrane crossing diffusion. The deprotonated hydroxyl group might be attracted to the
positively charged anode and encouraged the diffusion process as a result. Although guaiacol and
phenol were not detected in the anodic chamber, it is possible that these aromatic molecules
might have been oxidized and polymerized on the anode surface.181
At the present stage, adsorption study on highly porous surface remains challenge to
present spectroscopic techniques, we deduce the possible mechanism through various literature
results with our experimentation data. Hence, we took the data Ni (111) surface adsorption as the
ground of our analysis. Studies suggested that phenol and aniline both adsorbed through the piring interaction on metallic nickel surface178,

179

This suggests hydrogenation of the pi-bond

between the hydroxy group and methoxy group shown in scheme 3 is unfavorable as both
oxygenated group will disfavor the adsorption for such reaction to occur.
This could be due to the retention of the adsorption of the aromatic substrate, which
eventually got hydrogenated and desorbed from the electrode surface. A simple adsorption
experiment was conducted to investigate the amount of adsorption of phenol, the intermediate
substrate, on the electrode surface and the method is illustrated in Figure 3-11. A new Raney
nickel electrode was submerged into 30 ml of electrolyte solution containing 295.5 µmol (9.85
mM) of phenol without current passing for 5 minutes, and then washed with the borate buffer
electrolyte for 5 minutes. The electrode was then subjected to a new electrolyte bath for 30
minutes of electrolysis. After the initial adsorption step, only 228 µmol (7.6 mM) remained in
the original phenol electrolyte solution. The electrode was the washed with clean electrolyte, in
which 1.1 mM of phenol was washed away before subjecting to ECH. This purpose of this is to
allow the adsorbed phenol would undergo ECH to become cyclohexanol and desorbed from the

75

surface. During the electrolysis, cyclohexanol was detected from the solution, indicated the
amount of adsorbed phenol must have been hydrogenated on the electrode surface.

Figure 3-11: Adsorption study to investigate the cause of material loss.

The adsorption of aromatics is fall from accounting all the materials loss during the 6
hours guaiacol electrolysis shown in Figure 3-10b. Hence, a simple diffusion study of organic
molecule was studied. In this control experiment, a plain nickel with use to mimic the reactions
but to prevent aromatic hydrogenation. The results reviewed guaiacol and phenol were
vulnerable to diffusion across the Nafion membrane at 75 oC. After 5 hours of electrolysis of 20
mM (30 ml in volume) of guaiacol and 10 mM of phenol, only 14 mM of guaiacol and 9 mM of
phenol remained in the cathode compartment. 0.6 mM of guaiacol, and 1 mM of phenol were
found in the anode compartment during the beginning of the ECH but disappeared in the end.
This suggests both guaiacol and phenol diffused across the membrane in the beginning and was

76

oxidized on the anode surface. The diffusion studies accounts for the loss of materials, and
suggested Nafion membrane as a compartment separator should be reconsidered. Search for
other compartment separators is currently underway to improve the material balance due to

1.4

y = -0.8647x + 17.806

16

1.2

14

1.0

12
0.8

10
8

0.6

6

0.4

4

0.2

y = 0.8098e-1.524x

2
0

0.0
0.0

1.0

2.0
3.0
Time (hours)

4.0

12
1.4

y = -0.1714x + 9.8546

10

1.2

8

1.0
0.8

6

0.6
4
0.4
2

0.2

y = 1.4231e-2.663x

0

5.0

[Phenol] mM AnodeCompartment

18

[Phenoll] mM Cathode Compartment

20

[Guaiacol] mM Anode Compartment

[Guaiacol] mM Cathode Compartment

diffusion.

0.0
0.0

1.0

2.0
3.0
Time (hours)

4.0

5.0

Figure 3-12: Diffusion test on guaiacol and phenol under experimental condition. Raney
nickel electrode is substituted with a plain nickel bar to prevent the hydrogenation of the
substrates.

3.6.3 Current Efficiency
A different approach on the Columbic efficiency was attempted by plotting the charged
(in coulomb) spent on product generation verses the amount of charged passed to the reduction
shown in Figure 3-13.

77

Figure 3-13: Total charged passed of the reaction vs. amount of charge spent on guaiacol
reduction for the reaction shown in Figure 3-10.
Differentiation of “charge spent on product” over “charge passed” yields the function of
Columbic efficiency of the guaiacol ECH reaction describes in Figure 3-10. (Note: 8 and 6 moles
of electrons were consumed for each mole of cyclohexanol and phenol formed, respectively). A
2nd polynomial trend line (lime green) was applied to cover the data points of the experiment,
and gave an R2 value of 0.99, an excellent fitting index, and a function of y= 0.201 – 6.389 x 105 2

x . Differentiating function yields: dy/dx = 0.201 - 1.278 x10-4x. A set of data points were plotted

in red with the secondary axis on the right (red) showing the highest Columbic efficiency of the
process was 20.1%

78

3.6.4 Mechanistic Insight
The demethoxylation of guaiacol under such ambient condition was the key to the lignin
model compound upgrading. To understand the process better, we postulated several schemes
and conducted the study in deuterated solvent.

Figure 3-14:Proposed mechanism for the alkoxy-demethoxylation. Scheme 1 has been
eliminated with 1H-NMR study. Scheme 2 and 3 are the probable route for the reaction.
1

H-NMR showed presence of methanol, confirmed with sample spiking, with the singlet

peak at 3.49 ppm in the product mixture. The presence of non-deuterated methanol suggests
scheme 1 is an unlikely route as the formaldehyde produced from a basic concerted elimination
would have been hydrogenated to CH2DOD, which would shows up as a triplet.182
ECH of guaiacol was run in a similar condition as described in the experimental section
using D2O as a solvent. Reaction was only allowed to process 30 minutes instead of 6 to capture
all 3 components: cyclohexanol, phenol and guaiacol. Sample was analyzed by both 1H-NMR
and GC-MS to determine the possible mechanistic pathway by which guaiacol underwent
demethoxylation. Control experiments, where both guaiacol and phenol independently subjected
79

to 75 oC in an pH 10.5 borate buffer for 6 hours, showed no deuteration on the aromatic ring in
the absence of Raney nickel.
In principle, the deuterium installation on the phenol ring could provide valuable insight
of the demethoxylation since phenol generated from Figure 3-14 scheme 2 would result in diand tri-deuterated phenol in equal amount while Figure 3-14 scheme 3 will result solely dideuterated phenol. However, it occurred the formed phenol would undergo deuterium exchange
immediately at an unexpectedly high rate. The structural memory of the phenolic ring was erased
by the natural H/D exchange and led us to an indecisive conclusion of the demethoxylation
mechanism.
In addition to the deuterium study, we have also studied the reaction in the presence of pcresol, as a phenol competitor. Control experiment showed p-cresol and phenol hydrogenated at
approximated the same rate. 1:1 and 1:2 ratio of guaiacol to p-cresol was subjected to ECH, and
we observed the formation of cyclohexanol is sensitive to the present of p-cresol, which appears
to be competing with the phenol for the aromatic hydrogenation site. This indicates the formed
phenol must have desorbed from the surface, and competes with p-cresol for the hydrogenation
site.

80

Figure 3-15: Competition study of 10 mM guaiacol in the presence of various amount of pcresol. Upper left (a) guaiacol analysis, upper right (b) phenol analysis, lower left (c)
cyclohexanol analysis and lower right (d) p-cresol and 4-methylcyclohexanol analysis: With
20 mM of p-cresol added (----), with 10mM of p-cresol added (----), and no p-cresol added (---). Legends are same across all figures, each figure focuses on the specific compound of
interests specified on the vertical axis.
Interestingly, p-cresol also seems to be inhibiting guaiacol consumption as shown in
Figure 3-15. This leads us to believe p-cresol could be occupying the same reaction site that
achieves demethoxylation on guaiacol. If the two reactions: demethoxylation and aromatic
hydrogenation were taken place at two different sites, guaiacol depletion would remain
unaffected in the presence of p-cresol, and that is contrast to the evidence showed on Figure 3-15.
Along with the Figure 3-14 scheme 2 on the mechanistic equations, we postulate a mechanism
where guaiacol first adsorb on an active site and get partially hydrogenated, followed by the
release of the partially hydrogenated ring. The desorbed product is recirculated in the solution

81

and goes through a basic elimination to produce the corresponding alcohol, methanol in the case
of guaiacol, and restore the aromaticity to become phenol.
Cyclohexanol formation in the presence of 20 mM p-cresol formed significantly slower
than other two listed condition as shown in Figure 3-15c. Together with the information from
Figure 3-15b, it could be established that p-cresol was an effective phenol competitor substrate
for the adsorption/desorption study. Figure 3-15d was provided to illustrate the p-cresol was
converted into 4-methylcyclohexanol without reacting with other substrate in the reaction.
In further pursue of the mechanism, we have other similar substrates to explore the
demethoxylation process. Anisole did not undergo demethoxylation to produce benzene but
slowly goes through demethylation to produce phenol instead. and the stability of the resulting
anion, phenolate or methanolate, is crucial to the reaction. Moreover, this also implies the
presence of hydroxyl group is crucial to the demethoxylation process, leading to the involvement
of tautomerization during the reaction.
Similar

to

guaiacol,

o-anisidine

(IUPAC:

2-methoxyaniline)

also

underwent

demethoxylation producing aniline. However, the formed aniline was not hydrogenated to
aminocyclohexane183 as aniline is lesser favorable to hydrogenation compared to phenol with
nickel catalyst.183, 184 To support the foreign calculation data, all the possible pathway of the
guaiacol were mapped out. Base on the calculated hydroxyl-aromatic adsorption energy on the
Ni (111) surface along with the possible de-alkoxylation pathway, the cresol, and syringol results,
we proposal the mechanism goes through a stepwise hydrogenation sequence followed by an
alkoxy elimination to form phenol as shown in the following scheme:

82

Figure 3-16: Possible mechanism for the de-alkoxylation of 2-alkoxyphenol
3.7

Summary
This chapter described a mild electrocatalytic deoxygenation/hydrogenation process for

reduction of lignin model compounds in a simple, low-cost system that avoids the use of
precious metal or costly molecular catalysts. Remarkably, instead of arene reduction, the first
event in ECH of alkoxyphenols is the cleavage of the aryl-OR ether bond. The anode’s cobaltphosphate water oxidation catalyst selectively oxidizes water and prevents corrosion. The cobaltphosphate system served through our typical 6-hours reactions, and the 16-hour syringol trial
with no signs of degradation. Also, though Nafion is known to transport cations, no cobalt was
detected by EDX on the used Raney nickel cathode. On the other hand, the Raney nickel
cathodes were found to lose their catalytic hydrogenolysis activity over longer reaction runs. We
will report on our efforts to stabilize the Raney nickel ECH electrodes in a separate publication.
This upgrading scheme opens a new way to maximize yields from biomass-based
feedstocks via carbon-retentive energy upgrading using renewable electricity. In turn, it
represents a strategy for buffering demand-mismatched production of solar or wind energy by
storing it in a fungible chemical form.

83

To optimize efficiency and working lifetime of the system, areas of ongoing development
include improvements in cell design, energy and Columbic efficiency, and cathodic
electrocatalyst stability. The organic chemical transformations described here also have synthetic
potential. Studies to probe selectivity and mechanism of the aryl ether cleavages, and to extend
the range of substrates are also underway.

84

Chapter 4

Electrocatalytic Hydrogenation of Aliphatic Model Compounds in a Solid
Polymer Electrolyte Electrolyzer (SPEE)

4.1

Introduction
Reduction of the bio-oil lignin monomer, guaiacol, has been demonstrated in a

conventional dual compartment cell where a mildly basic borate buffer solution was employed as
a supporting electrolyte. The selective aromatic ether dealkoxylation observed under nearly
ambient conditions of pressure and temperature was encouraging. However, the need for added
electrolyte creates a barrier to ECH implementation in large scale processing; use of electrolyte
salts incurs materials and disposal costs; solubility of inorganics in bio-oil may be limiting; and
organic supporting electrolyte are often expensive. In order to make ECH attractive for fuel
production, the simplest process and cheapest chemicals possible should be used, while
expensive electrolytes should be avoided. Moreover, the energy input for fuel production should
be kept at minimum to make the processing energetically and economically favorable.
The solid polymer electrolyte electrolyzer (SPEE) project began with two thoughts: can
the two electrodes be placed right on the membrane surface a) to minimize the potential needed
to drive ion transport, and b) to eliminate the use of electrolyte by oxidizing water at the
membrane surface, generating protons for conduction through the membrane. This latter idea
was inspired in part by Professor Nocera’s artificial leaf design in which the two electrodes were
built on a silicon wafer to create a closed loop ion transport system for water electrolysis.
In many ways, the SPEE functions very similarly to a chemical fuel cell. In a chemical
fuel cell, for instance the hydrogen fuel cell, hydrogen gas passes through an electrode that is
loaded with hydrogen oxidization catalyst, typically platinum on carbon. The oxidized hydrogen
(proton, H+) then travels through a proton exchange membrane, typically Nafion, while the

85

electrons travel through the circuit. When the protons reach the other side, they react with
oxygen plus electrons to form water. Formation of water is exothermic, which means that the
reaction is thermodynamically favorable. For every water molecule that forms, four electrons are
“dragged” from the anode to cathode. The demand for electrons generates voltage, and therefore,
current to power electronic devices. Chemical fuel cells exploit the relative thermodynamic
stability of the reaction product, water in the case of H2 fuel cell, to power the movement of
electrons. In short, chemical fuel cells harvest electrical energy from exothermic reactions. (E.g.
H2(g) + 0.5 O2(g)  H2O(l) ΔH = -286 kJ/mol)
+ App.
e-

eH+

H+
nH+
nH2

H+
H+
H+
H+

Electrode
loaded with H2
oxidizing
catalyst

e-

e-

Proton
Exchange
Membrane

O2 + 4e- + 4H+

O2

2H2O

2H2O
Electrode
loaded with
O2 reduction
catalyst

Anode for
water
oxidation

H+
H+
H+
H+

2H2C=CH2 + 4H+ + 4e2CH2H-CH2H
Catalytically
active metal for
ECH

Proton
Exchange
Membrane

Figure 4-1: Schematic diagram to show the (Left) operating principle of a chemical (H2) fuel
cell and (Right) how the SPEE operates. Note: App. is short for electronic applications
Instead of transforming chemical bond energy into electrical energy, the SPEE takes
electrical energy and in this project, stores it as chemical energy by modifying the molecules
found in bio-oil via hydrogenation. The SPEE adopts the chemical fuel cell setup but uses
electrical energy for electrolysis. The close proximity of both electrodes to the membrane surface
allows ions to travel from anode to cathode without going through the reaction medium. This
enables the reaction to happen at a moderate voltage in an ion-deficient environment.
86

In the past, Ogumi et al. published a series of reports to demonstrate the use of the solid
polymer electrolyte electrolyzer (SPEE) in various organic electro-syntheses.185-191 In his work,
platinum was first deposited on the membrane using the Takenaka-Torika (T-T) method192
followed by electro-deposition of different metals such as gold187, copper189, zinc191, antimony191
and manganese190. A notable success in this series is the Kolbe reaction; this oxidative
decarboxylation and radical dimerization based on carboxylic acid was achievable on a platinum
modified membrane in a SPEE setting.186, 193 Soon after, the use of the SPEE in waste water
treatment such as organic dehalogenation194, NOx reduction195-199, water electrolysis200-202, and
organic synthesis191, 203-208 emerged in the literature.
The use of conventional hydrogen fuel cell membrane electrode assembly (MEA), made
of platinum catalyst carbon paper, had also been reported in food hydro-processing and more.209212

Recently, the SPEE has been utilized in demonstrating production of valuable chemicals from

bio-oil abundant compounds.213-215
It must be noted, however, that all of the examples mentioned above require the use of
platinum or other expensive metals for MEA construction. Seeing that the goal is to industrialize
bio-fuel production using bio-oil, the use of expensive metals should be avoided if possible. Of
the literature surveyed, only one example was free of precious metals; in this case, silver was
deposited instead of platinum on the membrane surface before the deposition of other metal
catalysts. The reaction hydrogenated benzaldehyde to benzyl alcohol with only a modest
Columbic efficiency (11 – 23%).216
Seeing the potential benefits SPEE can bring to renewable fuel production, this project
seeks a low cost metal that can catalyze organic hydrogenation with good catalytic activity and
longevity. The goal is to develop a scalable continuous flow style SPEE reactor that can serve as

87

an immediate upgrading unit at the bio-oil exhaust shown in Figure 4-2, and produce a partially
saturated bio-oil stable enough to be stored and transported.217 The primary function of the SPEE
will serve as a bio-oil stabilizing unit by hydrogenating carbonyl and organic acid reduction.
Catalysts capable of such reduction as well as aromatics hydrogenation should be targeted.

Ground
Biomass

Fast
Pyrolysis
Unit

Back feed
to pyrolysis
unit

Bio-oil

H2

RH-RH + 2OH-

2H+ + 2e-

R=R + 2H2O + 2e-

Saturated bio-oil

Storage or
centralized
processing to
fuels or
chemicals

2H2O --> O2 + 4e- + 4H+

(pH 7.0 Phosphate Buffer)

Figure 4-2: Solid Polymer Electrolyte Electrolyzer processing unit that connects to the fast
pyrolysis unit. Bio-oil can be upgraded immediately upon exit.

4.2

Preliminary Experiment I – SPEE Designs and Construction
The SPEE project has been ongoing for 2 years. Different versions of reaction cells have

been designed to address flaws and to optimize the Columbic efficiency of ECH. The project
began without awareness of the precedents of any SPEE or related work. The term solid polymer
electrolyte electrolyzer (reactor) was not known to the project until encountering the term from a
chemical fuel cell review. Hence this project began with the concept of constructing a reversed
fuel cell using low cost materials.
Unfortunately, all chemical fuel cells reported in the literature used research-grade
chemical fuel cell equipment, which are extremely costly and time consuming to make.
Moreover, these fuel cell reactors are designed to handle a small quantity of gas flow and hence
it is less practical to use them for solution phase hydrogenation reaction. Due to the under88

equipped and proto-type nature of this project, a simple small glass cell was made in the
chemical glass shop to serve as a proof-of-concept apparatus.
4.2.1 Preparation of the Membrane Electrode Assembly (MEA)
The MEA is the core part of the fuel cell where the reaction takes place. Hence the
preparation procedures followed a reported method. 218
The MEA was prepared by the conventional paint and dry method. First a dry piece of
carbon paper (2 x 2 cm, Toray, TGP-H-060) was weighed and then painted with the catalyst
solution, a suspension of 1 gram of nickel or copper powder (mesh 325) per 25 ml of
isopropanol. The painted carbon paper was then dried with a heat gun and reweighed. The
process was repeated until 5 mg cm-2 catalyst was loaded onto the catalyst surface. The treated
carbon paper was stored on a weighing boat inside of a dry dessicator until use.
The treated carbon paper was placed in the center of a dry Nafion N117 (2.5 x 2.5 cm)
membrane, with the catalyst loaded side facing the membrane. Then, they were sandwiched
between two layers of 5 x 5 cm of clean fiber glass reinforced silicon (commonly used as gas
diffusion gasket in a hydrogen fuel cell setup) to provide support and prevent the Nafion
membrane from melting onto the heat plates. The four sides of the gasket were carefully sealed
with masking tape without distorting the membrane-carbon paper position. The resulting packet
was heat-pressed at 135 oC at 1746 pounds (based on the dimension of the gasket 25 cm2 = 3.88
inch2, approximately 450 psi) for 3 minutes on each side. After that, the packet was removed
from the heat-presser, cooled to room temperature, and the carbon-Nafion composite was taken
out from the gasket support. The composite served as the MEA for the reaction.

89

4.2.2 Construction of the 1st Generation SPEE – The Prototype

Conductive
Carbon Paint

Sampling
Port &
Anode slot

• Leakage Issue
• Poor conductivity (Carbon
Paint)

Platinum
Mesh

Solution
Reservoir
Clip
Point

Circuitry

The Membrane Electrode Assembly (MEA)
Nafion N117 membrane hot-pressed on a
carbon cloth/papers loaded with catalyst

Figure 4-3: Schematic drawings of the 1st SPEE cell made of two identical pieces as shown in
the figure on the right, these two halves are held together with a cell metal clip.
The prepared MEA was placed in the reaction cell shown in Figure 4-3. The carbon paper
served as the cathode and its compartment was filled with 8 mL of 50 mM acetophenone
dissolved in deionized water. In the anode compartment, a platinum wire was placed
approximately 2 cm away from the Nafion membrane in 8 ml of 0.5 M sulfuric acid. An
arbitrarily picked target current, 100 mA (actual current fluctuate at 98 ± 2 mA, and the practice
is implied for the rest of the project unless it specified otherwise), was applied to the system.
After 3 hours of reaction time, the cathode solution was extracted with diethyl ether which was
evaporated under a N2 flow to yield a concentrated organic residue. A small drop of the residue
was then dissolved in d-chloroform and analyzed by 1H-NMR spectroscopy. The product yield
was to be determined from the ratio of the aromatic hydrogens of acetophenone and the
hydrogenated product, 1-phenylethanol. (Comparison between the aliphatic methyl groups is also

90

a viable option) Unfortunately, no reduction was observed, partly owing to the fact that the setup
suffered from several flaws:
a)

Solution from both compartments leaked easily because of the heterogeneous nature of
the carbon paint. The dried paint created an uneven surface which prevented effective
sealing, causing leakage in the compartment.

b)

The carbon paint, which served as the current collector, lacked conductivity. While the
carbon paint was good enough for SEM analysis that ran at several tens thousands
volts, it was not conductive enough for the SPEE that ran at less than ten volts.

c)

The Nafion membrane swelled when it was hydrated. The swelling caused the “crispy”
carbon paper to detach.

d)

The connection with alligator clips was not well considered. The cell lacked a defined
location to host the alligator clip. When the alligator clip was clipped to a little
exposure of the wet carbon paper during the reaction, the sharp teeth bit off the paper
easily.
Summarizing the flaws, the design required several modifications. The carbon paper

needed to be replaced with a more durable material such as carbon cloth. The carbon paint
current collector had to be replaced with a material that was stronger, more conductive and
would not enable leakage. The cell would preferably hold more than 8 mL solution so that minor
leakage would not impact the experiment as much. With these ideas for improvement in mind, a
new design, shown in Figure 4-4, was created.

91

4.2.3 Construction of the 2nd Generation SPEE
Platinum
mesh anode
eee-

Cathode Compartment

Anode Compartment

e-

Graphite plate cylinder
• Improves conductivity
• Improves clipping property
• Reduced Leakage
• Observed traces of ECH of
acetophenone to 1-phenylethanol
MEA
(Cathode)

Figure 4-4: Schematic drawings of the 2nd SPEE cell which combines conventional dual
compartment electrochemical cells with SPEE concepts together.
In the 2nd SPEE cell, the MEA cathode was prepared with the same procedure described
above except with carbon cloth (Vendor: CST, Model #: CF11X). A platinum mesh was
employed as the anode in 30 mL of 0.5 M sulfuric acid. The cathode compartment contained 30
mL of 50 mM acetophenone.
Solution leakage improved slightly with the graphite plate. The plate withstood the
alligator clip bite and improved the conductivity between carbon cloth and the alligator clip.
Replacing carbon paper with carbon cloth improved the durability of the MEA; however, the
used MEA had to be handled with extreme care because the cloth was only loosely attached to
the membrane after electrolysis again due to the membrane swelling.
In general, minor leakage at the rate of one drop every 2 – 3 minutes was observed in this
setup. Even though the 2nd SPEE cell was far from ideal, traces of product, 1-phenylethanol were
92

observed from 1H-NMR spectroscopy with the MEA prepared with copper powder. The result
needed improvement, but supported the SPEE concept as a viable option for bio-oil upgrading.
Even though the overall procedure was tedious and the setup in general took more than 30
minutes (as opposed to 10 minutes on the 4th SPEE cell), the observation of acetophenone
reduction was encouraging.
4.2.4 Construction of the 3rd Generation SPEE
After a vigorous literature search on chemical fuel cells seeking ideas for improvement,
the term “solid polymer electrolyte” came up in one of the ECH reviews. With a deeper search
on the term, all the research mentioned in the introduction came to light. Combining the new
knowledge and the lessons learned from the 2nd SPEE designs, the 3rd SPEE setup was created.
Two pieces of glass 5 x 5 x 0.5 cm with tracks (about 1 mm in depth created via the sand
blasting technique by Scott Bancroft, an experienced chemical glassblower) was employed to
host the MEA. The MEA was prepared as described above with the carbon cloth. Multiple layers
of Parafilm were used as a gasket to provide a physical buffer between the MEA and the glass
cell to reduce solution leakage. The platinum wire was placed inside of the anode compartment
track which was connected to a reservoir (approximately 200 ml) of 0.5 M sulfuric acid solution.
10 mL of 50 mM acetophenone cathode solution was also circulated by peristaltic pump at 0.8
mL min-1. The overall setup was held tightly with a carpentry clamp, which was modified with 2
holes to accommodate the inlet and outlet of the cathode and anode solutions. The whole setup
was depicted in Figure 4-5.

93

.
Pt Wire

Anode Side

Cathode
Clip

MEA
Parafilm (Gasket)

The setup is hold together
with a Carpenter Clamps

Solution track

Cathode Side
Solution IN
Solution OUT

Thick Glass Slides with tracks

Figure 4-5: Schematic drawings of the 3rd SPEE cell – 2 pieces of glass plate that host the
MEA supported by a pair of Parafilm gasket and Carpenter clamp.
The leakage issue was significantly improved compared to the previous designs.
Although minor leakage occurred occasionally when the MEA and Parafilm misaligned; the
setup provided a solid foundation for the SPEE project.
In an identical trial, ECH of acetophenone was conducted to compare the efficiency of 3rd
cell with the 2nd cell; unfortunately there was almost no reduction of acetophenone. It was
possible that the carbon cloth (0.005 inch) was too thick to allow any hydrogen ions to reach the
cathode solution from the membrane’s surface. Another reason could be that the hydrogen gas
from water reduction generated at the cathode surface was trapped inside of the tracks, and
blocked access of the cathode solution to the MEA surface. Unfortunately, when the cell was
placed sideways to vent the hydrogen gas, solution from both compartments leaked. So, even

94

though the cell was very promising, the combination of carbon cloth MEA and SPEE had very
limited success in the ECH of organic substrate.
4.3

Preliminary Experiment II – Exploring the 4thSPEE

4.3.1 Construction of the 4th Generation SPEE – The Current Version
Further searching of the literature on SPEE studies turned up an educational fuel cell
company, Horizon Fuel Cell Technologies (HFCT). HFCT sold affordable grade fuel cells to
teaching institutions for educational purposes. With a curious mindset and a “can’t hurt” attitude,
a Solar Hydrogen Science Kit was ordered from the HFCT. The kit came with a toy-grade solar
panel, a toy-grade electrical motor, measuring cylinder for the solutions, and most important of
all, a palm size electrolyzer. The electrolyzer was made of two pieces of acrylic, held together by
screws at the four corners. Inside of the cell, two silicone gaskets were placed between the
carbon MEA and the acrylic cases to prevent leakage,
The electrolyzer housing part was ideal - another 30 of the same electrolyzers were
ordered soon after, and these cells have been used for this research project since then. The 4th cell
(HFCT cell) had many advantages over the first three cells. First of all, the cell rarely leaked if it
was set up correctly. Second, the gaskets were held in places with the pre-implanted tracks. The
screws and bolts on the corners held the cells strongly together. Most importantly, the setup
typically took less than 5 minutes to assemble as opposed to almost 30 minutes for the previous
designs. Also, the acrylic castings allow for easy modification if necessary.
From the literature mentioned in the introduction, the idea that Nafion membrane could
be modified with deposited metals and serve as an electrode could significantly improve the
overall efficiency of the SPEE. Metal deposition on the membrane surface could bypass the
carbon support detachment issue due to membrane swelling. Moreover, if the metal catalyst was

95

deposited on the Nafion surface, it would be as intimately placed relative to the membrane
surface as possible, thereby minimizing the distance for the proton ion to travel.
4.3.2 Direct Metal deposition on Nafion Membrane
There are two methods to deposit metal on a Nafion membrane surface: the internal (ionexchange) reduction method (IR), and the Takenaka-Torika (T-T) method. For the IR method, a
dry membrane is first sand blasted to increase its surface area before it is soaked in an aqueous
metal ion solution. This allows the hydrogen ions from the sulfonic sites on membrane be
exchanged with the metal ions in the solution. The exchanged metal ions are then reduced by
soaking the membrane in a chemical reduction solution containing the reductants sodium
borohydride (NaBH4) or hydrazine (N2H4). Metals are reduced and deposited on both sides of the
membrane but usually not in the core. Weak conductivity at several tens of mega ohm level can
be observed across a face of the membrane but not through the membrane.
The T-T method only deposits metal on one side of the membrane by taking advantage of
the concentration gradient and the cationic permeability nature of the Nafion membrane. For T-T
deposition, a hydrated membrane is first mounted in a conventional dual compartment
electrochemical cell. Then one compartment is filled with the metal ion solution, while the other
is filled with the chemical reduction solution. The positively charged metal ions migrate through
the membrane because of the concentration gradient, and get reduced upon arrival the other side.
One precaution is that if the reductant concentration is too high, the reactant may migrate
through the membrane and reduce the metal ions in the stock solution.
In Chou’s work, the platinum deposited (using sodium borohydride) with the IR method
was shown to be more stable than the T-T method, which the Chou paper referred to as the
diffusion method. Compared to T-T method, the IR method created a more tightly bound

96

platinum surface was shown under SEM imaging and a higher current density was observed
during C.V. analysis suggesting the surface was better at generating H2 gas. However, when nonprecious metals such as nickel, lead and copper were used in place of platinum, the MEA did not
last long due to the low acid tolerance of those metals. Although reduction of benzaldehyde to
benzyl alcohol was observed with the MEA prepared with the earth abundant metals, their
surface erosion and resulting low conductivity necessitated exceedingly high voltage, and the
experiment had to be terminated before the intended experiment time.216
4.3.3 MEA Preparation with Different Reductants
Silver was picked to be the choice of metal foundation for the MEA preparation instead
of platinum because it is cheaper and it has a lower reduction potential (a less negative Eo),
which means it can be reduced easily. Although Chou’s results showed silver has the lowest
catalytic activity in benzaldehyde hydrogenation compared to nickel, copper and lead, the effect
might be modified if other catalytic metal layers were deposited on top of the silver, as in to
Ogumi’s work.189 Before examining the effects of different catalytic metals, it was important to
prepare a well-defined silver foundation.
Silver could be reduced easily with a common reductant such as sodium borohydride or
hydrazine. Though sodium borohydride (LD50: 160 mg/kg) is less toxic than hydrazine (LD50: 59
– 60 mg/kg), it is slightly more expensive and requires a basic environment to maintain its
stability or it will be hydrolyzed by water to form H2 gas: NaBH4 + 4H2O  4H2 + NaB(OH)4.
In low concentrations, hydrazine can be a safe and environmentally friendly chemical because it
emits only N2 as a by-product after it reacts. Therefore, it was reasonable to investigate the
possibility of replacing sodium borohydride with hydrazine.

97

The results from Chou’s studies suggested that the IR method preparation would give a
better surface than the T-T method. Accordingly, MEA with silver surface modification was
prepared by the IR method using both hydrazine and sodium borohydride. The prepared
membranes were used in the acrylic cell from Horizon Fuel Cell Technologies to hydrogenate
furfural, a relatively abundant compound found in bio-oil. The amount of product formed was
compared to see which reductant gave a better surface for ECH of organic substrates.
4.3.4 Experimental Procedures
Preparation of the silver-based MEA began with a dry Nafion membrane, cut to the
dimensions of 2.3 x 2.3 cm, which would fit the inside of the reaction cell (2.7 x 2.7 cm) after it
was hydrated. The membrane was then boiled for 30 minutes in 3% hydrogen peroxide followed
by 30 minutes of 50 mM sulfuric acid to remove impurities. The cleaned Nafion membrane was
stored in 5 mM sulfuric acid until use.219
The cleaned membrane was soaked in 10 mL of 5 mM silver nitrate solution overnight in
a 20 ml sample vial for silver ion exchange. Glass beads were used to displace some volume if
the solution could not cover the whole membrane. After that, the membrane was taken out and
washed with deionized water for 1 minute before being placed in either 1% (1:99 v./v.)
hydrazine solution or 0.1 M sodium borohydride solution (prepared in 0.1 M sodium hydroxide)
and soaked for 4 hours. A picture on the setup is shown in Figure 4-12 later in the chapter.
The electrochemical experiment was conducted as follow: through the cathode
compartment that holds about 0.2 mL, a reservoir of 10 mL of 10 mM furfural solution was
circulated using peristaltic pump, and a stainless steel mesh 314 (mesh #200) was used as a
current collector. For the anode compartment, a stainless steel mesh (mesh #200) was used as the
anode and current collector and was placed in solution of 0.5 mM Co(NO 3)2 dissolved in 0.1 M

98

pH 7.0 phosphate buffer (the Nocera catalyst recipe). Reactions were run for 2 hours at 100 mA
at room temperature with 0.25 ml of sample taken for analysis every 30 minutes. Samples were
diluted with an equal amount of deionized water prior to HPLC-UV analysis (isocratic flow at
0.5 ml ml-1; 30% methanol in 0.1 M pH 7.0 water buffer; room temperature; column: Aminex
HPX-87H, UV: 214 nm). The retention times for furfural and furfuryl alcohol were 37.6 and
26.0 minutes respectively.
After the experiment, the cells’ compartments were flushed and stored in deionized
water, and used for the same reaction the next day. Three identical experiments were run before
the cells were disassembled.
4.3.5 Results and Discussion
3.5

NaBH4 1st
NaBH4 2nd
NaBH4 3rd

3.0
2.5

Concentration (mM)

Concentration (mM)

3.5

2.0
1.5
1.0
0.5
0.0

Hydrazine 1st
Hydrazine 2nd
Hydrazine 3rd

3.0
2.5
2.0
1.5
1.0
0.5
0.0

0

2000

4000 6000
Time (s)

8000 10000

0

2000

4000 6000
Time (s)

8000 10000

Figure 4-6: Production of furfuryl alcohol from furfural ECH in SPEE. (Left) MEA was
prepared with sodium borohydride (Right) MEA was prepared with hydrazine.
Comparison between both trials in Figure 4-6, the MEA prepared with the sodium
borohydride recipe was more effective in catalyzing the hydrogenation of furfural than the one
prepared with hydrazine. However, interestingly, repeated use of the hydrazine-treated
membrane improved the catalytic effect and the product formation rate became comparable with
the borohydride membrane. This suggested that some activation process might be taking place on
the surface of the silver during electrolysis on the hydrazine-reduced membrane. Control
99

experiments showed that hydrazine could not reduce furfural alone, and furfural could not be
reduced in the absence of electricity with either one of the silver deposited membranes. In other
words, the reduction could not have happened due to any traces of hydrazine or sodium hydride
that might have remained in the membrane.

Figure 4-7: Used Ag-MEA after 3 uses of electrolysis. (Left) a used Ag-MEA prepared with
sodium borohydride, and (right) a used Ag-MEA prepared with hydrazine. Conductivity
measure was taken using multimeter for both membranes:

Another interesting observation was that the borohydride membrane shown in Figure 4-7
showed significant signs of erosion after the 3 trials whereas the hydrazine membrane was still
looking smooth and shiny. The fact that the furfural reduction rate on the borohydride membrane
was constant while the silver catalyst eroded was intriguing. One would expect the reduction rate
to fall during the erosion if the silver was responsible for the reduction. This phenomenon will be
examined in more detail in the next section.
In the above comparison, the membrane prepared via hydrazine reduction appeared to be
more durable than the membrane prepared with sodium borohydride, and the reduction rate
increase with repeated use was interesting. Even though the differences and mechanism were not
fully understood, the hydrazine recipe was chosen over the borohydride method of MEA
preparation because of its durability and improvement through multiple uses. Moreover, the
resistance measured on the MEA prepared with hydrazine is lower than that of the MEA
100

prepared with sodium borohydride. ~3MΩ and ~10MΩ were measured for the hydrazine and
borohydride membrane, respectively. Although both showed resistance in the mega-ohm level,
this resistivity is still lower than for a cleaned membrane ~16MΩ. The decrease in the resistance
suggests silver has been deposited on the surface, but the level of deposition perhaps is not as
uniform as expected. To improve the conductivity across the membrane, an additional metal was
plated on top of the silver, improving reduce the cross section resistance.
4.3.6 Formation of Ag-Cu Bimetallic Layer Catalyst
In Ogumi’s study, the nitro group hydrogenation of 4-bromonitrobenzene was improved
by plating a layer of copper on the MEA deposited with platinum.189 Compared to platinum,
copper has a greater hydrogen overpotential, which makes it favor catalytic reduction of organic
substrates over hydrogen evolution.94 For this reason, a layer of copper was plated onto the AgMEA to see if the same effect could be achieved. The plating strategy followed Ogumi’s report:
50 mM of copper sulfate solution was circulated in the cathode compartment and a 100 mA
current was applied for the copper deposition. The Ag-MEA with copper deposition was denoted
as Cu-Ag-MEA

Figure 4-8: Visual comparison between (left) before, and (right) after copper plating onto the
Ag-MEA

101

0

Voltage (V)

Furfuryl Alcohol (mM)

Ag 1st
Ag 2nd
Cu-Ag 1st
Cu-Ag 2nd

3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0

5000
Time (s)

10000

Ag 1st
Ag 2nd
Cu-Ag 1st
Cu-Ag 2nd

4.2
4.1
4.0
3.9
3.8
3.7
3.6
0

5000
Time (s)

10000

Figure 4-9: (Right) ECH of 10 ml of 10 mM furfural. Condition: Ag is deposited on Nafion
using 1% Hydrazine solution. Reactions were run at 98 ± 2 mA at 1 atm at room temperature
(22 ± 1 ⁰C) for 6 hours. Note: Cu-Ag is prepared by plating copper onto the Ag-MEA
electrochemically using 50 mM CuSO4 at 98 ± 2 mA for 5 minutes. (Left) the voltage log for
the reactions displayed on the right
From Figure 4-9 right, the copper plating did not have significant impact on product
formation as the data points for Ag and Cu-Ag almost overlapped which suggested the copper
plating had no significant effect on the reduction. With the copper plating (Cu-Ag 1st), the
voltage needed for 100 mA reduced noticeable, but rises again after several hours. This rise in
voltage suggests the copper may be eroded during the reaction, and interestingly, the change in
copper layer does not have any effect on the reduction of furfural. Perhaps the reduction was
taking place elsewhere e.g. on the current collector which would also explain why the product
formation improved after the 1st use on Figure 4-6. The Ag-MEA prepared with the borohydride
recipe appears to be a loose powdery surface, similar to Chou’s observation for the platinum, and
hence did not block the proton transport to the surface. On the other hand, the hydrazine recipe
formed a smooth silver surface which blocked the proton transport. The acidic site might have
been liberated due to minor erosion after the first use, which improves the proton transport and
thereby increases the reduction rate. In principle the voltage log could give hints as to whether
silver was blocking proton transport; blockage would be expected to result in higher voltage

102

during constant current electrolysis. However, the reaction area is less than 5 cm2. So, with data
measured to only 2 significant figures, the change may be too small to be observable. To verify
the proposition, a set of control experiments were conducted to confirm where exactly did the
reaction took place – the current collector or the membrane surface.
The experimental procedures in the following set of experiments evolved with minor
modifications: the trials were ran for 6 hours instead of 2 to observe the reaction longer, and
acetol (IUPAC: 1-hydroxypropan-2-one) was used as the model compound instead of furfural.
This is because the furfural stock solution began to form dark precipitate and turn brown after 2
weeks of storage. The change in the furfural stock appeared to be resulting from furfural
condensation in the aqueous environment.220 Compared to furfural, aqueous acetol was more
stable. It showed no signs of polymerization or oxidation after weeks of storage. HPLC analysis
of acetol and propylene glycol (hydrogenated acetol) took less time and effort compared to
furfural. Hence, acetol was used as a model compound for further studies exploring the project.
To verify the importance of the current collector, current collectors with the center cut
out were utilized to look for changes in the reduction rate. The area of the current collectors was
reduced approximately 54.6% (1.7 x 1.7 cm was cut out from 2.3 x 2.3 cm).

103

CC (No Ag Foundation)
CC (with Ag Foundation)
Cut CC (No Ag Foundation)
Cut CC (with Ag Foundation)

Propylene Glycol (M)

0.12
0.10
0.08
0.06
0.04
0.02
0.00
0

5000

10000
Time (s)

15000

20000

Figure 4-10: Reduction of acetol in examining the necessity of the silver foundation and the
importance of the current collector.
Figure 4-10 shows that the cut current collectors have a lower reduction rate of acetol
both with and without silver foundation. This is a clear indication that there is a relationship
between the ECH rate and the current collector, which in turns suggests that reaction must have
been happening on the current collectors rather than on the silver surface. It is interesting that
these findings are somewhat in contrast to the conventional prospective, in which hydrogenation
takes place on the platinum catalyst surface deposited on the membrane. It is, however, it is
important to note that here, silver was used to replace platinum. Perhaps, compared to platinum,
silver was less catalytically active relative to the stainless steel current collector, and hence the
reduction reactions took place on the stainless steel surface instead.
The important lesson learned from the above preliminary data was that ECH can be run in the
absence of a supporting electrolyte as long as the electrodes are in close contact with the
membrane surface. Also, the catalyst metal does not have to be deposited on the membrane
surface. In fact, metal deposition on the membrane may block proton transport. Under constant
current electrolysis, the influx of proton from the anode compartment might have dissolved the
silver sheet formed on the cathode surface and becomes H2. Protons transport improved as the
104

silver surface erodes; increase in proton transport led to improvement of organic reduction. This
proposition is in consistence with the results displayed in Figure 4-6, where the metal modified
MEA improves after the first use.
4.4

New Perspective on the SPEE and C.E. % Improvement Effort
Seeing that the silver foundation on the membrane was unnecessary, the SPEE setup was

simplified to two meshes closely pressed up to the membrane surface, which resembled the setup
of a micro-gap electrolyzer207. After this discovery, the project focused on how the cathode mesh
and electrolyte environment affected the reduction rate. Current efficiency was used as a measure
to evaluate the effectiveness of the system in catalyzing hydrogenation.
The first objective was to select a metal mesh that had a good catalytic effect. A literature
survey found copper and nickel were the two most common low cost catalysts for
electrocatalytic hydrogenation.98, 119, 120, 221 Unfortunately, copper and stainless steel are the only
common metals available in mesh form from commercial vendors. Although nickel mesh could
be made upon request, the cost was unexpectedly high and hence was not purchased.
4.4.1 Experimental Procedures
Copper meshes (mesh #200) and stainless steel 314 (mesh #200) were cut to in the
dimensions of 2.3 x 2.3 cm and were stored in hexane until use. The cathode solution was 10 mL
of 5% acetol with 5% acetic acid in water and the anode solution was a reservoir of 200 mL of
0.1 M pH 7.0 phosphate buffer solution. Reaction was run at room temperature.

105

Copper Mesh to
Cathode Port
Acrylic Cell
Nafion
N 117
Anode
Mesh

Copper wire extension
to Anode mesh

Catholyte
Inlet

Catholyte
Exit

Silicon
Gasket

Copper
Mesh

Figure 4-11: Solid Polymer Electrolyte Electrolyzer (SPEE) parts, and the setup

Peristaltic
Pump

Catholyte
Reservoir

Anolyte
Reservoir

The SPEE

Figure 4-12: (Left) typical experimental setup of a SPEE. (Right) close-up to a SPEE with
alligator clips and silicon tubing connected
The materials shown in Figure 4-11 (left) were prepared and assembled in order starting
from the anode compartment (marked with a positive sign)  silicon gasket  anode mesh 
silicon gasket  Nafion membrane  cathode mesh  silicon gaskets  cathode compartment
(market with a negative sign).
Cathode solution was circulated by a peristaltic pump at 0.8 ml min-1. 0.25 mL of sample
was taken hourly from the reservoir, and was diluted with equal amounts of deionized water
before HPLC-RI analysis (Isocratic flow at 0.5 mL ml-1; 5 mM of sulfuric acid; Room
temperature; Column: Aminex HPX-87H, Waters 410 RI detector setting: internal temperature
35 oC, Sensitivity: 128)

106

Table 4-1: 10 ml of 5% acetol (924.5 mM) dissolved in deionized water undergo ECH to
propylene glycol after 6 hours at a constant current 98 ± 2 mA, at room temperature and
ambient pressure. All current collectors are mesh 200 in size, and had been cleaned with
hexane, acetone and water prior to use. Stainless steel 304 (Mesh 200) was used in a 0.1M pH
7.0 phosphate buffer as an anode for all trials. Material balance was calculated by summing
acetol and propylene glycol alone. *Initial acetol concentration for the “copper with 5% acetic
acid” trial was 716 mM
Propylene
Acetol
Current Material
glycol
remaining
efficienc balance End pH
formed
(mM)
y (%)
(%)
(mM)
Stainless Steel
876
0
0
94.8
6.34
(Control with no epassed)
274
117
10.4
42.3
11.23
Stainless Steel
Nickel plated on
12
319
28.4
35.8
12.28
Stainless Steel
114
398
32.9
55.4
10.88
Copper
Copper with 5% Acetic
335
152
13.8
68.0
5.23
Acid*

The catalytic effect of three different metals: copper, nickel and stainless steel (SS) were
compared in Table 4-1. In the absence of acetic acid, the reaction rate decreased in the following
order Cu > Ni >> SS. Stainless Steel showed relatively low catalytic activity towards acetol
hydrogenation, but the situation was changed when it was plated with nickel in a Watts plating
bath 1 .155 This nickel plating confirmed the current collector mesh was the reaction surface.
Copper had the highest Columbic efficiency of the three metals, and it also had the greatest
material balance relative to the other two electrolysis trials. However, it must be noted that the
mass balance of all the trials was surprisingly low 35.8 – 55.4%. Based on the control
experiment using stainless steel, approximately 5% of the material loss was due to membrane
crossover. The remaining losses were likely to be associated with the high pH of the cathode
1

-1

-1

Nickel plating was done in a Watts bath at 100 mA for 5 minutes (H 3BO3 30g L , NiCl2.6H2O, 45 g L ,
-1
NiSO4.6H2O 300g L )

107

solution after 6 hours of electrolysis. The cathode solution changed from colorless to yellow
which suggested acetol might have gone through aldol condensation in the increasingly basic
conditions.
To verify whether acetol condensed in the basic environment, 0.1 mL of 1 M NaOH was
added to the 10 ml 5% acetol solution and was left overnight at room temperature without
electrolysis. The mixture turned a mild yellow color that resembled the color of the cathode
solution after ECH. The sample was then analyzed by liquid phase MS (Note: no
chromatography involved). Higher molecular weight compounds (presumably oligomers) were
detected from the control experiment and confirmed that acetol did condense in the basic
environment. The higher final pH from the nickel plated mesh trial suggested that nickel favored
water reduction (Scheme 4-1, equation d) forming hydroxyl anion. Paired with the potassium ion
diffusing from the anode chamber through the membrane into the cathodic chamber, it is clear
that the pH increased as the reaction proceeded.
Although the condensation can lead to the formation of molecules with a greater carbon
number, feature preferred for fuel purposes (typical carbon number of gasoline ranged from C 5 –
C8, whereas the majority of bio-oil molecules typically range from C2-C6), uncontrollable
condensation is undesirable, especially in real bio-oil where hundreds of organics are present
leading to formation of unusable tar. Knowing there was already insufficient carbon to cover the
US transportation sector usage from biomass, it is important to maximize the utilization and
minimize losses of carbon during the upgrading process.

108

a) 2H+ + 2e- ---> H2
b) Acetol + 2e- + 2H+ ---> Propylene Glycol
c) Acetol + 2H2O + 2e- ---> Propylene Glycol + 2OHd) H2O + 2e- ---> 2OH- + H2
Scheme 4-1: Possible cathodic reactions: a) Hydrogen evolution b) acetol undergoes ECH to
propylene glycol using electrons and protons c) acetol undergoes ECH to propylene glycol
using water and electron as a reductant. d) cathodic water reduction forming hydroxide anion.

Based on the molecular weight information gathered from direct MS analysis of the
solution, the processes shown in
Figure 4-13 were proposed to explain the material losses during the ECH experiments.
Understanding how condensations could lead to material loss could be helpful in deriving an
upgrading strategy that maximizes carbon retention. Compound structures were proposed based
on detected MS parent peaks with knowledge of aldol condensation, and estimated quantity was
given based on their relative height.

Figure 4-13: Proposed reaction pathways of acetol. Candidates are proposed based on the
observed molecular weight from aqueous phase MS analysis. Percentages of the compounds
are given based on the relative intensity of the peak.

109

Electrolysis using an ionic salt buffer as the anolyte can increase pH which can lead to
material loss, but on the other hand, such phenomena may be an option in neutralizing the biooil’s acidity. It must be kept in mind, however, that if potassium ion (from anode buffer) is the
cation flowing from anode to cathode compartment, it may defeat the purpose of using the SPEE,
which was proposed as a strategy to upgrade bio-oil in the absence of a supporting electrolyte.
However, if the phosphate buffer is replaced with any protic solution such as sulfuric acid, the
stainless steel will corrode quickly and terminate the reaction – only platinum can withstand the
acidic oxidative environment in electrolysis. There was no obvious solution to the problem,
either the project would proceed with an expensive platinum mesh anode to carry on with the
SPEE principle, or continue to explore the possibility of using the stainless steel and see if it is
possible to improve its lifetime under the anodic oxidative environment. Since platinum mesh
was a costly investment for this relatively new project, investigation of stainless steel was
preferred until the project matured.
4.4.2 ECH of acetol in the presence of acetic acid
Seeing that aldol condensations led to material loss, 5% (v. /v.) of acetic acid was added
to the acetol cathodic solution. Addition of 5% acetic acid served two purposes a) to mimic the
acidic nature of bio-oil, and b) to verify whether the acidic cenvironment could improve mass
balance. A control experiment analyzed by HPLC-RI showed acetol did not undergo
polymerization in the presence of 5% acetic acid over 15 hours at room temperature, and the
acetol stock solution showed no signs of changes after months of storage.
When 5% (v./v.) acetic acid was added, the C.E. for propylene glycol formation dropped
by almost 20% (Table 4-1, without acetic acid 32.9 % vs. with acetic acid 13.8 %). This is
expected as greater H+ concentration favors H2 formation rather than acetol hydrogenation.

110

Formation of H2 occurred at the expense of H+ from acetic acid, so the average pH after 6 hours
of electrolysis increased from 2.89 to 5.23. On the other hand, the mass balance also improved
12.6 % (without acetic acid 55.4 % vs. with acetic acid 68 %), which is significant but still
incomplete. HPLC-RI analysis showed no side products or polymerized products formed in the

Current Efficiency (%)

reaction. Therefore, the 30% loss of material may be due to membrane crossover.
50
45
40
35
30
25
20
15
10
5
0

1st

0

5000

10000

15000

2nd

20000

3rd

4th

25000

Time (s)

Figure 4-14: ECH of 10 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added. Condition:
copper mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 ⁰C) for 6 hours. Copper
mesh is being reused multiple times and compared with the Monel 400 mesh 200 Note:
reaction is evaluated using the cumulative current efficiency, each point accounts for the
product accumulated previously. For equation, see Scheme 2-2

Interestingly, when the copper mesh was reused again for the same reaction, the
Columbic efficiency trend changed significantly. The Columbic efficiency of new copper mesh
started low at approximately 3.7% and rose to 13.8% whereas the used copper mesh started at ca.
30% then dropped slowly during the course of reaction. The trend suggested that perhaps the
copper was being activated during its 1st use. Upon disassembly of the reaction cell, a black layer
was found on the surface of the copper. Examination under SEM with EDX analysis revealed the
black layer resulted from morphological changes on the cathode surface. As shown in Figure
4-15, nickel and copper were detected (EDX analysis: 55% Ni and 30% Cu) with minor carbon
111

and oxygen impurities. The nickel likely originated from the stainless steel anode mesh that was
slowly undergoing anodic oxidation; this notion was confirmed by the SEM images of the used
stainless steel mesh in Figure 4-15 (bottom left). Knowing stainless steel contains 8 – 10%
nickel, and seeing the SEM observation of the surface erosion of the stainless steel, it is
reasonable to conclude that the nickel deposition originated from the stainless steel anode.

Figure 4-15: SEM examination on copper mesh before (Top, Left) and after (Top, Right) 6
hours of electrolysis. New Stainless Steel (Bottom, Left) and used stainless steel (Bottom,
right). Samples were washed with water, acetone and chloroform to remove organic residues.
Then they were dried in vacuum desiccator for 2 hours before SEM examination at 12kV and
10 mm working distance under ultrahigh vacuum.
The positively charged nickel ions (Ni2+) presumably travelled through the membrane
and re-deposited on the copper mesh. This newly formed nickel-copper surface may be
catalytically active towards organic hydrogenation. Moreover, it appears rough as seen in Figure
112

4-15 (top right) which would have increased the surface area thereby improving the ECH rate.
Interestingly, even though the stainless steel anode mesh was undergoing oxidation, it lasted for
30 hours, 5 trials of 6 hours, of electrolysis without obvious macroscopic signs of degradation.
4.4.3 Effect of Anolyte on Current Efficiency Improvement
Given the improvement in Columbic efficiency triggered by the nickel re-deposition, it
was proposed that a nickel-copper alloy might possess a special advantage towards catalytic
hydrogenation. Hence, Monel 400 2 , a commercially available nickel-copper alloy mesh, was

Current Efficiency (%)

purchased to test out the effect of acetol hydrogenation.
SS in Phosphate Buffer
SS in 0.5 H2SO4
Pt in 0.5 H2SO4
Pt in H2O

50
45
40
35
30
25
20
15
10
5
0
0

5000

10000
15000
Time (s)

20000

25000

Figure 4-16: ECH of 10 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added. Condition:
Monel 400 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 ⁰C) for 6 hours. The anode
compartment conditions are listed in the legend
When the copper mesh cathode was replaced with the Monel 400 mesh, the efficiency
reached the well over 30% (Figure 4-16 - [SS in phosphate buffer]), which seemed like that
nickel-copper alloy was responsible for the catalytic enhancement. However, when the Monel
400 mesh was used in conjugation with platinum mesh, which gave no nickel ions, the efficiency
2

Monel 400 is an alloy consisting of 63% Ni min., 28 – 34 % Cu, 2.5% Fe, and 2% Mn. Note: the
400 associated with Monel describes its composition, and is not a mesh number. The mesh number for
the Monel 400 was 200, same as other meshes used.

113

decreased dramatically to less than 5% (Figure 4-16 – [Pt in H2O]). Therefore, the re-deposition
of the nickel ions, which were leached from the stainless steel anode, must have been a
contributing factor towards the catalytic enhancement. To verify the importance of nickel, in a
separate trial of 10 mM of nickel chloride was added to the catholyte with platinum used as the
anode. As expected, the initial efficiency improved (Figure 4-16, [Pt in H2O with 10 mM NiCl2],
10.6% vs. [Pt in H2O] 3.1%). The fall of the efficiency is likely owing to the small quantity of
nickel ions that were then plated out in the 10 ml electrolyte. In 10 ml of 10 mM of nickel ion
solution, there is only 0.1 mmol of nickel ions, which would require 19.3 coulombs to plate out.
Given a 6 hours electrolysis would supply roughly of 2000 coulombs, and assuming of 30% went
into acetol hydrogenation, there is still 70% of the 2000 coulombs (2000 x 70% = 1400 C) going
towards H2 formation and nickel plating. Thus it is reasonable to understand the 10 ml of 10 mM
nickel solution does not have enough nickel ions for test its effect. When the solution volume
increased to 100 mL, there was 10 times more nickel ions in the solution to improve the supply
of nickel ions. In a sufficient nickel supply environment, the reaction gave a Columbic efficiency
trend became more steady as shown in Figure 4-17.
The effect of anolyte pH on Columbic efficiency was also examined. When the stainless
steel anode was used with 0.5 M H2SO4, the Columbic efficiency dropped significantly
compared to the trials with the potassium phosphate buffer. Apparently, when there was a high
influx of protons from the anode compartment, the cathode surface concentration of protons
increases, favoring H2 formation which in turn decreased the Columbic efficiency. In other
words, a non proton flux would improve the Columbic efficiency by lowering the proton
concentration on the cathode surface. This model was supported by the high Columbic efficiency
boost using potassium phosphate buffer anolyte. When the stainless steel mesh anode was used

114

in phosphate buffer solution, the initial Columbic efficiency was 9.5% higher than using it in 0.5
M sulfuric acid. The differences increased to 17.6% after 6 hours of electrolysis shown in Figure
4-16. In the beginning, the influx of potassium ions decreased the proton concentration on the
cathode surface, and reduced the rate of H2 formation. The higher efficiency after 6 hours of
electrolysis may result from the potassium ion influx replacing H+ ions and thus increasing the
pH of the cathode solution; higher pH disfavors H2 formation, and forces reaction to favor
organic ECH under the constant current electrolysis setting.
4.4.4 Influence of Metal Ion Re-deposition
Aside from nickel, the Columbic efficiency improvement due to other metals that could
have been leached from the stainless steel mesh was also examined. Stainless steel (314) mesh is
made of 0.03% sulfur, 0.05% phosphorus, 0.08% carbon, 0.10% of nitrogen, 0.75% silicon, 2.0%
Manganese, 8 – 10% nickel, 18 – 20% chromium, and (remaining) 67.3 – 71.8% of iron. Of the
listed composition, only nickel, chromium and iron could be re-deposited on the cathode.
Therefore, chromium (II) acetate222 and chromium (III) acetate (from air oxidation of chromium
(II) acetate) and iron (III) chloride and several different concentrations of nickel (II) chloride
were tested and compared for their improvement on Columbic efficiency. The chloride salts were
preferred over nickel acetate because of their high solubility in the aqueous electrolyte. The
presence of 5% acetic acid made the choice of counter ion, chloride or acetate, unimportant.

115

100 mM Nickel Chloride

20.0

20 mM Nickel Chloride

18.0

10 mM Nickel Chloride

Current Efficiency (%)

16.0
14.0

10 mM Iron Chloride

12.0

10 mM Chromium (III) Chloride

10.0

No Metal (Control)

8.0
6.0
4.0
2.0
0.0
0

2

4

6
Time (hours)

8

10

12

Figure 4-17: ECH of 100 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added.
Condition: Monel 400 mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 ⁰C) with a
platinum mesh anode in deionized water for 6 – 12 hours. Metal salts were added at the
beginning of the experiment.
Only the nickel salts had a positive impact on the Columbic efficiency. Of the three
concentrations tested 10, 20, and 100 mM, the 20 mM of nickel chloride trial reached its highest
Columbic efficiency to 18%, and declined slowly. The 10 and 100 mM trial maintained a
reasonable flat line after reaching its plateau. The increases in efficiency between 10 to 20 mM
of nickel chloride is reasonable, double the concentration led to approximately 2.5 times greater
efficiency. However, when the nickel ion concentration got too high in the 100 mM trial, the
efficiency was lower than that of 20 mM, this probably resulted from competition between nickel
ions and the organic substrate for reduction.
Compared to the nickel (II) ions, chromium (III) and iron (III) ions decreased the
Columbic efficiency. This can be rationalized through the concept of standard reduction potential
of these metals. Chromium (III) (+0.74v vs. SHE.) and iron (III) (+0.44v vs. SHE) have greater
positive potential compared to nickel (+0.25v vs. SHE) which means they are more likely to stay

116

in an oxidized form under mildly reducing environment. Hence, EDX surface analysis showed
only nickel was deposited on the copper mesh, but not the other two metals. The lack of plating
of iron and chromium supports the nickel deposition was indeed the key factor towards catalytic
improvement on acetol hydrogenation. Similar observation was made by Navarro is his
conventional two undivided cell, in which a sacrificial nickel enhanced the efficiency of carbonyl
groups hydrogenation. The constant nickel deposition was proposed to have a positive effect on
organic substrate hydrogenation. 98, 120
Knowing that potassium ions influx and nickel plating were responsible for the Columbic
efficiency enhancement, the two factors were tested together to see if they could be used to
improve the Columbic efficiency further.

35

Current Efficiency (%)

30
10 mM Nickel Chloride (Phosphate
Buffer anolyte)

25

10 mM Nickel Chloride (Deionized
Water anolyte)

20
15
10
5
0
0

2

4

6

8

10

Time (Hour)

Figure 4-18: ECH of 100 ml of 5% acetol with 5% (v. /v.) glacial acetic acid added.
Condition: Monel 400 mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 ⁰C) with
a platinum mesh anode in different anolytes for 12 hours.

117

12

100
90

10

70

[Nickel] mM

8

60
6

50
40

4

30

[Potassiumm] mM

80
Nickel (Phosphate Buffer anolyte)
Nickel (Deionized Water anolyte)
Potassium (Phosphate Buffer)

20

2

10
0

0
0

2

4

6

8

10

Time (Hour)

Figure 4-19: ICP-OES analysis of nickel and potassium in the cathode solution for trials
described in Figure 4-18. Samples were diluted 125 times with 2% nitric prior analysis (0.2
ml of sample in 25 ml of 2% nitric acid).

From Figure 4-18, the use of phosphate buffer anolyte had a significantly higher
Columbic efficiency compared to deionized water anolyte – a boost of almost 25%. With
phosphate buffer, the efficiency started at above 30% and declined slowly, whereas the
Columbic efficiency for the deionized water trial slowly rose. The difference in these trends
could be explained by combining the information from Figure 4-18 and Figure 4-19. Compared
to the water anolyte, the nickel ion concentration dropped quicker in the presence of phosphate
buffer anolyte. This suggests that a high influx of protons, generated via water splitting, from the
anolyte was discouraging the nickel deposition and thus decreasing the Columbic efficiency.
The fact that the Columbic efficiency did not correlate with the steady rise of potassium
ion concentration, shown in Figure 4-19, suggests the potassium ion has no direct effect in aiding
the reduction. Thus, the rise in conductivity, due to the potassium influx, had no effect on the
reduction. The constant influx of potassium might have created a steady state concentration of
118

potassium ions near the cathode mesh surface which reduced the local proton concentration, and
in turns favored nickel deposition. In other words, the potassium influx made the cathode surface
less acidic and so did not favor H2 formation as much. In short summary, the combination of
potassium influx and nickel ion deposition are the two major factors in boosting the Columbic
efficiency.
4.4.5 Mesh Size Effect on Current Density

0.47
0.94
2.31
2.19

178
43
418
214

154
197
162
244

Material
Balance (%)

1.7
2.57
3.69
3.63

Current
Efficiency (%)

3.59
2.72
1.60
1.66

Acetol
Remaining
(mM)
Propylene
glycol Formed
(mM)

Copper / Nafion
Exposure ratio

67.9
51.47
30.25
31.36

Copper
Exposure (cm2)

0.011
0.012
0.0045
0.0022

Nafion
Exposure (mesh
open area) (cm2)

Wire Thickness
(inch)

16
22
100
200

% Open Area

Copper Mesh
Size

Table 4-2: ECH of 10 ml of 5% acetol with 5% (v/v) glacial acetic acid added. Condition:
copper mesh with various mesh size at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 ⁰C)
for 6 hours. Note: mesh size 2.3 x 2.3 cm = 5.29 cm2

13.2
17.3
14.5
21.9

47.1
34.0
72.2
69.1

The effect of mesh size was also examined. Based on Table 4-2 there was no simple
correlation between mesh number and C.E. found. However, the mesh size had some impact on
the material balance – the lower the mesh number, the greater the open area (the Nafion
exposure), which increases the area for organic diffusion through the Nafion membrane. Mesh
16 and 22 only have 34% and 47% material balance respectively while mesh 100 and 200 have
72.2% and 69.1% respectively. Note: the copper exposure of mesh 100 and 200 are same, 0.27
cm2, the wire from mesh 200 was half as thick as that of mesh 100. The result suggests lower
mesh number (with greater opening) leads to lower material balance because of membrane
crossing of organic substrates.

119

4.4.6 The Effect of Current Density on Current Efficiency

Acetol remained
(mM)

P.G. formed
(mM)

C.E. (%)

Material Balance
(%)

29280
14400
7200

Average applied
potential (V)

9.21
18.87
36.86

Charge passed (C)

Current Density
(mA cm-2)

48.7
99.8
195.0

Time (s)

Average
Applied Current
(mA)

Table 4-3: ECH of 10 ml of 5% acetol with 5% (v/v) glacial acetic acid added. Condition:
Monel 400 mesh 200 at different current density (mA cm-2) 1 atm at room temperature (22 ±
1 ⁰C). Note: charge passed was calculated by the summation of multiplying current level
(recorded hourly) with the time interval between sampling time. Note: current density was
calculated based on the cut dimension of 2.3 x 2.3 cm = 5.29 cm2

1422.9
1438.7
1404.0

2.67
2.72
3.85

429
214
420

49
244
121

6.7
21.9
16.7

67.9
69.1
76.9

Two other current settings were attempted to explore the effect of current density using
the Monel 400 mesh 200 and the results are summarized in Table 4-3. 200 mA has a lower
Columbic efficiency, might be due to the higher current density was favoring H2 formation over
the ECH of organics. The average applied voltage for 200 mA was 3.85 (average of initial and
final voltage); 41.5 % higher than the average applied voltage of the 100 mA trial. The increase
in voltage is required to drive the higher current which means in turn, that more reactions must
take place on the electrode surface and hence a higher voltage is required to generate a thicker
double layer. The material balance was also higher for the 200 mA trial. This is most likely due
to a higher concentration of H2 gas formed on the surface, and the bubbles increase the shielding
between the cathodic solution and the membrane. In short, the 200 mA trial gave an 8%
improvement on the material balance compared to the 100 mA; but required a 41.5% greater
voltage. Hence, 98 ± 2 mA (targeted at 100 mA) was a better current setting for the reaction.

120

When 50 mA was applied to the reaction, the applied potential got slightly lowered but
the C.E. % decreased dramatically down to 6.7%. From the standpoint of the current, a lower
applied current would lead to less adsorbed hydrogen on the surface and the adsorbed organics
would be less likely to encounter an adsorbed hydrogen atom to achieve hydrogenation. From
the perspective of voltage, a lower current led to lower applied voltage, which means a decrease
in cathodic voltage. When the cathodic voltage gets too low (or too high), the electron energy
level does not match with the LUMO of the substrate pi orbital: this causes an inefficient transfer
of the electron to the π* orbital of the organics substrate.
B

Organic is
closer to an
adsorbed H

Surface is
saturated with H

H H H H H H H H H H H H

π*

H X = Y H H

H

H

Electrode Surface

eπ*

Potential (V)

ePotential (V)

Electrode - Cathode

H

Electrode Surface

Electrode Surface

Potential (V)

Organic is far
from adsorbed
H

C

H H H H

H H H H X = Y

Electrode - Cathode

X = Y

Electrode - Cathode

A

e-

π*

Figure 4-20: graphical interpretation of A) at low current, hydrogen adsorption is low,
resulting in low hydrogenation efficiency. when the applied potential is too low (or too high),
energy of the electron and the LUMO is mismatched, causing an inefficient transfer and lower
the C.E. B) at moderate level of current, surface is filled with enough adsorbed hydrogen and
can fit in organics to achieve hydrogenation. C) at high current, surface is saturated with
adsorbed hydrogen and disfavors organic adsorption. H2 formation becomes dominated result.
energy of the transfer electron matches the LUMO of the pi bond to give an efficient
electron(s) transfer

121

4.4.7 Isopropyl Alcohol Study

Current Efficiency (%)

80
70

Water with 5% Acetic Acid
Isopropyl Alcohol
Isopropyl Alcohol with 5% Acetic Acid
Water

60
50
40
30
20
10
0

Time (s)

Figure 4-21: ECH of 10 ml of 5% acetol in water or isopropanol, with and without acetic
acid. Condition: Monel 400 mesh 200 at 98 ± 2 mA at 1 atm at room temperature (22 ± 1 ⁰C)
for 6 hours.

Despite the fact that the cell could operate in the absence of supporting electrolyte, the
SPEE required water to achieve ECH (equation c in Scheme 4-1). When water was replaced with
isopropanol in the cathode compartment, the C.E. dropped to less than 10 % and the C.E. was the
same with and without addition 5% glacial acetic acid. In the absence of water, the nickel ion
leached from stainless steel could not be plated on the surface of the cathode mesh, and the
potassium ion was not soluble in isopropanol. The combination of these two factors resulted in
near-zero Columbic efficiency. Interestingly, gas evolution was still observed from the cathode
mesh for both isopropanol trials, suggesting H+ might have diffused through the Nafion
membrane, and been immediately reduced to H2 on the mesh. At the same time, it might be
possible that IPA (pKa ~15, which is roughly same as water) assisted the H+ transportation from
the Nafion surface for hydrogen evolution. The important lesson from this part of the study is
that an aqueous environment is critical to the hydrogenation efficiency, because it is central to
122

the metal ions transportation. The lack of metal ion replenishment on the cathode surface, as
discussed in the previous section, will disfavor organic substrate hydrogenation, and H2
evolution the dominant reaction.
4.5

Summary
After almost 3 years of work, the mechanism of the SPEE system is finally understood. The

ongoing deposition of nickel and the mildly basic environments in the catholyte are the two most
important factors in boosting the Columbic efficiency. Admittedly, the use of the stainless steel
mesh anode started out with a different purpose; it was supposed to support the cobalt phosphate
water oxidation catalyst as an ongoing water oxidizing anode similar to the guaiacol reduction.
Unfortunately, the optimal current density for the cobalt phosphate catalyst was 1.17 mA cm-2
while the SPEE operates at almost 20 mA cm-2, a difference of almost 20 times. In a long run, an
non-oxidizable anode of a high surface materials for the water oxidation catalyst is desirable.
After all, stainless steel would be a costly replacement on a large scale.
In summary, the anode surface is first oxidized and releases nickel ions. The nickel ions,
potassium ions, and hydrogen ions travel through the Nafion membrane to reach the cathode
surface. The influx of potassium ions created a less acidic catholyte solution which favors the
deposition of nickel. Freshly deposited nickel in turn favors organic hydrogenation and improves
the Columbic efficiency. Although the SPEE functions in the absence of aqueous electrolyte, it
offers a low Columbic efficiency because organic solvents disfavor ion transport. The overall
mechanistic scheme is depicted in Scheme 4-2.

123

e- +

e- e-

e0.1 M pH 7.0
Potassium Phosphate Buffer

ee-

e-

H+
H+
H+

K+
H+

H2O
Ni2+ + 2e-

H+
H+
Ni2+

K+

H+

H+

Ni2+ + 2eH2
H+ + 2e-

H+

RHRH’ + 2OH-

H+
H+

H+

R=R’ + 2H2O + 2e-

K+

K+

Stainless
Steel
314Mesh

H+ 2+
Ni

K+

H+

[Ni]

[Ni]

K+

Ni2+

O2 + 4H+ + 4e-

Deionized Water or
inert organic solvent

Monel 400
Mesh

Nafion 117
Membrane

Scheme 4-2: Revised mechanism for the SPEE cell that accounts for all observed reactions.
Reaction mechanism is proposed based on the conditions described in the diagram.
The current density also has impact on the Columbic efficiency; an optimal current
density for acetol is approximately 20 mA cm-2.
Although cyclic voltammetry (CV) could be used to determine the reduction potential on
high H2 overpotential metals, mercury for instance, copper and nickel cannot benefit from CV
analysis because the H2 evolution peak current will over shadow the organic reduction peak.
And, unfortunately, the reduction potential values of organic substrates are different between
metals, and so there is no direct way to probe the optimal reduction voltage for copper or nickel
but through experimentation.
From an energy standpoint, the SPEE is a better system than the conventional
electrolyzer because it has a much smaller internal resistance. The classical dual compartment
124

cell requires a minimal 5.5 volts to run at 50 mA at 75 oC (almost 9 volts at room temperature in
0.1 M borate buffer supporting electrolyte) whereas a SPEE cell needs less than 3 volts to run at
100 mA at room temperature in a supporting electrolyte free environment. The energy aspect
definitely makes the SPEE system more appealing for catalytic hydrogenation in fuel production.
Although the nickel ion deposition and use of the potassium ion influx for Columbic efficiency
enhancement defeated the initial purpose of SPEE, the trace quantity of nickel ion dopant is
consumed by the cathode and the potassium salt can be removed easily with aqueous extraction.
So SPEE is still arguably an efficient and cost effective strategy for bio-oil upgrading.
The important aspects of Columbic efficiency enhancement are summarized in Table 4-4.
Estimations are given based on comparison between trials that have only one changed variable.

Presence of
acetic acid

Nickel ion
redeposition

Acidic
anolyte

Mesh Size

Current
Density

Enhance
C.E.%
Qualitative
Estimate
Estimate %
Sources

K+ influx to
lower
surface
acidity

Table 4-4: Factors that enhance current efficiency towards organic substrate hydrogenation

Yes

No

Yes

No

N.A.

Depends

High

High

Moderate

Moderate

N.A.

Moderate

20 – 25
Figure
4-16
(p.113)
Figure
4-18
(p.117)

20
Table 4-1
(p.107)
Figure
4-21
(p.122)

10 - 15
Figure
4-16
(p.113)
Figure
4-17
(p.116)

15 – 20
Figure 4-16
(p.113)

N.A.
Table 4-2
(p.119)

Depends
Table 4-3
(p.120)

The percentage values are only given as a crude estimation on the efficiency because the
precise influence may depend on the actual condition. In general, the surface acidity appears to
have the greatest impact; lower pH tends to favor more H2 formation than organic hydrogenation
as expected. The situation can be averted with salted anolyte to provide an aprotic cation influx
125

to discourage H2 formation. However, the presence of acetic acid is a great challenge toward
hydrogenation; the formation of H2 gas will not only “waste” electrons but it also creates a thin
gas layer that may shield the catholyte from the membrane surface. Although, technically, the
formation of H2 is not entirely wasteful, because H2 is a valuable commodity that carries high
fuel value, which can be used to fuel the pyrolysis process shown in Figure 4-2.
4.5.1 Miscellaneous Ideas and Future Directions
Plating strategies are being developed to create a ruthenium surface which opens to the
pathway towards aromatic hydrogenation. In the past, several thoughtful attempts on
incorporating the guaiacol Raney-nickel chemistry, which is presented in chapter 4, has been
tried but to no avail. The Raney-nickel plating strategy creates a thick fragile surface that cannot
be fitted in to the small compartment of the current SPEE design. Packing Raney-Nickel powder
in the compartment either led to powder leakage or high pressure that the peristaltic pump cannot
withstand. An innovative Raney-nickel deposition strategy is required to achieve the aromatic
hydrogenation the SPEE. Lessard et el reported a different Raney-nickel deposition strategy, in
which mixture of the powder ingredients were pressed in an electrode shape before activation.151
The electroless preparation perhaps gives control for the size and shape of the electrode and is
worth a try. At the same time, the ruthenium catalyst deposition on the copper mesh has also
been attempted, and the proto-type results are reported in the final chapter. The simplicity of
SPEE opens limitless opportunities for modifications. The flexibility can be another advantage of
in bio-oil hydrogenation aside from electrolyte free catalysis; the SPEE components can be
modified easily to tailor the specific needs for bio-oil, a complex mixture whose composition
varies depending on feedstocks and upgrading. Furthermore, easy implementations and parts

126

substitutions are great advantage in industrial synthesis as these processes often require regular
part replacement.
Although acetol is the only model compound tested and it may not reflect the overall
hydrogenation chemistry happening during the upgrading process, understanding the mechanism
of SPEE marks the critical first step towards an efficient and effective hydrogenation. Aside
from developing a new mesh surface for aromatic hydrogenation, other abundant molecules such
as furfural and 2,3-butanedione will soon be examined in the SPEE system. The Columbic
efficiency data of reducing these molecules will be helpful to understand ECH of different
functional groups, and lead to better predictions on how bio-oil will behave during the upgrading
reaction.

127

Chapter 5
5.1

Pedagogy: Electrocatalytic Hydrogenation of Acetol with US. Coins

Introduction
The addition of hydrogen across pi-bonded systems is a category of transformation

central to organic synthesis. But textbook alkene hydrogenations involve costly noble metal
catalysts, H2 gas handling, and often elevated pressures and temperatures.223 Catalytic
hydrogenations are thus relatively unattractive as pre-college or undergraduate classroom
experiments. However, organic hydrogenation has existed for over a century and is one of the
most widely applied reactions in the history of organic syntheses. For carbonyl pi systems,
hydrogenations are usually effected with hydride delivery reagents such as NaBH 4 or LiAlH4.
Such reactions are easy and quick and therefore readily adapted to use in a teaching lab. On the
other hand, the reagents can be hazardous and costly, they have poor atom economy, and the
required post-reaction workups generate substantial waste streams.
Compared to H2 molecular hydrogenation (MH) and carbonyl reduction by hydride
reagents (HR), electrocatalytic hydrogenation (ECH) is a relatively uncommon technique. With a
direct current at modest voltage, hydrogenation of organic substrates can be readily achieved on
the surface of a cathode of suitably chosen catalytic metal. ECH typically runs at ambient
pressure and modest temperatures, conditions mild enough to make it straightforward to
implement as a pre-college or undergraduate laboratory experiment.97, 221
This experiment is designed for students equipped with entry level of organic chemistry.
Participants should be made aware of the harsh condition of conventional H2 hydrogenation so
they may appreciate ECH as be a mild alternative. We believe this experiment can be a good
complementary study to the introduction of catalytic H2 hydrogenation or pi-bond addition. Or, it

128

can serve as an extra laboratory activity to help student appreciate chemical transformations that
can happen with ambient objects such as battery, coins and filter papers.
Since ECH is outside of the usual coverage of the ACS bachelor’s degrees program,
instructors should introduce this exercise where they see fit. In our case, we introduced this
experiment to students as seeking the part of their “honors option” during their sophomore
organic chemistry class. Some high school students from local institutions also tested it. A total
of 30 undergraduate sophomore and 2 high school students participated in teams of 2 – 3; all
participants completed the experiment within 2 hours under supervision of a graduate assistant.
A typical setup for ECH can be as simple as a container of conductive solution with two
electrodes coupled with a direct current power supply. Building on this simplicity, we
demonstrate herein that ECH can be practiced even with ordinary familiar materials: coins and
batteries!

129

a

Adsorbed
Hydrogen

H

X

=Y
H

H

H

H

Electrode Surface

H HX

X

H

=Y

c

- YH

H

H

Electrode Surface

Hydrogenated
Organic
Compound

H

b

Adsorbed
Organic
Compound

H

H

HX
H

Electrode Surface

- YH
H

H

H

d

Electrode Surface

e

f

Hydrogen
Evolution

H2
H

H

H

-H

H

H

Electrode Surface

Electrode Surface

Figure 5-1: Schematic diagram of ECH of unsaturated organic molecules (a-d) and hydrogen
evolution (e -f)
On a catalytic metal cathode in aqueous electrolyte, reduction of protons from solution
forms surface-bound hydrogen atoms. When an unsaturated organic molecule is adsorbed
nearby, it can undergo hydrogenation by bonding with these hydrogens (Figure 5-1a). However,
hydrogen gas may also form via either the Tafel reaction, in which two adsorbed hydrogen atoms
come together, bond and desorb (Figure 5-1b), or the Heyrovski reaction, where a solution
proton reacts with a surface-bound hydride. With more than one reaction consuming the
delivered current, it is important to quantify the electrochemical selectivity for formation of the
desired product, a measure known as Columbic efficiency.
As shown in Scheme 5-1, Acetol, a bio-derived substrate, can undergo ECH to propylene
glycol without an external hydrogen supply, using batteries and coins with modest heating. The
purpose of this exercise is to show how catalytic chemical transformations can be performed
130

with familiar household objects, highlighting the idea that chemistry is not just the province of
experts with access to specialized resources. Properly probed, chemical behavior can be found
and easily understood in everyday things.
5.2

Hazards and Disposals
Reagents used or formed in this study--acetol, propylene glycol, and ammonium chloride--

are generally harmless and non-corrosive, making it suitable for undergraduate or high school
students. The Benedict’s solution requires neutralization for disposal, but is otherwise nonhazardous. Nevertheless standard safety protections such as lab coat safety gloves should be
worn at all times during experiments. After the experiments, the used coins should be cleaned
with de-ionized water and gently brushed to remove any surface residue before being returned to
circulation.
5.3

Experimental Procedures

5.3.1 Materials and Preparations
All procedure in the manuscript is written for instructor preview, student should be working with
the stepwise procedure found in the supporting info section.
Preparation from the instructor:
Recommended quantities of supplies for each team (2-3 students): 2 x 1.5 v batteries, multimeter,
5-6 plastic bags, 4-5 electrical wires with banana clips and alligator clips, 12 test tubes, 10 pennies and 6
nickels and 8 office binder clips.
Quantitative Benedict’s (QB) solution224 can either be purchased or made from common reagents:
Solution 1: Dissolve 100 g of sodium citrate, 32.5 g of anhydrous sodium carbonate, and 62.5 g of
potassium thiocyanate in 400 ml of de-ionized water. Solution 2: Dissolve 9 g of copper sulfate
pentahydrate in 50 ml of deionized water. Add solution 2 to solution 1 with rapid stirring. Add 0.13 g
potassium hexacyanoferrate to the mixture, then dilute to 500 ml. Dilute 35 ml of the resulting solution to

131

100 ml with water. The stock solution does not require a hazard warning label, but its final disposal calls
for a labeled hazardous waste container.
Preparation from the Students

Figure 5-2: (left) Experimental setup of ECH of acetol with coins. A copper coin anode is
clipped with red wire, and the nickel coin cathode is clipped with the black wire. An extra
coin is added on each side to act as a current collector and provide enough pressure for the
clamp (right). Setup is in a small sample bag to prevent filter paper drying.
Standards preparation for analysis: cut out ten 1”x 2” rectangles of filter paper; fold each
into a 0.5”x 0.5” square. Prepare 20 ml each of 5% v/v acetol (0.67 M) and 5% v/v propylene
glycol (0.67 M) in 0.25 M of ammonium chloride electrolyte. Use these stocks to make standard
5 ml solutions of 5:0, 4:1, 3:2, 2:3, 1:4, and 0:5 acetol: propylene glycol ratios. In each standard,
soak one filter paper for 5 minutes, then remove and squeeze out the excess electrolyte with 2
clean coins. Clamp each of the soaked filter papers as shown in Figure 5-2 without passing
electricity through them. These control experiments are to be left in a sealed bag for 10 minutes
(or time equivalent to actual experimental runs). Then submerge each filter paper in a separate 10
ml aliquot of QB solution, and proceed with the experimental trials.
5.3.2 Experiment
Clean the coins by boiling them in 5% CH3COOH solution for 10 minutes and rinse them
thoroughly with distilled water. Prepare 10 mL of 1 M acetol in 0.25 M NH4Cl electrolyte
solution. Cut out a 1”x 2” filter paper and fold it into a 0.5”x 0.5” square and measure its weight,
132

then soak it in the prepared electrolyte for 5 minutes. Squeeze out the excessive electrolyte with
the dried, cleaned coins and reweigh the filter paper. The amount of acetol absorbed into filter
paper can be calculated from the weight difference and known concentration of the solution.
Sandwich the filter paper between the cathode coin and anode coin with a binder clip carefully
insulated with masking tape to prevent a short circuit. To minimize electrolyte evaporation,
enclose the coin and filter paper assembly in the bag as shown in Figure 5-2.

Figure 5-3: Electronic setup of the experiment. Note: Drawing is not to scale
Connect the setup with a pre-configured power supply set at 3 V or a pair of 1.5 V
batteries connected in series, as shown in Figure 5-3. Be sure to connect multimeter in ammeter
mode. Upon closing the circuit, record the current reading every 10 seconds for the first five
minutes. After that, take the current reading at 1 minute intervals until the reaction time reaches
10 minutes. The amount of charge passed through the circuit can be computed from the area
under the curve in a current vs. time plot. After reaction, take out and separate the coins, put the
filter paper into a test tube, and pour in 10 ml of QB solution. Heat the set of QB-treated
standards prepared earlier along with these experiment sets for 20 minutes in a boiling water
bath. Be sure the solution is loosely capped with a rubber stopper to minimize evaporation during
heating. The resulting change of color of the solutions should approximately resemble Figure
5-4. Allow solutions to cool to room temperature, filter them individually through Pasteur
pipettes packed with a small piece of cotton at the tip to remove the light-scattering particles, and
133

then analyze them spectrophotometrically. Calibrate spectrophotometer, Spec 20 or equivalent,
with the standard solutions at λ = 670 nm using de-ionized water as a 100% transmittance
calibrant. Plot the absorbance (y) vs. percent acetol (x) in a standard plot (Figure 5-5), which can
be used to determine the concentration of the remaining acetol from experimental samples.

Figure 5-4: Color changes of Q.B. solution in response to concentration of acetol. From left
to right: 100:0, 80:20, 60:40…..0:100 acetol (5% v/v): propylene glycol (5% v/v)

Absorbance

Absorbance of acetol at 670 nm at different concentration

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

y = -0.0043x + 0.5834
R² = 0.9347

0

20
40
60
80
% Dilution of the 5% acetol solution

100

Figure 5-5: Change in absorbance of QB solution in response to % of acetol in standard
solution. Note: students can take readings in transmittance and calculate the absorbance
using the equation: A = 2 – log (%T)
5.4
Current Efficiency Analysis
eq. 1

2H+ + 2e- → H2

eq. 2

Scheme 5-1: Acetol hydrogenation to propylene glycol electrocatalytically. and
hydrogenation evolution

134

Table 5-1: Results of ECH of acetol in various conditions. All trials were run using a penny as
the anode at room temperature. Note: results were determined by 1H-NMR peak integration to
verify the formation of propylene glycol (See Figure 5-6). 1) 3 V d.c. power supply; 2) 3 V
(two AA cells in series) batteries; 3) 3 V d.c. power supply with bag/coins in an ice bath.
1
2
3

Cathode
Penny
Nickel
Penny
Nickel
Penny
Nickel

Charge Passed (mC)
4.95
4.99
5.97
7.97
4.83
7.02

Mole of P.G.
formed (µmol)
3.5
7.9
5.8
12.8
3.0
9.7

% C.E.
13.5
30.6
18.9
31.0
12.2
26.5

Figure 5-6: 1H -NMR spectrum (Red) propylene glycol standard (Orange) ECH of acetol
with penny for 10 minutes; low resolution of the orange spectrum was due to Cu2+ ion
contamination. (Green) acetol standard (Blue) experiment 10 minutes with no electricity
passed
The fraction of electrons passed that are spent in the desired reaction is termed
“Columbic efficiency”. Two reactions occur on the cathode surface as shown in equations 1 and
2. The total charge passed can be estimated by integrating the current vs. time plot shown in
Figure 5-7. The Columbic efficiency is calculated as follows:

135

( )

(

)

where Mol. = moles of propylene glycol formed. F = faraday's constant, 96485 C mol -1, n = number of electrons per reaction.

Figure 5-7: An example of current-time profile. Total charge passed (Ctotal) can be estimated
by calculating area of the curve. Area under the curve can be estimated by simple summing of
the trapezoidal areas.
5.5

Results and Discussion

A layer of copper black can be seen on the cathode coin surface after the reaction; Figure
5-8 shows analyses of this surface by SEM and EDX spectroscopy. The copper black layer was
generated by re-deposition of copper ions that came from the surface of the sacrificial penny
anode3

Figure 5-8: (left) SEM spectroscopic analysis of plain penny surface, and (right) the black
copper layer after reaction

136

In general, the nickel cathodes show greater Columbic efficiency for ECH than the
penny. The nickel coin surface is an alloy of copper and nickel metal,225 so the results suggest
that the catalytic effect of nickel metal doped in copper is greater towards ECH than that of
copper alone. In an ice bath (trial 3), the Columbic efficiency of the run does not change
significantly at low temperature. The 3 V battery trial (trial 2) shows a lower conversion with the
penny cathode and greater conversion with nickel cathode; these variations may reflect slightly
different battery voltages, but the Columbic efficiency agrees fairly well with the other trials,
suggesting that catalytic function remains constant.
The analytical method was designed to accommodate laboratory that have limited access
to analytical instruments such as GC, LC and 1H-NMR. Noting that the Beer’s law related
experiment are very common among entry level analytical chemistry courses, we proposed a
non-instrument intensive analytical scheme that only requires Benedict’s solution and a Spec 20
UV-vis spectrophotometer. Admittedly, this scheme does not measure the formation of
propylene glycol and student should be focusing on the formation of product (P.G.) rather than
disappearance of acetol. However, to the best of our knowledge, there is no simple means to
determine small quantities of propylene glycol without employing sophisticated instruments.
Despite the success and simplicity of the above organic conversion, there are a few
potential issues worth mentioning. a) A relatively high concentration of acetol, 5% v./v. (0.67 M)
is needed to ensure extraction efficiency for analysis. Volume to volume type concentration is
chosen specifically to minimize students waiting time at a balance to weigh reagents. b)
Excessive electrolyte must be squeezed out as much as possible to prevent alternative conduction
paths that could short circuit the cell. Consistency in the squeezing is crucial to the results as it
determines the amount of acetol left in the filter paper. As the same time, filter paper should not

137

be too dried out for the electrochemical reaction. Admittedly, hand squeezing of the electrolyte
may seem subjective. However, this error can be avoided via determining the weight of the
liquid (electrolyte) absorbed into the filter paper.
As for the reference standards preparation, the weighing part can be omitted because the
line of best fit (Figure 5-9) compensates for the uncertainty due to filter paper squeezing,

Absorbance

assuming the filter paper squeezing process is consistence throughout the preparation.
0.6
0.55
0.5
0.45
0.4
0.35
0.3
0.25
0

20

40
60
% of Acetol

80

100

Figure 5-9: Change in absorbance of QB solution in response to % of acetol in standard
solution.
Also, the reaction should be run at room temperature or below: higher temperatures lead
to filter paper dehydration and changes in the resistance of the system. c) Teams of 2-3 students
are highly recommended. d) A short reaction time (10 minutes) is recommended to minimize
oxidative damage to the coin surfaces. e) Accuracy of the 10 ml QB solution addition is crucial
to the experiment results; small changes in the observed transmittance can result in large
variations in the Columbic efficiency calculations. A mechanical pipette, which we used, is
recommended to maximize the consistency of each aliquot transfer. f) Absorbance measured
digitally on typical modern UV-vis instruments can help minimize variability from transmittance
or absorbance conversion. g) The amount of current passed during the reaction depends on the

138

cleanness of the wires and alligator clips. Though Columbic efficiency is an intrinsic property of
the reaction, it is best to keep all equipment as clean as possible to maximize reproducibility.
This experimental procedure was tested by a group of undergraduate students (class of
30) in groups of 2-3. Average lab times ranged from 150-180 minutes. Lab time can be shortened
to 2 hours if QB solution, filter papers, boiling water bath, and coins are prepared by the
instructor in advance. Special care should be taken to ensure consistency of the QB solution
preparation and addition, Spec 20 calibration, amount of electrolyte carried in filter paper (hand
squeezed as hard as possible), electronic connections of the electrodes, and multimeter setup. For
consistency and time efficiency, the instructor may prepare the standard mixtures, QB solutions,
data sheet templates, pre-calibrated spectrophotometry, and pre-cleaned coins in advance. In
general, reproducibility problems are typically due to variations in preparation of the filter paper
(saturation with electrolyte) and addition of QB solution.
The aim of this experiment is to introduce the concept that organic transformations can be
performed electrochemically with familiar objects, even with coins. Unlike molecular
hydrogenation (H2) or chemical hydrogenation (CH), Electrocatalytic Hydrogenation (ECH)
does not require harsh conditions, purchase or preparation of a specific reducing reagent or
handling of H2 gas, making it attractive for entry level laboratory experiments. The use of U.S.
coins in this experiment highlights the point that organic chemistry does not necessarily require
elaborate setups or reagents. In addition, acetol is an oxygenate like many bio-based organic
compounds of relevance to future technologies in which refineries will make chemicals and fuels
from renewable plant-derived carbon compounds instead of fossil petroleum.

139

5.6

Summary
We have reported a simple electrochemical experiment that employs only common

abundant objects and harmless chemicals. This experiment is designed to introduce mild
electrochemical hydrogenation to students as a complimentary reaction to the conventional H2
hydrogenation. Though electrocatalytic hydrogenation (ECH) of acetol to propylene glycol (PG)
may seem trivial but one should be reminded that PG (the product) is vital product that impact
our daily lives heavily. PG can be found in antifreeze reagent, plastic precursor, preservatives in
food and tobacco products. Although the current industrial production of propylene glycol is
achieved differently, Acetol is an environmentally friendly reagent that is formed in biomass fast
pyrolysis, which converts biomass to bio-oil. In other words, production of propylene glycol, can
be achieved from biomass via electrocatalysis, a method potentially powered by renewable
electricity source such as solar and wind. In summary, this experiment was designed to
demonstrate a “green” chemical transformation that can be achieved with daily used objects. As
a bonus, we hope this work can also raise the awareness on sustainability.

140

Chapter 6
6.1

Conclusions

Project I - ECH of Bio-oil Model Compound, Guaiacol to Chemicals and Fuel
Stabilization and energy upgrading of the model bio-oil components, guaiacol and its

congeners, has been achieved electrocatalytically. Hydrodeoxygenation of guaiacol to phenol
and cyclohexanol was demonstrated using a Raney-nickel electrocatalyst under ambient
conditions. Although cyclohexanol still requires further deoxygenation before it reaches the fuel
level energy, it is far more stable than guaiacol in the presence of other organic substrates. The
phenolic ring of guaiacol can easily be oxidized or undergo polymerization whereas the saturated
ring of cyclohexanol is stable under reasonable storage condition. With this technique developed,
it is possible to envision bio-oil getting stabilized in a local electrocatalytic upgrading station,
stored, and transported to a centralized refinery for deoxygenation (via dehydration) before it is
distributed as fuel.

Ideally, of course, it would be better if the final deoxygenation (e.g.

cyclohexanol to cyclohexane) could be achieved electrocatalytically.
Improving Columbic efficiency remains a major future challenge. Although Raney-nickel
was an excellent low cost substitution for platinum group metals, it preferentially generated
hydrogen on its surface leading to low Columbic efficiency during organic hydrogenation.
Doping Raney Nickel with low hydrogen activity metals (See Chapter 2, volcano plot) to create
new catalyst surface may improve the situation. Furthermore, the stability of Raney-nickel is far
from satisfactory – its reactivity decreased significantly after use or exposure to air. Doping the
Raney-nickel with foreign metals such as chromium, similar to stainless steel manufacture, might
be a good strategy to improve its stability. Following up on the material described above, several
experiments are proposed to provide a good starting point for the next project participant.

141

6.1.1 Improvement I – Electrode Material Synthesis
Raney Nickel electrodes have been a good catalyst towards aromatic bonds
hydrogenation, which is usually done with platinum group metal. In Chapter 4, aromatic alkoxycleavage occurred in a mildly basic environment using a Raney-Nickel cathode prepared with the
Lessard electro co-deposition method. Guaiacol to phenol conversion was followed by aromatic
hydrogenation to yield cyclohexanol. Raney-nickel catalyzed the reaction with moderately good
Columbic efficiency and excellent selectivity shown in Figure 6-1.

Figure 6-1: ECH of guaiacol using Raney-Nickel cathode. Reaction proceeds with excellent
selectivity to form cyclohexanol as the sole product, and does not form the planar 2methoxycyclohexanol
However, compared to precious platinum, Raney-Nickel is sensitive to air oxidation and
acidic environments because nickel can be easily oxidized and dissolved in acid. Seeing the
acidic nature of bio-oil, it is important to develop a skeletal catalyst that can tolerate mild acidity
while still giving good catalytic effect towards hydrogenation. Moreover, the skeletal catalyst
would preferably be air-stable to minimize the storage precautions. Hence, the following three
starting directions are proposed to develop an improved skeletal metal.
Nickel-Aluminum-[Metal] synthesis: a significant number of studies have explored the
dopant effects of other transition metals such as chromium, molybdenum, tin, lanthanum226 into
the nickel aluminum alloy. These studies are typically conducted to improve the catalytic effect
on hydrogen oxidation for H2 fuel cell application. There is yet to be a modified skeletal nickel
142

study specifically focused on organic ECH. The natural approach would be to synthesize alloys
of known composition from literature and examine their catalytic effects in organic ECH.
Separately, cyclic voltammetry (CV) can be used to probe the effects of the dopant on the
oxidative potential of the modified skeletal metal catalyst. Successful development of an acid
resistant skeletal metal catalyst would be a valuable asset to the bio-oil hydrogenation.
“Skeletalize” known acid resistive alloy: a large number of metal alloys are known to
have acid resistive property, such as Monel 400, Hasteloy C, Inconel 600, stainless steel 304
(316), and many more. If aluminum could be fused to the surface of these alloys followed by
basic solution treatment to create a porous surface, this could revolutionize the fundamental
preparation technique for creating high surface area catalyst.
This strategy is fairly promising, because in his first publication, Murray showed Ni-Al
alloy can be synthesized by simply dissolving nickel (mp 1453 oC) into molten aluminum (mp
660 oC) under coal fire heating. In other words, the vast difference in melting point is not an
impediment to the synthesis, which therefore should allow a wide substrate survey range.
Success of this approach would allow any individual to create a porous highly active catalyst of a
well-studied metal alloy with desired properties.
Electroless metal coating on Raney-Nickel electrode: freshly activated Raney-Nickel
comes with excellent reducing power on its surface owing to the aluminum leaching. In principle,
treating a freshly activated Raney-Nickel catalyst in a metal ion salt solution should achieve
electroless deposition of a thin metal coating on the catalyst surface. This thin coating technique
has several merits: a) the preparation procedure is simple, b) the amount of coating will
automatically be thin and uniform because the reducing power will be gone once the metal is

143

plated, and c) a precious metals skeletal electrode could take advantage of this strategy because
the technique requires minimal electrolyte for plating.
6.1.2 Improvement II – Substrate Scope and Mechanistic Studies
Being able to hydrogenate guaiacol to cyclohexanol is an encouraging step in upgrading
similar aromatic compounds found in bio-oil. Syringol (2,6-dimethoxyphenol) reacted similarly
to guaiacol to produce cyclohexanol as the final product. The current step is to expand the
substrate scope to other aromatics such as eugenol, vanillin, sinapyl alcohol and many more.
Understanding how mixtures of different substrates hydrogenate is critical to address the
complexity of bio-oil during its upgrading – the mechanism of hydrogenation may change in the
presence of multiple substrates.
The current proposed mechanism for guaiacol demethoxylation is informative but not yet
conclusive. Although some proposals have been ruled out by designed experiments, the exact
mechanism (e.g. how guaiacol adsorbs on the nickel surface) is not clear. Probing guaiacol
adsorption on the electrode surface may explain and enable control over the poor selectivity seen
in the 3- and 4-methoxyphenol reactions, which showed only partial ether cleavage, leading to
similar quantities of the corresponding methoxycyclohexanol and cyclohexanol itself.
6.2

Project II – Development of SPEE system for Bio-oil Hydrogenation
The success of the SPEE cell in ECH of simple aliphatic compounds in the absence of

electrolyte was encouraging. However, bio-oil contains a fair number of aromatic compounds
that require a more active catalyst for hydrogenation. Ideally, it would be feasible to merge the
Raney-nickel chemistry with the SPEE cell setup, using a Raney-nickel cathode to hydrogenate
the guaiacol or aromatic compounds. So far, several attempts in this direction have been
unsuccessful as discussed in the conclusion section of chapter 5.

144

Ruthenium has been employed frequently in various research projects in the Jackson
Group. It is a natural idea to develop a strategy to deposit a ruthenium surface which opens a
path towards aromatic hydrogenation in the SPEE cell. As a prototype experiment, a multiple
layer plating scheme was developed to deposite ruthenium onto the stainless mesh surface to
carry out phenol hydrogenation. Starting from the support, layers were as follows: Stainless Steel
--> Nickel --> Palladium --> Ruthenium. Although the deposition of these multiple layers seems
cumbersome, the resulting system did reduce phenol, an aromatic molecule in low yield, in the
absence of electrolyte at slightly elevated temperature (75 oC). The use of ruthenium has great
potential in the bio-oil upgrading project, and it is a priority to improve the current plating
strategy to ensure reproducibility as much as efficiency for the deposition.
6.2.1 Improvement I – Deposition of Ruthenium on Stainless Steel Mesh
Procedures: a 2.5 x 2.5 cm stainless steel was cleaned with hexane to remove wiring oil
residue prior the plating. First, the cleaned stainless steel mesh surface was activated by anodic
oxidation in 30 oC 10% H2SO4 for 30 seconds at 500 mA using a RVC carbon as cathode. After
that, the mesh was rinsed briefly with deionized water and plated with nickel in a Watts Strike
bath for 2 minutes at 75 mA using a nickel bar as anode. After the Watts Strike bath plating, the
mesh is immediately soaked into a 60 oC Woods plating bath and is plated for 15 minutes at 200
mA using a nickel bar as anode. The resulting mesh is denoted as SS-Ni.
The SS-Ni was activated with palladium by placing in contact for 2 minutes in a 5%
PdCl2 solution that was prepared in 2M Hydrochloride acid. The resulting mesh was denoted as
SS-Ni-Pd. Then the SS-Ni-Pd mesh is electrocatalytically plated with ruthenium in a dual
compartment cell, where the catholyte was prepared with 0.1236 g of RuCl3.xH2O in 50 ml of

145

solution while the anolyte was 0.1 M HCl. The plating took take for an hour for 100 mA using a
platinum wire as anode.
Each stage of the plating processing, nickel, palladium and ruthenium, was imaged with
SEM for surface morphology changes examination, and the results are shown below:

Ruthenium – the catalytic layer
• 1 hr of plating in Ru/HCl bath
• Catalyze aromatic ECH
• 4-alkoxy ether cleavage
• Cheapest in the PGM family

Palladium activation layer
• 2 mins of Pd sol’n treatment
• Bond well with Ru and Ni

Nickel – Intermediate layer
• 20 mins – Strike and Plate
• Well studied metal - deposition
• Cheap
• Possible for Ru deposition

Stainless Steel Mesh 200
• Readily available
• Cheap
• Long lasting

Figure 6-2: Multiple layers of metals plated onto stainless steel to support the deposition of
ruthenium.
The complexity of this plating procedure arises from the lack of direct electrochemical
deposition of ruthenium onto stainless steel surface. Nickel is a common metal that can be plated
on stainless steel after the Woods and Watts bath treatment,227 and can alloy well with palladium
because they belong to the same group.228 Ruthenium has been reported to deposit on palladium
chemically, but this strategy requires elaborate preparation of the reagent solution229, and hence
electrochemical plating was attempted. Fortunately, ruthenium deposition was successful as
shown in SEM images, and the EDX analyses.
146

Table 6-1: atomic % from EDX analysis on the surface of the electrodes, SS-Ni-Pd and SSNi-Pd-Ru, showing palladium and ruthenium were deposited onto the surface. Note: the
remaining percentage is made of nickel and chloride.
Catalyst Surface
Ruthenium
Palladium
SS-Ni-Pd
Not detected
38.755 ± 5.082
SS-Ni-Pd-Ru
13.680 ± 1.653
4.452 ± 2.078
Both SS-Ni-Pd and SS-Ni-Pd-Ru were used as the electrode for ECH of 10 ml of 50 mM
of guaiacol in SPEE at 75 oC for hours at 100 mA. Preliminary results showed 22.1% Columbic
efficiency for cyclohexanol formation and no signs of 2-methoxycyclohexanol, an unprecedented
hydrogenation of aromatic substrate in an electrolyte free environment.
6.2.2 ECH of Aqueous Bio-oil using Ru-SPEE
Seeing the excitement in aromatic reduction, an aqueous fraction of bio-oil was attempted.
An aqueous fraction of bio-oil samples preparation was prepared as earlier published.217 Bio-oil
is an organic mixture comprising hundreds of organic components, and so it is easier to address a
portion at a time in the beginning. The aqueous portion was chosen the groups had earlier studied
hydrogenation of bio-oil in a conventional cell using ruthenium on active carbon cloth, and 0.1
M sodium chloride was used as a supporting electrolyte.
A 10 ml sample of aqueous bio-oil fraction was circulated in the cathode compartment at
0.8 ml min-1 in the SPEE cell equipped with the SS-Ni-Pd-Ru cathode and a stainless steel anode
in 0.1 M pH 7.0 phosphate buffer. Reaction was run for 6 hours at 75 oC at 100 mA, and the
results are shown in Figure 6-4.

147

Figure 6-3: GC-MS results on the aqueous fraction of bio-oil before hydrogenation in SPEE
cell. Temperature program: 50 oC hold for 3 minutes, then 15 oC/minute until 270 oC and
hold for 5 minutes

Propylene Glycol
Levoglucosan

1,2-Ethylene Glycol

Furanmethanol

1,4 Butanediol
Phenol

1,2-butanediol

Syringol

Guaiacol

Creosol

Figure 6-4: GC-MS results on the aqueous fraction of bio-oil after hydrogenation in SPEE
cell. Temperature program: 50 oC holds for 3 minutes, then 15 oC/minutes until 270 oC and
hold for 5 minutes
After 6 hours of electrolysis, the pH rose from 2 to 11, as tested by universal pH paper.
The color of the bio-oil changed from clear light yellow to clear dark brown, and no precipitation
was observed. Another distinctive change was that the barbecue smell of the bio-oil disappeared
after the electrolysis. Presumably the rise in basicity caused bio-oil components to polymerize
which accounts for the darkened color and more importantly, the disappearance of many peaks in

148

Figure 6-4. From Figure 6-4, formation of various diols and furanmethanol proved organic
hydrogenation happened during the electrolysis as expected. However, there were still noticeable
amounts of aromatic substrate remaining. The greatest concern is the disappearance of the many
peaks compared to the starting spectrum shown in Figure 6-3 due to basic polymerization,
similar to the acetol condensation. For these reasons, two strategies were proposed to resolve the
basicity issue: a) the use of an anion exchange membrane, and b) use of platinum to supply
protons via water splitting. The use of an anion exchange membrane allows the hydroxide
formed during cathodic reduction to be transported away and hence maintaining the cathodic
acidity constant. In the case of acetol hydrogenation in SPEE: Acetol + 2H2O + 2e- --->
Propylene Glycol + 2OH-. Although the use of platinum mesh in deionized water can provide H+
to neutralize the basicity, the use of platinum would raise the cost of the system, and so it should
be considered as the last resort.
An important consideration for this project is to decrease the acidity of bio-oil by
converting acetic acid to neutral organics such as ethanol. So far, there is yet to be a reported
reaction in literature demonstrating reduction of acetic acid, however, acetic acid oxidative
condensation (Kolbe reaction) has been in an SPEE setup and is worth to consider. Otherwise,
addition of ammonium hydroxide would be required to neutral the acetic acid before it can be
transported or stored.
6.3

New Cell Design – The High Pressure Tube Reactor

Sparked by the continuous flow nature of the SPEE cell, another cell design came to light – the
HPLC-like flow tube reactor.

149

Platinum wire
Nafion Tube that
hosts the Pt wire
(inside)

Catalyst Powder
filter ring/trap

Copper Tube

T-junction Joint

Figure 6-5: HPLC-Flow style cell anatomy
Peristaltic Pump provides
catholyte circulation
inside of the copper tube

Peristaltic Pump provides
Anolyte circulation inside
of the Nafion tube inside
of the copper tube

Parts Vendor: Globalfia

Cathode Alligator Clip
Catholyte Reservoir

Anolyte Reservoir

Anode Alligator Clip
connected to the Pt
wire
Cell Building & photo: Andrew Henika

Figure 6-6: the general overall setup of the Flow cell reactor
The flow cell reactor is designed to explore if a higher pressure environment could
improve the efficiency of ECH. There is only one study published regarding the effect of
150

elevated pressure in ECH of phenanthrene using Raney-Nickel with N2 as an inert gas.230 The
result showed elevated pressure had a negative impact on the Columbic efficiency under constant
current electrolysis, evidently due to immobilization of the formed hydrogen in the pores of the
electrode. However, when the current is switched to periodic mode, elevated pressure at 20.4 atm
has a positive effect. The positive effect could be due to a pressure of hydrogen in the pores
insuring by reversal of reaction, a higher concentration of chemisorbed hydrogen and the
replenishment of the chemisorbed hydrogen consumed in the hydrogenation reaction.
In the flow cell reactor, the copper tube body serves as a high pressure electrolyte
container and the cathode. Inside the copper tube, a Nafion membrane tube houses a platinum
wire that serves as the anode. During the reaction, an HPLC pump is employed to circulate the
catholyte at moderate flow rates from 0.1 – 1 ml min-1. The end tiny pores trap shown in Figure
6-5 serves as a catalyst powder trapper (future direction – high surface with trapped catalyst
powder) and creates back pressure. The higher pressure may boost the absorption for both
hydrogen and organics, thereby improving the Columbic efficiency. So far, standard pressure
electrolysis has been performed to ensure the functionality of the tube flow cell. In the flow cell,
10 ml of 5% acetol solution dissolved in 0.25 M of ammonium acetate (as a supporting
electrolyte) catholyte was circulated in the flow system at 1.0 ml min-1 at room temperature. A
50 mA constant current (estimated current density 117.9 mA cm-2, based on inner cylinder area
of 0.424 cm-2) was applied to the system for 6 hours. The anode was a platinum wire inside of a
Nafion tube hold a reservoir of 0.1 M H2SO4 solution. In addition, an identical experiment was
also performed in a non-circulated mode in which 800 ml of cathode was pumped in a “singlepass” motion to simulate the actual industrial style. The comparison was shown in Figure 6-7.

151

C.E. (%)

60
50
40
30
20
10
0

Flow Style (Single
pass/nonrecirrculated)

Batch Style
(Recirculated)

0

5000

10000
Time (s)

15000

20000

Figure 6-7: Prototype testing of the flow cell reactor. Reaction condition: 5% of acetol
dissolved in 0.25 M of ammonium acetate, 50 mA (117.9 mA cm-2), room temperature, 1 ml
min-1. Anolyte was the platinum wire inside of 0.1 M H2SO4 solution. 0.01 ml of sample was
taken hourly and was mixed with 0.75 ml of D2O for H-NMR analysis. The absolute
concentration of product was calculated by the internal concentration reference, 1,4-dioxane,
that was pre-dissolved in the D2O solvent. Note: the current efficiency is calculated based on
the charge passed and product formed hourly, NOT accumulatively.
In general, the reaction proceeded with good efficiency, forming propylene glycol as the
sole product. A single pass experiment showed a small decline in Columbic efficiency
suggesting the cathode surface, inside of the copper tube, was slowly deactivating during the
reaction. The batch style flow system started out at a higher Columbic efficiency (extrapolating
the line back to zero), but reached the same level as the single pass reactor after an hour, and the
decreased substantially. This was likely due to the lack of reactant on the electrode surface, and
in turns forced the electrode surface to form hydrogen, which is not included in the Columbic
efficiency calculation.
An elevated pressure may benefit the electrocatalytic hydrogenation by having the
formed H2 gas potentially be incorporated into the molecules through catalytic hydrogenation.
Furthermore, a higher pressure allows the system to reach a higher temperature which may
improve the Columbic efficiency. Several attempts on increasing the pressure of the system led

152

to catholyte leakages, and current effort is being spent on modifying seals and pressure valves to
ensure consistent system performance.

6.4

Overall Progress Summary
Raney Nickel Project (Chapter 4)
•
•

•
•

Low Specific Energy
Easily oxidizable or reactive
with other organics

Moderately High Specific Energy
Highly stable in the presence of
other organics

Stage II
(Good for Storage and
Transportation

Bio-oil
•
•

Low Specific Energy
Easily oxidizable or reactive
with other organics

Fuel
•
•

Low Specific Energy
Highly Stable for storage and
transpiration

SPEE Project (Chapter 5)

Figure 6-8: Overall time summary of the progress of this research project.
Compared to the SPEE project, the Raney nickel project is relatively closer to the fuel
stage because of the guaiacol demethoxylation. Unfortunately, the mechanism of this process is
not applicable to cyclohexanol, which typically requires dehydration to deoxygenate. In this
context, dehydration is unlikely because ECH typically runs in an aqueous environment, and
even if an organic solvent were available (neglecting the expensive cost of organic electrolyte),
the acidic environment would favor H2 evolution and perhaps worse, dissolve the electrode.
Therefore, cyclohexanol dehydration would be more readily achieved in a centralized location.
However, an acid tolerant catalytically active metal should still be pursued; bio-oil is acidic due
to the large quantity of acetic acid. Another improvement would be to find other water splitting
153

catalysts also made of earth abundant materials but effective at higher current density. The
simplicity of cobalt phosphate and the neutral pH performance range are remarkably attractive,
but the low current density (1.17 mA cm-2) demands a huge surface area support for adequate
current output. Furthermore, use of phosphate buffer will decrease acidity for bio-oil which is
good at first but will induce polymerization in the long run beyond neutral pH. Neutralization of
bio-oil should be achieved via hydrogenation of the acetic acid, not introducing aprotic cations.

6.5

Final Remarks – The Carbon Balance
Both projects, bio-oil aromatic model compounds upgrading and the SPEE systems, have

been studied extensively to understand the mechanism of how the reaction occurred. Various
side projects have been devised to advance the fuel production of the project. It is important
however, to remember that even with perfect Columbic efficiency and perfect biomass carbon
conversion to fuel, there is still not enough carbon by 2030 to address the carbon demand solely
for the transportation sector today. Therefore, in order to make biomass fuel sustainable, bio-oil
conversion to fuel cannot be the only effort; an alternative fuel consumption strategy is critically
needed to maximize the energy extraction from hydrocarbon fuel. The current internal
combustion engine (ICE) can, at best, be 40% energy efficient. However, fuel cell technology
could boost the energy extraction efficiency to 80%. If ICE could be replaced by hydrocarbon
fuel cell, it would effectively reduce the US fuel dependence by half.
This section is not meant to discuss hydrocarbon or related chemical fuel cell in detail,
but to keep readers reminded that creating a sustainable renewable fuel cycle will require more
than just making hydrocarbons. Renewable fuel production is only the first half of the
sustainable fuel cycle; the other half is the development of a more efficient energy extraction
system to lower fuel demand. As noted, the projected 2030 non-food grade biomass will only

154

have enough carbon to cover approximately 2/3 of the current consumption today. Hence a
different approach for extracting hydrocarbons’ energy is needed to complete the overall
equation. It need not be a fuel cell, though that appears most promising today. As much as a
world free from fossil carbon use would be wonderful, hydrocarbon will still be needed as the
primary energy carriers that are safe, well-established, and high in specific energy. Until a new
energy source is identified with equal or greater merits, hydrocarbons will still be the primary
fuels for transportation. By replacing fossil with biomass carbon and a devising more energy
efficient extraction systems, a future that travels using a truly sustainable renewable fuel recycle
can be envisioned.

155

REFERENCES

156

REFERENCES

1. BTS Table 1-11: Number of U.S. Aircraft, Vehicles, Vessels, and Other Conveyances.
http://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/national_transportation_statist
ics/html/table_01_11.html (accessed 12 Mar 2014).
2. EIA How many alternative fuel and hybrid vehicles are there in the U.S.?
http://www.eia.gov/tools/faqs/faq.cfm?id=93&t=4 (accessed 1 Mar 2014).
3. Thomas, G. Overview of Storage Development DOE Hydrogen Program.
http://www1.eere.energy.gov/hydrogenandfuelcells/pdfs/storage.pdf (accessed 1 Mar 2014).
4. Musk, E.; Straubel, J. B. Model S Efficiency and Range.
http://www.teslamotors.com/blog/model-s-efficiency-and-range (accessed 1 Mar 2014).
5. EERE Ethanol Fueling Station Locations.
http://www.afdc.energy.gov/fuels/ethanol_locations.html (accessed 1 Mar 2014).
6. Tesla Tesla Charging | Tesla Motors. http://www.teslamotors.com/charging#/calculator
(accessed 1 Mar 2014).
7. Sherif, S. A.; Barbir, F.; Veziroglu, T. N., Wind energy and the hydrogen economy—review
of the technology. Solar Energy 2005, 78 (5), 647-660.
8. DeLuchi, M. A., Hydrogen vehicles: an evaluation of fuel storage, performance, safety,
environmental impacts, and cost. Int. J. Hydrogen Energy 1989, 14 (2), 81-130.
9. Winkler, W.; Nehter, P., Thermodynamics of Fuel Cells. In Modeling Solid Oxide Fuel Cells,
Bove, R.; Ubertini, S., Eds. Springer Netherlands: 2008; pp 13-50.
10. O’Garra, T.; Mourato, S.; Pearson, P., Analysing awareness and acceptability of hydrogen
vehicles: A London case study. Int. J. Hydrogen Energy 2005, 30 (6), 649-659.
11. Stopher, P.; Stanley, J., Introduction to Transport Policy: A Public Policy View. 1st ed.;
Edward Elgar Publishing Limited: Cheltenham, UK, 2014; p 360.
12. Cadwallader, L. C.; Herring, J. S. Safety Issues with Hydrogen as a Vehicle Fuel; Idaho
National Engineering and Environmental Laboratory: Idaho Falls, 1999; p 78.
13. Kikukawa, S.; Yamaga, F.; Mitsuhashi, H., Risk assessment of Hydrogen fueling stations for
70MPa FCVs. Int. J. Hydrogen Energy 2008, 33 (23), 7129-7136.
14. Bockris, J., A Hydrogen Economy. Science 1972, 176 (4041), 2.

157

15. Peplow, M. Hydrogen's False Economy.
http://www.rsc.org/chemistryworld/2013/03/hydrogen-economy-clean-energy (accessed 1 Mar
2014).
16. EERE Benefits and Challenges. https://www.fueleconomy.gov/feg/fcv_benefits.shtml
(accessed 1 Mar 2014).
17. Dillich, S. Hydrogen Production & Delivery.
http://www.hydrogen.energy.gov/pdfs/review12/pd_plenary_dillich_2012_o.pdf.
18. Siler, S. Pump it Up: We Refuel a Hydrogen Fuel-Cell Vehicle.
http://www.caranddriver.com/features/pump-it-up-we-refuel-a-hydrogen-fuel-cell-vehiclehydrogen-filling-stations-are-still-rare-page-3 (accessed 2 Mar 2014).
19. Bossel, U.; Eliasson, B. Energy and the Hydrogen Economy; Alternative Fuels Data Center U.S. Department of Energy: 2003.
20. Çengel, Y. A.; Boles, M. A., Thermodynamics: An Engineering Approach. McGraw-Hill
Professional Publishing: 2004.
21. EERE Ethanol Production and Distribution.
http://www.afdc.energy.gov/fuels/ethanol_production.html (accessed 27 Mar 2014).
22. Nopmongcol, U.; Griffin, W. M.; Yarwood, G.; Dunker, A. M.; MacLean, H. L.; Mansell, G.;
Grant, J., Impact of dedicated E85 vehicle use on ozone and particulate matter in the US. Atmos.
Environ. 2011, 45 (39), 7330-7340.
23. Yang, Y.; Bae, J.; Kim, J.; Suh, S., Replacing Gasoline with Corn Ethanol Results in
Significant Environmental Problem-Shifting. Environmental Science & Technology 2012, 46 (7),
3671-3678.
24. Hill, J.; Nelson, E.; Tilman, D.; Polasky, S.; Tiffany, D., Environmental, economic, and
energetic costs and benefits of biodiesel and ethanol biofuels. Proceedings of the National
Academy of Sciences 2006, 103 (30), 11206-11210.
25. Donner, S. D.; Kucharik, C. J., Corn-based ethanol production compromises goal of reducing
nitrogen export by the Mississippi River. Proceedings of the National Academy of Sciences 2008,
105 (11), 4513-4518.
26. Fargione, J.; Hill, J.; Tilman, D.; Polasky, S.; Hawthorne, P., Land Clearing and the Biofuel
Carbon Debt. Science 2008, 319 (5867), 1235-1238.
27. Kim, S.; Dale, B. E., Life cycle assessment of various cropping systems utilized for
producing biofuels: Bioethanol and biodiesel. Biomass Bioenergy 2005, 29 (6), 426-439.

158

28. Steynberg, A. P.; Espinoza, R. L.; Jager, B.; Vosloo, A. C., High temperature Fischer–
Tropsch synthesis in commercial practice. Appl. Catal., A 1999, 186 (1–2), 41-54.
29. de Klerk, A., Fischer–Tropsch Process. In Kirk-Othmer Encyclopedia of Chemical
Technology, John Wiley & Sons, Inc.: 2000.
30. Espinoza, R. L.; Steynberg, A. P.; Jager, B.; Vosloo, A. C., Low temperature Fischer–
Tropsch synthesis from a Sasol perspective. Appl. Catal., A 1999, 186 (1–2), 13-26.
31. Demirbas, A., Biorefineries: Current activities and future developments. Energy Conversion
and Management 2009, 50 (11), 2782-2801.
32. Fatih Demirbas, M., Biorefineries for biofuel upgrading: A critical review. Applied Energy
2009, 86, Supplement 1 (0), S151-S161.
33. Jitaru, M., Electrochemical Carbon Dioxide Reduction - Fundamental and Applied Topics
(Review). Journal of the Univeristy of Chemical Technology and Metallurgy 2007, 42 (4), 12.
34. Azuma, M.; Hashimoto, K.; Hiramoto, M.; Watanabe, M.; Sakata, T., Electrochemical
Reduction of Carbon Dioxide on Various Metal Electrodes in Low‐Temperature Aqueous
 KHCO 3 Media. J. Electrochem. Soc. 1990, 137 (6), 1772-1778.
35. Hori, Y., Electrochemical CO2 Reduction on Metal Electrodes. In Modern Aspects of
Electrochemistry, Vayenas, C.; White, R.; Gamboa-Aldeco, M., Eds. Springer New York: 2008;
Vol. 42, pp 89-189.
36. Haynes, L. V.; Sawyer, D. T., Electrochemistry of carbon dioxide in dimethyl sulfoxide at
gold and mercury electrodes. Anal. Chem. 1967, 39 (3), 332-338.
37. Bewick, A.; Greener, G. P., The electroreduction of CO2 to glycollate on a lead cathode.
Tetrahedron Lett. 1970, 11 (5), 391-394.
38. Bewick, A.; Greener, G. P., The electroreduction of CO2 to malate on a mercury cathode.
Tetrahedron Lett. 1969, 10 (53), 4623-4626.
39. Kuhl, K. P.; Cave, E. R.; Abram, D. N.; Jaramillo, T. F., New insights into the
electrochemical reduction of carbon dioxide on metallic copper surfaces. Energy Environ. Sci.
2012, 5 (5), 7050-7059.
40. Gattrell, M.; Gupta, N.; Co, A., A review of the aqueous electrochemical reduction of CO2 to
hydrocarbons at copper. J. Electroanal. Chem. 2006, 594 (1), 1-19.
41. Kondratenko, E. V.; Mul, G.; Baltrusaitis, J.; Larrazabal, G. O.; Perez-Ramirez, J., Status and
perspectives of CO2 conversion into fuels and chemicals by catalytic, photocatalytic and
electrocatalytic processes. Energy Environ. Sci. 2013, 6 (11), 3112-3135.

159

42. Gibbins, J.; Chalmers, H., Carbon capture and storage. Energy Policy 2008, 36 (12), 43174322.
43. Metz, B.; III., I. P. o. C. C. W. G., IPCC Special Report on Carbon Dioxide Capture and
Storage. Cambridge University Press for the Intergovernmental Panel on Climate Change: 2005.
44. Rochon, E.; Bjureby, E.; Johnston, P.; Oakley, R.; Santillo, D.; Schulz, N.; Goerne, G. v.
False Hope - Why Carbon Capture and Storage won't save the climate; The Netherlands, 2008.
45. Pires, J. C. M.; Martins, F. G.; Alvim-Ferraz, M. C. M.; Simões, M., Recent developments on
carbon capture and storage: An overview. Chem. Eng. Res. Des. 2011, 89 (9), 1446-1460.
46. EIA Petroleum & Other Liquids.
http://www.eia.gov/dnav/pet/pet_cons_psup_dc_nus_mbblpd_a.htm (accessed 26 Mar 2014).
47. Tverberg, G. Why is US Oil Consumption Lower? Better Gasoline Mileage?
http://ourfiniteworld.com/2013/01/31/why-is-us-oil-consumption-lower-better-gasoline-mileage/
(accessed 2 Mar 2014).
48. EIA How much petroleum does the United States import and from where?
http://www.eia.gov/tools/faqs/faq.cfm?id=727&t=6 (accessed 1 Mar 2014).
49. James.H.Butler The NOAA Annual Greenhouse Gas Index.
http://www.esrl.noaa.gov/gmd/aggi/ (accessed 25 Apr 2011).
50. Tans, P. ESRL Global Monitor Division - Global Greenhouse Gas Reference Network.
http://www.esrl.noaa.gov/gmd/ccgg/trends/ (accessed 4 Mar 2014).
51. EPA Overview of Greenhouse Gases.
http://www.epa.gov/climatechange/ghgemissions/gases/co2.html (accessed 1 Mar 2014).
52. Statistics, B. o. T. National Transportation Statistics.
http://www.bts.gov/publications/national_transportation_statistics/ (accessed 20 Apr 2011).
53. Downing, M.; Eaton, L. M.; Graham, R. L.; Langholtz, M. H.; Perlack, R. D.; Turhollow Jr,
A. F.; Stokes, B.; Brandt, C. C. U.S. Billion-Ton Update: Biomass Supply for a Bioenergy and
Bioproducts Industry; ORNL/TM-2011/224; Other: BM0102030; CEBM007; TRN:
US201118%%1083 United States10.2172/1023318Other: BM0102030; CEBM007; TRN:
US201118%%1083Thu Jun 06 09:43:04 EDT 2013ORNLEnglish; 2011; p Medium: ED.
54. Joosten, J.; Gross, P.; Kydes, A.; Maples, J. D. The Impact of Increased Use of Hydrogen on
Petroleum Consumption and Carbon Dioxide Emissions; National Energy Information Center:
2008; p 20.

160

55. Zhu, X.-G.; Long, S. P.; Ort, D. R., What is the maximum efficiency with which
photosynthesis can convert solar energy into biomass? Curr. Opin. Biotechnol. 2008, 19 (2),
153-159.
56. Govindjee; Govindjee, R. What is Photosynthesis?
http://www.life.illinois.edu/govindjee/whatisit.htm (accessed 26 Mar 2014).
57. Blaikie, D. Most Efficient Solar Panels. http://sroeco.com/solar/most-efficient-solar-panels.
58. Bridgwater, A. V.; Meier, D.; Radlein, D., An overview of fast pyrolysis of biomass. Org.
Geochem. 1999, 30 (12), 1479-1493.
59. Elliott, D. C.; Beckman, D.; Bridgwater, A. V.; Diebold, J. P.; Gevert, S. B.; Solantausta, Y.,
Developments in Direct Thermochemical Liquefaction of Biomass - 1983-1990. Energy Fuels
1991, 5 (3), 399-410.
60. Bridgwater, A. V., Production of high grade fuels and chemicals from catalytic pyrolysis of
biomass. Catal. Today 1996, 29 (1-4), 285-295.
61. Bridgwater, A. V., Principles and practice of biomass fast pyrolysis processes for liquids. J.
Anal. Appl. Pyrolysis 1999, 51 (1-2), 3-22.
62. Bridgwater, A. V.; Peacocke, G. V. C., Fast pyrolysis processes for biomass. Renew. Sust.
Energ. Rev. 2000, 4 (1), 1-73.
63. Bridgwater, A. V., Fast Pyrolysis of Biomass 2: A Handbook. CPL Scientific Publishing
Services, Limited: 2002.
64. Czernik, S.; Bridgwater, A. V., Overview of Applications of Biomass Fast Pyrolysis Oil.
Energy Fuels 2004, 18 (2), 590-598.
65. Nowakowski, D. J.; Bridgwater, A. V.; Elliott, D. C.; Meier, D.; de Wild, P., Lignin fast
pyrolysis: Results from an international collaboration. J. Anal. Appl. Pyrolysis 2010, 88 (1), 5372.
66. Bridgwater, A. V., Review of fast pyrolysis of biomass and product upgrading. Biomass
Bioenergy 2012, 38 (0), 68-94.
67. Campbell, M. M.; Sederoff, R. R., Variation in lignin content and composition - Mechanism
of control and implications for the genetic improvement of plants. Plant Physiol. 1996, 110 (1),
3-13.
68. Mosier, N.; Wyman, C.; Dale, B.; Elander, R.; Lee, Y. Y.; Holtzapple, M.; Ladisch, M.,
Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresour.
Technol. 2005, 96 (6), 673-686.

161

69. Elliott, D. C., Historical Developments in Hydroprocessing Bio-oils. Energy Fuels 2007, 21
(3), 1792-1815.
70. Elliott, D. C.; Hart, T. R., Catalytic Hydroprocessing of Chemical Models for Bio-oil. Energy
Fuels 2008, 23, 631-637.
71. Singh, S. K.; Mukherjee, A. K.; Singh, M. M., Corrosion characteristics of mild steel in
aqueous solution of formic acid containing some acetic acid. Indian J. Chem. Technol. 2008, 15
(2), 174-179.
72. Dickerson, T.; Soria, J., Catalytic Fast Pyrolysis: A Review. energies 2013, (6), 25.
73. Driveclean Ethanol (E85) Flex Fuel.
http://www.driveclean.ca.gov/Search_and_Explore/Technologies_and_Fuel_Types/Ethanol_(E8
5)_Flex_Fuel.php (accessed 27 Mar 2014).
74. Zhao, C.; Kou, Y.; Lemonidou, A. A.; Li, X.; Lercher, J. A., Highly selective catalytic
conversion of phenolic bio-oil to alkanes. Angew. Chem. 2009, 121, 4047-4050.
75. Zhao, C.; Kou, Y.; Lemonidou, A. A.; Li, X.; Lercher, J. A., Hydrodeoxygenation of bioderived phenols to hydrocarbons using RANEY® Ni and Nafion/SiO2 catalysts. Chem. Commun.
2010, 46, 412-414.
76. Song, C., An overview of new approaches to deep desulfurization for ultra-clean gasoline,
diesel fuel and jet fuel. Catal. Today 2003, 86 (1–4), 211-263.
77. Babich, I. V.; Moulijn, J. A., Science and technology of novel processes for deep
desulfurization of oil refinery streams: a review. Fuel 2003, 82 (6), 607-631.
78. Brunet, S.; Mey, D.; Pérot, G.; Bouchy, C.; Diehl, F., On the hydrodesulfurization of FCC
gasoline: a review. Appl. Catal., A 2005, 278 (2), 143-172.
79. Grange, P., Catalytic Hydrodesulfurization. Catalysis Reviews 1980, 21 (1), 135-181.
80. Lee, J.; Kim, Y. T.; Huber, G. W., Aqueous-phase hydrogenation and hydrodeoxygenation of
biomass-derived oxygenates with bimetallic catalysts. Green Chem. 2014, 16 (2), 708-718.
81. Talmadge, M. S.; Baldwin, R. M.; Biddy, M. J.; McCormick, R. L.; Beckham, G. T.;
Ferguson, G. A.; Czernik, S.; Magrini-Bair, K. A.; Foust, T. D.; Metelski, P. D.; Hetrick, C.;
Nimlos, M. R., A perspective on oxygenated species in the refinery integration of pyrolysis oil.
Green Chem. 2014, 16 (2), 407-453.
82. Zacher, A. H.; Olarte, M. V.; Santosa, D. M.; Elliott, D. C.; Jones, S. B., A review and
perspective of recent bio-oil hydrotreating research. Green Chem. 2014, 16 (2), 491-515.

162

83. Mortensen, P. M.; Grunwaldt, J.-D.; Jensen, P. A.; Jensen, A. D., Screening of Catalysts for
Hydrodeoxygenation of Phenol as a Model Compound for Bio-oil. ACS Catalysis 2013, 3 (8),
1774-1785.
84. Mortensen, P. M.; Grunwaldt, J. D.; Jensen, P. A.; Knudsen, K. G.; Jensen, A. D., A review
of catalytic upgrading of bio-oil to engine fuels. Appl. Catal., A 2011, 407 (1–2), 1-19.
85. Pstrowska, K.; Walendziewski, J.; Luzny, R.; Stolarski, M., Hydroprocessing of rapeseed
pyrolysis bio-oil over NiMo/Al2O3 catalyst. Catal. Today 2014, 223, 54-65.
86. Saidi, M.; Samimi, F.; Karimipourfard, D.; Nimmanwudipong, T.; Gates, B. C.; Rahimpour,
M. R., Upgrading of lignin-derived bio-oils by catalytic hydrodeoxygenation. Energy Environ.
Sci. 2014, 7 (1), 103-129.
87. Ruddy, D. A.; Schaidle, J. A.; Ferrell, J. R.; Wang, J.; Moens, L.; Hensley, J. E., Recent
advances in heterogeneous catalysts for bio-oil upgrading via "ex situ catalytic fast pyrolysis":
catalyst development through the study of model compounds. Green Chem. 2014, 16 (2), 454490.
88. Popov, A.; Kondratieva, E.; Mariey, L.; Goupil, J. M.; El Fallah, J.; Gilson, J. P.; Travert, A.;
Mauge, F., Bio-oil hydrodeoxygenation: Adsorption of phenolic compounds on sulfided (Co)Mo
catalysts. J. Catal. 2013, 297, 176-186.
89. Wisniak, J.; Hershkowitz, M.; Stein, S., Ind. Eng. Chem. Prod. Res. Dev. 1974 13, 232-236.
90. Zhang, Z.; Jackson, J. E.; Miller, D., J. Appl. Catal. A 2001, 219, 89-98.
91. Dalavoy, T. S.; Jackson, J. E.; Swain, G. M.; Miller, D. J.; Li, J.; Lipkowski, J., Mild
electrocatalytic hydrogenation of lactic acid to lactaldehyde and propylene glycol. J. Catal. 2007,
246, 15-28.
92. Ilikti, H.; Rekik, N.; Thomalla, M., J. Appl. Electrochem. 2002, 32, 603-609.
93. Ilikti, H.; Rekik, N.; Thomalla, M., J. Appl. Electrochem. 2004, 34, 127-136.
94. Lund, H., A Century of Organic Electrochemistry. Journal of Electrochemical Society 2002,
149, 21-33.
95. White, H. L., Introduction to Industrial Chemistry. Wiley: 1986.
96. Vilar, M.; Oliveira, J. L.; Navarro, M., Appl. Catal., A. 2010, 372, 1-7.
97. Navarro, D. M. A. F. N.; Navarro, M., Catalytic Hydrogenation of Organic Compounds
without H2 Supply: An Electrochemical System. J. Chem. Educ. 2004, 81, 1350-1352.

163

98. Santana, D. S.; Melo, G. O.; Lima, M. V. F.; Daniel, J. R. R.; Areias, M. C. C.; Navarro, M.,
Electrocatalytic hydrogenation of organic compounds using a nickel sacrificial anode. J.
Electroanal. Chem. 2004, 569 (1), 71-78.
99. Ilikti, H.; Rekik, N.; Thomalla, M., Electrocatalytic hydrogenation of alkyl-substituted
phenols in aqueous solutions at a Raney nickel electrode in the presence of a non-micelleforming cationic surfactant J. Appl. Electrochem. 2004, 34, 127-136.
100. Ilikti, H.; Rekik, N.; Thomalla, M., Electrocatalytic hydrogenation of phenol in aqueous
solutions at a Raney nickel electrode in the presence of cationic surfactants. J. Appl. Electrochem.
2002, 32 (6), 603-609.
101. Anderson, M. T.; Martin, J. E.; Odinek, J. G.; Newcomer, P. P., Effect of Methanol
Concentration on CTAB Micellization and on the Formation of Surfactant-Templated Silica
(STS). Chem. Mater. 1998, 10 (6), 1490-1500.
102. Shi, Z.; Zhou, Y.-H.; Cha, C.-S., Influence of perfluorinated surfactants on the
electrochemical behaviour of a lead electrode in sulfuric acid solution. J. Power Sources 1998,
70 (2), 214-221.
103. Beraud, V.; Thomalla, M.; Lessard, J., The influence of a non-micelle-forming surfactant on
the electrocatalytic hydrogenation of carvone and limonene in aqueous medium at Raney nickel
electrodes. Can. J. Chem. 1997, 75 (11), 1529-1535.
104. Chambrion, P.; Roger, L.; Lessard, J.; Béraud, V.; Mailhot, J.; Thomalla, M., The influence
of surfactants on the electrocatalytic hydrogenation of organic compounds in micellar, emulsified,
and hydroorganic solutions at Raney nickel electrodes. Can. J. Chem. 1995, 73 (6), 804-815.
105. Cifuentes, A.; Bernal, J. L.; Diez-Masa, J. C., Determination of Critical Micelle
Concentration Values Using Capillary Electrophoresis Instrumentation. Anal. Chem. 1997, 69
(20), 4271-4274.
106. Swann, S.; Nelson, G. H., The Electrolytic Reduction of Acetophenone. Transactions of
The Electrochemical Society 1935, 67 (1), 201-207.
107. Laursen, A. B.; Varela, A. S.; Dionigi, F.; Fanchiu, H.; Miller, C.; Trinhammer, O. L.;
Rossmeisl, J.; Dahl, S., Electrochemical Hydrogen Evolution: Sabatier’s Principle and the
Volcano Plot. J. Chem. Educ. 2012, 89 (12), 1595-1599.
108. de Chialvo, M. R. G.; Chialvo, A. C., Hydrogen evolution reaction: Analysis of the VolmerHeyrovsky-Tafel mechanism with a generalized adsorption model. J. Electroanal. Chem. 1994,
372 (1–2), 209-223.
109. Schmickler, W.; Santos, E., Interfacial Electrochemistry. Springer: 2010.

164

110. Cleghorn, S. J. C.; Pletcher, D., Investigations of the electrocatalytic hydrogenation of
organic molecules at palladium on nickel cathodes. Electrochim. Acta 1993, 38 (18), 2683-2689.
111. Antonietta Casadei, M.; Pletcher, D., The influence of conditions on the electrocatalytic
hydrogenation of organic molecules. Electrochim. Acta 1988, 33 (1), 117-120.
112. Cleghorn, S. J. C.; Pletcher, D., The mechanism of electrocatalytic hydrogenation of
organic molecules at palladium black cathodes. Electrochim. Acta 1993, 38 (2–3), 425-430.
113. Polcaro, A. M.; Palmas, S.; Dernini, S., Electrochemical reduction of carbonyl compounds
at modified carbon felt electrodes. Electrochim. Acta 1993, 38 (2–3), 199-203.
114. Gultyai, V. P.; Strelets, V. V., Electrochemical hydrogenation of citral. Russian Chemical
Bulletin 1999, 48 (1), 202-204.
115. Schwientek, M.; Pleus, S.; Hamann, C. H., Enantioselective electrodes: stereoselective
electroreduction of 4-methylbenzophenone and acetophenone. J. Electroanal. Chem. 1999, 461
(1–2), 94-101.
116. Brennan, M. P. J.; Brown, O. R., Mechanism of the cathodic reduction of acetophenone in
acidic aqueous-methanolic media. Journal of the Chemical Society, Faraday Transactions 1:
Physical Chemistry in Condensed Phases 1973, 69 (0), 132-142.
117. Chapuzet, J. M.; Lasia, A.; Lessard, J., Electrocatalysis. Wiley-VCH: New York, 1998.
118. Couper, A. M.; Pletcher, D.; Walsh, F. C., Electrode materials for electrosynthesis.
Chemical Reviews 1990, 90 (5), 837-865.
119. Santana, D. S.; Lima, M. V. F.; Daniel, J. R. R.; Navarro, M., Electrocatalytic
hydrogenation of organic compounds using current density gradient and sacrificial anode of
nickel. Tetrahedron Lett. 2003, 44, 4725-4727.
120. Vilar, M.; Oliveira, J. L.; Navarro, M., Investigation of the hydrogenation reactivity of some
organic substrates using an electrocatalytic method. Appl. Catal., A 2010, 372 (1), 1-7.
121. Raney, M., Method of preparing catalytic material. Google Patents: 1925.
122. Raney, M., Method of producing finely-divided nickel. Google Patents: 1927.
123. Krupp, H.; Rabenhorst, H.; Sandstede, G.; Walter, G.; McJones, R., Raney Type Transition
Metals as Fuel Electrode Catalysts. J. Electrochem. Soc. 1962, 109 (7), 553-557.
124. Augustine, R. L., Heterogeneous Catalysis for the Synthetic Chemist. Marvel Dekker, Inc.:
New York, 1995.

165

125. Fouilloux, P., The nature of raney nickel, its adsorbed hydrogen and its catalytic activity for
hydrogenation reactions (review). Applied Catalysis 1983, 8 (1), 1-42.
126. Mozingo, R.; Spencer, C.; Folkers, K., Hydrogenation by Raney Nickel Catalyst without
Gaseous Hydrogen. J. Am. Chem. Soc. 1944, 66 (11), 1859-1860.
127. Besson, M.; Djaouadi, D.; Bonnier, J. M.; Hamar-Thibault, S.; Joucla, M., Structure and
Catalytic Properties in Hydrogenation of Valeronitrile of Raney Nickel Prepared from Cr and
Mo Doped Ni2Al3 Alloys. In Stud. Surf. Sci. Catal., M. Guisnet; J. Barrault; C. Bouchoule; D.
Duprez; G. Pérot R. Maurel; Montassier, C., Eds. Elsevier: 1991; Vol. Volume 59, pp 113-120.
128. Birry, L.; Lasia, A., Studies of the Hydrogen Evolution Reaction on Raney Nickel—
Molybdenum Electrodes. J. Appl. Electrochem. 2004, 34 (7), 735-749.
129. Robin, D.; Comtois, M.; Martel, A.; Lemieux, R.; Cheong, A. K.; Belot, G.; Lessard, J., The
electrocatalytic hydrogenation of fused poly cyclic aromatic compounds at Raney nickel
electrodes: the influence of catalyst activation and electrolysis conditions. Can. J. Chem. 1990,
68, 1218-1227.
130. Kokes, R. J.; Emmett, P. H., The Role of Hydrogen in Raney Nickel Catalysts. J. Am. Chem.
Soc. 1959, 81 (19), 5032-5037.
131. Filippov, D. V.; Ulitin, M. V.; Korosteleva, P. O., The thermodynamic characteristics of
processes in Raney nickel surface layers with the participation of adsorbed hydrogen forms. Russ.
J. Phys. Chem. 2008, 82 (5), 835-840.
132. Endoh, E.; Otouma, H.; Morimoto, T.; Oda, Y., New Raney nickel composite-coated
electrode for hydrogen evolution. Int. J. Hydrogen Energy 1987, 12 (7), 473-479.
133. Choquette, Y.; Brossard, L.; Lasia, A.; Menard, H., Study of the Kinetics of Hydrogen
Evolution Reaction on Raney Nickel Composite‐Coated Electrode by AC Impedance Technique.
J. Electrochem. Soc. 1990, 137 (6), 1723-1730.
134. Smith, H. A.; Chadwell, A. J.; Kirslis, S. S., The Role of Hydrogen in Raney Nickel
Catalyst. The Journal of Physical Chemistry 1955, 59 (9), 820-822.
135. Freel, J.; Robertson, S. D.; Anderson, R. B., The structure of Raney nickel: III. The
chemisorption of hydrogen and carbon monoxide. J. Catal. 1970, 18 (3), 243-248.
136. Renouprez, A. J.; Fouilloux, P.; Coudurier, G.; Tocchetti, D.; Stockmeyer, R., Different
species of hydrogen chemisorbed on Raney nickel studied by neutron inelastic spectroscopy.
Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed
Phases 1977, 73 (0), 1-10.
137. Konvalinka, J. A.; Van Oeffelt, P. H.; Scholten, J. J. F., Temperature programmed
desorption of hydrogen from nickel catalysts. Applied Catalysis 1981, 1 (3–4), 141-158.
166

138. Kokes, R. J.; Emmett, P. H., Adsorption Studies on Raney Nickel. J. Am. Chem. Soc. 1961,
83 (1), 29-31.
139. Tétényi, P.; Babernics, L., Investigations on the chemisorption of benzene on nickel and
platinum. J. Catal. 1967, 8 (3), 215-222.
140. Hauptmann, H.; Walter, W. F., The Action of Raney Nickel on Organic Sulfur Compounds.
Chemical Reviews 1962, 62 (5), 347-404.
141. Hochard, F.; Jobic, H.; Massardier, J.; Renouprez, A. J., Gas phase hydrogenation of
acetonitrile on Raney nickel catalysts: reactive hydrogen. J. Mol. Catal. A: Chem. 1995, 95 (2),
165-172.
142. Kleiderer, E. C.; Kornfeld, E. C., Raney Nickel as an Organic Oxidation-Reduction Catalyat.
J. Org. Chem. 1948, 13 (3), 455-458.
143. Bawane, S. P.; Sawant, S. B., Hydrogenation of p-nitrophenol to metol using Raney nickel
catalyst: Reaction kinetics. Appl. Catal., A 2005, 293 (0), 162-170.
144. Mozingo, R.; Wolf, D. E.; Harris, S. A.; Folkers, K., Hydrogenolysis of Sulfur Compounds
by Raney Nickel Catalyst. J. Am. Chem. Soc. 1943, 65 (6), 1013-1016.
145. Anantharaman, V.; Pintauro, P. N., The Electrocatalytic Hydrogenation of Glucose: I .
Kinetics of Hydrogen Evolution and Glucose Hydrogenation on Raney Nickel Powder. J.
Electrochem. Soc. 1994, 141 (10), 2729-2741.
146. Fouilloux, P.; Martin, G. A.; Renouprez, A. J.; Moraweck, B.; Imelik, B.; Prettre, M., A
study of the texture and structure of Raney nickel. J. Catal. 1972, 25 (2), 212-222.
147. Chang, N.-S.; Aldrett, S.; Holtzapple, M. T.; Davison, R. R., Kinetic studies of ketone
hydrogenation over Raney nickel catalyst. Chem. Eng. Sci. 2000, 55 (23), 5721-5732.
148. Lohrberg, K.; Kohl, P., Preparation and use of Raney-Ni activated cathodes for large scale
hydrogen production. Electrochim. Acta 1984, 29 (11), 1557-1561.
149. Hu, W., Electrocatalytic properties of new electrocatalysts for hydrogen evolution in
alkaline water electrolysis. Int. J. Hydrogen Energy 2000, 25 (2), 111-118.
150. Belot, G.; Desjardins, S.; Lessard, J., Electrocatalytic hydrogenation of organic compounds
on Devarda copper and Raney nickel electrodes in basic media. Tetrahedron Lett. 1984, 25 (47),
5347-5350.
151. Cyr, A.; Chiltz, F.; Jeanson, P.; Martel, A.; Brossard, L.; Lessard, J.; Ménard, H.,
Electrocatalytic hydrogenation of lignin models at Raney nickel and palladium-based electrodes.
Can. J. Chem. 2000, 78, 307-315.
167

152. Mahdavi, B.; Chambrion, P.; Binette, J.; Martel, E.; Lessard, J., Electrocatalytic
hydrogenation of conjugated enones on nickel boride, nickel, and Raney nickel electrodes. Can.
J. Chem. 1995, 73 (6), 846-852.
153. Mahdavi, B.; Lafrance, A.; Martel, A.; Lessard, J.; Ménard, H.; Brossard, L.,
Electrocatalytic hydrogenolysis of lignin model dimers at Raney nickel electrodes. J. Appl.
Electrochem. 1997, 27 (5), 605-611.
154. Martel, A.; Mahdavi, B.; Lessard, J.; Ménard, H.; Brossard, L., Electrocatalytic
hydrogenation of phenol on various electrode materials. Can. J. Chem. 1997, 75 (12), 1862-1867.
155. Dang, A.; Capuano, G. A.; Chapuzet, J. M.; Lessard, J., The coelectrodeposition of raney
nickel alloy powder on a stainless steel grid. Int. J. Hydrogen Energy 1993 18 (11), 941-944. .
156. Kanan, M. W.; Nocera, D. G., In Situ Formation of an Oxygen-Evolving Catalyst in Neutral
Water Containing Phosphate and Co2+. Science 2008, 321, 1072-1075.
157. Lutterman, D. A.; Surendranath, Y.; Nocera, D. G., A Self-Healing Oxygen-Evolving
Catalyst. J. Am. Chem. Soc. 2009, 131 (11), 3838-3839.
158. Kanan, M. W.; Surendranath, Y.; Nocera, D. G., Cobalt-phosphate oxygen-evolving
compound. Chem. Soc. Rev. 2009, 38 (1), 109-114.
159. Esswein, A. J.; Surendranath, Y.; Reece, S. Y.; Nocera, D. G., Highly active cobalt
phosphate and borate based oxygen evolving catalysts operating in neutral and natural waters.
Energy Environ. Sci. 2011, 4 (2), 499-504.
160. Reece, S. Y.; Hamel, J. A.; Sung, K.; Jarvi, T. D.; Esswein, A. J.; Pijpers, J. J. H.; Nocera,
D. G., Wireless Solar Water Splitting Using Silicon-Based Semiconductors and Earth-Abundant
Catalysts. Science 2011, 334 (6056), 645-648.
161. Yagi, M.; Kaneko, M., Molecular Catalysts for Water Oxidation. Chemical Reviews 2000,
101 (1), 21-36.
162. Rüttinger, W.; Dismukes, G. C., Synthetic Water-Oxidation Catalysts for Artificial
Photosynthetic Water Oxidation†. Chemical Reviews 1997, 97 (1), 1-24.
163. Braun, R. D., Introduction to chemical analysis. McGraw-Hill Ryerson, Limited: 1982.
164. Navarro, D. M. A. F. N.; Navarro, M., J. Chem. Educ. 2004, 81, 1350-1352.
165. Santana, D. S.; Lima, M. V. F.; Daniel, J. R. R.; Navarro, M., Tetrahedron Lett. 2003, 44,
4725-4727.

168

166. Menini, R.; Martel, A.; Ménard, H.; Lessard, J.; Vittori, O., Electrochim. Acta. 1998, 43,
1697-1703.
167. Mahdavi, B.; Los, P.; Lessard, M. J.; Lessard, J., Can. J. Chem. 1994, 72, 2268-2277.
168. Cheong, A. K.; Lasia, A.; Lessard, J., Electrochem. Soc. 1994, 141, 975-982.
169. Dang, A.; Capuano, G. A.; Chapuzet, J. M.; Lessard, J., Int. J. Hydrogen Energy 1993 18,
941-944. .
170. Robin, D.; Comtois, M.; Martel, A.; Lemieux, R.; Cheong, A. K.; Belot, G.; Lessard, J.,
Can. J. Chem. 1990, 68, 1218-1227.
171. Gang, W.; Liyin, B.; Rongchun, X., Electrochimica Acta. 2010, 55, 4030-4038. .
172. Cry, A.; Chiltz, F.; Jeanson, P.; Martel, A.; Brossard, L.; Lessard, J.; Ménard, H., Can. J.
Chem. 2000, 78, 307-315.
173. Pivovar, B. S., Polymer 2006, 47, 4194-4202.
174. Sergeev, A. G.; Hartwig, J. F., Selective, Nickel-Catalyzed Hydrogenolysis of Aryl Ethers.
Science 2011, 332 (6028), 439-443.
175. Kelley, P.; Lin, S.; Edouard, G.; Day, M. W.; Agapie, T., Nickel-Mediated Hydrogenolysis
of C–O Bonds of Aryl Ethers: What Is the Source of the Hydrogen? J. Am. Chem. Soc. 2012, 134
(12), 5480-5483.
176. Tobisu, M.; Chatani, N., Catalytic Hydrogenolysis of C-O Bonds in Aryl Ethers.
ChemCatChem 2011, 3 (9), 1410-1411.
177. Fedorov, A.; Toutov, A. A.; Swisher, N. A.; Grubbs, R. H., Lewis-base silane activation:
from reductive cleavage of aryl ethers to selective ortho-silylation. Chem. Sci. 2013, 4 (4), 16401645.
178. Delle Site, L.; Alavi, A.; Abrams, C. F., Adsorption energies and geometries of phenol on
the (111) surface of nickel: An ab initio study. Physical Review B 2003, 67 (19), 193406.
179. Kishi, K.; Chinomi, K.; Inoue, Y.; Ikeda, S., X-ray photoelectron spectroscopic study of the
adsorption of benzene, pyridine, aniline, and nitrobenzene on evaporated nickel and iron. J.
Catal. 1979, 60 (2), 228-240.
180. Aller, B. V., Raney cobalt hydrogenation catalysts. III. Applications and promoter effects.
Journal of Applied Chemistry 1958, 8 (8), 492-495.

169

181. Bao, L.; Xiong, R.; Wei, G., Electrochemical polymerization of phenol on 304 stainless
steel anodes and subsequent coating structure analysis. Electrochim. Acta 2010, 55 (12), 40304038.
182. Wheelhouse, R. T.; Stevens, M. F. G., DECOMPOSITION OF THE ANTITUMOR DRUG
TEMOZOLOMIDE IN DEUTERIATED PHOSPHATE BUFFER - METHYL-GROUP
TRANSFER IS ACCOMPANIED BY DEUTERIUM-EXCHANGE. J. Chem. Soc.-Chem.
Commun. 1993, (15), 1177-1178.
183. Cyr, A.; Huot, P.; Belot, G.; Lessard, J., The efficient electrochemical reduction of
nitrobenzene and azoxybenzene to aniline in neutral and basic aqueous methanolic solutions at
devarda copper and raney nickel electrodes: electrocatalytic hydrogenolysis of N=O and N=N
bonds. Electrochim. Acta 1990, 35 (1), 147-152.
184. Adkins, H.; Cramer, H. I.; Connor, R., THE RATE OF HYDROGENATION OF
ACETOACETIC ESTER, DEHYDROACETIC ACID, BENZENE, PHENOL AND ANILINE
OVER NICKEL AT PRESSURES FROM 27 TO 350 ATMOSPHERES. J. Am. Chem. Soc.
1931, 53 (4), 1402-1405.
185. Ogumi, Z.; Nishio, K.; Yoshizawa, S., Denki-Kagaku 1981, (49).
186. Ogumi, Z.; Yamashita, H.; Nishio, K.; Takehara, Z.-I.; Yoshizawa, S., Application of the
SPE method to organic electrochemistry—III. Kolbe type reactions on Pt-SPE. Electrochim.
Acta 1983, 28 (11), 1687-1693.
187. Ogumi, Z.; Nishio, K.; Yoshizawa, S., Application of the SPE Method to Organic
Electrochemistry - II. Electrochemical Hydrogenation of Olefinic Double Bonds. Electrochim.
Acta. 1981, 26, 1779-1782.
188. Ogumi, Z.; Ohashi, S.; Takehara, Z., Application of the SPE method to organic
electrochemistry—VI. Oxidation of cyclohexanol to cyclohexanone on Pt-SPE in the presence of
iodine and iodide. Electrochim. Acta 1985, 30 (1), 121-124.
189. Ogumi, Z.; Inaba, M.; Ohashi, S.-i.; Uchida, M.; Takehara, Z.-i., Application of the SPE
method to organic electrochemistry—VII. The reduction of nitrobenzene on a modified Pt-nafion.
Electrochim. Acta 1988, 33 (3), 365-369.
190. Ogumi, Z.; Inatomi, K.; Hinatsu, J. T.; Takehara, Z.-i., Application of the SPE method to
organic electrochemistry—XIII. Oxidation of geraniol on Mn,Pt-Nafion. Electrochim. Acta 1992,
37 (7), 1295-1299.
191. Ogumi, Z.; Yasuzawa, M.; Kunugi, A.; Inaba, M., Electroreduction of acetophenone on PtNafion composite electrodes. Denki Kagaku 1994, 62 (12), 5.
192. Takenaka, H.; Torikai, E., Kokai Tokkyo Koho (Japan Patent) 1980, 55, 38934.

170

193. Hicks, M. T.; Fedkiw, P. S., Kolbe Electrolysis of Acetic Acid in a Polymer Electrolyte
Membrane Reactor. J. Electrochem. Soc. 1998, 145 (11), 3728-3734.
194. Heyl, A.; Jörissen, J., Electrochemical detoxification of waste water without additives using
solid polymer electrolyte (SPE) technology. J. Appl. Electrochem. 2006, 36 (11), 1281-1290.
195. Machida, M.; Shono, E.; Kimura, M.; Yamauchi, S., Electrochemical lean-deNOx catalyst
using H+-conducting solid polymer electrolyte. Catal. Commun. 2003, 4 (12), 631-635.
196. Wang, K.; Yan, H.; Liu, J.; Sun, X.; Wang, E. f., The Electrochemical Behavior and the
Quantitative Determination of Nitrogen Dioxide at a Solid Polymer Electrolyte Supported Pt-Au
Electrode. Electroanalysis 2004, 16 (16), 1318-1323.
197. Hasnat, M. A.; Alam, M. S.; Karim, M. H. M.-u.; Rashed, M. A.; Machida, M., Divergent
catalytic behaviors of Pt and Pd films in the cathode of a sandwiched type membrane reactor.
Applied Catalysis B: Environmental 2011, 107 (3–4), 294-301.
198. Hasnat, M. A.; Agui, R.; Hinokuma, S.; Yamaguchi, T.; Machida, M., Different reaction
routes in electrocatalytic nitrate/nitrite reduction using an H+-conducting solid polymer
electrolyte. Catal. Commun. 2009, 10 (7), 1132-1135.
199. Hasnat, M. A.; Ishibashi, I.; Sato, K.; Agui, R.; Yamaguchi, T.; Ikeue, K.; Machida, M.,
Electrocatalytic Reduction of Nitrate Using Cu-Pd and Cu-Pt Cathodes/H+-Conducting Solid
Polymer Electrolyte Membrane Assemblies. Bull. Chem. Soc. Jpn. 2008, 81 (12), 1675-1680.
200. Takenaka, H.; Torikai, E.; Kawami, Y.; Wakabayashi, N., Solid polymer electrolyte water
electrolysis. Int. J. Hydrogen Energy 1982, 7 (5), 397-403.
201. Choi, P.; Bessarabov, D. G.; Datta, R., A simple model for solid polymer electrolyte (SPE)
water electrolysis. Solid State Ionics 2004, 175 (1–4), 535-539.
202. Nuttall, L. J., Conceptual design of large scale water electrolysis plant using solid polymer
electrolyte technology. Int. J. Hydrogen Energy 1977, 2 (4), 395-403.
203. Jörissen, J., Electro-organic synthesis without supporting electrolyte: Possibilities of solid
polymer electrolyte technology. J. Appl. Electrochem. 2003, 33 (10), 969-977.
204. Jörissen, J., Ion exchange membranes as solid polymer electrolytes (spe) in electro-organic
syntheses without supporting electrolytes. Electrochim. Acta 1996, 41 (4), 553-562.
205. He, P.; Watts, P.; Marken, F.; Haswell, S. J., Electrolyte free electro-organic synthesis: The
cathodic dimerisation of 4-nitrobenzylbromide in a micro-gap flow cell. Electrochem. Commun.
2005, 7 (9), 918-924.

171

206. Inaba, M.; Ogumi, Z.; Takehara, Z. i., Application of the Solid Polymer Electrolyte Method
to Organic Electrochemistry: XVII . Indirect Electrochemical Debromination Using Viologens as
Microscopic Phase‐Transfer Mediators. J. Electrochem. Soc. 1994, 141 (10), 2579-2586.
207. Horcajada, R.; Okajima, M.; Suga, S.; Yoshida, J.-i., Microflow electroorganic synthesis
without supporting electrolyte. Chemical Communications 2005, (10), 1303-1305.
208. Kunugi, A.; Fujioka, M.; Yasuzawa, M.; Inaba, M.; Ogumi, Z., Electroreduction of 2cyclohexen-1-one on metal–solid polymer electrolyte composite electrodes. Electrochim. Acta
1998, 44 (4), 653-657.
209. An, W.; Hong, J. K.; Pintauro, P. N., Current efficiency for soybean oil hydrogenation in a
solid polymer electrolyte reactor. J. Appl. Electrochem. 1998, 28 (9), 947-954.
210. An, W.; Hong, J.; Pintauro, P.; Warner, K.; Neff, W., The electrochemical hydrogenation of
edible oils in a solid polymer electrolyte reactor. I. Reactor design and operation. J. Am. Oil
Chem. Soc. 1998, 75 (8), 917-925.
211. Warner, K.; Neff, W. E.; List, G. R.; Pintauro, P., Electrochemical hydrogenation of edible
oils in a solid polymer electrolyte reactor. Sensory and compositional characteristics of low
Trans soybean oils. J. Am. Oil Chem. Soc. 2000, 77 (10), 1113-1118.
212. An, W.; Hong, J.; Pintauro, P. N.; Warner, K.; Neff, W., The electrochemical hydrogenation
of edible oils in a solid polymer electrolyte reactor. II. Hydrogenation selectivity studies. J. Am.
Oil Chem. Soc. 1999, 76 (2), 215-222.
213. Green, S. K.; Tompsett, G. A.; Kim, H. J.; Kim, W. B.; Huber, G. W., Electrocatalytic
Reduction of Acetone in a Proton-Exchange-Membrane Reactor: A Model Reaction for the
Electrocatalytic Reduction of Biomass. ChemSusChem 2012, 5 (12), 2410-2420.
214. Green, S. K.; Lee, J.; Kim, H. J.; Tompsett, G. A.; Kim, W. B.; Huber, G. W., The
electrocatalytic hydrogenation of furanic compounds in a continuous electrocatalytic membrane
reactor. Green Chem. 2013, 15 (7), 1869-1879.
215. Qiu, Y.; Xin, L.; Chadderdon, D. J.; Qi, J.; Liang, C.; Li, W., Integrated electrocatalytic
processing of levulinic acid and formic acid to produce biofuel intermediate valeric acid. Green
Chem. 2014, 16 (3), 1305-1315.
216. Yi-Lin, C.; Tse-Chuan, C., Metals and alloys bonded on solid polymer electrolyte for
electrochemical reduction of pure benzaldehyde without liquid supporting electrolyte. J.
Electroanal. Chem. 1993, 360 (1–2), 247-259.
217. Li, Z.; Garedew, M.; Lam, C. H.; Jackson, J. E.; Miller, D. J.; Saffron, C. M., Mild
electrocatalytic hydrogenation and hydrodeoxygenation of bio-oil derived phenolic compounds
using ruthenium supported on activated carbon cloth. Green Chem. 2012, 14 (9), 2540-2549.

172

218. Sompalli, B.; Gasteiger, H.; Mathias, M. F.; Scozzafava, M., Methods of preparing
membrane electrode assemblies. Google Patents: 2003.
219. Fırtına, İ.; Güner, S.; Albostan, A., Preparation and characterization of membrane electrode
assembly (MEA) for PEMFC. International Journal of Energy Research 2011, 35 (2), 146-152.
220. Dudley, J. R., Furfural condensation products. Google Patents: 1950.
221. Li, Z. L.; Kelkar, S.; Lam, C. H.; Luczek, K.; Jackson, J. E.; Miller, D. J.; Saffron, C. M.,
Aqueous electrocatalytic hydrogenation of furfural using a sacrificial anode. Electrochim. Acta
2012, 64, 87-93.
222. Housecroft, C. E.; Sharpe, A. G., Inorganic Chemistry. Pearson Prentice Hall: 2005.
223. Smith, M. B.; March, J., March's Advanced Organic Chemistry: Reactions, Mechanisms,
and Structure. Wiley: 2007.
224. Lab Manual in Biochemistry: Immunology and Biotechnology. Tata McGraw-Hill Pub.:
2007.
225. The United States Mint About Us.
http://www.usmint.gov/about_the_mint/index.cfm?action=coin_specifications (accessed
02/01/2011).
226. Gruen, D. M.; Mendelsohn, M. H.; Dwight, A. E., Lanthanum nickel aluminum alloy.
Google Patents: 1979.
227. Gay, R. N.; Raymond, W. K., Process for electroplating nickel over stainless steel. Google
Patents: 1988.
228. Mroczkowski, R. S., Electronic Connector Handbook: Theory and Applications. McGrawHill: 1998.
229. Mallory, G. O.; Hajdu, J. B.; Electroplaters, A.; Society, S. F., Electroless Plating:
Fundamentals and Applications. The Society: 1990.
230. Menini, R.; Martel, A.; Ménard, H.; Lessard, J.; Vittori, O., The electrocatalytic
hydrogenation of phenanthrene at Raney nickel electrodes: the influence of an inert gas pressure.
Electrochim. Acta. 1998, 43 (12-13), 1697-1703.

173