DESIGN AND ENGINEERING OF BIO-BASED COATING FOR USE ON PAPER AND
OTHER SUBSTRATES
By
Kyle Earl Thompson

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Chemical Engineering – Master of Science
2013

ABSTRACT
DESIGN AND ENGINEERING OF BIO-BASED COATING FOR USE ON PAPER AND
OTHER SUBSTRATES
By
Kyle Earl Thompson
An economic infrastructure based on crude oil coupled with dwindling oil reserves has
led to extensive research in recent years into bio-based products, minimizing the dependency on
petroleum-based products.

Soybean oil is a renewable, bio-based material that is readily

available domestic crop. Expanding the products they’re used for could improve the local
economy of rural communities and also enhance the domestic industrial base when used in
value-added industrial products. Utilizing the basic unsaturated triglyceride structure of soybean
oil and the water-resistant characteristics it demonstrates provides a suitable material for
hydrophobic coatings when used in conjunction with siloxane chemistry.
Reacting low-saturated soybean oil and vinyl silanes via a high temperature, high
pressure alkyl-substituted reaction results in a silylated soy oil product. When this material is
exposed to moisture, a chain of Si-O-Si siloxane bonds are formed between the silylated soy oil
creating a viscous liquid suitable for coating on various substrates of which provides enhanced
hydrophobic properties. For this project, silylated soy oil was evaluated for use on untreated
paper for the production of bio-based, biodegradable, waterproof trash bags. Specifically, a
portion of the project involved scale-up trials for consideration as a manufactured product.
Analysis of the synthesized silylated soy product, assessment of the water resistance
provided by the silylated soy oil coatings, and the results and estimated economic model from
scale-up trials comprise this thesis.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................................ vii
LIST OF FIGURES ....................................................................................................................... ix
Chapter 1: Introduction to Project Scope .................................................................................. 1
1.1 Thesis Organization .................................................................................................................. 1
1.2 Demand for Bio-Based & Biodegradable Products .................................................................. 2
1.3 Objectives ................................................................................................................................. 4
1.4 Background ............................................................................................................................... 5
1.4.1 Soy Oil................................................................................................................................ 5
1.4.2 Silane and Silicone Chemistry ........................................................................................... 8
REFERENCES ............................................................................................................................. 11
Chapter 2: Synthesis and Analysis of Silylated Soy Oil .......................................................... 13
2.1 Alder-Ene Reaction................................................................................................................. 13
2.2 Reaction Conditions ................................................................................................................ 14
2.2.1 Catalysts and Enophiles ................................................................................................... 14
2.2.2 Reaction Conditions & Reactor Selection........................................................................ 15
2.2.3 Reactor Size...................................................................................................................... 17
2.2.4 Reactants & Reactant Ratios ............................................................................................ 18
2.3 Experimental ........................................................................................................................... 20
2.3.1 Materials ........................................................................................................................... 20
2.3.2 Method – Silylated Soy Oil .............................................................................................. 20
2.4 Product Analysis ..................................................................................................................... 21
2.4.1 TGA & Graft Percentage.................................................................................................. 22
2.4.2 Viscosity ........................................................................................................................... 23
2.4.3 Volatile Organic Compound Measurements .................................................................... 24
2.4.4 NMR ................................................................................................................................. 25
2.5 Results & Discussion – 2 Liter Parr Reactor .......................................................................... 25
2.5.1 TGA .................................................................................................................................. 25
2.5.2 Graft Percentages ............................................................................................................. 28
2.5.3 Viscosity ........................................................................................................................... 30
iii

2.5.4 Volatile Organic Compound Measurements .................................................................... 32
2.6 Results & Discussion – 5 Gallon Parr Reactor ....................................................................... 33
2.6.1 TGA & Graft Percentages ................................................................................................ 33
2.6.2 Viscosity ........................................................................................................................... 37
2.7 Results & Discussion – Varying Silane Reactants and Reactant Ratios ................................ 39
2.7.1 Mixture B – 1:1 Molar Ratio Soy Oil:VTMOS ............................................................... 39
2.7.2 Mixture C – 1:3 Molar Ratio Soy Oil:VTEOS ................................................................ 40
2.7.3 Mixture D – 1:1 Molar Ratio Soy Oil:VTEOS ................................................................ 40
REFERENCES .......................................................................................................................... 43
Chapter 3: Silylated Soy Oil Coatings ...................................................................................... 45
3.1 Coating Background & Cure Chemistry ................................................................................. 45
3.2 Silicone Coatings & Cure Chemistry...................................................................................... 46
3.3 Coating Variables.................................................................................................................... 48
3.3.1 Coating Weight ................................................................................................................ 48
3.3.2 Cure Temperature & Time ............................................................................................... 48
3.3.3 Catalyst and Moisture Concentrations ............................................................................. 49
3.3.4 Paper Types ...................................................................................................................... 50
3.3.5 Additional Factors ............................................................................................................ 51
3.3.5.1 Pot Life Time ............................................................................................................. 51
3.3.5.2 Blend (Mixing) Time ................................................................................................. 52
3.3.5.3 Tack-Free Time ......................................................................................................... 52
3.4 Materials ................................................................................................................................. 53
3.5 Product Analysis ..................................................................................................................... 54
3.5.1 Cobb Test ......................................................................................................................... 54
3.5.2 Water Contact Angle ........................................................................................................ 55
3.6 Experimental ........................................................................................................................... 57
3.6.1 Varying Paper Samples with Low Basis Weight ............................................................. 57
3.6.1.1 Method – Varying Paper Samples with Low Basis Weight ...................................... 57
3.6.1.2 Results & Discussion – Varying Paper Samples with Low Basis Weight ................ 58
3.6.2 – Coat Weight Variations ................................................................................................. 61
3.6.2.1 Method – Coat Weight Variations ............................................................................. 61
3.6.2.2 Results & Discussion – Coat Weight Variations ....................................................... 61
3.6.3 Cure Time & Temperature ............................................................................................... 63
3.6.3.1 Method – 4hr Cure Time with Cure Temps at 60°C and Room Temp ..................... 63
iv

3.6.3.2 Results & Discussion – 4hr Cure Time with Cure Temps at 60°C and Room Temp 64
3.6.4 – Reduced Cure Times at 60°C and Room Temp ............................................................ 66
3.6.4.1 Method – Reduced Cure Times at 60°C and Room Temp ........................................ 66
3.6.4.2 Results & Discussion – Reduced Cure Times at 60°C and Room Temp .................. 66
3.6.5 – 24hr Cure Time, 60°C vs. 80°C Cure Temp ................................................................. 68
3.6.5.1 Method – 24hr Cure Time, 60°C vs. 80°C Cure Temp ............................................. 68
3.6.5.2 Results & Discussion – 24hr Cure Time, 60°C vs. 80°C Cure Temp ....................... 69
3.6.6 – Varying Paper Samples with High Basis Weight.......................................................... 69
3.6.6.1 Method – Varying Paper Samples with High Basis Weight ..................................... 69
3.6.6.2 Results & Discussion – Varying Paper Samples with High Basis Weight ............... 71
3.7 Optimal Coating Conditions ................................................................................................... 72
REFERENCES .......................................................................................................................... 75
Chapter 4: Scale-Up Trials and Economics.............................................................................. 76
4.1 Scale-Up Trials ....................................................................................................................... 76
4.1.1 Gravure Coating System .................................................................................................. 77
4.1.2 Line Speed ........................................................................................................................ 77
4.1.3 Oven Temperature ............................................................................................................ 78
4.1.4 Curing Formulation .......................................................................................................... 78
4.1.4.1 Method – Varying Water Concentration for Pre-Coat Mixture ................................ 78
4.1.4.2 Results & Discussion – Varying Water Concentration for Pre-Coat Mixture .......... 79
4.2 Scale-Up Trial Results and Discussion ................................................................................... 81
4.2.1 Grease Resistance of Scale-Up Trial Coatings ................................................................ 82
4.2.1.1 Method – Grease Resistance Measurements of Scale-Up Trial Coatings ................. 82
4.2.1.2 Results & Discussion – Grease Resistance Measurements of Scale-Up Trial Coatings
............................................................................................................................................... 82
4.3 Scale-Up Trial Economic Evaluation ..................................................................................... 83
REFERENCES .......................................................................................................................... 89
Chapter 5: Recommended Future Work .................................................................................. 90
5.1 Proposed Synthesis Experiments ............................................................................................ 90
5.1.1 Variation of Reactant Molar Ratios ................................................................................. 90
5.1.2 Variation of Reaction Time and Temperature.................................................................. 91
5.2 Proposed Coating Experiments ............................................................................................... 92
5.2.1 DBTDL & Water Concentrations .................................................................................... 92
v

5.2.2 Use of Paper GN354872 with High Basis Weight ........................................................... 93
5.2.3 Use of Secondary Coating with Silylated Soy Oil ........................................................... 94
5.3 Proposed Manufacturing Experiments .................................................................................... 94
5.3.1 Different Coating System ................................................................................................. 94
REFERENCES .......................................................................................................................... 98

vi

LIST OF TABLES

Table 2.1 Alder-Ene reactant ratios used in 2 L Parr Reactor for synthesis of silylated soy oil. 21
Table 2.2 TGA results of calibration standards of unreacted low-sat soy oil:VTMOS mixtures 26
Table 2.3 Graft percentages of silylated soy oil products run in 2 L Parr reactor at 280°C for 16
hours. Graft percentages were calculated using Eq. 2.1. ............................................................. 29
Table 2.4 Summary of graft percentage and associated viscosity from Reactions 1-9............... 31
Table 2.5 VOC results of convection oven and TGA heating methods. Values indicate weight
percent loss for both silylated soy oil and pre-coating mixtures containing 1wt% water and
DBTDL. ........................................................................................................................................ 33
Table 2.6 Graft percentages of silylated soy oil products run in 5 gal. Parr reactor at 250°C, 20
hours calculated using Eq. 2.1. ..................................................................................................... 34
Table 2.7 Summary of graft percentage and associated viscosity from Reactions 10-19. .......... 38
Table 3.1 Comparison of uncoated paper characteristics provided by Wausau Paper Co. ......... 58
Table 3.2 Cobb Values at both two and ten minute intervals of coated and uncoated papers. ... 60
Table 3.3 Water contact angles of various paper types of both coated and uncoated samples. .. 60
Table 3.4 Cobb Values of four hour cure time at combinations of 60°C and room temperature.
....................................................................................................................................................... 63
Table 3.5 Coat weights and Cobb Values of samples cured at various time intervals ranging
from 15 minutes to 4 hours. Each sample was cured for equal amounts of time at both 20°C
(room temp) and 60°C. ................................................................................................................. 68
Table 3.6 Comparison of sample sets cured at 60°C and 80°C for one hour of 24 hour cure time.
....................................................................................................................................................... 69
Table 3.7 Uncoated high basis weight paper characteristics provided by Wausau Paper Co. .... 70
Table 3.8 Two minute Cobb Values of coated and uncoated high basis weight papers. ............ 71
Table 4.1 Coat weights and Cobb Values of coatings prepared at different mixing times using
0.1wt% water in pre-coating mixture............................................................................................ 80
Table 4.2 Equipment settings and Cobb Values of coatings from first two trials........................ 81

vii

Table 4.3 Coating and bag production specifications including coating settings from full-scale
trials, estimated production volume, bag sizes, and paper requirements. ..................................... 83
Table 4.4 Estimated paper costs including purchasing and transportation to coating facilities
based off of production demands highlighted in Table 4.3. ......................................................... 84
Table 4.5 Estimated raw material, processing, and production costs associated with the synthesis
of silylated soy oil based off of production demands outlined in Table 4.3. ................................ 85
Table 4.6 Estimated raw material costs for curing step during coating process. Quantities were
based off of production demands outlined in Table 4.3................................................................ 86
Table 4.7 Estimated coating production costs based off of production demands in Table 4.3. ... 86
Table 4.8 Estimated manufacturing, synthesis, and processing costs for production of one
million silylated soy oil trash bags................................................................................................ 87

viii

LIST OF FIGURES

Figure 1.1 The basic chemical structure of a soy oil triglyceride molecule. The R-groups can be
any combination of several fatty acid chains. ................................................................................. 6
Figure 1.2 Chemical structure of the five most common fatty acids found in soy oil. .................. 7
Figure 1.3 Chemical structures of VTMOS and VTEOS used in the synthesis of silylated soy oil.
......................................................................................................................................................... 9
Figure 2.1 Reaction mechanism of a basic Alder-Ene reaction. ................................................. 13
Figure 2.2 Mechanism showing grafted VTMOS onto unsaturated fatty acid chains of soy oil
via an Alder-Ene reaction. ............................................................................................................ 14
Figure 2.3 Effects of reaction time and temperature on the reactivity of VTMOS and soy oil [8].
....................................................................................................................................................... 16
Figure 2.4 Relationship between bulk viscosity and graft percentages of silylated soy oil
achieved via Alder-Ene reactions carried out in 2 L Parr reactor at 280°C for 16 hours. ............ 19
Figure 2.5 TGA results of low-sat soy oil and VTMOS blanks measuring wt% loss and dwt% as
a function of temperature. Dashed lines are VTMOS results and solid lines are soy oil results. 27
Figure 2.6 TGA calibration curve of unreacted low-sat soy oil:VTMOS mixtures. .................. 28
Figure 2.7 Overlay of TGA plots from Reactions 1-9 using 5°C/min ramp up to 450°C under
nitrogen gas. .................................................................................................................................. 30
Figure 2.8 Overlay of TGA plots from Reactions 10-19 using 5°C/min ramp up to 450°C under
nitrogen gas. .................................................................................................................................. 35
Figure 2.9 Final product of Reaction 11 from partially cured silylated soy oil. The solid
material on the stirrer is fully cured silylated soy oil. A small amount of viscous liquid silylated
soy oil was collected, characterized, and analyzed but was not used for further coating tests and
applications. .................................................................................................................................. 36
Figure 2.10 Viscosity measurements of 5 gal. reactor products with respect to graft percentage.
....................................................................................................................................................... 38
Figure 3.1 Polymer Condensation reaction resulting in a cured silicone polymer with OR
representing siloxane substituents (e.g. methoxy or ethoxy groups in VTMOS and VTEOS,
respectively) and R substituents representing vinyl groups which graft to the unsaturated double
bonds of soy oil during an Alder-Ene reaction. ............................................................................ 47
ix

Figure 3.2 Single Roll Euclid Lab Coater used for all coatings made throughout duration of
research. Pneumatic pressure gauge allowed for precise coat weights. The design and small size
of the coater required the use of minimal coating material. ......................................................... 54
Figure 3.3. Cobb cylinder, blotting paper, and 10-kg roller used for measuring Cobb Values. . 55
Figure 3.4 Surface contact angle represented as θC with respect to interfacial energies [8]....... 56
Figure 3.5 Variation in coat weights of silylated soy oil when varying blade pressure on ECS. 62
Figure 3.6 Cobb Values when varying the coat weight of silylated soy oil on GN505320 paper.
....................................................................................................................................................... 63
Figure 4.1 Viscosity measurements with respect to mixing time using 1.0wt% and 0.1wt%
water for pre-coating mixture formulations. ................................................................................. 79

x

Chapter 1: Introduction to Project Scope
1.1 Thesis Organization
This thesis is organized into five separate and distinguishable chapters. Each chapter
provides further insight into the different segments of the project that were addressed during the
course of the research. Finally, the focal points from each chapter are summarized to present the
current and future status of the project.
Chapter One discusses the desire for bio-based and biodegradable goods in today’s
market and the benefits they have over current petroleum-based materials. This chapter also
discusses the purpose of the project and provides necessary details and background needed to
understand the chemistry and engineering aspects of the project.
Chapter Two describes the process for synthesizing silylated soy oil.

The reaction

chemistry and settings are introduced including several reaction variables that were examined.
The product analysis methods used during the production of silylated soy oil are highlighted
along with appropriate interpretations of each experimental result.
Chapter Three describes the paper coating process on a small, lab bench scale. The
coating method used is explained and other areas including pot life time, tack-free time, cure
time, cure temperature, and mixing time are also described. The results and methods of analysis
for each coating variation are presented and summarized.
Chapter Four details the progress made in the large-scale trials that have occurred during
this project. This includes common issues experienced during scale-up projects and potential
problems specific to this project. Additionally, variables that are not a factor in small scale paper
1

coating processes such as coating speed, coating method, and drying time and temperature are
introduced and discussed.
Chapter Five summarizes the purpose of the project while emphasizing its importance
and practical uses. Significant data and results found throughout the course of the research
project are organized and presented. Also, the status of the project will be stated as well as
potential future work that could further the development of the current product.

1.2 Demand for Bio-Based & Biodegradable Products
In recent years there has been an increasing demand for bio-based products in today’s
marketplace. Rising oil prices coupled with diminishing oil reserves have resulted in increased
investigation of alternatives to fuel and other petroleum-based products. Similarly, regulations
on corporations’ carbon footprints and disposal methods of petroleum-based plastics have
become stricter over the past two decades, resulting in a demand for more eco-friendly products.
The diminishing of the world’s oil reserves can be summarized in a simple supply and
demand model. Formation of fossil fuels occurs over millions of years and everyday products
that are made from these fossil fuels, such as plastic bottles, can take several hundreds of years to
decompose in the environment [1]. This replenishment of fossil fuels is a process that cannot be
sped up; it is fixed and thus accounts for a fixed available supply.
The use of biobased materials to replace fossil resources offers an intrinsic value
proposition and serves to reduce the product carbon footprint. One of the principle values of
using biobased materials arises from the use of short, sustainable and in balance biological
carbon where the rate and time scales of sequestration into plant biomass as CO2 is in balance
with its use as organic materials. The use of fossil carbon, on the other hand, is sequestered over
2

millions of years but it is released back into the atmosphere in a much shorter time frame.
Consequently the use of fossil resources is unsustainable and leads to a net greenhouse gas
accumulation in the atmosphere with all the negative aspects associated with this phenomenon.
World crude oil consumption in 2011 was estimated at over 87 million barrels per day
[2]. This results in a worldwide crude oil consumption of more than 30 billion barrels per year.
Additionally, crude oil consumption has increased 38% over the last thirty years. Current
estimates of the top 17 countries with proven oil reserves indicate a total of 1,234 billion barrels
remaining [3]. Therefore assuming crude oil consumption remains at 2011 levels, these reserves
will be exhausted in approximately 40 years. This trend of increased consumption of fossil fuels
only seems to be growing while the oil supply remains fixed. As a result, the demand for crude
oil products will far outweigh the supply in the very near future.
Based on these statistics, replacing petroleum-based products with similar products
derived from renewable resources, such as plant-based products, has become a highly researched
area. Several new uses are being discovered for the different parts of cash crops, specifically
soybean plants including applications in cosmetics, cleaning materials, particle board, adhesives,
and pharmaceuticals [4].

Since soybean plants are a major domestic crop, expanding the

products they’re used for can improve the local economy of rural communities and also enhance
the domestic industrial base when used in value-added industrial products. Using a bio-based
product, such as soy oil, to replace existing petroleum-based products, while maintaining similar
product characteristics and functionality, could help diminish the dependency on fossil fuels. In
addition to the fact that soy is readily biodegradable in the environment [5], it provides a
renewable resource that could replace petroleum-based products, thus alleviating the supply and
demand issues that arise with increased petroleum consumption with limited petroleum reserves.
3

Using functional groups commonly associated with soy oil as reaction sites, it is possible
to synthesize several new biomaterials that replicate common everyday products that are not biobased. More specifically, several natural oils are excellent coating materials, and when exposed
to air, form a tough scratch-free film [6]. The synthesis, chemistry, and characterization of
modified soy oil for use as coating material will be discussed in this thesis. Additionally coating
applications, coating formulations, as well as product development for bio-based, biodegradable
non-plastic bags will also be discussed.

1.3 Objectives
The focus of this research project is to create a bio-based alternative to current petroleumbased plastic bags commonly used for waste disposal. Creating a bio-based and biodegradable
coating that could be applied to untreated brown paper used for trash bags was the focus of this
project. Using soy oil as the primary ingredient while utilizing commonly practiced silane cure
chemistry used in several different areas of coating industries, it is possible to form short
siloxane chains by hydrolysis of silicon alkoxy bonds (Si-O-C) upon exposure to atmospheric
moisture and the formation of stable siloxane (Si-O-Si) networks [7].
Under high temperature and pressure, the vinyl groups in the silane molecules react with
the unsaturated fatty acid chains of soy oil’s triglycerides using a peroxide catalyst. The product
of this reaction is a modified soy oil grafted with the silane molecule at the site of an
unsaturation in the fatty acid chain. Subjecting this product to moisture leads to hydrolysis of
alkoxy silanes and subsequent condensation of silanols, creating a crosslinked polymer chain
consisting of Si-O-Si bonds. This condensation reaction further increases the molecular weight
of the product and creates a viscous liquid suitable for paper coating.

4

Several variables were considered and investigated throughout this project.
(1) Determine optimal reaction conditions for producing the desired modified soy oil. Such
factors that were examined include variations of the reaction time, temperature, reactant
ratios, and type of reactants.

The products from each of these variations were

characterized and analyzed for their chemical and physical properties.
(2) Analyze the coating performance of each modified soy oil product. The appropriate
conditions that meet the required coating weights, pot life time, tack-free time, mixing
time, and cure time were determined. These factors are of critical importance when
operating at a large manufacturing scale paper coating facility.
(3) Define process conditions of industrial standards. The most favorable coating speed,
coating thickness, and curing (or drying) temperature were determined in a pilot plant
facility. In these trials, the choice of desired paper grade and weight used for making
final product was evaluated. The desired result of this project is to have a product that
can be mass-manufactured.

While each of the mentioned variables affects product

performance, it is also important to consider product variations from a fiscal perspective
as well.

1.4 Background
Before describing the details of this project, it is essential to first understand the
fundamental properties of the materials and instruments that were used throughout the research
process.
1.4.1 Soy Oil
Soy oil is extracted from the seed, or the bean, of a soybean plant. Soybeans are
comprised of approximately 19% oil with the remainder being proteins, carbohydrates, and
5

moisture. To isolate the soy oil, the beans are sliced into flakes and immerged in hexane which
extracts the oil from the bean. After removing the remaining bean solids, the hexane and soy oil
are separated using a distillation or evaporation process.
The basic chemical structure of soy oil is a mixture of triglycerides that are composed of
glycerol attached to three fatty acids as shown in Figure 1.1. A fatty acid is a molecule with a
carboxylic acid and an aliphatic tail. These aliphatic tails can contain double bonds (unsaturated)
or be free of double bonds (saturated). Typically, soy oil contains 4.3-4.6 double bonds on the
fatty acid chains per mole depending on the origin of the seed [8]. The fatty acid molecules
found in soy oil are linoleic acid [51%], oleic acid [23%], palmitic acid [10%], α-linolenic acid
[7%], and stearic acid [4%] (see Figure 1.2). These five fatty acids comprise a total of 95% of
the fatty acid chains commonly found in soy oil. For this project, the double bonds of the
unsaturated fatty acids were targeted as the reaction sites to achieve the desired silylated soy oil
product.

O
H2C

O

C

R

O
HC

O

C

R

O
H2C

O

C

R

Figure 1.1 The basic chemical structure of a soy oil triglyceride molecule. The R-groups can
be any combination of several fatty acid chains.

O

HO
6

Linoleic Acid (51%)
O

HO

Oleic Acid (23%)
O

HO

Palmitic Acid (10%)

O

HO
α-Linolenic Acid (7%)
O

HO

Stearic Acid (4%)
Figure 1.2 Chemical structure of the five most common fatty acids found in soy oil.

Soy oil is of great interest in this project due to its chemical properties along with its low
cost and high availability. For a demonstration of oil’s hydrophobic properties, one must simply
pour a cup of soy oil into a glass of water resulting in two distinct immiscible layers of water and
oil. This observed bilayer is a result of water’s polarity. Polarity is an uneven arrangement of
charges within a molecule. It is well known that polar molecules are only able to dissolve in
polar solvents. Because soy oil is a non-polar molecule it does not dissolve when mixed with
water. Therefore soy oil’s water-repellent characteristics could, in the right circumstances, be
7

used in waterproof applications. However, if unmodified soy oil was used directly as a coating it
would not demonstrate useful properties due to its low viscosity and inability to crosslink, or
form chemical bonds with adjacent soy oil molecules. Using soy oil as a substrate in well
established silane chemistry techniques, though, allows its hydrophobic properties to be of much
greater use for coating purposes.
1.4.2 Silane and Silicone Chemistry
Silicon is a chemical element that is found in the same column of the periodic table as
Carbon and as such, it is a tetravalent metalloid. Due to its four valence electrons, silicon is able
to achieve many different chemical bonding formations. Silane, for instance, utilizes silicon’s
four valence electrons to bond to four hydrogen molecules much like carbon would in a methane
molecule.
However, in many ways silicon behaves very differently from carbon. For example, a SiO bond is longer and more flexible than a C-O bond due to the larger atomic size of silicon
compared to carbon. An Si-O bond also has a lower barrier to rotation than a C-O bond. These
differences are some of the factors that contribute to the flexible structures and low glass
transition temperatures often displayed by several silicone materials [9]. Because of these
unique physical properties, silicone materials derived from silicon are often utilized in polymer
science. Silicones are organosilicon polymers and are most commonly used in sealant, adhesive,
and lubricant chemistry. One unique structure commonly used in silicone chemistry is siloxane
which exhibits repeating Si-O-Si bonds that form the backbone of the molecule.
The repeating Si-O-Si backbone has an associated bond energy of ~450 kJ/mol compared
to ~360 kJ/mol for a repeating C-O-C bond (ether). Bond energy is the measure of energy
required to break a molecule into its separate components. The significance of such a difference
8

between the two bonds energies is the ability for silicones to better resist hemolytic bond
cleavage compared to ether compounds.

The larger bond energy of a silicone backbone

translates to a more thermally-stable molecule that is more resistant to temperature extremes.
Larger bond energy, along with the longer Si-O bonds and lower rotational barrier compared to
C-O bonds contribute to lower glass-transition temperatures (temperature at which a material
changes from a viscous liquid to a fixed crystalline solid).

H3CO

Si

OCH3

H3CH2CO

Si

OCH2CH3

OCH3

OCH2CH3

Vinyltrimethoxysilane (VTMOS)

Vinyltriethoxysilane (VTEOS)

Figure 1.3 Chemical structures of VTMOS and VTEOS used in the synthesis of silylated soy oil.

For the purpose of this project, vinyltrimethoxysilane (VTMOS) and vinyltriethoxysilane
(VTEOS) were used in the synthesis of silylated soy oil (Figure 1.3). These compounds were
selected for two main reasons. First, these vinyl compounds do not unergo thermal or radical
polymerization due to the presence of the silicon atom that stabilizes free radicals. Second, the
alkoxy groups attached to the silicon atom are readily hydrolyzed in the presence of moisture and
the resulting silanols condense to yield siloxane linkages. This double bond, when reacted with
low-saturated soy oil in the presence of a peroxide catalyst at high temperatures, is grafted onto

9

the fatty acid in the soy oil at the site of the double bonds on the fatty acid chains. This reaction
occurs via an Alder-Ene mechanism.

10

REFERENCES

11

REFERENCES

[1] U.S. National Park Service; Mote Marine Lab, Sarasota, FL. U.S. National Park Service.
[Online] September 2012.
http://des.nh.gov/organization/divisions/water/wmb/coastal/trash/documents/marine_debris.pdf.
[2] U.S. Department of Energy. U.S. Energyl Information Administration. [Online] September
2012.
http://www.eia.gov/cfapps/ipdbproject/iedindex3.cfm?tid=5&pid=5&aid=2&cid=ww,&syid=200
7&eyid=2011&unit=TBPD.
[3] U.S. Energy Information Administration. U.S. Energy Information Administration. [Online]
September 2012. http://www.eia.gov/countries/
[4] Pedersen, Palle. Iowa State University. Soybean Extension and Research Program. [Online]
September 2012. http://extension.agron.iastate.edu/soybean/uses_soyproducts.html.
[5] Tran, P.; Graiver, D.; Narayan, R., Biocomposites synthesized from chemically modified soy
oil and biofibers. J Appl Polym Sci 2006, 102 (1), 69-75
[6] Kumar, A.; Vemula, P. K.; Ajayan, P. M.; John, G., Silver-nanoparticle-embedded
antimicrobial paints based on vegetable oil. Nat Mater 2008, 7 (3), 236-241
[7] Jones, D. Dow Corning Corp. [Online] 1997.
http://www.dowcorning.com/content/publishedlit/25-877.pdf.
[8] Tran, P.; Seybold, K.; Graiver, D.; Narayan, R., Free radical maleation of soybean oil via a
single-step process. J Am Oil Chem Soc 2005, 82 (3), 189-194
[9] Dow Corning Corporation. Discovery Center: Dow Corning Corp. Dow Corning Corp.
[Online] October 2012. http://www.dowcorning.com/content/discover/discoverchem/si-vsorganic.aspx.

12

Chapter 2: Synthesis and Analysis of Silylated Soy Oil
2.1 Alder-Ene Reaction
An Alder-Ene reaction involves an alkene hydrocarbon with an allylic hydrogen and a
compound with a multibond (alkene or alkyne) acting as an enophile, a molecule able to donate
an extra pair of electrons from its double bond. This reaction results in a substituted alkene with
the double bond shifted to the allylic position [1]. This reaction is classified as a percyclic
reaction, a reaction in which the intermediate is a cyclic ring, and can be further described as a
rearrangement reaction due to the shift of the double bond on the alkene hydrocarbon (Figure
2.1).

H

H

Figure 2.1 Reaction mechanism of a basic Alder-Ene reaction.
Specifically, an Alder-Ene reaction involving unsaturated soy oil and VTMOS allows for
VTMOS to act as an enophile and graft onto a double bond of an unsaturated fatty acid chain of
soy oil’s triglyceride structure (Figure 2.2). Also, because vinyl silanes do not polymerize via
free radicals like other organic vinyl monomers (i.e. styrene or acrylates) due to the stabilization
of the silicon atom, VTMOS and VTEOS are ideal compounds for reactions carried out at high
temperatures.

This stabilizing affect inhibits the free radicals formed on the double bond

adjacent to the silanes to not result in homopolymers.

As a result, this simple addition

modification incorporates silicone properties into soy oil that it would otherwise not
demonstrate. Incorporating widely used silicone coating technology into a hydrophobic media,
13

such as soy oil, provides a unique bio-based material than can be readily used in several different
water-resistant coating applications.
O

O

R

R
HO

HO
H

Si
H3CO

Si

OCH3

H3CO

OCH3

OCH3
OCH3

Figure 2.2 Mechanism showing grafted VTMOS onto unsaturated fatty acid chains of soy oil
via an Alder-Ene reaction.
2.2 Reaction Conditions
2.2.1 Catalysts and Enophiles
The Alder-Ene reaction has become a widely used application in organic synthesis
because of its ability to control the absolute regio- and stereochemistry during the carbon-carbon
bond formation [2]. However, due to the reaction’s pericyclic group transfer and high activation
energies, typically high temperatures and highly activated substrates are required [3]. Often
times, Lewis acids are used as catalysts since they are able to form complexes with the electrondeficient enophiles. Selection of the appropriate Lewis acid strongly depends on the enophile
being used. For instance, studies have shown that acrylate and propriolate esters require very
strong Lewis acids [4-6], whereas reactions with more reactive aldehydes need only a mild
Lewis acid, such as iron (III) chloride [7].
Previous work [8-9] involving the synthesis of silylated soy oil used 2,5-Bis(tertbutylperoxy)-2,5-dimethylhexane, Luperox 101 (L101), as a catalyst for the synthesis of silylated
14

soy oil. This peroxide is often used as a free radical polymerization initiator, but has also been
used for the degradation of polypropylene [10]. Other reports of hydrosilylation of fatty acid
methyl-esters (FAMEs) and triacylglycerols (TAGs) have been reported in the presence of
platinum catalysts and radical initiators [8, 11-13]. The most commonly utilized hydrosilanes
from these reports are triethoxy- and trichlorosilanes. Of these reports, chlorosilanes in the
presence of a mixture of peroxides or platinum-based catalysts are the only reactants that have
been successfully used in which non-terminal double bonds were targeted as the reaction site.
Achieving a reaction mechanism that can utilize internal double bonds as a reaction site for
silane-additions becomes very valuable for the purpose of this project because the most abundant
fatty acid chains found in soy oil do not have terminal double bonds. If non-terminal double
bonds were not reactive via an Alder-Ene reaction, the yield of the desired silylated soy oil
product would be too low to be of any practical use. Only one process has been found claiming
to use vinylsilanes and unsaturated naturally-occurring vegetable oils as a method for
synthesizing silylated oils onto non-terminal double bonds using a high temperature addition
reaction [14].
2.2.2 Reaction Conditions & Reactor Selection
The high temperatures required to overcome the activation energy of an Alder-Ene
reaction presents a number of conditions that must be taken into consideration when selecting the
appropriate vessel for a reactor. First, it is critical that the reactor be able to maintain an inert
atmosphere throughout the entirety of the reaction to prevent any undesired side reactions as a
result of curing that may occur from moisture in the air. At a lab bench scale, the most practical
setup is a closed batch reactor equipped with inlet and outlet gas valves used for purging the
headspace prior to reaction commencement. Second, a closed batch reactor being used at high
temperatures will be subject to very high pressures, typically in excess of 100 psi. Pressures this
15

high remove the possibility of using a glass-lined reactor. Therefore a small stainless steel
reactor that can accommodate high pressure/high temperature reactions is best suited for an
Alder-Ene reaction.

Figure 2.3 Kinetics of reaction time and temperature on reactivity of VTMOS and soy oil [8].

Parr Instrument Company has several models of reactors that cater to the specifications
required of a typical Alder-Ene reaction. Initially a 2 L stainless steel Parr reactor was used for
synthesizing silylated soy oil for early trials. The reactor was easily able to withstand reaction
temperatures of 280°C and pressures between 100-150 psig. Additionally, the reactor was
equipped with gas inlet and outlet valves used for purging the headspace of air that may
16

otherwise result in undesired side reactions. Another useful feature of the selected reactor was
the sampling port that allowed samples to be collected in mid-reaction for kinetics data and
analysis.
Determining the optimal reaction temperature and time to achieve a silylated soy oil
product was investigated in previous studies [8-9]. Initial work showed that the conversion of
silane increases as the temperature of the reaction is increased. Likewise, longer reaction times
resulted in higher conversion of silanes, though conversation rates slow considerably after the
first two hours of the reaction (Figure 2.3)
Further research of this same reaction used a reaction temperature of 280°C to expedite
the reaction process. Once at this temperature, a reaction time of 12 hours was used resulting in
graft percentages between 75-80% [9]. For this project, a reaction temperature of 280°C was
used, however it was found that increasing the reaction time from 12 hours to 16 hours increased
the grafting percentage of VTMOS which also resulted in a desired increase in viscosity of the
final product.
2.2.3 Reactor Size
The primary goal of this research was to develop a silylated soy oil material suitable for
coating at a large industrial-scale output. Because scaling up a process includes incorporating
several other unforeseen factors and variables into the coating process, it was necessary to
perform test trials at an industrial-scale facility.

Coating paper at a larger scale required

significantly more coating material which necessitated a larger reactor.
While a 2 L Parr reactor was used for small-scale trials, a 5 gal. Parr reactor of similar
design was used for producing silylated soy oil for large-scale test trials. Unlike the 2 L reactor,
17

the larger reactor exhibited a decreased maximum heat capacity of the heating mantle.
Consequently, a reaction temperature of 250°C was used for all reactions performed in the 5 gal.
reactor. To adjust for the decreased temperature, the reaction time was extended to achieve
similar product properties performed in the 2 L reactor. Ultimately, all reactions carried out in
the 5 gal. reactor were run at 250°C for 20 hours.
2.2.4 Reactants & Reactant Ratios
Another factor investigated in previous studies was the reactant ratio of soy oil to
VTMOS. Several different molar ratios were evaluated ranging from 1:15 to 1:3 moles of soy
oil:moles of VTMOS. The results suggested a more balanced reactant ratio yielded higher graft
percentages. This is advantageous for two reasons. Firstly, higher graft percentages result in
increased viscosities (Figure 2.4) and therefore provide better properties for surface coatings.
Secondly, from a manufacturing perspective, because VTMOS is the cost-determining reactant
in this Alder-Ene reaction reducing the amount used per volume of product provides a more
economically competitive product.
Two different reaction ratios were investigated and compared for performance properties.
A 1:3 molar ratio soy oil:VTMOS (900 g-mol soy oil:445 g-mol VTMOS) was used for paper
coating test trials and product development. Reactions were also carried out using a 1:1 molar
ratio of soy oil:VTMOS (900 g-mol soy oil:148 g-mol VTMOS) and evaluated with respect to
the 1:3 molar reactions.
In addition to varying molar ratios of soy oil and silane used during the Alder-Ene
reactions, the selected silane enophile was also varied to determine the impacts on cure and
coating properties. Stricter environmental regulations on coating industries have made it highly

18

desirable to incorporate bio-renewable materials into current technologies [15]. Along with
incorporating bio-based, bio-renewable materials into the coating industry, it is also of great

600

Silylated Soy Oil Viscosity (cP)

500

400

300

200

100

0
75

80

85
90
95
VTMOS Graft Percentage (%)

100

Figure 2.4 Relationship between bulk viscosity and graft percentages of silylated soy oil
achieved via Alder-Ene reactions carried out in 2 L Parr reactor at 280°C for 16 hours.

importance to limit the amount of hazardous materials that are used. One area highly regulated
is volatile organic compound (VOCs) emissions, commonly associated with coating and paint
industries. Section 12 of the Clean Air Act provides a full list of Hazardous Air Pollutants
(HAPs) that are regulated in all U.S. manufacturing facilities. Among this list is methanol [16], a
byproduct of the proposed curing mechanism in silylated soy oil coatings when using VTMOS as
the silane reactant. Ethanol, however, is not on the HAP list. Therefore, using VTEOS as the
19

silane enophile in the Alder-Ene reaction, thus emitting ethanol instead of methanol during the
curing process, provides a way or producing a more environmentally-friendly silylated soy oil.
If products from both VTEOS and VTMOS reactions result in similar coating performance
properties, the implementation of VTEOS in place of VTMOS in the Alder-Ene reaction would
provide a more environmentally friendly product and safer working conditions.
Ultimately, the most desired silylated soy oil would minimize the ratio of soy:silane, thus
requiring less amounts of the price-determining silane material. Furthermore, VTEOS would be
the preferred silane reactant to VTMOS due to the Clean Air Act’s HAP list. However, it is
necessary for VTEOS silylated soy oil to demonstrate analogous hydrophobic properties to that
of VTMOS silylated soy oil.

2.3 Experimental
2.3.1 Materials
Low-saturated soybean oil was provided by Zeeland Farm Services, Inc., Zeeland, MI.
Vinyltrimethoxysilane (VTMOS) [CAS 2768-02-7] and vinyltriethoxysilane (VTEOS) [CAS 7808-0] were purchased from Gelest, Inc., Morrisville, PA. Luperox 101 (L101) [CAS 78-63-7],
2,5-Bis(tert-butylperoxy)-2,5-dimethylhexane (technical grade, 90%) was purchased from
Sigma-Aldrich Co., St. Louis, MO. All reagents were used as received without any additional
purification.
2.3.2 Method – Silylated Soy Oil
As mentioned in the “Reactant & Reactant Ratios” section (2.2.4), two molar ratios of
soy oil:silane were tested, 1:1 and 1:3 moles. Additionally, both VTMOS and VTEOS were
tested as enophiles in the Alder-Ene reaction.

20

Table 2.1 Alder-Ene reactant ratios used in 2 L Parr Reactor for synthesis of silylated soy oil
Mixture
A
B
C
D

Soy Oil:Silane (molar ratio)
1:3
1:1
1:3
1:1

Silane
VTMOS
VTMOS
VTEOS
VTEOS

Soy Oil (g)
900
900
900
900

Silane (g) L101 (g)
445
9
148
9
571
9
190
9

For each reaction, adequate amounts of low-sat soy oil and silane were weighed in a 2 L
Erlenmeyer flask (Table 2.1). L101 was weighed at 1.0 weight percent of soy oil and added to
the Erlenmeyer flask containing the soy oil-silane mixture. The contents of the flask were
quickly swirled to ensure a homogeneous mixture and were poured into the Parr reactor. The
Parr reactor was closed and, with both the gas inlet and outlet valves open, the reactor was
purged for approximately two minutes with nitrogen gas. After purging, both gas valves were
immediately closed and contents were mixed at approximately 60 rpm throughout the entirety of
the heating, reaction, and cooling process.
The reaction was heated at a high rate until it reached 200°C. Once the reactor reached
200°C, the heating rate was slowed until the final reaction temperature of 280°C was achieved.
The reaction time began once the reaction temperature of 280°C was achieved and was allowed
to continue for 16 hours. In some cases, intermediate samples were taken via the sampling port
both during the heating period and throughout the course of the reaction to evaluate the reaction
kinetics. All intermediate samples were collected in small 10 mL scintillation vials and were
capped immediately to prevent any unreacted contents from escaping.

2.4 Product Analysis
Several analytical techniques were used to evaluate the results of the reactions, the
products that were obtained, and the physical properties of the final products. The following is a
21

list of each technique used, a brief description what each technique is used for, and how the
results are applicable for the analysis of synthesized silylated soy oil.
2.4.1 TGA & Graft Percentage
Thermogravimetric analysis (TGA) is an instrument that measures the change in weight
of a sample by varying the temperature in a controlled environment. Increasing the temperature
to a point where one of the components of the sample evaporates or thermally degrades results in
a weight loss. The temperature and magnitude of weight loss that occurs can provide insight into
the composition of the material being tested. For the purpose of this study, TGA was used to
determine if a reaction had occurred between low-sat soy oil and the silane compound used in the
reaction.
For example, if known quantities of unreacted soy oil and VTMOS were mixed and
subjected to TGA using a simple ramp heating method of 10°C/min, it can be expected to see all
of the VTMOS completely evaporate by its boiling temperature of 123°C. Likewise, if soy oil
and VTMOS undergo an Alder-Ene reaction, it should be expected that only the unreacted
VTMOS would evaporate by 123°C. If the mass of each reactant used in the reaction is known,
the amount of silane that reacts, or grafts, onto the unsaturated double bonds of soy oil can be
easily calculated (Eq. 2.1 for VTMOS and Eq. 2.2 for VTEOS). As shown in Figure 2.4, once
the graft percentage of the silylated soy oil is known, one can estimate other relevant properties

Graft % =

related to its coating potential.

(Eq. 2.1)

wt%VTMOS −wt% loss@123°C
wt%VTMOS
22

∗ 100%

(Eq. 2.2)

Graft % =

wt%VTEOS −wt% loss@160°C
wt%VTEOS

∗ 100%

TGA is also very useful in illustrating the reproducibility of a reaction by comparing the
change in weight percentage (wt% loss) with respect to temperature change. Additionally,
plotting derivative weight (dwt%) loss with respect to temperature provides another method of
further insuring a reaction’s reproducibility. This trend highlights the greatest rates of change
(inflection points) on the wt% loss curve. Comparing plots using the same temperature profile
should provide identical overlaying trends if the same product was achieved during the reaction.
Therefore, running multiple products from the same reaction on TGA using the same heating
method should result in superimposed wt% and dwt% trend lines if the reaction is reproducible.
2.4.2 Viscosity
The viscosity of a coating material can be a critical factor when considering a scale-up
process. The viscosity limits of a coating material can vary dramatically from small bench scale
coaters to large industrial scale coating processes. Viscosities that are too low absorb, or bleed,
into the coating substrate resulting in a product that doesn’t demonstrate the desired coating
properties. Alternatively, high viscosities can prove problematic for coating instruments and
machinery.
Using a small benchtop coater it’s possible to make uniform coatings using extremely
viscous liquids. At small scales, factors such as pot life and instrument operation limits are not
of any significant concern. Conversely, most large scale coating equipment has very defined
viscosity ranges that they’re able to operate at. Surpassing these limits could result in equipment
malfunction that could shut down an entire plant or coating facility for hours or days at a time.
Therefore it is important to determine the lower and upper viscosities of the coating material that
23

the coating instruments will be subject to during the coating and curing process. Additionally,
comparing viscosities amongst products provides another way of verifying the reproducibility of
a reaction.
2.4.3 Volatile Organic Compound Measurements
Volatile organic compounds (VOCs) are organic chemicals that exhibit a high vapor
pressure at room temperature. The displayed high vapor pressure is a result of a low boiling
point which makes these particular substances easier to evaporate into the air. While many
VOCs pose little health threat, there are numerous anthropogenic VOCs that are highly regulated
under law by the Clean Air Act as Hazardous Air Pollutants (HAPs).
VOC content (ASTM D2369) is measured by weighing 0.2 gram samples into a preweighed aluminum dish and placing the samples into a vented oven set to 110°C ± 5°C for one
hour. Samples are weighed before and after heating and the weight change is recorded. Using
this change in weight, the VOC percentage of silylated soy oil can be found using Eq. 2.3. All
materials still present in the sample after heating are considered non-volatile and are deemed
“solids”.
The maximum allowable VOC levels vary depending on the organic compound and the
product category they’re being used for (i.e. contact adhesives, general purpose cleaners,
insecticides, etc.). Several products in the coating industry are water or solvent-based resulting
in elevated VOC levels. Because soy oil has a very high flash point however (approximately
320°C), and is considered a solid for VOC measurements, silylated soy oil demonstrates much

VOC % = 100 − ��

� ∗ 100�

lower VOC levels than other commonly used solvent-based coatings.

(Eq. 2.3)

24

W2 −W1
SA

Where: W2 = weight of dish and sample after heating
W1 = weight of dish
SA = weight of sample before heating

2.4.4 NMR
Nuclear magnetic resonance (NMR) is a very common and powerful tool used to
determine various physical and chemical properties of an atom or molecule. This analytical
technique exploits the magnetic properties given off by all atoms and molecules which, in turn,
make it possible to determine the chemical structure of a molecule. While any number of
chemicals can be tested using NMR, it is most commonly utilized for organic molecules.
For this project, NMR was used to highlight the discrepancies between the molecular
structures of unreacted low-sat soy oil and silylated soy oil. Successfully grafting silane onto
unsaturated fatty acid chains results in unique chemical bonding properties that are easily
recognizable and detectable in an NMR spectrum. Performing NMR spectroscopy on samples
before and after Alder-Ene reactions provides a method of verification as to the extent of the
Alder-Ene reaction.

2.5 Results & Discussion – 2 Liter Parr Reactor
2.5.1 TGA
In order to properly evaluate results obtained from TGA it’s necessary to run blanks to
generate a baseline for all other analyses. Figure 8 shows two overlaid TGA results of pure lowsat soy oil and pure VTMOS. For both tests, a ramp to 150°C at a rate of 10°C/minute was used.
As expected, soy oil exhibited no weight loss (solid line in Fig. 2.5) when the temperature

25

reached 150°C due to its high flashpoint. In contrast, VTMOS experienced a 100% weight loss
prior to its boiling point of 123°C (dashed line in Fig. 2.5).
Control samples of unreacted soy oil:VTMOS at different ratios were also evaluated
using TGA. Mixtures of unreacted soy oil:VTMOS (25:75, 50:50, and 75:25 respectively) were
evaluated using a ramp method up to 450°C at a rate of 5°C/minute. This heating method was
chosen to ensure a more precise wt% loss plot. The mixture was heated to 450°C to ensure that
weight loss due to low-sat soy oil was also shown. Results from Figure 2.5 suggest that a weight
percentage loss at 123°C would be seen totaling the total amount of VTMOS in each respective
mixture.
Table 2.2 TGA results of calibration standards of unreacted low-sat soy oil:VTMOS mixtures
wtVTMOS (g)
0.2505
0.5013
0.7499

wtsoy oil (g)
0.7541
0.5020
0.2507

wt%VTMOS
24.94
49.97
74.95

wt%loss @ 123°C
21.07
44.67
71.06

∆wt%actual-observed
3.87
5.30
3.89

The results of the wt% calibration line of 25%, 50%, and 75% VTMOS-containing
mixtures are very accurate compared to total weight percentage of VTMOS used in each mixture
(Table 2.2). However, none of the mixtures’ observed weight percentages account for the total
weight percent of VTMOS used in the analyses (Figure 2.6). This is due to the volatility of
VTMOS at room temperature. To highlight this effect, an isothermal TGA was run at room
temperature with pure VTMOS for a total of 40 seconds (time between loading sample and
commencement of TGA analysis). Adding the 40 second wt% loss of VTMOS at room temp to
each calibration point resulted in actual wt% losses within 1.0 wt% of the theoretical weight loss
percentages.

26

120

80

Weight Percentage (wt%)

60
80
50
40

60
Soy Oil wt%

30

40
VTMOS wt%
20

Soy Oil dwt%
20

VTMOS dwt%

Derivative Weight Percentage (dwt%)

70
100

10
0

0
0

20

40

60
80
100
Temperature (°C)

120

140

160

Figure 2.5 TGA results of low-sat soy oil and VTMOS blanks measuring wt% loss and dwt% as
a function of temperature. For interpretation of the references to color in this and all other
figures, the reader is referred to the electronic version of this thesis.
Ultimately, using a TGA calibration curve and performing an isothermal TGA at room
temperature on pure VTMOS provided an accurate, acceptable calibration line. Therefore, it was
determined that analysis of silylated soy oil samples could be accurately performed using
5°C/min ramp on TGA. As a result, graft percentages from each Alder-Ene reaction could be
accurately calculated using this same method.

27

0.01% (Pure Soy Oil)
100

Weight Percent (wt%)

21.07% (25% VTMOS)
80

44.67% (50% VTMOS)

60

40
71.06% (75% VTMOS)
20
99.63% (Pure VTMOS)
0
20

40

60

80
100
Temperature (°C)

120

140

Figure 2.6 TGA calibration curve of unreacted low-sat soy oil:VTMOS mixtures.
2.5.2 Graft Percentages
For initial research, Alder-Ene reactions were performed in a 2 L Parr reactor with a
reaction temperature of 280°C and a reaction time of 16 hours. Graft percentages from reaction
using Mixture A reactant ratios (Table 2.1) are shown in Figure 2.7 and Table 2.3. Graft
percentages were calculated using Eq. 2.1.
For manufacturing purposes, repeatability is of the highest priority. Therefore it’s critical
to be able to control and accurately predict the silane graft percentage by varying the reaction
time or temperature.

Similarly, it is important that each silylated soy product fall within

predetermined limits for each analytical method used for product evaluation. This ensures that
the each reaction product meets certain acceptable criteria thus exemplifying identical properties
28

from previous reactions. As seen in Table 2.3, the average graft percentage for nine reactions
run under the same conditions was found to be 90.3% ± 2.5%. Because the calculated graft% of
reactions 6, 8, and 9 exceeded one standard deviation unit (2.5%) they were not used for any
coating studies or evaluation.
Table 2.3 Graft percentages of silylated soy oil products run in 2 L Parr reactor at 280°C for 16
hours. Graft percentages were calculated using Eq. 2.1.
Reaction wt%VTMOS wt%loss123°C graft%
1
2
3
4
5
6
7
8
9
Average
Std. Dev.

33
33.6
33.5
33.5
33.6
33.5
32.9
33.3
32.8

3.059
2.827
3.138
3.179
2.809
2.004
2.877
4.863
4.143

90.7
91.6
90.6
90.5
91.6
94
91.3
85.4
87.4
90.3
2.5

The graft percentage of each reaction also provides insight into the viscosity of the
silylated soy product (Fig. 2.4). Higher viscosity silylated soy products result in shorter pot-life
times. Pot life time is the time in which a catalyzed thermosetting resin retains sufficiently low
viscosity for processing [17]. For initial lab work, pot life is not of major concern because the
coating instrument used is not restricted by the viscosity of the coating material. In industrial
coating processes however, pot life plays a major factor in determining the feasibility of using a
certain coating matrix. A pot life under three hours is generally considered too short for use and
implementation at large-scale industrial applications.

As a result, achieving precise and

consistent silane graft percentages, like shown in Table 2.3, allows for accurate predictions of

29

pot life during the coating process making the determined graft percentage calculated for each
batch incredibly useful.

15.0

100
90
80
Weight Percentage (wt%)

11.0
70
9.0

60

7.0

50
40

5.0

30
3.0
20

Derivative Weight Percentage (dwt%)

13.0

1.0

10
0

-1.0
0

100

200
300
Temperature (°C)

400

500

Figure 2.7 Overlay of TGA plots from Reactions 1-9 using 5°C/min ramp up to 450°C under
nitrogen gas.
2.5.3 Viscosity
Each product was evaluated for relative viscosity using a Brookfield DV-E viscometer
with an LV spindle set at a rate of 30 rpm. Results from Reactions 1-9 are presented in Table
2.4. All viscosities were measured at room temp (22°C) and were taken within 24 hours after
removing products from the reactor following reaction completion.
30

Table 2.4 Summary of graft percentage and associated viscosity from Reactions 1-9.
Reaction Graft %
Viscosity (cP)
1
90.7
307
2
91.6
296
3
90.6
162
4
90.5
193
5
91.6
530
6
94.0
308
7
91.3
465
8
85.4
106
9
87.4
116
Avg.
90.3
275.9
Std. Dev.
2.5
148.7

While reactions 6, 8, and 9 were not used for coating purposes they did provide valuable
insight regarding the relationship between graft percentage and viscosity. Reactions 8 and 9
exhibited slightly lower graft percentages but drastically lower viscosities. Similarly, Reactions
2, 5, 6, and 7 showed graft percentages slightly above average, but had drastically increased
viscosities. This data suggests that even slight variations of silane graft percentage will have
drastic impacts on the resulting viscosity.
Excluding Reactions 6, 8, and 9 from the viscosity statistical evaluation, the standard
deviation becomes 146.1 cP despite reducing the number of data points by 33%. This large
standard deviation is most likely due to partial hydrolysis, or crosslinking, of some products (i.e.
Reactions 5 and 7) during the reaction. Partially hydrolyzed samples would drastically affect the
measured viscosity of silylated soy oil.

31

2.5.4 Volatile Organic Compound Measurements
VOCs were measured using two separate techniques. The first method was performed in
a digitally controlled mechanical convection oven. The second method was done using a TGA
heating method. Samples of uncoated, uncured silylated soy oil were tested using both heating
methods.

All tests were done in triplicate and followed ASTM D2369.

For comparison

purposes, mixtures of silylated soy oil, water, and dibutyltin dilaurate (DBTDL) were also
analyzed to determine the impact gelation may have on VOC levels. Gelation is the process of a
polymer cross-linking until which it exhibits no flow when in steady-state. This is another way
of describing a polymer’s curing behavior and properties [18]. Silylated soy oil, water, and
DBTDL mixtures were prepared by adding 1.0 wt% each (based on weight of silylated soy oil)
of water and DBTDL. All VOC calculations were performed using Eq. 2.3.
The convection heating method resulted in lower VOCs than the TGA method. This was
due to smaller sample weights used in the TGA method. Because TGA is only capable of
evaluating sample sizes well below 0.20 g, the sample surface areas using TGA would be higher
and therefore easier to evaporate at elevated temperatures. Despite, this difference in sample size
between the two heating methods, the results indicate VOCs of approximately 9.5 wt% for pure
silylated soy oil and 10.5 wt% for the silylated soy mixture used for coatings. In each heating
method, the silylated oil mixtures exhibit higher VOCs than pure silylated oil by approximately
1.0 wt%. This is attributable to the addition of water in the mixtures which has a boiling point
below the testing temperature (110°C).
Under the Clean Air Act, waterproof coatings have an allowable content limit of 600
grams of VOC per liter of coating material [19]. The average density from all silylated soy
products was found to be 0.992 g/mL using a 25 mL specific gravity picometer. Using the
32

results from Table V, for silylated soy oil mixtures, the VOC equates to 99.8 grams of VOC per
liter of coating material, well below the regulated 600 gram per liter limit. Similarly, pure
silylated soy oil has a VOC content of 88.9 grams per liter of coating. Due to the high solids
content as a result of the high soy oil concentration, silylated soy oil coatings demonstrate very
low VOC contents.

Specifically, the magnitude at which silylated soy oil falls below the

allowable VOC limit suggests the coating to be extremely environmentally friendly in regards to
air quality.
Table 2.5 VOC results of convection oven and TGA heating methods. Values indicate weight
percent loss for both silylated soy oil and pre-coating mixtures containing 1wt% water and
DBTDL.
Convection Oven Heating Method VOC %
TGA Heating Method VOC %
Silylated Oil with 1 wt%
Silylated Silylated Oil with 1 wt%
Sample
Silylated Oil
Water & DBTDL
Oil
Water & DBTDL
A
9.06
10.01
10.22
11.05
B
9.12
10.01
9.97
10.68
C

8.70

10.16

10.03

10.83

avg.

8.96

10.06

10.07

10.85

std. dev.

0.23

0.09

0.13

0.19

2.6 Results & Discussion – 5 Gallon Parr Reactor
2.6.1 TGA & Graft Percentages
Large amounts of silylated soy oil were required for manufacturing-scale trials.
Consequently, a much larger reactor was needed to achieve the required amount of coating
material necessary. Manipulation of a reactor’s dimensions and volume drastically impacts a
reaction’s kinetics resulting in variations in heating, cooling, and reaction times. As a result,
several reaction conditions were changed between 2 L and 5 gal. reactors in order to meet the
hydrophobicity, viscosity, and pot life specifications required of most coating facilities.
Additionally, the heating mantle for the 5 gal. Parr reactor was limited to 250°C, further
33

complicating the heating, cooling, and reaction times needed to provide the desired
characteristics of the final product.

Table 2.6 Graft percentages of silylated soy oil products run in 5 gal. Parr reactor at 250°C, 20
hours calculated using Eq. 2.1.
Reaction wt%VTMOS wt%loss123°C
10
33.1
7.053
11
33.1
0.842
12
33.1
6.646
13
33.1
4.981
14
32.8
6.619
15
33.6
5.675
16
33.3
6.792
17
33.6
6.959
18
33.6
6.796
19
33.6
6.576
Average
Std. Dev.

Graft %
78.7
97.5
79.9
84.9
79.8
83.1
79.6
79.3
79.8
80.4
82.3
5.7

Ultimately, reactions were carried out for 20 hours to compensate for the reduced
reaction temperature and larger reactor volume. Mixture A reactant ratios (Table 2.1) were used
for all reactions performed in the 5 gal. Parr reactor. TGA results and graft percentages are
shown in Figure 2.8 and Table 2.6, respectively. Graft percentages were calculated using Eq.
2.1.

34

8.0

80

7.0

70

6.0

60

5.0

50

4.0

40

3.0

30

2.0

20

1.0

10

0.0

0

Derivative Weight Percentage (dwt%)

9.0

90

Weight Percentage (wt%)

100

-1.0
0

100

200
300
Temperature (°C)

400

500

Figure 2.8 Overlay of TGA plots from Reactions 10-19 using 5°C/min ramp up to 450°C under
nitrogen gas.
Like the results from Reactions 1-9 carried out in the 2 L Parr reactor, the average graft
percentage and standard deviation were found to be 82.3% ± 5.7%. Reaction 11, however, was
subject to partial curing (Figure 2.9) during the reaction causing a gelatinous substance to form
within the reactor. The partial curing inside the reactor was most likely due to residual water
present in the reactor causing the reacted silane molecules to undergo intermolecular crosslinking. Subsequently this also accounts for its drastically elevated high graft percentage. Like
Reactions 6, 8, and 9 from the 2 L reactor, Reaction 11 exceeded one standard deviation unit for

35

graft percentage and was not used for any coating purposes. Excluding Reaction 11 from the
total data resulted in an average graft percentage of 80.6% ± 2.0%.

Figure 2.9 Final product of Reaction 11 from partially cured silylated soy oil. The solid
material on the stirrer is fully cured silylated soy oil. A small amount of viscous liquid silylated
soy oil was collected, characterized, and analyzed but was not used for further coating tests and
applications.
Similar to results highlighted in Figure 2.7, the reproducibility of the desired silylated soy
oil is shown in the TGA results found in Figure 2.8. The overlapping wt% loss and derivative
wt% loss curves indicate products with very comparable thermal degradation properties, thus
supporting the reaction’s reproducibility in large volume reactors. Establishing reproducibility
of the desired product in a larger reactor is a critical component in developing the scale-up
process for the synthesis and application of any coating material.
36

2.6.2 Viscosity
Each product was evaluated for bulk viscosity using a Brookfield DV-E viscometer with
an LV spindle set and a rate of 30 rpm. It was important that products from Reactions 10-19
show lower viscosities with less variability to demonstrate pot life times exceeding three hours.
Products with lower viscosities require much more time to develop a high viscosity. Because the
maximum working viscosity of most industrial coating processes is 800 cP (for gravure coating
technique), silylated soy oil with viscosities in 40-80 cP range drastically improve the pot life
potential. Therefore, it was desired that Reactions 10-19 demonstrate lower viscosities than
Reactions 1-9 while still exhibiting graft percentages large enough to allow the silylated soy oil
to completely cure. Table 2.7 summarizes the graft percentages and viscosities of reactions
performed in the 5 gal. reactor.
With the exception of Reaction 11, which partially cured, the standard deviation was
substantially smaller for reactions performed in the 5 gal. reactor versus those performed in the 2
L reactor.

Excluding reactions exceeding one standard deviation, average viscosity was

calculated to be 41.2 cP ± 4.5 cP. As a result, the trend between graft percentage and viscosity
was much more predictable (Figure 2.10). Plotting the viscosities of Reactions 10-19 with
respect to graft percentage of each product provides a very definitive polynomial equation that
allows for accurate predictions of viscosity. This plot, in conjunction with pot life experimental
data discussed in chapter three, provide an accurate method for estimating pot life of silylated
soy oil using only graft percentage and viscosity measurements. This allows for a reliable, yet
rudimentary, method for accepting or rejecting a reaction’s products for coating use.

37

Table 2.7 Summary of graft percentage and associated viscosity from Reactions 10-19.
Reaction
10
11
12
13
14
15
16
17
18
19
Avg.
Std. Dev.

250

Viscosity (cP)
39
260
40
41
36
52
40
44
39
40
63.1
69.3

y = 0.8244x2 - 133.73x + 5461
R² = 0.9946

200
Viscosity (cP)

Graft %
78.7
97.5
79.9
84.9
79.8
83.1
79.6
79.3
79.8
80.4
82.3
5.7

150
100
50
0
75

85
95
Graft Percentage (%)

Figure 2.10 Viscosity measurements of 5 gal. reactor products with respect to graft percentage.

38

2.7 Results & Discussion – Varying Silane Reactants and Reactant Ratios
In an effort to improve coating performance while minimizing production costs, different
silane molecules and reaction ratios were investigated. As shown in Table I, Alder-Ene reactions
were carried out using four different reactant mixtures. Though Mixture A was the primary
focus throughout the entirety of the project; reactions using Mixtures B-D were performed and
analyzed for comparison purposes.
2.7.1 Mixture B – 1:1 Molar Ratio Soy Oil:VTMOS
Multiple reactions were performed in a 2L Parr reactor using a 1:1 molar ratio of lowsaturated soy oil to VTMOS.

Each reaction resulted in a viscous liquid product.

The

consistencies of the products were similar to that of jelly or jam and with viscosities ranging
from 1800-42,000 cP. The increased viscosity seen in the 1:1 molar ratio of soy oil:VTMOS
compared to 1:3 molar ratio results from the degree of curing which took place while the large
range in viscosity can be attributed to the degree of curing that occurred during the course of the
reaction. Similarly, the graft percentages of each reaction exceeded 95%, again, as a result of the
curing process.
As may be expected, the physical properties resulting from a 1:1 molar reaction pose
several new issues not seen by the products of the 1:3 molar reaction, mostly due to the dramatic
viscosity increase. An initial viscosity of 1800 cP already far exceeds the upper viscosity limit
of gravure roller coating capabilities. Adding water and DBTDL to promote further crosslinking
would only increase the already high viscosity of 1:1 molar silylated soy oil. As a result, an
alternative coating process to using a gravure roller would need to be investigated. Additionally,
with an increased viscosity comes a decreased pot-life. Because 1:1 molar silylated soy oil has a
very high viscosity, it’s pot-life time is substantially reduced even if using other coating
techniques with higher operational viscosity limits. As a result, this particular reaction ratio was
39

not investigated much beyond initial physical property analysis and preliminary coating
evaluations. It is possible that varying the reaction time or temperature may provide a usable
silylated soy oil product for coating purposes, however, further investigation would be required
to determine if this reactant ratio is feasible and provides a usable product.
2.7.2 Mixture C – 1:3 Molar Ratio Soy Oil:VTEOS
One reaction was performed using VTEOS at a 1:3 molar ratio of soy oil and VTEOS in
place of VTMOS. The result of the reaction was very comparable to reactions using Mixture A
(VTMOS of the same molar ratio). The graft percentage was calculated to be 75.1% with a
viscosity of 62 cP.

Additionally, coated samples using VTEOS showed very similar

hydrophobic properties to coatings using Mixture A silylated soy oil. For instance, measuring
the Cobb Values, a method of quantitatively measuring water resistance, for Mixture C products
shows water absorption to be 19.5 grams per square meter (gsm). This value is considered to be
quite hydrophobic in the paper industry. These results are discussed in greater detail in Chapter
3: Silylated Soy Oil Coatings. While the majority of research was done using a Mixture A
silylated soy oil, it can be safely assumed from the results obtained from TGA, viscosity, and
measured Cobb Values that a VTEOS formulation could easily be used in place of a 1:3 molar
ratio of soy oil and VTMOS.
2.7.3 Mixture D – 1:1 Molar Ratio Soy Oil:VTEOS
Much like the results seen in Mixture B silylated soy oil, using VTEOS in a 1:1 molar
ratio also exhibits a substantial increase in viscosity. Though the viscosity of the Mixture D
product has a somewhat lower viscosity at 336 cP with a graft percentage of 81.1%, the elevated
viscosity would still prove problematic in most industrial coating settings.

40

If Mixtures B and D were to be used for coatings, one way to extend their pot life would
be to reduce the concentrations of water and DBTDL used during the pre-coating mixing and
curing process. Reducing the water and DBTDL concentrations used for the pre-coating mixture
may provide extended pot life despite these mixtures exhibiting significantly elevated viscosities.
Investigating these variables as well as reaction temperature and time may provide a product
similar in physical properties to those seen in Mixtures A and C. Also, Cobb Values, which
quantitatively measure water resistance, provide no additional hydrophobic benefit when using
Mixtures B and D for coatings.

41

REFERENCES

42

REFERENCES

[1] Alder, K.; Pascher, F.; Schmitz, A., Absorption of maleic acid-anhydride and azodicarbon
acid-ester in simple unsaturated carbohydrogens. Information on the substitution processes in the
allyl position. Ber Dtsch Chem Ges 1943, 76, 27-53.
[2] Oppolzer, W., Asymmetric Diels-Alder and Ene Reactions in Organic-Synthesis.
Angewandte Chemie-International Edition in English 1984, 23 (11), 876-889.
[3] Snider, B. B., Lewis-Acid-Catalyzed Ene Reactions. Accounts Chem Res 1980, 13 (11), 426432.
[4] Akermark, B.; Ljungqvist, A., Thermal Formation of Bialkyls from Alkylmetals .2. Some
Reactions of Neophylnickel Complexes. J Organomet Chem 1978, 149 (1), 97-112.
[5] Fienemann, H.; Hoffmann, H. M. R., Cyclobutenecarboxylic Esters Via Aluminum-Chloride
Induced [2+2] Cycloadditions of 2-Propynoic Esters to Cyclic Olefins. J Org Chem 1979, 44
(15), 2802-2804.
[6] Snider, B. B.; Duncia, J. V., Stereoselective and Regioselective Ene Reactions of Methyl
Alpha-Chloroacrylate. J Am Chem Soc 1980, 102 (18), 5926-5928.
[7] Gill, G. B.; Marrison, K.; Parrott, S. J.; Wallace, B., Stereoselection in the Alcl3-Catalyzed
Ene Additions of Chloral to 1,2-Dialkyl Ethylenes. Tetrahedron Lett 1979, (50), 4867-4870.
[8] Srinivasan, Madhusudhan. "Synthesis, Properties and Applications of Bio-Based Materials."
Michigan State University, 2010. United States -- Michigan: ProQuest. Web. 4 Apr. 2013.
[9] Zhuang, Yue. “Novel Synthetic Strategy To Produce Biobased Silylated Soybean Oil
Coating Material.” Michigan State University, 2011. United States – Michigan: ProQuest.
Web, 4 Apr. 2013.
[10] Sigma-Aldrich Co. Products: Sigma-Aldrich Co. Sigma-Aldrich Co. Website. [Online]
http://www.sigmaaldrich.com/catalog/product/aldrich/388092?lang=en&region=US.
[11] Behr, A.; Naendrup, F.; Obst, D., Platinum-catalysed hydrosilylation of unsaturated fatty
acid esters. Adv Synth Catal 2002, 344 (10), 1142-1145
[12] Saghian, N.; Gertner, D., Hydrosilylation of Long-Chain Unsaturated Fatty-Acid Esters. J
Am Oil Chem Soc 1974, 51 (8), 363-367.
[13] El Kadib, A.; Castel, A.; Delpech, F.; Riviere, P., Silylation of triacylglycerol: an easy route
to new biosiloxanes. Chem Phys Lipids 2007, 148 (2), 112-120

43

[14] R. Narayan, D. Graiver, K.W. Farminer, M. Srinivasan. (US Patent 8,110,036, 2012).
[15] Chen, Z. G.; Chisholm, B. J.; Patani, R.; Wu, J. F.; Fernando, S.; Jogodzinski, K.; Webster,
D. C., Soy-based UV-curable thiol-ene coatings. J Coat Technol Res 2010, 7 (5), 603-613.
[16] Environmental Protective Agency. U.S. Clean Air Act. United States Government Printing
Office. [Online] http://www.gpo.gov/fdsys/pkg/USCODE-2008-title42/pdf/USCODE-2008title42-chap85.pdf.
[17] Komprex Permormance Composites. Glossary: Komprex. Komprex Composites Website.
[Online] http://www.komprex.com/Glossary/index.htm.
[18] Ferry, John D. Viscoelastic Properties of Polymers. New York: Wiley, 1980.
[19] U.S. Environmental Protection Agency. Volatile Organic Compounds Detailed Fact Sheet.
U.S. Environmental Protection Agency. [Online]
http://www.epa.gov/ttnatw01/183e/aim/aimfact.pdf.

44

Chapter 3: Silylated Soy Oil Coatings
3.1 Coating Background & Cure Chemistry
Coatings are coverings that are applied to a surface, or substrate, that enhances the
surface properties of the substrate. They can have several different functionalities including
adhesive, wear resistance, corrosion resistance, release properties, and appearance adjustments,
as well as several other applications. Much like the numerous coating functionalities, there are
many different coating techniques used to apply a coating to a substrate. Coatings can be applied
from solutions, emulsions, or directly from the bulk providing the viscosity is sufficient.
Additionally, coatings can be synthesized from a multitude of different media such as water,
solvents, and acrylics.
Furthermore, the multitude of uses of coatings ranging from protective functions such as
anticorrosive paint applied to cars, to decorative properties for things such as camouflaging
military equipment or color coding which indicates a container’s contents, or even antibacterial
coatings used in sterile environments such as hospitals, provide another physical requirement
that must be met. Such a broad spectrum of requirements explains why no single coating can
simultaneously satisfy every demand. Providing coating materials for the durable protection,
decoration, and improvement of objects made of wood, metal, plastic, or paper at reasonable
prices can only be met by adopting different formulations using a range of material
combinations. Each combination targets a limited field of substrates, a selected application
method, and a specific profile of film properties [1].
When the different coating functionalities, application methods, and formulations are
considered, there are many ways in which a coating can be produced. Therefore, the quality
45

achieved during the coating process is not only the function of the material used, but is
dependent on the coating material and design, the appropriate applications process, and the
substrate that is used.

Manipulating any of these variables can drastically impact the

performance properties of the final coated product. Thus, it is critical to investigate all aspects of
the coating process in order to determine the optimal conditions that result in the desired coated
product.

3.2 Silicone Coatings & Cure Chemistry
The incorporation of siloxane molecules into the unsaturated double bonds in soy oil
allows the modified soy oil to demonstrate properties that are typical of repeating Si-O-Si bonds
associated with siloxane coating chemistry. While several other coating substrates can be used
as supplied, silicone polymers must be crosslinked (or cured) to achieve their final properties [2].
This process requires a crosslinking agent which promotes the reaction of a silicone molecule to
react with a functional site of an adjacent silicone molecule. When a silicone molecule has more
than one functional group, it can react with multiple silicone molecules, thus forming a complex
chain or branched network. The most commonly used cure reactions are free-radical, addition,
and condensation reactions. Condensation reactions are often used in sealants and waterproofing
coatings and were used as the curing mechanism for this project.

Crosslinking using a

condensation reaction typically uses an organometallic tin catalyst.

Specifically, dibutyltin

dilaurate (DBTDL) and dibutyltin dioctoate (DBTDO) are commonly used as a catalyst in many
condensation curing reactions. In addition to a tin catalyst, the curing process is also dependent
on the presence of moisture which allows the silicone molecules to form a crosslinked network.
This resulting network is considered to be a cured product (Figure 3.1).

46

R

R
R

RO

Si

OR + RO

Si

OR
RO

OR

+

OR

H2O

OR

+

OR

Si

OR + RO

Si

Si

Si

OR

+ 4 ROH

O

Sn catalyst
Si

O

Si

OR

OR
R

R

O

O
RO

RO

R

R

R

Figure 3.1 Polymer Condensation reaction resulting in a cured silicone polymer with OR
representing siloxane substituents (e.g. methoxy or ethoxy groups in VTMOS and VTEOS,
respectively) and R substituents representing vinyl groups which graft to the unsaturated double
bonds of soy oil during an Alder-Ene reaction.

The cure rate of a condensation reaction is dependent on the amount of catalyst and
moisture present and is independent of curing, or drying, temperature. However, because an
alcohol is formed during the curing process of silylated soy oil, increased temperatures help
remove the undesired alcohol byproducts more rapidly resulting in a shorter tack-free time.
Tack-free time is the time required for a coated sample to be dry to the touch once the coating
has been applied. Exposing a silylated soy oil coating to increased temperatures can result in a
more complete cure with minimal alcohol byproducts. Much like the variety of ways that a
coating can be manufactured, adjusting coating variables such as cure temperature, cure time,
catalyst and moisture concentrations, paper grade, and coating weight all have impacts on the
properties of a coating. In essence, it is necessary to view the final product, not as a series of
separate processing steps, but as an integrated system. Furthermore, the functional performance
of a silicone coating is affected by the physical properties of the silicone as well as the
application process and the performance characteristics required of the final product [3].

47

3.3 Coating Variables
As mentioned, several variables must be considered when developing a silicone coating
that meets certain physical property criteria. Manipulation of coat weights, cure temp, cure time,
water content, catalyst concentration, and paper grades were investigated.
3.3.1 Coating Weight
The amount of coating placed on a substrate can substantially affect the desired physical
properties. For this project, a high degree of hydrophobicity is desired for a few hours, yet over
several weeks, the coating must be biodegradable under aqueous conditions. A coat weight too
low wouldn’t demonstrate the needed water resistance due to the limited amount of silylated soy
oil that can interact with the surface of the paper. Specifically, a low coat weight may not
maximize the interactions between the fibers of the paper and the silylated soy coating. This
coating-paper interaction depends greatly on the distribution of the fiber bundles in the paper
being used.
In contrast, a coat weight too high can result in saturation of the coating material into the
paper, thus impacting the strength and hydrophobicity of the coated product. Additionally, using
excess silylated soy oil would negatively impact production costs while not providing any
additional hydrophobicity beyond a certain point. However, small increases in coat weight can
provide improved hydrophobic properties. Therefore, determining a coat weight that is high
enough to meet the needed water resistant specifications, yet low enough to be coated on paper
without being absorbed while minimizing production costs is critical to product development.
3.3.2 Cure Temperature & Time
Due to the moisture-cured siloxane reaction used in this project, the resulting byproduct
of the crosslinking is an alcohol. The presence of alcohol on the surface of the coating has a
substantial negative impact on hydrophobicity. Therefore, the alcohol must be removed to
48

achieve the maximum water resistivity. Heating the finished coatings allows for any residual
alcohol on the coating surface to be removed quickly and effectively. However, exposing the
paper coatings to temperatures too high may diminish the structural integrity of the coating or
paper itself, making it brittle or flaky and rendering it useless.
The time required for the coating to completely cure is varied by the catalyst
concentration for silicone condensation reactions. While it is not necessary for a coating to be
fully cured in order to demonstrate drastically improved hydrophobicity, the amount of time a
coated product is allowed to cure varies its water-resistant properties. Likewise, cure time and
tack-free time are directly influenced by the cure temperature that samples are subjected to. A
shorter tack-free time would allow a coating facility to operate at a higher coating speed resulting
in higher production rates.
While operation and installation costs for a dryer would be expensive, the increase in
production volume by significantly reducing cure and tack-free times would drastically improve
production costs.

Determining the relationship between cure temp, cure time, and

hydrophobicity provide insight into the optimal coating conditions that result in the best
hydrophobic coating while minimizing energy demands associated with a dryer.
3.3.3 Catalyst and Moisture Concentrations
Two important variables that affect the cure kinetics of a condensation reaction are the
concentrations of catalyst and water reacted. As a result, varying the concentrations of the
catalyst used in the cure process will have a direct effect on the viscosity and, ultimately, the potlife of a condensation cure. While viscosity and pot-life are not of primary concern at a small,
development-scale volume, they become an increasingly important variable to consider when
coating large volumes associated with mass manufacturing. Therefore determining the optimal
49

formulations of water and catalyst and how they affect cure rate becomes a critical component of
the coating development.

While it is important to be able to meet predetermined pot-life

requirements, it is equally as important to meet the required coating performance criteria. Also,
due to the cost of DBTDL, reducing the amount used during the cure process can drastically
improve production costs.
3.3.4 Paper Types
There are several variables in paper production that can substantially impact the physical
properties, and ultimately, the applications that each paper type is used for. These variables
include basis weight (the weight per unit area, commonly expressed in gsm’s), fiber source
(material used to make the pulp that is developed into paper), fiber orientation (the alignment of
the fibers in the paper), filler content (minerals added to the pulp to enhance desired properties),
formation (the uniformity of the fibers throughout the paper), and several other factors that
contribute to the functionality of the paper [4]. Each of these factors can affect how well a
coating interacts with the surface fibers of a paper. This makes the paper selection for this
project important when considering the property performance of the final coated product.
For example, a paper with a low degree of formation could negatively impact the waterresistance due to the inconsistent distribution of fiber bundles that interact with the coating.
Similarly, low degree of formation can decrease a paper’s tensile strength [5]. Because this
project is focused on creating coated paper for trash bag production, it is necessary for the
product to demonstrate both uniform water-resistance and a high tensile strength. Therefore,
using a paper with good construction characteristics should improve the strength and
hydrophobicity of the coated product. However, evaluating and considering the several other

50

variables of paper production that affect a paper’s physical properties can also provide a greatly
improved coated product.
3.3.5 Additional Factors
When developing a coating such as silylated soy oil, it is essential to view the final
product, not as a series of separate steps and components, but as an integrated system. The
functional performance of the silicone is affected not only by the physical properties of the
silicone, but also by the application process and performance characteristics required of the final
product [3].

Therefore factors such as pot life, mixing, and tack-free times must also be

considered in development of the silicone coating “system”. While the hydrophobicity of the
coating and tensile strength of the paper are of utmost importance, it is equally important to
demonstrate a long pot life and a short tack-free time. Therefore, improving water resistance is
equally as important as reducing a coatings tack-free time. For instance, improving a coating’s
hydrophobicity while neglecting tack-free time would result in an unusable product at an
industrial scale because re-rolling the paper after the coating application would result in the felt
side (coated side of the paper) adhering to the wire side (back side of the paper) creating a solid
block of paper that couldn’t be un-rolled. Similarly, improving tack-free time while neglecting a
coating’s water resistance would also be undesirable. Consequently, creating a coating system
that satisfies both the coating performance and the application process requirements presents the
most difficult challenge of this project.
3.3.5.1 Pot Life Time
Pot life time is the length of time from when the curing materials have been mixed to the
time where the viscosity increases to the point where the coating material is no longer usable due
to extensive curing. When using a gravure roll coater, the upper viscosity limit is approximately
800 cP. For a coating to be considered usable at an industrial scale, the pre-coating mixture must
51

not exceed the upper viscosity limit for at least three hours. To extend the pot life of the precoating mixture, DBTDL and water concentrations can be varied. As a result, the concentrations
of DBTDL and water were varied to determine optimal concentrations which would satisfy both
coating performance and pot life requirements.
3.3.5.2 Blend (Mixing) Time
Blend time, or mixing time, is closely associated with pot life time. Blend time is the
amount of time to achieve a predefined level of homogeneity of a tracer in a mixing vessel. For
this project, it is defined as the amount of time between the combination of the curing reactants
(silylated soy oil, DBTDL, and water) and the time in which this pre-coating mixture is applied.
Varying the blend time impacts the viscosity of the coating mixture which, in effect, impacts
how well the coating coats to the paper’s surface.

As previously mentioned, coating

performance varies dramatically with how well the coating interacts with the surface of the
paper. Because a certain degree of strength is needed for use as a trash bag, and because a
saturated coating diminishes a paper’s tensile strength, it is important to have a good surface
coating while maintaining a viscosity below 800 cP. Like pot life, blend time is also dependent
on viscosity limits while also meeting the desired coating properties.
3.3.5.3 Tack-Free Time
Tack-free time is commonly associated with the cure process and is considered to be the
time that has elapsed between mixing and the point of time when the mixture is no longer wet, or
tacky, to the touch [6]. It is commonly associated with the cure process itself. Tack-free time is
especially critical at larger scales where rolls of paper are used as opposed to single page
(8.5”x11”) samples. Having a long tack-free time at a large scale would result in the felt side of
the coating adhering to the wire side of the roll causing the coated roll of paper to adhere to

52

itself. More importantly, however, a long tack-free time requires a slow coating rate which
increases production costs.
Tack-free time can be varied by adjusting the coat weight and cure temperature. Lower
coat weights and higher cure temperatures promote a quicker cure (dry) time whereas heavier
coat weights and lower temperatures will exhibit longer tack-free times. While increased coat
weights, to a certain point, exhibit improved hydrophobicity, the resulting increase of tack-free
time can make the coating system “unbalanced”. Also, while higher cure temperatures provide
shorter tack-free times, temperatures too high can result in degradation of the coating itself.
Therefore finding a balance between coat weight, cure temp, and tack-free time must be taken
into consideration when developing the coating system.

3.4 Materials
Silylated soy oil was used from experiments outlined in Chapter 2. Dibutyltin dilaurate
(DBTDL) [CAS 77-58-7] was purchased from Sigma-Aldrich Co., St. Louis, MO and was used
as received without any modifications. All papers used for coating purposes were supplied by
Wausau Paper, Mosinee, WI.
Coatings were performed using a Euclid Coating Systems Single Roll Lab Coater (ECS).
This is a small, pneumatic, bench-top blade-roller coating system that allows the operator to
adjust the contact pressure at the blade-roller interface (Figure 3.2). In doing so, a range of coat
weights can be achieved.

53

Figure 3.2 Single Roll Euclid Lab Coater used for all coatings made throughout duration of
research. Pneumatic pressure gauge allowed for precise coat weights. The design and small size
of the coater required the use of minimal coating material.
3.5 Product Analysis
Coated samples were evaluated for water-resistance using two specific methods that
provide insight into the water repellent properties provided by the silylated soy oil coatings. The
following section describes each analytical technique used, a brief description of the technique,
and how the results were applicable for the silylated soy oil coatings.
3.5.1 Cobb Test
Cobb Testing provides a way of quantitatively measuring the absorption of water (or
water resistance) on the surface of paper. A substrate’s water absorptiveness is a function of
various paper characteristics such as sizing, porosity, and formation, along with several other
factors.
Cobb Tests were performed following TAPPI T441 test method by cutting samples into
12.5 cm x 12.5 cm squares. Each sample was weighed and placed on a neoprene mat and
54

2

clamped in place by a 100 cm Cobb Cylinder using a crossbar and tightening knobs (Figure
3.3). One hundred milliliters of water was poured on the sample surface. Timing began at the
moment pouring commenced. After one minute and forty-five seconds, the water was poured
from the cylinder using care not to drop any water on the portion of the sample outside of the
100 cm2 ring. The crossbar was loosened and the sample was placed, with wetted side up, on a
sheet of blotting paper. After two minutes, a second blotting paper was placed on top of the
sample and was rolled once forward and once back using a 10-kg roller to remove any excess
water. No downward force was exerted when using the roller. After rolling, the sample was
removed from blotting sheets, folded in half, and reweighed [7]. The total water absorption, or
Cobb Value, measured in grams per square meter (gsm) was calculated using Eq. 3.1.

(Eq. 3.1)

Cobb Value (gsm) =

paper massfinal −paper massinitial
0.01 m2

Figure 3.3. Cobb cylinder, blotting paper, and 10-kg roller used for measuring Cobb Values.
3.5.2 Water Contact Angle
Contact angle measures the orientation of a liquid droplet at the liquid-vapor interface
when on a solid surface. This observed angle denotes the degree of hydrophobicity but is also
55

affected by the roughness of the surface and the type of liquid used. The angle made between
the liquid and the solid surface can be quantified using Young’s Equation. Young’s Equation
(Eq. 3.2) considers a thermodynamic equilibrium between the liquid droplet (L), the solid surface
(S) and the gas, or air, surrounding the droplet (G). Representing each interfacial energy using γ
and the equilibrium contact angle as θC, the Young’s Equation states:

(Eq. 3.2)

0 = 𝛾 𝑆𝐺 − 𝛾 𝑆𝐿 − 𝛾 𝐿𝐺 cos 𝜃 𝐶

The contact angle of a droplet on a solid surface is measured using a sessile drop method
(Fig. 3.4). This uses a goniometer, an instrument used to measure angles, and an optical system
to measure and record the liquid-solid interface. Typically, water contact angles measuring
under 90° are considered hydrophilic, while contact angles above 90° are deemed hydrophobic.
While surfaces such as metal normally exhibit contact angles near 0°, or a complete spread of the
water droplet on the metal surface, it is common for polymers to behave as hydrophobic surfaces
with contact angles over 100°. Surfaces exhibiting low surface energy may even display water
contact angles above 120° [8].

Figure 3.4 Surface contact angle represented as θC with respect to interfacial energies [8].
56

For this project, water contact angles were measured at time intervals of 60, 120, 180,
and 300 seconds for both coated and uncoated samples. The angle recorded from both the right
and left side of each water droplet was averaged to give the contact angle at each time point.
Each paper sample was measured four times at each time interval and the average water contact
angle was calculated. Data were summarized for each paper type (coated and uncoated) and time
interval.

3.6 Experimental
3.6.1 Varying Paper Samples with Low Basis Weight
3.6.1.1 Method – Varying Paper Samples with Low Basis Weight
Different paper samples (Table 3.1) were tested to evaluate how paper selection would
affect the coating performance of silylated soy oil. Each coating was made by mixing 5.0 g of
silylated soy oil (Mixture A, Table 1.1), with 4 wt% water (0.2 g), and 1 wt% DBTDL (0.05 g).
This was mixed for approximately two minutes on a magnetic stir plate at 1500 rpm. With paper
already secured to the ECS roller and blade pressure set at 40 psi, 2-3 mL of mixed coating was
spread evenly across the blade-roller interface. The roller was turned one complete revolution
and the paper was then removed from the roller. Coat weights were calculated, two minute Cobb
Values, ten minute Cobb Values, and water contact angles were measured for each sample. All
Cobb Tests were done in triplicate and all data points for water contact angle were done four
times. Blank uncoated samples were also tested for Cobb Values and water contact angles.

57

Table 3.1 Comparison of uncoated paper characteristics provided by Wausau Paper Co.
Uncoated Paper Characteristics
GN553713 GN612899 GN703742 GN354872
2

Basis Weight (lb./3000ft )
2

Basis Weight (g/m )
Caliper (mils)
2
Cobb Size, 1min (g/m )
Porosity (sec/100 ml)
Tensile CD (lb./in.)
Tensile MD (lb./in.)
Tensile Ratio
Tear CD
Tear MD
Smoothness, Felt
Smoothness, Wire

55.0

61.0

70.0

35.0

89.5
6.0

99.3
5.0

114.0
6.4

57.0
2.9

28
17
21
55
2.6
137
101
384
356

26
23
29
60
2.1
161
136
133
133

35
28
36
58
1.6
107
102
226
187

65
35
16
41
2.6
44
33
116
87

3.6.1.2 Results & Discussion – Varying Paper Samples with Low Basis Weight
Uncoated Cobb Values provided by Wausau Paper Co. (Table 3.1) varied dramatically
between each paper type. This is likely attributable to the basis weights and porosities of each
paper. For instance, GN354872 has a low basis weight and a high porosity, resulting in a high
uncoated Cobb Value compared to the other paper types. A higher basis weight is typically a
result of a more dense arrangement of fiber bundles in the paper. Therefore the fibers are more
compact in the paper, resulting in less fiber surface area that is subjected to water exposure. This
contributes to lower Cobb Values. Similarly, a paper’s porosity is a result of how quickly water
absorbs, or transmits, water through its surface. The lower a paper’s porosity, the more water
resistant it is.
Two minute Cobb Values show substantial hydrophobicity improvements; GN354872
and GN703742 in particular. While GN354872 demonstrated the worst uncoated Cobb Values,
it displayed the lowest two minute Cobb Value and the second lowest ten minute Cobb Value.
58

This is due to the paper’s low-density fiber arrangement. Having a lower fiber content would
allow the molecules of the silylated soy oil to better interact with the fiber surface, and thus form
a better hydrophobic network across the entirety of the paper; much for the same reason
uncoated paper exhibits poor water resistance.
It’s also important to consider the difference between the two minute and ten minute
Cobb Values. Though all coated samples show acceptable degrees of water resistance for two
minutes, none of the ten minute Cobb Values measurements show a high degree of
hydrophobicity. Because the trash bags made from the silylated soy oil coatings will likely be
exposed to aqueous material in excess of ten minutes, it is critical to demonstrate significantly
improved water resistance for time intervals of ten minutes or longer. One potential solution is
to use a paper with a higher basis weight. This would result in a lower initial Cobb Value and
provide increased hydrophobicity over longer periods of time.
Unexpectedly, results from water contact angle (Table 3.3) did not coincide with the
Cobb Values (Table 3.2). Where GN354872 demonstrated the lowest two minute Cobb Values,
the coated samples using this paper had the lowest contact angle of the four paper types
evaluated, suggesting the worst water-resistance. GN354872 samples did, however, exhibit the
largest contact angle improvements between coated and uncoated samples.

Though no

recordable contact angles were able to be measured for uncoated samples of GN354872 due to
complete water absorbance in less than sixty seconds, coated samples maintained a contact angle
above 80° for all time intervals. This increase in water contact angle is due to the reduced
surface tension, or interface tension, provided by the silylated soy oil coating.

59

Table 3.2 Cobb Values at both two and ten minute intervals of coated and uncoated papers.
2 Minute Cobb Values (gsm)
10 Minute Cobb Values (gsm)
Cobb std. Blank Cobb
std.
Cobb
std. Blank Cobb std.
Paper Type Value dev.
Value
dev.
Value dev.
Value
dev.
GN354872
19.1
0.1
62.4
3.0
39.1
8.2
62.5
0.8
GN553713
20.4
0.2
26.2
0.8
36.4
0.3
40.1
2.7
GN612899
24.5
0.2
28.4
0.3
42.3
0.3
43.6
0.6
GN703742
22.2
0.2
91.7
15.3
44.7
0.6
74.6
0.0

Table 3.3 Water contact angles of various paper types of both coated and uncoated samples.
COATED SAMPLE AVERAGES
Paper
Type
60s
120s
180s
300s
GN354872 87.4
85.4
83.8
81.3
GN553713 108.4
103.7
102.4
98.2
GN612899 104.0
101.6
99.6
96.2
GN703742 107.7
106.4
104.6
102.1

UNCOATED SAMPLE AVERAGES
60s
NA
114.7
122.9
64.4

120s
NA
111.9
122.2
57.9

180s
NA
110.8
121.6
46.8

300s
NA
107.3
119.1
NA

Coated samples of GN703742 also showed significant hydrophobic improvements using
contact angle, yet it demonstrated one of the highest Cobb Values; a much better indicator of true
water resistance. While the results from the water contact angles were contradictory to the Cobb
Test results, the data still provided a useful method of comparing the hydrophobicity
improvements between coated and uncoated samples. Ultimately, for trash bag production,
Cobb Values provide a better analytical method of evaluating water resistance, while water
contact angle can be useful to support the overall improvements seen between the coated and
uncoated samples.
In summary, every paper type demonstrated improved water resistance when coated with
silylated soy oil. Using Cobb Values and contact angle measurements, paper GN354872 showed
the most hydrophobic improvements comparing coated and uncoated samples. Additionally,
60

GN354872 had the lowest two minute and ten minute Cobb Values. While GN354872 did show
several promising hydrophobic improvements, a lower ten minute Cobb Value is essential for the
development and implementation of silylated soy oil coatings for trash bag purposes.
3.6.2 – Coat Weight Variations
3.6.2.1 Method – Coat Weight Variations
Silylated soy oil was mixed with 1.0 wt% DBTDL and 4.0 wt% water for approximately
two minutes and applied to paper GN505320 using a Euclid Coating System (ECS) at blade
pressures between 25-60 psi in 5 psi increments. Each pressure was repeated twice and the coat
weight for 8.5” x 11” paper was calculated using Eq. 3.3.

Coating Weight (gsm) =

(Eq. 3.3)

masspost−coating −masspre−coating
0.0603 m2

Cobb Tests were performed on all samples. Results of blade pressure, coat weight, and
Cobb Values were evaluated to determine how coat weight affects the water-resistance of the
silylated soy oil coatings.
3.6.2.2 Results & Discussion – Coat Weight Variations
Varying blade pressure of the ECS impacts the coat weight of each sample (Figure 3.5).
Subjecting each coating to a higher blade pressure removes additional, excess coating material
from the surface of the paper providing very low coat weights. However, coatings made with
blade pressures between 45-60 psi have little difference in coat weight. This coat weight
corresponds to the minimal amount of coating needed to cover the surface of the paper.
Therefore, it can be assumed that the minimum amount of silylated soy oil required to make a

61

surface coating is approximately 3.5 gsm which can be accomplished on the ECS using a blade
pressure of 45psi.
While there is a clear trend between blade pressure of the coating system and coat weight,
there is much less distinguishable difference between coat weights and Cobb Values. Coatings
made using a blade pressure of 25 psi resulted in an average coat weight of 6.0 gsm, whereas
coatings made using a blade pressure of 60 psi resulted in an average coat weight of 3.5 gsm.
Despite the coat weight being more than 70% higher, coatings made at 25 psi show no
statistically significant Cobb Value improvement over coatings made with lower coat weights.
6.5
6.0

Coat Weight (gsm)

5.5
5.0
4.5
4.0
3.5
3.0
20

30
40
50
Coating Blade Pressure (psi)

60

Figure 3.5 Variation in coat weights of silylated soy oil when varying blade pressure on ECS.
Summarizing the data from both Figure 3.5 and Figure 3.6, it is most efficient to operate
manufacturing scale coaters that result in a coat weight between 3.0-3.5 gsm. Using a higher
coat weight only uses excess coating material while providing limited hydrophobic
62

improvements.

By setting coating equipment to achieve a coat weight near 3.5 gsm, the

economics of the coating system is fully optimized while not sacrificing any water resistance
performance. This is important when considering any future large-scale silylated soy oil coating
operation.

Cobb Value (gsm)

24.0

23.0

22.0

21.0
3.0

4.0
5.0
Coat Weight (gsm)

6.0

Figure 3.6 Cobb Values when varying the coat weight of silylated soy oil on GN505320 paper.
3.6.3 Cure Time & Temperature
Several experiments were performed to determine the effects of cure time and cure
temperature have on the performance of silylated soy oil coatings.

The method for each

experiment and the results are included in the following sections.
3.6.3.1 Method – 4hr Cure Time with Cure Temps at 60°C and Room Temp
Each coating was made by mixing 5.0 g of silylated soy oil (Mixture A, Table 1.1), with
4 wt% water (0.2 g), and 1 wt% DBTDL (0.05 g). This was mixed for approximately two
63

minutes on a magnetic stir plate at 1500 rpm. With paper (GN354872) already secured to ECS
roller and blade pressure set at 40 psi, 2-3 mL of mixed coating was spread evenly across the
blade-roller interface. Roller was turned one complete revolution and paper was removed from
the roller.
Each sample was subjected to four hours of total cure time using a combination of 60°C
and room temp (approximately 20°C) cure temperatures. Five samples were cured at 60°C for
all four hours, five samples were cured for two hours at 60°C and then removed from the oven
and allowed to cure at room temp for two hours, and the remaining five samples were cured for
four hours at room temp (Table 3.9). Once the pre-determined curing time was complete,
samples were tested for two minute Cobb Values. Blank, uncoated samples were also tested.
3.6.3.2 Results & Discussion – 4hr Cure Time with Cure Temps at 60°C and Room Temp
The average coating weight for all samples was 4.1 ± 0.3 gsm; considered a light
coat weight by most industry standards. However, even with a low coat weight, the waterresistance improves dramatically as shown in Table 3.4. Silylated soy oil coatings provided
water resistant improvements in excess of 160% for samples cured at room temp and 220% for
samples cured at 60°C for four hours. Because all coatings had nearly identical coat weights, the
20% Cobb Value improvement between coatings cured at 60°C versus those cured at room temp
can be directly attributed to cure temperature.
The improved Cobb Values seen at 60°C cure temps is likely due to the removal of
undesired VOCs, specifically methanol, from the coating surface. Subjecting silylated soy oil
coatings to temperatures that approach or exceed the boiling point of methanol (65°C) promote
the evaporation and removal of this undesired organic byproduct. Therefore when cure times are
restricted, Cobb Values improve with increased cure temperatures. Subjecting silylated soy oil
64

coatings to elevated cure temperatures may also improve the cure kinetics resulting in a more
complete cure reaction and, thus, providing a better crosslinked network which would provide an
additional explanation for Cobb Value improvement.
Table 3.4 Cobb Values of four hour cure time at combinations of 60°C and room temperature.
SAMPLE
A1
A2
A3
A4
A5
average
std. dev.
B1
B2
B3
B4
B5
average
std. dev.
C1
C2
C3
C4
C5
average
std. dev.
BL1
BL2
BL3
BL4
BL5
average
std. dev.

Wt0 (g)
0.9700
0.9718
0.9633
0.9734
0.9642

WtF (g)
1.1571
1.1564
1.1492
1.1579
1.1519

0.9580
0.9544
0.9618
0.9568
0.9766

1.1512
1.1364
1.1590
1.1440
1.1679

0.9838
0.9852
0.9751
0.9669
0.9693

1.2093
1.1950
1.2072
1.1938
1.2196

0.9297
0.9336
0.9083
0.9153
0.9002

1.5337
1.5412
1.5093
1.5207
1.4889

Cobb Value
(gsm)
18.71
18.46
18.59
18.45
18.77
18.6
0.1
19.32
18.20
19.72
18.72
19.13
19.0
0.6
22.55
20.98
23.21
22.69
25.03
22.9
1.5
60.40
60.76
60.10
60.54
58.87
60.1
0.7

65

Room Temp Cure
Time (min)
0
0
0
0
0

60°C Cure
Time (min)
240
240
240
240
240

120
120
120
120
120

120
120
120
120
120

240
240
240
240
240

0
0
0
0
0

NA
NA
NA
NA
NA

NA
NA
NA
NA
NA

The difference in Cobb Values between samples cured at room temp and samples cured
at 60°C led to additional experiments to investigate cure temperatures above the boiling point of
methanol. Variations of cure times at elevated cure temperatures were also examined.
3.6.4 – Reduced Cure Times at 60°C and Room Temp
3.6.4.1 Method – Reduced Cure Times at 60°C and Room Temp
Each coating was made by mixing 5.0 g of silylated soy oil (Mixture A, Table 1.1), with
4 wt% water (0.2 g), and 1 wt% DBTDL (0.05 g). This was mixed for approximately two
minutes on a magnetic stir plate at 1500 rpm. With paper (GN354872) already secured to the
ECS roller and blade pressure set at 40 psi, 2-3 mL of mixed coating was spread evenly across
the blade-roller interface. The roller was turned one complete revolution and the paper was
removed from the roller.
Fifteen samples were cured at various time intervals ranging from 15 minutes to 4 hours.
Each sample was cured for equal amounts of time at both 20°C (room temp) and 60°C cure
temperatures. Each time point was repeated in triplicate. After each cure time was reached,
samples were evaluated using two minute Cobb Values. Uncoated blanks were also tested.
3.6.4.2 Results & Discussion – Reduced Cure Times at 60°C and Room Temp
The average coat weight for all samples was 4.0 ± 0.3 gsm (Table 3.5). While the coat
weights were very consistent, the Cobb Values varied greatly. Uncoated blanks resulted in Cobb
Values in excess of 60 gsm (62.9 ± 1.0 gsm). Samples allowed to cure for a total of four hours
showed Cobb Values similar to those in Table 3.7 (19.7 ± 0.9 gsm) which also had a total cure
time of four hours. Reducing the total cure time from four hours to two hours resulted in a 1.0
gsm increase in Cobb Value on average. This difference is statistically insignificant, however,
due to the standard deviations of both sample sets. As a result, it is not necessary to allow

66

samples with coat weights near 4.0 gsm to exceed two hours of total cure time in order to reach
optimal Cobb Values.
On the contrary, comparing 30 minute total cure times (26.1 ± 0.7 gsm average Cobb
Value) to two hour total cure times (20.8 ± 0.5 gsm average Cobb Value) provides significantly
different Cobb Values; a 27% improvement. Samples with 10 minute total cure times exhibit
comparable Cobb Values (26.4 ± 1.2 gsm average Cobb Values) to those allowed to cure for 30
minutes with equal amounts of oven and room temp cure time. Because there is very little
difference in these Cobb Values, it can be deduced that total cure times occurring at 60°C and
room temp should fall between 30 minutes and four hours in length.
Samples 13-15 were cured at 60°C for 30 minutes. Comparing the Cobb Values of
samples 13-15 with samples 7-9 which have the same cure time but different cure temperatures
show a 3.5 gsm decrease in Cobb Value, or a 15% hydrophobicity improvement. Using the data
from Table 3.4 and Table 3.5 suggests that silylated soy coatings demonstrate lower Cobb
Values when cured at elevated temperatures for 30 and 240 minutes. While shorter cure times
demonstrate dramatically improved water resistance compared to uncoated samples, the
maximum hydrophobic properties of silylated soy coatings are only achieved at cure
temperatures near, or above, the boiling point of methanol for at least 30 minutes. Exceeding
cure times of four hours is unnecessary, as subjecting samples to cure intervals longer than this
shows no improvement in coating performance.

67

Table 3.5 Coat weights and Cobb Values of samples cured at various time intervals ranging
from 15 minutes to 4 hours. Each sample was cured for equal amounts of time at both 20°C
(room temp) and 60°C.

Sample
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Blank1
Blank2
Blank3

Coat
Weight
(gsm)
3.48
3.48
3.54
3.81
3.81
3.98
4.15
4.15
4.31
4.31
4.31
3.98
4.31
4.31
4.15
0.00
0.00
0.00

Cure
Condition
(°C)
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
NA
NA
NA

Cure
Time
(min)
120
120
120
60
60
60
15
15
15
5
5
5
0
0
0
NA
NA
NA

Room Temp
Cure Time
(min)
120
120
120
60
60
60
15
15
15
5
5
5
30
30
30
0
0
0

Cobb Value
(gsm)
18.7
20.0
20.4
20.9
20.3
21.3
25.6
26.9
25.9
25.4
26.0
27.8
24.3
22.4
21.4
64.1
62.4
62.3

Std.
Dev.
0.9

0.5

0.7

1.2

1.5

1.0

3.6.5 – 24hr Cure Time, 60°C vs. 80°C Cure Temp
3.6.5.1 Method – 24hr Cure Time, 60°C vs. 80°C Cure Temp
Each coating was made by mixing 5.0 g of silylated soy oil (Mixture A, Table 1.1), with
4 wt% water (0.2 g), and 1 wt% DBTDL (0.05 g). This was mixed for approximately two
minutes on a magnetic stir plate at 1500 rpm. With paper (GN354872) already secured to the
ECS roller and blade pressure set at 40 psi, 2-3 mL of mixed coating was spread evenly across
the blade-roller interface. The roller was turned one complete revolution and the paper was
removed from the roller.

68

Eight samples were subjected to 24 hours of total cure time, with the first hour of cure
being at either 60°C or 80°C. After one hour at the designated oven cure temp, samples were
removed and the cure process completed at room temp for the remaining 23 hours. At the end of
the 24 hour cure period, samples were tested for two minute Cobb Values.
3.6.5.2 Results & Discussion – 24hr Cure Time, 60°C vs. 80°C Cure Temp
The average coating weight was 3.4 ± 0.1 gsm for all coated samples. Similar to the
results from the four hour cure time experiment (3.6.3.2), Cobb Values for all samples were
greatly improved from uncoated blanks. Samples cured at 60°C for the first hour resulted in an
average Cobb Value of 20.9 ± 0.1 gsm, while samples cured at 80°C for the first hour showed an
average Cobb Value of 18.6 ± 0.6 gsm (Table 3.6).
Table 3.6 Comparison of sample sets cured at 60°C and 80°C for one hour of 24 hour cure time.
60°C Samples
80°C Samples
20.9
18.6
Cobb Average (gsm)
0.1
0.6
Std. Dev. (gsm)
3.57
3.81
Coating Weight Avg. (gsm)

While the difference in Cobb Values for samples cured at 60°C versus 80°C is only 1.3
gsm, it equates to a 12% difference. Though increasing the cure temp slightly above the boiling
point of methanol does not result in substantial differences in Cobb Value as may be expected,
the increased cure temp could provide a shorter tack-free time at an industrial scale.
3.6.6 – Varying Paper Samples with High Basis Weight
3.6.6.1 Method – Varying Paper Samples with High Basis Weight
Most paper trash bags have basis weights ≥75gsm, which is commonly considered a high
basis weight. To better assess and simulate the effectiveness and performance properties of

69

silylated soy oil coatings for use as trash bag coatings, multiple high basis weight papers (Table
3.7) were coated and tested for two minute Cobb Values.
Each coating was made by mixing 5.0 g of silylated soy oil (Mixture A, Table 1.1), with
4 wt% water (0.2 g), and 1 wt% DBTDL (0.05 g). This was mixed for approximately two
minutes on a magnetic stir plate at 1500 rpm. With paper already secured to the ECS roller and
blade pressure set at 40 psi, 2-3 mL of mixed coating was spread evenly across the blade-roller
interface. The roller was turned one complete revolution and the paper was removed from the
roller.
Table 3.7 Uncoated high basis weight paper characteristics provided by Wausau Paper Co.
Uncoated Paper Characteristics
GN503832 GN503944 GN505320
Basis Weight (lb./3000ft2)

50.0

50.0

50.0

Basis Weight (g/m )
Caliper (mils)

81.4
5.3

81.4
6.0

81.4
4.0

Cobb Size, 1min (g/m2)
Porosity (sec/100 ml)
Tensile CD (lb./in.)
Tensile MD (lb./in.)
Tensile Ratio
Tear CD
Tear MD
Smoothness, Felt
Smoothness, Wire

21
34
26
52
2.0
95
88
375
357

19
14
25
58
2.3
124
106
409
397

21
50
24
56
2.3
88
69
150
95

2

Eight samples were subjected to 24 hours of total cure time, with the first hour of cure
being at 80°C. After one hour at oven temp, samples were removed and allowed to finish the
cure process at room temp for the remaining 23 hours. At the end of the 24 hour cure period,

70

samples were tested for two minute Cobb Values. Each different paper type was tested six times.
Uncoated blanks for each paper type were also evaluated.
3.6.6.2 Results & Discussion – Varying Paper Samples with High Basis Weight
Each paper type tested had the same basis weight, comparable uncoated Cobb Values,
and similar tensile strength properties. The most distinguishable difference between the papers
was the porosity of each. Additionally, GN503944 and GN503832 had fluorinated additives in
the paper which provide additional water resistance, whereas GN505320 contained no additives.
As a result, samples with fluorinated additives show very low two minute Cobb Values for both
coated and uncoated samples. Specifically, GN503832 showed excellent hydrophobic properties
when coated with Cobb Values below 10 gsm (Table 3.8). However, the tack-free time of
GN503832 was considerably longer (> 1 min) even when exposed to 80°C conditions.
Table 3.8 Two minute Cobb Values of coated and uncoated high basis weight papers.
2 Minute Cobb Values (gsm)
Paper
Coated Cobb
Std. Dev.
Blank Cobb
Std. Dev.
GN505320
25.5
0.4
43.2
3.2
GN503944
14.9
0.2
13.3
0.4
GN503832
9.6
0.6
25.3
0.4

The environmental impacts of using a paper containing fluorinated additives for use in an
environmentally-friendly coating must also be considered.

It is well documented that

fluorocarbons have negative environmental impacts. While the coated samples of the fluorinated
papers demonstrate increased water resistance, the superior property performance these papers
exhibit is somewhat negated by the use of fluorocarbons and their inherent environmental
impacts.

71

Comparing the water resistant results of GN505320 and GN354872 (Table 3.8 and Table
3.2, respectively) indicate that both experience significant hydrophobic improvements of coated
samples when using two minute Cobb Values to evaluate hydrophobicity. Paper GN354872 has
a lower two minute Cobb Value than GN505320 (19.1 versus 25.5 gsm). However, GN354872
has a basis weight too low to be used as a trash bag. One option to consider that was not tested
would be to manufacture GN354872 at a higher basis weight (≥80 gsm) suitable for use in trash
bags. This would provide a higher basis weight that would increase tensile strength while
exhibiting lower Cobb Values compared to the other non-additive paper of GN5050320.

3.7 Optimal Coating Conditions
Summarizing the results from the coating experiments (section 3.6), it is clear that cure
temperature, cure time, coat weight, and paper grade are very important to the performance
properties and production costs of silylated soy oil coatings. Results from sections 3.6.1 and
3.6.6 indicate that a paper’s porosity and basis weight have a significant impact on the coating
performance of silylated soy oil. Specifically, using a paper with high porosity (>30s/100mL)
and high basis weight (>80gsm) provides better Cobb Values than papers with low porosity and
basis weights. Additionally, the use of a fluorinated paper can significantly improve the waterresistance of a coating, however using such a paper grade can increase costs and has negative
environmental impacts.
Similar to paper with high basis weight and porosity, elevated cure temperatures and
increased cure times also improve water-resistance of silylated soy oil coatings. Results from
Table 3.5 emphasize the significance of how cure time affects the water-resistance properties of
silylated soy oil coatings. Furthermore, results summarized in Table 3.4 and Table 3.6 show that
increased cure temperatures provide significant improvements in Cobb Values.
72

Therefore,

allowing silylated soy oil coatings to cure for a minimum of four hours at temperatures of 80°C
or higher provide the most water resistant product.
Section 3.6.2 investigating the effect of coat weight on measured Cobb Values shows no
statistical trend. Because a higher coat weight doesn’t result in a more hydrophobic coating, it is
most beneficial to operate at the lowest possible coat weight while ensuring that the entirety of
the surface is coated. In doing so, coating production is maximized without requiring more
coating material, thus reducing the production costs per bag.
Moisture and catalyst concentrations were also mentioned as factors in the rate of cure for
silylated soy oil coatings. The effects of varying water and catalyst concentrations also impact
the performance properties and pot-life time of silylated soy oil coatings and will be discussed
further in Chapters 4 and 5.

73

REFERENCES

74

REFERENCES

[1] BASF Handbook on Basics of Coating Technology. AJ Goldschmidt, HJ Streitbeger. 2003.
[2] Dow Corning Corporation. Discovery Center: Dow Corning Corp. Dow Corning Corp.
[Online] January 2013. http://www.dowcorning.com/content/discover/discoverchem/curesystems.aspx.
[3] Blackwood, WR of Dow Corning Corp. Published Literature. Dow Corning Corp. [Online]
November 2004. http://www.dowcorning.com/content/publishedlit/26-1302-01.pdf.
[4] Neenah Paper Inc. Resources. Neenah Paper Inc. [Online]
http://www.neenahpaper.com/Resources/GlossaryTerms.
[5] MM Nazhad, EJ Harris, CTJ Dodson, RJ Kerekes. The Influence of Formation On Tensile
Strength of Paper Made From Mechanical Pulps.
[6] E. Epple & Co. Dictionary. E. Epple & Co. Website. [Online] http://www.epplechemie.de/en/dictionary/tack-free-time/.
[7] Westpak, Inc. Material Analysis Division. Westpak Website. [Online]
http://www.westpak.com/material-analysis/cobb.asp.
[8] de Gennes, P.G. (1985). "Wetting: statics and dynamics". Reviews of Modern Physics 57:
827–863.

75

Chapter 4: Scale-Up Trials and Economics
Performing coatings using silylated soy oil in a small lab-scale setting is an essential step
in creating a business plan for developing any potential large scale processes. While initial work
was performed to provide proof of concept data, the majority of the work in this thesis was
generated to provide better operating conditions for mass manufacturing operations of silylated
soy oil coatings. For the manufacturing scale-up process two criteria must be met for silylated
soy oil to be considered for use in a large-scale coating facility. First, the final product must
exhibit a certain degree of water resistance while being synthesized in large volumes. Second,
the product must be manufactured in a cost efficient manner.
Previous chapters have shown that silylated soy oil can be synthesized and coated
effectively at small scales. However, when considering large-scale production, several economic
factors must be considered such as raw material costs, utility costs, equipment costs, operation
costs, as well as several other factors. Trials at an industrial scale were performed to determine
product performance using significantly larger coating systems and to better evaluate the
potential economical model for producing a silylated soy oil coating.

4.1 Scale-Up Trials
To date, two scale-up test trials were been performed at Sierra Coating Technologies,
LLC in De Pere, WI. The initial trial was done to determine if similar performance properties
were achieved for both small and large-scale coatings.

The second trial tested product

performance when varying coat weight and line speed. Though several of the same synthesis,
analysis, and coating techniques were used at both lab and industrial scales, there were numerous
differences between each coating system that were considered.

76

4.1.1 Gravure Coating System
A gravure coating process has become one of the most popular types of coating
techniques used today due to its ability to apply less coating with a high degree of precision [1].
Gravure coatings differ from other commonly used coating techniques in that the coating roller is
patterned with a surface engraving, of which both the shape and size of the pattern can be varied
to achieve the final coat weight and specifications of the desired coating [2]. These engravings
vary in shape and size, thus providing different coat weight parameters. For a given gravure
pattern, there is only a small range which the coating thickness can be varied through changing
process parameters. Therefore, selecting the correct specification of the gravure pattern at the
outset is important [2].
Because a different coating method was used for large-scale trials, it was necessary to
determine how gravure coatings would perform with similar coat weights to studies performed
using a roller and blade coating apparatus with coat weights ranging from 3.5-6.0 gsm.
Additionally, coatings with heavier coat weights were performed to determine if significant
increases resulted in improved hydrophobicity.
4.1.2 Line Speed
Line speed is defined as the amount of material coated per time usually expressed in feet
per minute (fpm). Coatings made using the ECS were not tested for coating speed because only
one sheet of paper was coated at a time. At large scales however, rolls of paper thousands of feet
in length are used and coating speeds can be adjusted to vary coating properties and
manufacturing time. Most coating facilities provide pricing on a per time basis, thus operating at
faster line speeds allows more product to be made in less time, resulting in cheaper coating costs.
For both trials, line speed was set at 76 fpm. However, it is important to evaluate Cobb Values
with respect to line speed should in future trials to optimize coating production costs.
77

4.1.3 Oven Temperature
As discussed in sections 3.3.2 and 3.3.5.3, varying the cure or dry temperature can
significantly reduce the tack-free time of a coating. At small scales, dry temperatures at room
temperature, 60°C, and 80°C were evaluated. However, if line speeds are drastically increased,
coatings must exhibit a very short tack-free time. To achieve this, oven temperatures can be
increased to promote the removal rate of any undesired alcohol byproducts on the coating
surface. For both trials, oven temperatures were set to 125°C. When evaluating line speed in
future trials, it may be necessary to increase oven temperatures in an effort to reduce the tackfree time required of coatings made at faster line speeds.
4.1.4 Curing Formulation
Variation of the amount of catalyst and water can affect both pot-life and performance
properties of silylated soy oil coatings (section 3.3.3). Experiments done for all small-scale
studies used a formulation of 1wt% DBTDL and 4wt% water (weight percentage is in reference
to amount of silylated soy oil used). However, this formulation was observed to solidify in less
than an hour, rendering it unsuitable for large-scale coatings. Therefore, formulations with
reduced water concentrations were studied to determine how this would affect pot life. Water
concentrations of 1wt% and 0.1wt% with 1wt% DBTDL were measured for viscosity changes
over time.
4.1.4.1 Method – Varying Water Concentration for Pre-Coat Mixture
All pre-coating mixtures were made using 80g of silylated soy oil and 0.8g of DBTDL.
Water concentrations of 1wt% and 0.1wt% (0.8g and 0.08g, respectively) were measured for
viscosity at various time intervals until measurements exceeded 800 cP. Once each pre-coating
mixture was prepared, samples were placed on a magnetic stir plate mixing at approximately 500
rpm. At each measured time interval, stirring was stopped, the sample was manually stirred for
78

approximately five seconds to assure no surface film had developed and the viscosity was
measured using a Brookfield DV-E viscometer with an LV spindle set at a rate of 30 rpm. Each
concentration was done in triplicate and the viscosities for each time point were averaged.
4.1.4.2 Results & Discussion – Varying Water Concentration for Pre-Coat Mixture
Viscosities of pre-coating mixtures using 1.0wt% water show steady increases in
viscosity. The 800cP upper viscosity limit was reached in 16 minutes (Figure 4.1). Achieving
such an elevated viscosity in such a short time makes a 1.0wt% water formulation unusable for
large-scale trials. Conversely, pre-coating mixtures using 0.1wt% water maintain significantly
lower viscosities for much longer time. In fact, the average viscosity after three hours, the
minimum required pot life time for large-scale trials, was well below 300 cP.

900
800

Viscosity (cP)

700
600

1.0wt% Water

500

0.1wt% Water

400
300
200
100
0.00

1.00

2.00
Time (hours)

3.00

Figure 4.1 Viscosity measurements with respect to mixing time using 1.0wt% and 0.1wt%
water for pre-coating mixture formulations.
79

To demonstrate that the silylated soy oil demonstrated water resistant properties after
extended mixing times, coatings were made with the pre-coating mixtures at 0h, 1.5h, and 3h
mixing time points. All Coatings were cured for one hour at 80°C and 23 hours at room
temperature. After samples were cured, two minute Cobb Values were measured. Results from
Cobb Value measurements indicate that increased mixing time using a 0.1wt% water formulation
still exhibits improved water resistance (Table 4.1). Similarly, Cobb Values of coatings at
various mixing times show no statistically significant difference. Consequently, large-scale
coating trials were performed using 1.0wt% DBTDL and 0.1wt% water to ensure a low
viscosity. Furthermore, coatings could be made knowing coating performance would not change
with respect to mixing time.
Table 4.1 Coat weights and Cobb Values of coatings prepared at different mixing times using
0.1wt% water in pre-coating mixture.
0.1wt% Water using 5/29/12 Silylated Oil
SAMPLE
C1
C2
D1
D2
Average
Std. Dev.
E1
E2
F1
F2
Average
Std. Dev.
G1
G2
H1
H2
Average
Std. Dev.

Wt0 (g)
0.9756
0.9760
0.9654
0.9667

WtF (g)
1.1469
1.1519
1.1440
1.1457

0.9834
0.9648
0.9863
0.9758

1.1600
1.1414
1.1621
1.1570

0.9811
0.9866
0.9756
0.9691

1.1509
1.1583
1.1494
1.141

Cobb Value (gsm)
17.1
17.6
17.9
17.9
17.6
0.4
17.7
17.7
17.6
18.1
17.8
0.2
17.0
17.2
17.4
17.2
17.2
0.2
80

Mix Time (hrs)
0.0
0.0
0.0
0.0

1.5
1.5
1.5
1.5

3.0
3.0
3.0
3.0

Coat Weight (gsm)
3.48
3.48
3.81
3.81
3.65
0.19
3.81
3.81
3.81
3.81
3.81
0.00
3.81
3.81
3.98
3.98
3.90
0.10

4.2 Scale-Up Trial Results and Discussion
The initial scale-up trial was used as a preliminary test to determine if coating silylated
soy oil at an industrial level was achievable and also to determine if water resistant properties
were comparable to previous studies done at small scale lab experiments. This trial used paper
GN354872 (see Table 3.4 for properties). The second scale-up trial investigated how larger coat
weights would affect the performance properties of the coatings. The second trial used paper
GN505320 (see Table 3.10 for properties). Table 4.2 summarizes the manufacturing settings and
two minute Cobb Values of the coatings made at both trials.
Visual inspection of both coat weights from Trial #2 indicated that the coatings exhibited
partial saturation into the paper. The partial saturation is likely due to not allowing the precoating mixture to develop enough viscosity prior to coating. Much like applying pure soy oil
onto a piece of paper would result in saturation, or bleeding, into the substrate, silylated soy oil
that hasn’t had enough time to develop a pseudo-crosslinked network will also exhibit some
saturation. As a result of the partial saturation, the Cobb Values of the coatings from Trial #2
were notably higher.
Table 4.2 Equipment settings and Cobb Values of coatings from first two trials.
Trial 1
Trial 2
Trial 2
Low Coat Weight
High Coat Weight
Coating Type
Line Speed (fpm)
Oven Temp (°C)
Paper Grade
Coat Weight (gsm)
Cobb Value (gsm)
Cobb Value Std. Dev.

Direct Gravure
76
125
GN354872
4.0
18.9
2.1

Direct Gravure
76
125
GN505320
4.8
23.3
0.1

81

Direct Gravure
76
125
GN505320
8.0
24.0
0.6

4.2.1 Grease Resistance of Scale-Up Trial Coatings
Grease resistance testing is used to measure the repellency of paper to grease, oil, and
waxes. Typically this test is used for products treated with fluorochemical sizing agents which
impart both organophobic and hydrophobic characteristics to paper through a reduction in the
surface energy of the sheet [3, 4].

The test was developed to allow paper producers to

qualitatively measure the application of fluorochemicals into a sheet. By testing reagents with
slight variations in “aggressiveness”, the highest numbered solution that remains on the surface
of the paper without exhibiting any absorption provides the “kit rating” of the paper. The higher
a paper’s rating, the more grease resistance it demonstrates. For trash bag production, it’s
important to consider the grease resistance rating due to the contents that it may be exposed to.
Grease absorption, like water absorption, can reduce the strength of the paper making it more
susceptible to ripping, tearing, or breaking. For this reason, grease resistance was measured for
all scale-up trials.
4.2.1.1 Method – Grease Resistance Measurements of Scale-Up Trial Coatings
Grease resistant reagents and analysis were done according to TAPPI T 559. Each
coating was repeated five times. The highest reagent number that resulted in no evidence of
saturation was considered as “passed. The average grease resistance value was calculated to the
nearest 0.5 units.
4.2.1.2 Results & Discussion – Grease Resistance Measurements of Scale-Up Trial Coatings
All coatings from both trials showed no grease resistance (0.0 units). Thus, it may be
necessary to use a secondary, outer coating in conjunction with a silylated soy oil base coating in
order to achieve the desired water and grease resistance for trash bag production. Alternatively,
incorporating fluorochemical additives to the paper source directly could also provide improved
grease resistance, eliminating the need for a secondary coating. However, as mention briefly in
82

section 3.6.6.2, fluorinated additives in paper are associated with negative environmental
impacts. Therefore, using a non-fluorinated secondary coating may provide the best option for
achieving a high degree of water and grease resistance while having very little negative
environmental impacts.

4.3 Scale-Up Trial Economic Evaluation
Economic evaluations were completed for the processes in the production of silylated soy
oil coated trash bags including paper, silylated soy synthesis, coating, and bag production costs.
Calculations were made assuming one million bags were manufactured each year.

Bag

dimensions were set at 18”x12”x36” with a coat weight of 7.0 gsm. As a result, 1.5 million
square meters of paper are required (Table 4.3). Paper prices were quoted at $0.15/m2 from
Wausau Paper resulting in a paper cost of $220,000. Assuming a transportation cost of $0.11/kg,
and a paper basis weight of 80 gsm, the total paper costs were calculated to be approximately
$240,000 (Table 4.4).
Table 4.3 Coating and bag production specifications including coating settings from full-scale
trials, estimated production volume, bag sizes, and paper requirements.

COATING AND BAG SPECIFICATIONS
coating type
line speed (fpm)
oven temp (°C)
roll width (in.)
# of bags/yr
bag dimensions

direct gravure
76
125
60
1000000
18"x12"x36"

bag size (SA sq. in.)
paper/year (sq. in.)
paper/year (sq. ft.)
paper/year (sq. m.)
roll length needed (in.)
roll length needed (ft.)

2376
2376000000
16500000
1532900
39600000
3300000

83

Table 4.4 Estimated paper costs including purchasing and transportation to coating facilities
based off of production demands highlighted in Table 4.3.

PAPER COSTS
cost of paper ($/sq. yd)
cost of paper ($/sq m)
total paper cost
transportation of coated paper to bag maker ($/kg)
mass of paper (kg)
total transportation cost to bag maker ($)

$
$
$
$
$

0.12
0.15
223,803
0.11
122632
13,518

TOTAL PAPER COSTS

$

237,321

Knowing the total paper needed for bag production and assuming a coat weight of 7.0
gsm, the amount of silylated soy oil required per year is almost 11,000 kg. Based off of the mass
ratios for a 1:3 molar ratio of soy oil:VTMOS shown in Table 2.1, the required volume of each
reactant was calculated. Costs for VTMOS were provided by Gelest, Inc., soybean oil costs
were provided by Zealand Farms, and L101 costs were provided by Sigma Aldrich. Raw
material costs associated with silylated soy oil synthesis were estimated to be $120,000 (Table
4.5). Assuming a processing cost of $1000/100 gallons produced, total synthesis costs were
calculated to be $150,000, with the cost of VTMOS accounting for more than $75,000. Because
VTMOS accounts for half of the synthesis costs, reducing the reactant ratio would greatly reduce
material costs. Incorporating the cost of DBTDL needed for curing (Table 4.6), the total raw
material costs amount to approximately $175,000.

84

Table 4.5 Estimated raw material, processing, and production costs associated with the synthesis
of silylated soy oil based off of production demands outlined in Table 4.3.

SILYLATED SOY OIL SYNTHESIS COSTS
moles soy oil
mass soy oil/mole (kg)
moles silane
mass VTMOS/mole (kg)
mass L101/rxn (kg)
low sat soy oil price ($/kg)
soy oil wt% of rxn mixture
low sat soy oil needed (kg)
low sat soy oil cost total ($)
VTMOS price ($/kg)
VTMOS wt% of rxn mixture
VTMOS needed (kg)
VTMOS cost total ($)
L101 price ($/kg)
L101 wt% of rxn mixture
L101 needed (kg)
L101 cost total ($)
Processing @ $1000/100gal ($/kg)
total processing cost @ $1000/100gal
TOTAL SYNTHESIS COSTS ($)

$
$
$

1
0.9
3
0.148
0.009
1.81
66.48%
7134.0
12,896
22
32.85%
3524.9
77,548
370.58
0.66%
71.3
26,437
2.90
31,153

$

148,035

$

$

$

The coating manufacturing price for this project was estimated based off of settings and
results from the trials done at Sierra Coating Technologies. The coating price was quoted at
2

$0.20/m resulting in a coating cost over $300,000 (Table 4.7). Ultimately, coating production
was found to be the most costly step in the bag production process. Running the coatings at an
increased line speed could reduce coating production costs, however, significantly influencing
the overall manufacturing costs.

Therefore, evaluating the production limits and coating

performance properties at higher line speeds should be investigated.

85

Table 4.6 Estimated raw material costs for curing step during coating process. Quantities were
based off of production demands outlined in Table 4.3.

SILYLATED SOY CURING COSTS
wt% DBTDL used for curing
mass DBTDL used for curing (kg)
wt% water used for curing
mass water used for curing (kg)
DBTDL price ($/kg) [sigma]
DBTDL needed (kg)
DBTDL cost total ($)
TOTAL CURE COSTS ($)

$

1%
107.3
4%
429.2
340.00
107.3
36,483.01

$

36,483.01

$

Table 4.7 Estimated coating production costs based off of production demands in Table 4.3.

COATING COSTS
coating price ($/1000 sq. in.)
coating price ($/sq. m.)
coating weight (gsm)
coating material needed/yr (g)
coating material needed/yr (kg)
coating cost/year ($)
transportation of coating to coater ($/kg)
total transportation cost to coater ($)

$
$

$
$
$

0.13
0.20
7
10730297
10730
306,266
0.11
1,183

TOTAL COATING COSTS

$

307,449

Costs for bag production were quoted at $41/1000 bags by Wisconsin Converting Inc.
Therefore yearly bag production costs would be $41,000. Accounting for paper costs, raw
material costs, silylated soy oil synthesis, coating costs, and bag production, production of one
million bags is estimated to be $800,000, or $0.81 per bag (Table 4.8). Though these are
estimated costs, reducing the amount of VTMOS used for silylated soy oil synthesis and
increasing the coating line speeds could significantly improve the proposed economic model.

86

Table 4.8 Estimated manufacturing, synthesis, and processing costs for production of one
million silylated soy oil trash bags.

RAW MATERIAL & PRODUCTION COSTS
Paper Costs
Silylated Soy Synthesis Costs
Coating Manufacturing Costs
Curing Costs
Bag Production
Total Costs
Total Costs w/ 5% waste

$
$
$
$
$
$
$

237,320.96
148,035.16
307,449.06
36,483.01
41,000.00
770,288.19
808,802.59

COST PER BAG

$

0.81

87

REFERENCES

88

REFERENCES

[1] Davis-Standard, LLC. Publications & Documents. Davis-Standard Website. [Online]
http://www.er-we-pa.de/public_html/Company/pubs/gravureCoater.html.
[2] Kapur, N, et al. A Review of Gravure Coating Systmes. s.l. : Convertech & e-Print, 2011 .
2185-6931 .
[3] Rengel, G.L. and Young, R.C., “Internal Sizing with Fluorochemicals for Oil Resistance,” in
Swanson, J.W. (ed.), TAPPI Monograph 33 – Internal Sizing of Paper and Paperboard: 170-188,
TAPPI, NY (1971).
[4] Char, R.M., and Schwartz, C.A., “Fluorochemical Sizing,” in Reynolds, W.F. (ed.), The
Sizing of Paper (2nd ed.): 87-101, TAPPI Press, Atlanta (1989).

89

Chapter 5: Recommended Future Work
Because of the numerous aspects of this project that came into consideration when
developing a marketable finished product, many ideas were not fully investigated due to time
constraints. While the research highlighted in this thesis provides an outline for synthesizing,
coating, and curing silylated soy oil in an effective and efficient manner, there are several small
experiments that could be investigated that may provide further insight into enhancing the
performance, economics, energy, and time demands in the development of silylated soy coatings.
Proposed investigations for synthesis, coating, and scale-up projects have been included, each of
which would have a direct impact on the performance, cost, or time required to produce and coat
silylated soy oil more efficiently.

5.1 Proposed Synthesis Experiments
5.1.1 Variation of Reactant Molar Ratios
Multiple reactions were done in a 2L Parr Reactor using a 1:1 molar ratio of soy oil and
VTMOS (900g soy oil and 148g VTMOS).

Each reaction resulted in a product with a

significantly elevated graft percentage and viscosity. As a result of the high viscosities, the
products were unable to be used for large scale use due to the constraints of the gravure coating
system used for initial trials. If a different coating system was used for large-scale production
trials that was capable of handling higher viscosity demands, this 1:1 molar product could be a
viable option for drastically reducing raw material costs by cutting VTMOS use by 66%
compared to the 1:3 molar ratio. Most importantly though, would be to demonstrate similar
coating performance properties, as well as reproducibility at a small scale prior to large-scale
trials.

90

Interestingly, the theoretical minimal molar ratio that could result in a complete curing of
silylated soy oil is 3:1 molar ratio soy oil:VTMOS (900g soy oil and 49g VTMOS). This is
because of the triglyceride structure of soybean oil. Because each soy oil molecule has three
fatty acid chains, it is only necessary to graft one vinyl silane molecule to any one of the fatty
acid chains in order to achieve a completely crosslinked network. This reactant ratio would
provide an 89% reduction in VTMOS required if it fell within viscosity limitations and was
shown to provide similar coating properties. Reducing the amount of the cost-determining
reactant in the synthesis process would drastically reduce production costs.
5.1.2 Variation of Reaction Time and Temperature
Earlier work investigating the synthesis of silylated soy oil evaluated the affects of
varying time and temperature (see Figure 2.3), however this study was done using a 1:7 molar
ratio of soy oil:VTMOS [1, 2]. The same principles and kinetics should apply to different
reactant ratios though. Therefore, one way to counteract the high viscosities associated with 1:1
molar reactions could be to reduce the reaction temperature or time. This would be beneficial for
two reasons. First, it would reduce reaction time allowing more product to be made in a shorter
amount of time. Second, a lower temperature would reduce the energy demands required of such
a lengthy reaction. Operating at lower temperatures would help minimize processing costs.
Several settings varying the reaction conditions would have to be done to provide
sufficient data, though. And because the reaction would still be rather lengthy, this could be
quite time consuming since multiple reactions would need to be performed for each reach
condition. However, proper evaluation of the affects of reaction time and temperature could
greatly improve operational costs for any large-scale production.

The major risk of reducing

the reaction time and temperature is that the product would have a very low graft percentage.
91

When used for coatings, this could result in a lower crosslink density which could reduce the
water resistance. Ultimately, in order to properly evaluate how reaction temperature and reaction
time impact coating performance properties, numerous reactions must be carried out.

5.2 Proposed Coating Experiments
5.2.1 DBTDL & Water Concentrations
As was discussed in section 3.3.3 and 4.1.4, the concentrations of water and catalyst
significantly impact the crosslinking rate of silylated soy oil (see Figure 4.1). However, due to
the lengthy pot-life time required for scale-up trials, each data point for catalyst and water
concentrations took three hours to collect. Additionally, each water and catalyst concentration
was repeated a minimum of three times. As a result, thorough analysis of the affects of both
DBTDL and water concentrations would take a considerable amount of time.
Analysis of how varying the concentrations of DBTDL and water would be beneficial for
several reasons. Being able to accurately predict viscosity as a function of both [DBTDL] and
[water] would allow silylated soy oil coatings to be mixed and coated on any coating system,
regardless of its viscosity limitations. Having pot-life data with respect to [DBTDL] and [water]
would allow manipulations to be made to the pre-coating mixture that would satisfy the viscosity
limits and pot-life time for any coating system.
Another benefit of analyzing the affects of [DBTDL] and [water] used in the curing
process is how this changes water and grease resistant coating properties. Making coatings with
different pre-coating mixture concentrations may result in different Cobb and grease resistance
results.

Consider the affects of using a high water concentration with a low DBTDL

concentration. The high water content would likely increase crosslink density, while a low

92

DBTDL content would cause the cure process to occur more slowly. While this is theoretical,
the potential benefits of varying DBTDL and water concentrations are very intriguing. Likewise,
reducing the amount of DBTDL used in the pre-coating mixture would substantially reduce raw
materials costs.
5.2.2 Use of Paper GN354872 with High Basis Weight
As shown in section 3.6.1, paper GN354872 provides drastic water resistivity
improvements when coated with silylated soy oil. For instance, comparing two minute Cobb
Values of both GN354872 and GN505320, the paper used for all large-scale trials, it can be seen
that GN354872 experiences a Cobb Value improvement over 220% (62.4 gsm to 19.1 gsm)
whereas GN505320 demonstrated a Cobb Value improvement of 70% (43.2 gsm to 25.5 gsm).
Though GN354872 provides significantly better hydrophobic improvement than GN505320, due
to the low basis weight at which it was manufactured, the paper was not further investigated for
implementation as a coated trash bag product. However, if GN354872 was manufactured at a
higher basis weight suitable for use in trash bag production, this may provide the desired strength
properties needed for trash bags while also providing better hydrophobic properties when
compared to GN505320.
Another factor that must be considered is the difference in paper costs between
GN354872 and GN505320.

If the cost of using GN354872 is significantly higher than

GN505320, having slightly improved water resistant trash bags may not be justified from a
monetary standpoint. While it’s important to make a high quality product, it is equally important
to make it economically.

93

5.2.3 Use of Secondary Coating with Silylated Soy Oil
One issue that has become a concern in the development and testing of silylated soy oil
coatings is the lack of grease resistance that it provides. This is of particular concern due to the
contents that the coating would likely be subjected to when used for trash bags. A common
practice in coating industries, however, is to use different coatings in conjunction with one
another, each serving to improve one desired performance property. Therefore, applying several
different coatings provides a final product that exhibits all of the desired physical characteristics.
Using silylated soy oil with another coating that provides grease barrier properties could help
satisfy both the water and grease resistance desired of the final product. Some work has been
done by Michelman, Inc. evaluating the coating performance of silylated soy oil when used with
other coatings, but a much more in-depth study would be required to determine optimal coating
conditions when using two or more coatings in conjunction with one another.

5.3 Proposed Manufacturing Experiments
5.3.1 Different Coating System
All lab-scale coatings have been made using a blade coating system. The benefit of using
this simple coating method is little coating material is required, the coating weight can be very
easily controlled, and the maintenance and clean-up is quick with the use of a solvent such as
acetone or methanol. All full-scale coatings were made using a gravure coating system. This
method of coating has become widely used due to the ability to coat very thin layers at high line
speeds. The concept of a gravure roll coating system stems from the engravings on the roller
itself. A gravure roller traps and retains the coating material in its cells, or grooves. Excess
coating material is removed using a primary blade [3]. When the substrate comes into contact
with the roller, only the material captured in the grooves of the roller is applied to the surface.
Varying the dimensions of the grooves on the gravure roller allows for incredibly accurate ranges
94

in coat weight. However, due to the nature of silylated soy oil and its quick cure rate, and thus
quick viscosity increase in the presence of water and DBTDL, the grooves of a gravure roll can
become clogged and filled with gelled silylated soy oil. This results in an inconsistent coat
weight throughout the paper roll. Furthermore, it creates downtime in the manufacturing process
because the gravure roller has to be cleaned to remove any buildup of cured silylated soy oil in
the roller’s cells. As a result, the use of a gravure coating system may not be the most ideal
coating method for large scale coating production.
As mentioned in Chapters 3 and 4, pot-life also plays a significant role in determining the
feasibility of scaling up a coating material. Recent experiments have shown that high viscosity
silylated soy oil provides not only a better surface coating, with no leakage into the paper fibers,
but also provides better water resistance. Therefore, generating increased viscosities becomes
important to the performance of the coatings and also to the future research and development of
this project. Gravure coating systems are typically set up so that the coating material is being
pumped onto the roller directly. This arrangement forces the silylated soy oil, water, and catalyst
to be pre-mixed before being applied to the roller, which accounts for the concern over viscosity
and pot-life issues. However, if a blade and roller technique was used at large scale, increased
viscosities become much less of a concern. In fact, the use of a blade coating system would
allow the necessary amount of water to be added at the blade-roller interface in a continuous
process alleviating the concern of increased viscosity and pot-life. While a buildup of cured
silylated soy oil may accumulate on the blade itself, this could be quickly and easily removed by
simply scraping off the cured, unused excess silylated soy oil. Additionally, implementing largescale manufacturing using this coating technique would serve to closer replicate any results
experienced in lab using the ECS coater.
95

REFERENCES

96

97

REFERENCES

[1] Srinivasan, Madhusudhan. "Synthesis, Properties and Applications of Bio-Based Materials."
Michigan State University, 2010. United States -- Michigan: ProQuest. Web. 4 Apr. 2013.
[2] Zhuang, Yue. “Novel Synthetic Strategy To Produce Biobased Silylated Soybean Oil
Coating Material.” Michigan State University, 2011. United States – Michigan: ProQuest.
Web, 4 Apr. 2013.
[3] Powell, C. A.; Savage, M. D.; Gaskell, P. H., Modeling the meniscus evacuation problem in
direct gravure coating. Chem Eng Res Des 2000, 78 (A1), 61-67.

98