、 
 
 
 

SINGLE ION CONDUCTORS BASED ON POLYMER-MODIFIED
NANOPARTICLES: SYNTHESIS AND CHARACTERIZATION
By
Hui Zhao

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Chemistry
2012

ABSTRACT
SINGLE ION CONDUCTORS BASED ON POLYMER-MODIFIED
NANOPARTICLES: SYNTHESIS AND CHARACTERIZATION
By
Hui Zhao

Lithium ion battery has been widely used nowadays. Occurrence of polarization
has been a problem in the currently used electrolyte, especially in high-current
application such as hybrid vehicles. In this work we modified silica nanoparticles
with polyelectrolytes containing anion groups, lithium cation is the counterion. To
get the lithium conductor, the polymer modified nanoparticles were blended with
poly(ethylene oxide) oligomer. Movement of the anions were largely restricted by
surface anchored polymer backbones, lithium ion became the only ion that is
conducting current.
Four different polymer structures were explored in this study, aiming at
improving conductivity of the material. FTIR and TGA were used to prove
successful synthesis of the polymer modified particles. AC impedance was used
to measure the conductivity.

 
 
 
 

ACKNOWLEDGEMENTS

I would like to thank my advisor Professor Gregory Baker, who was always
patient and taught me all the basic knowledge of polymer synthesis. Many thanks
to my guidance committee, Dr. William Wulff, Dr. Mitch Smith and Dr. Merlin
Bruening, for helpful discussions.
Many thanks to Dan Holmes, Rui Huang, Kathy  Severin, Alacia Pastor for
teaching me different characterizing techniques. I would like to thank past and
current group members Qin, Sampa, Erin, Tom, DJ, Gina, Quanxuan, Wen, Heyi,
Yiding, Zhe and Greg. They are wonderful labmates and I enjoyed doing research
with them.
Last but not least, none of this would be possible without my Mom and Dad.
They have been very supportive and patient along the way. This thesis is
dedicated to my parents.

iii 
 

TABLE OF CONTENTS
 

LIST OF TABLES ................................................................................................. V
LIST OF FIGURES .............................................................................................. VI
LIST OF SCHEMES ............................................................................................ IX
LIST OF ABBREVIATIONS ................................................................................. XI
1. Introduction ....................................................................................................... 1
1.1 Lithium Ion Batteries.. ................................................................................... 1
1.2 Polymer Electrolytes ..................................................................................... 6
1.3 Conductivity Measurements........................................................................ 10
1.4 Lithium Transference Number .................................................................... 15
1.5 Electrolytes with High tLi+ in Literature ........................................................ 18

2. Results and Discussion .................................................................................. 27
2.1. Particles Modified with Lithium 4-Styrenesulfonate.. .................................. 29
2.2. Particles Modified with Lithium 4-Styrenesulfonate and PEGMA ............... 37
2.3. LiTFSI Analogure Polymer Modified Particles............................................ 44
2.4. Phosphonic Acid Modified Particles ........................................................... 53
3. Conclusions .................................................................................................... 60
4. Experimental Section ...................................................................................... 61
5. References ..................................................................................................... 71

iv 
 

LIST OF TABLES
 

Table 1. Conductivities of lithium salts used in lithium ion battery research ......... 3
Table 2. Examples of synthetic lithium salts ......................................................... 5
Table 3. Typical polymer electrolytes and their conductivities ............................ 10
Table 4. Weight content of LiTFSI in LiTFSI/PEGDME-500 at various O/Li ratios
-2

(2.4 x 10

g Li/ g LiTFSI) ................................................................................... 14

Table 5. Lithium transference numbers for typical lithium salts .......................... 16
Table 6. Particle weight content for electrolytes prepared from Si-PSSLi (2.4 
-2

10

g Li/g nanoparticles) dispersed in PEGDME-500........................................ 35

Table 7. Particle weight fraction for Si-PSSLi-PEGMA (1.56 x 10

-2

g Li/g sample)

dispersed in PEGDME-500 ................................................................................ 42
Table 8. Particle weight content for electrolytes prepared from Si-TFMA-Li (2.0 x
-2

10

g Li/g sample) dispersed in PEGDME-500 ................................................. 51

Table 9. Particle weight content for electrolytes prepared from Si-PPLi (3.45 x 10

-2

g Li/g sample) dispersed in PEGDME-500 ......................................................... 57
 

v 
 

LIST OF FIGURES
 

Figure 1. A schematic diagram of a lithium ion battery using metal oxide cathode
and graphite anode............................................................................................... 2

Figure 2. Side and top views of the 72 helix of PEO ............................................. 7

Figure 3. A typical Nyquist plot obtained from AC impedance spectroscopy ...... 12
Figure 4. Temperature dependent conductivity for LiTFSI/PEGDME-500
electrolytes at various O/Li ratios........................................................................ 14
Figure 5. Conductivity vs. O/Li for LiTFSI/PEGDME-500 at various temperatures
........................................................................................................................... 15
Figure 6. Polarization during discharge of a lithium ion battery .......................... 18
Figure 7. TEM image of Snowtex-S silica nanoparticles (samples were dispersed
in water dropped on a copper grid, and dried before taking the TEM image) ..... 29
Figure 8. FTIR spectra of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
grafted silica particles, and (c) Si-PSSLi, silica particles grafted with poly(lithium
4-styrene sulfonate) ............................................................................................ 31
Figure 9. Thermogravimetric Analysis (TGA) curves of (a) Snowtex-S silica
particles; stabilized by CTAB; (b) Si-Initiator, silica particles grafted with initiator,
and (c) Si-PSSNa, silica particles grafted with poly(sodium 4-styrene sulfonate),
and (d) Si-PSSLi, poly(lithium 4-styrene sulfonate) grafted silica particles ......... 32
Figure 10. TEM image of Si-PSSLi (samples were dispersed in water, dropped on
a copper grid, and dried before taking the TEM image) ...................................... 35
Figure 11. Temperature dependant conductivity for Si-PSSLi/PEGDME-500 at
various O/Li ratios............................................................................................... 36

vi 
 

Figure 12. Room temperature conductivity as a function of the O/Li ratio for
Si-PSSLi/PEGDME-500 electrolytes .................................................................. 37
Figure 13. FTIR spectra of (a) Snowtex-S silica particles, stabilized by CTAB; (b)
Si-Initiator, silica particles grafted with initiator, and (c) Si-PSSLi-PEGMA; silica
particles grafted with poly(lithium 4-styrenesulfonate) and PEGMA ................... 40
Figure 14. TGA curves of (a) Snowtex-S silica particles, stabilized by CTAB; (b)
Si-Initiator, silica particles grafted with initiator, (c) Si-PSSNa-PEGMA; silica
particles grafted with poly(sodium 4-styrenesulfonate) and PEGMA and (d)
Si-PSSLi-PEGMA; silica particles grafted with poly(lithium 4-styrenesulfonate) . 41
Figure 15. TEM image of Si-PSSLi-PEGMA (samples were dispersed in water,
dropped on a copper grid, and dried before taking the TEM image)................... 42
Figure 16. Temperature dependant conductivity of Si-PSSLi-PEGMA/
PEGDME-500 electrolytes at various O/Li ratios .............................................. 43
Figure 17. Room temperature conductivity vs O/Li ratio for Si-PSSLi-PEGMA
/PEGDME-500 .................................................................................................... 44
Figure 18. FTIR spectra of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
modified silica particles; (c) Si-TFMA-Li, lithiated poly(trifluoromethane sulfonic
aminoethylmethacrylate) modified silica particles ............................................... 48
Figure 19. FTIR spectra of (a) Si-TFMA, poly(trifluoromethane sulfonic
aminoethylmethacrylate) modified silica particles; (b) Si-TFMA-Li, lithiated
poly(trifluoromethane sulfonic aminoethylmethacrylate) modified silica particles ...
........................................................................................................................... 49
Figure 20. TGA curves of (a) Snowtex-S silica particles, stabilized by CTAB; (b)
Si-Initiator, silica particles grafted with initiator, (c) Si-TFMA, poly(trifluoromethane
sulfonic aminoethylmethacrylate) grafted silica particles; and (d) Si-TFMA-Li,
lithiated poly(trifluoromethane sulfonic aminoethylmethacrylate) grafted silica
particles .............................................................................................................. 50
vii 
 

Figure 21. Temperature dependant conductivity of Si-TFMA-Li/PEGDME-500
electrolytes at various O/Li ratios ...................................................................... 51
Figure 22. Room temperature conductivity as a function of the O/Li ratio for
Si-TFMA-Li/PEGDME-500 .................................................................................. 52
Figure 23. TGA curves of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
modified silica particle; (c) Si-PDEVBP, silica particle modified by poly(diethyl
4-vinylbenzylphosphonate); (d) Si-PPLi, poly(lithium 4-vinylbenzylphosphonate)
........................................................................................................................... 56
Figure 24. FTIR spectra of (a) Si-PDEVBP, silica particle modified by poly(diethyl
4-vinylbenzylphosphonate); (b) Si-PPLi, poly(lithium 4-vinylbenzylphosphonate)
........................................................................................................................... 57
Figure 25. Temperature dependent conductivity of Si-PPLi/PEGDME-500 at
various O/Li ratios............................................................................................... 58
Figure 26. Room temperature conductivity vs. O/Li ratio for Si-PPLi/
PEGDME-500 electrolytes .................................................................................. 59

Images in this dissertation are presented in color.

viii 
 

LIST OF SCHEMES
 

Scheme 1. Lithium transport mechanism in PEO described by Ratner et al ........ 8
Scheme 2. Typical aza-ether and boron derivatives in McBreen’s work. ............ 19
Scheme 3. A typical calixarene described by Wieczorek .................................... 19
Scheme 4. Boroxine containing polymer described by Mehta ............................ 20
Scheme 5. Two pre-polymer structures (top) and the boron crosslinker (bottom)
described by Kerr ............................................................................................... 22
Scheme 6. Polysiloxane comb polymers synthesized by Shriver ....................... 23
Scheme 7. Monomers described by Endo .......................................................... 24
Scheme 8. Synthetic approach to polyelectrolyte-decorated nanoparticles ........ 28
Scheme 9. Immobilization of initiators on silica particles and the following surface
ATRP with sodium 4-styrene sulfonate ............................................................... 30
Scheme 10. Lithiation of silica particles grafted by poly(sodium 4-styrenesulfonate)
(Si- PSSNa) ........................................................................................................ 31
Scheme 11. Quantitative determination of the Li content in nanoparticles ......... 34
Scheme 12. Copolymerization of sodium 4-styrenesulfonate and PEGMA from
silica Nanoparticles ........................................................................................... 38
Scheme 13. Method used to calculate the copolymer content in
Si-PSSNa/PEGMA modified nanoparticles ......................................................... 39
Scheme 14. Lithiation of Si-PSSNa-PEGMA ...................................................... 39
Scheme 15. A trifluoromethane sulfonic aminoethylmethacrylate (TFMA) inspired
by the chemical structure of LiTFSI .................................................................. 46

ix 
 

Scheme 16. Synthesis of 2-(trifluoromethanesulfonamido)ethyl methacrylate
(TFMA) ............................................................................................................... 46
Scheme 17. ATRP of trifluoromethane sulfonic aminoethylmethacrylate (TFMA)
from silica particles ............................................................................................. 47
Scheme 18. Lithiation of Si-TFMA to Si-TFMA-Li ............................................... 48
Scheme 19. Synthesis of diethyl 4-vinylbenzylphosphonate (DEVBP) ............... 53
Scheme 20. ATRP of diethyl 4-vinylbenzylphosphonate (DEVBP) from particle
surfaces .............................................................................................................. 54
Scheme 21. Lithiation procedure for poly(diethyl 4-vinylbenzylphosphonate)
grafted silica ....................................................................................................... 55

x 
 

LIST OF ABBREVIATIONS

σ

conductivity

AC

alternating current

ATRP

atom transfer radical polymerization

Bpy

bipyridine

CTAB

etyltrimethylammonium bromide

DEVBP

diethyl 4-vinylbenzylphosphonate

DMF

N, N-dimethylformamide

DSC

differential scanning calorimetry

EtOAc

ethyl acetate

FTIR

Fourier transform infrared spectroscopy

GPC

gel permeation chromatography

LIB
LiTFSI

Lithium Ion Batteries
lithium bis(trifluoromethane sulfone)imide

ppm

parts per million

PDI

polydispersity index

PDEVBP
PEGMA

poly(diethyl 4-vinylbenzylphosphonate)
poly(ethylene glycol) methyl ether methacrylate

PEG

poly(ethylene glycol)

PEO

poly(ethylene oxide)
xi 

 

PEGDME-500

poly(ethylene oxide) dimethyl ether (average
molecular weight 500)

PEM
PMDETA

proton exchange membranes
N, N, N’, N’, N’’-Pentamethyldiethylenetriamine

PPOH
PPLi

poly(4-vinylbenzylphosphonic acid)
poly(lithium 4-vinylbenzylphosphonate)

PSSLi

poly(lithium 4-styrenesulfonate)

PSSNa

poly(sodium 4-styrenesulfonate)

PTFMA

poly(trifluoromethane sulfonic aminoethylmethacrylate)

PVDF
Si-PDEVBP

polyvinylidene fluoride
silica particle modified by
poly(diethyl 4-vinylbenzylphosphonate)

Si-PPOH

Si-PPLi

silica particle modified by
poly(4-vinylbenzylphosphonic acid)
silica particle modified by
poly(lithium 4-vinylbenzylphosphonate)

Si-PSSLi

Si-PSSLi-PEGMA

silica particle modified by
poly(lithium 4-styrenesulfonate)
silica particle modified by co-poly(lithium
4-styrenesulfonate)-poly(ethylene glycol)
methyl ether methacrylate

Si-PSSNa

Si-PSSNa-PEGMA

silica particle modified by
poly(sodium 4-styrenesulfonate)
silica particle modified by co-poly(sodium
4-styrenesulfonate)-poly(ethylene glycol)
xii 

 

methyl ether methacrylate
Si-TFMA

Si-TFMA-Li

silica particle modified by poly(trifluoromethane
sulfonic aminoethylmethacrylate)
silica particle modified by lithiated poly
(trifluoromethane sulfonic aminoethylmethacrylate)

r.t.

room temperature

TEM
TFMA

transmission electron microscopy
trifluoromethane sulfonic aminoethylmethacrylate

TGA

thermal gravimetric analysis

Tg

glass transition temperature

TMSBr

trimethylbromosilane

TMSCl

trimethylchlorosilane

VFT

Vogle-Fulcher-Tamman

xiii 
 

1. Introduction
1.1.

Lithium Ion Batteries

Since rechargeable lithium batteries were proposed in the 1970s, the
development of electrolytes appropriate for lithium ion batteries has been an
1

active research field. High energy density batteries based on lithium is attractive
since it is the lightest metal (equivalent weight M = 6.94 g/mol, and specific gravity
3

ρ = 0.53 g/cm ) and the most electropositive metal (-3.04 V versus standard
hydrogen electrode). Initially, the emphasis was on batteries with metallic lithium
anodes to maximize the battery capacity, but metallic lithium posed significant
safety issues. To avoid the use of lithium metal, the “rocking-chair” concept

2

was

introduced, where ions are shuttled between two interaction compounds, usually
lithium intercalated graphite as the anode and a metal oxide as cathode. (Figure
1).

1 
 

Figure 1. A schematic diagram of a lithium ion battery using metal oxide cathode
and graphite anode.
Electrolytes in commercial lithium ion batteries are normally comprised of
lithium salts dissolved in highly polar solvents (high dielectric constant), such as
ethylene carbonate (EC), propylene carbonate (PC), dimethyl carbonate (DMC),
and derivatives of tetrahydrofuran (THF). The basic requirement for a salt is that
it has a low lattice energy, which results in a high degree of dissociation in the
solvent.

Both

the

salt

and

solvent

must

chemically,

thermally,

and

electrochemically stable throughout the voltage window of the cell. Parameters of
commonly used lithium salts are listed in Table 1.

2 
 

3

a

o

Tm ( C) σ (mS/cm in PC)

σ (mS/cm in
1:3 EC/DMC)

LiClO4

236

5.5

8.4

LiBF4

293

4.6

4.9

LiPF6

340

5.8

10.7

LiAsF6

340

5.7

11.1

LiCF3SO3

>300

2.4

2.8

Li[N(SO2CF3)2] (TFSI)

234

5.1

7.2

Li[N(SO2CF2CF3)2]

>350

4.4

a

7.8

a

All measurements at room temperature; PC = propylene carbonate, EC =

ethylene carbonate, DMC = dimethyl carbonate; all salt concentrations are 1M.
3

Table 1. Conductivity of lithium salts used in lithium ion battery research.
The Ionic conductivity (σ) at a specific temperature is often expressed as

　  =n  q   　
where n is the number of charge carriers per unit volume, q is charge of the carrier
and  is the mobility of the ions. Ion mobility is related to the degree of
dissociation, ion size, charge distribution, and the degree to which ions aggregate.
For lithium salts, these requirements are usually satisfied with large anions where
the charge is highly delocalized. A number of such anions have been synthesized,
as shown in Table 2. Most of the anions in Table 2 are large and have electron
3 
 

withdrawing groups that are designed to delocalize the negative charge, decrease
the electrostatic interaction between lithium cation and anion, and increase the
degree of dissociation.
The use of organic solvents has some limitations. Degradation of organic
solvents at electrode surfaces forms the solid-electrolyte-interphase (SEI). If the
SEI is mechanically stable and has high ionic conductivity, degradation is limited
to the cell’s first few cycles. However, unstable SEIs can lead to consumption of
the electrolyte and decreased cell capacity. Volatile degradation byproducts can
cause an increase in cell pressure necessitating high-pressure packaging. Also,
some electrolytes are corrosive and react with seals and containers.

4 
 

Table 2. Examples of synthetic lithium salts.

5 
 

1.2.

Polymer electrolytes

The use of solid electrolytes such as poly (ethylene oxide) (PEO) could address
the above issues of using organic solvents. Solid-state electrolytes would
eliminate the need for polypropylene or polyethylene mesh separators that
required with liquid electrolytes, reducing cost and potentially increasing the
energy density of the battery. Polymer electrolytes may inhibit lithium dendrite
formation and provide flexibility in processability. Also, since mass transport of
polymers is minimal, polymers should form robust SEIs. Wright and coworkers
reported ionic conductivity for Na and K salts complexes of poly(ethylene oxide)
4

(PEO) in 1973, but the technological importance of this solid electrolyte was not
realized until 1978, when Armand described a solid state lithium cell with a
PEO-based electrolyte.

5

PEO, poly(ethylene glycol) (PEG), and poly(oxyethylene) (POE) are common
names for a polymer with the same repeating unit, -OCH2CH2-. Ring opening
polymerization of ethylene oxide by anionic or cationic processes provides PEO of
predictable molecular weights, based on the monomer to initiator ratio. At room
temperature, PEO is a waxy white solid material comprised of an amorphous and
a crystalline phase. The glass transition temperature (Tg) of high molecular weight
amorphous PEO is ~ -60 oC, while the crystalline phase forms a 72 helix (Figure
2), seven ethylene oxide repeating units completing two turns in the fiber period of

6 
 

6

1.93 nm. The melting point (Tm) of PEO depends on the molecular weight. The
Tm for high molecular weight PEO is ~ 67 oC, but as the molecular weight
decreases, the PEO melting point decreases and low molecular PEO is a viscous
liquid at room temperature. Also, the effect of the polymer end groups becomes
more important in low molecular PEO.

1.93

 
Top View

6

Figure 2. Side and top views of the 72 helix of PEO. Reprinted with permission
from ref 6. Copyright 2012 American Chemical Society. (For interpretation of the
references to color in this and all other figures, the reader is referred to the
electronic version of this thesis.)
+

The generally accepted model for of Li in PEO transport is that of Ratner and
co-workers. Lithium cations coordinate to the lone pairs of electrons in oxygen
7 
 

atoms, forming a 4-8 oxygen chelate. When an external voltage is applied, lithium
ions hop to a neighboring chelation site, leading to a net flow of charge (Scheme
1.). Because the movement of ions is directly coupled to polymer chain mobility,
this

behavior

is

thermally

activated

and

is

often

described

by

8-10

Vogel-Tammann-Fulcher (VTF) equation

σ0=σ0exp-Ea(T-T0)
where σ0 is the pre-exponential factor related to number of charge carriers, Ea is
the apparent activation energy for ion transport, and T0 is a parameter related to
chain mobility of the polymer. For polymers, low Tgs should correlate to fast
relaxation and high conductivity. Since meaningful lithium transport is limited to
the amorphous regions, the research described in this thesis used poly (ethylene
oxide) dimethyl ether (Mn = 500) as a model for PEO.

7

Scheme 1. Lithium transport mechanism in PEO described by Ratner et al.

The ionic conductivity of solid polymer electrolytes such as PEO/LiClO4 or
-7

-5

PEO/LiTFSI (Lithium Bis(trifluoromethane sulfonyl)imide) are 10 -10

8 
 

S/cm at

room temperature, several orders of magnitude lower than conventional liquid
electrolytes (0.001-0.1 S/cm). The low conductivities should be expected since
ion mobility is coupled to polymer chain mobility. To obtain high conductivity in
“polymer-like” materials, gel polymer electrolytes were developed where a liquid
electrolyte is encapsulated in a non-conducting polymer host. These two-phase
materials appear to be solid state electrolytes but have the conductivity of liquid
electrolytes.
Alternatives to PEO have been synthesized, and representative examples are
shown in Table 3. Oxygen or nitrogen atoms are included in these polymers to
provide suitable sites for Li+ coordination, and branching is often used to prevent
crystallization. Using low Tg polymer backbones such as phosphazenes and
siloxanes tends to enhance conductivity. Polymethacrylates and acrylates comb
architectures also have low Tgs when the “teeth” are PEG side chains.

9 
 

Polymer Host

poly(ethylene oxide)
(PEO)

Repeat Unit

O

poly(propylene oxide)
(PPO)

(PEO)8-LiClO4

10

n

O

(POM)-LiClO4

10

(PPO)8-LiClO4

10

DMS-LiClO4

10

O

poly[methoxy poly(ethylene
oxide) methacrylate]

n
Si

O n

n

O

PMG22-LiCF3SO3
EO:Li = 18:1

O
O

poly(ethylene imine)
(PEI)

Conductivity
(S/cm), 20 °C

n

poly(oxymethylene)
(POM)

poly(dimethyl siloxane)
(PDMS)

Polymer Electrolyte

-8
-8

-8

-4

3*10

-5

x

H
N
n

PEI-LiClO4

10

-3

2
O
poly[bis-2-(2-methoxyethoxy)
ethoxyphosphazene] (MEEP)

2

1.3.

2*10

(MEEP)4-LiTFSI

P N n
O

Table 3.

-5

(MEEP)4-LiBF4

5*10

(MEEP)4-LiC(CF3SO2)2

Typical polymer electrolytes and their conductivities.

10

-5

-4

11

Conductivity measurements

Ohm’s law describes the potential (E) as a function of current (I) and resistance
(R)

E  IR
10 
 

2

It is know that in a bounded area with area A (in cm ) and length l (in cm) the
resistance is defined as

R  l / A
Where ρ is the resistivity of the electrolyte material in Ωcm, and

  R( A / l )
conductivity σ is the reciprocal of resistivity

 1/ 
so

  l / RA
unit of conductivity σ is 1/ Ωcm, which is S/cm (S represents Siemens).
Ionic conductivity of polymer electrolytes (or polyelectrolytes) was measured
using Alternating Current (AC) Impedance Spectroscopy. In direct current (DC)
circuit, resistance was used to describe the ability of circuit to resist the flow of
electrical current; impedance extended this concept to AC circuit. In a real AC
Impedance experiment, the electrolyte material was sandwiched between two
electrodes, a sinusoidal voltage was applied to this sample cell, a sinusoidal
current would be correspondingly detected. Frequency of the applied voltage
ranged from 5 mHz to 1.3 MHz in this work. The sample cell here (including
electrolyte material and two electrodes) responded to the applied AC voltage by
generating a current, which was contributed from two parts: electrolyte material
itself (as a resistor), and the interface between electrolyte and electrode (as a
11 
 

capacitor). Since electrodes used to build up the sample cell are much more
conductive then the electrolyte material, resistance of electrodes was ignored. If
there was only resistor in this circuit, the resulting current would always be in
phase with the applied sinusoidal voltage, and the magnitude of the impedance
was given by resistance, which is independent of frequency; if there was only
capacitor in the circuit, voltage lags behind the current by 90o. For the
convenience of calculation, impedance could be expressed as a complex

Z  Z ' jZ ''
Z’ is the real part ( Z '  Z cos ), Z’’ is the imaginary part ( Z ''  Z sin  ) and j is
the imaginary symbol. When data were plotted using X axis as real part, Y axis as

Figure 3. A typical Nyquist plot obtained from AC impedance spectroscopy.
12 
 

imaginary part, the diagram was called a Nyquist plot. Frequency was not
included in Nyquist plot, which was a major shortcoming. The bulk resistance of
material (R in Ω) could be obtained from the Nyquist plot at the specific frequency
where imaginary component is zero, which corresponds to the point where the
curve intercepts X axis (Figure 3).
Figure 4 shows typical temperature-dependent conductivity data for
electrolytes formed by dissolving lithium bis(trifluoromethane sulfonyl)imide
(LiTFSI) in polyethylene glycol dimethyl ether (PEGDME-500, average Mn ~500)
at various O/Li ratios. As expected for thermally activated transport, the
conductivity increases with temperature over the entire O/Li ratio range. The
physical properties of the electrolyte also change with the O/Li ratio. For example,
electrolytes at high O/Li ratios are low viscosity liquids at room temperature while
a sample with a O/Li ratio of 8 (45 wt% of LiTFSI) is very viscous. LiTFSI has a
molecular weight as high as 287 g/mol, this means every 287 g of LiTFSI has 7 g
of lithium, giving a lithium content of 2.4 x 10

-2

g Li/ g LiTFSI. This number is used

to convert O/Li ratio to the corresponding LiTFSI content (wt%) in each mixture,
as show in Table 4.

13 
 

O/Li Ratio

8

16

32

64

96

LiTFSI Content (wt%) 44.9 29.0 16.9 9.2 6.4

128 256
4.8

2.5

Table 4. Weight content of LiTFSI in LiTFSI/PEGDME-500 at various O/Li ratios
-2

(2.4 x 10

g Li/ g LiTFSI).

Figure 4. Temperature dependent conductivity for LiTFSI/PEGDME-500
electrolytes at various O/Li ratios.

14 
 

Figure 5. Conductivity as a function of the O/Li ratio for LiTFSI/PEGDME-500
electrolytes at various temperatures.
Figure 5 shows the same data set from Figure 4 plotted as conductivity as a
function of the O/Li ratio. The data show that the conductivity increases with the
lithium salt concentration, and then decreases after O/Li ratio of 8. These trends
are often ascribed to decreased chain mobility at high salt concentrations and
increased polymer Tg. (Attempts to prepare samples with an O/Li ratio of 4 (62 wt%
LiTFSI) did not result in useful samples since the LiTFSI did not fully dissolve in
the PEGDME-500 matrix.)
1.4 . Lithium transference number
In a lithium ion battery, the electrolyte is comprised of a lithium salt dissolved in
a solvent. As the battery is discharged, electrons flow from the anode to the
15 
 

cathode through the external circuit, while inside the cell, lithium cations migrate
from the anode to the cathode and intercalate into the cathode. Typical
measurements of the Li+ transference number (tLi+) in PEO-based electrolytes
range from 0.2-0.4 (Table 5), which indicates the anion as the dominant species in
Li+ transport.
Lithium
Salt

Polyether (or
solvent system)

O/Li
(concentration)

Temperature

LiClO4

PEO1000

20:1

30 oC

a

LiCF3SO3

PEO1000

24:1

30 oC

a

LiBF4

PEO500

20:1

70 oC

b

LiPF6

PC/EC/DMC
(1:1:1)

1M

Room
Temperature

a
b

t+
0.23
0.21
0.32

b

0.38±0.04

data collected using Tubandt Technique
data are collected using galvanostatic polarization method
Table 5. Lithium transference numbers for typical lithium salts.
Determining accurate lithium transference values from the alternating current,

direct current polarization, and Hittorf/Tubandt methods has been challenging.
Recently, two methods were introduced, the DC polarization method developed
by Bruce and Vincent,

12

and pulsed field gradient (PFG) measurements. The

latter provides ion diffusion rates, which can differ from transference numbers
obtained from the DC polarization method.
The DC polarization experiment, also called steady-state method, uses a

16 
 

sample cell comprised of the electrolyte sandwiched by two lithium metal
non-blocking electrodes. A potentiostat is used to apply a small DC polarization
voltage (△V) across the sample, and the initial (I0) and final steady-state (Is)
currents are determined. AC impedance spectroscopy is used to determine the
electrolyte resistances before (R0) and after (Rs) the potentiostatic measurements.
The lithium cation transference number (tLi+) of the electrolyte samples is
calculated from the following equation.

The significance of a low t+ is polarization of the lithium ion cell during charging
and discharging. Clustering of cations and anions forms a concentration gradient
that limits the current during charge and discharge (Scheme 2). The resulting
internal resistance generates heat during the charge and discharge process,
limiting the power density of the battery and often affects the chemical stability of
electrolyte. Thus, the effective voltage during discharge is V (cell voltage) - Vp
(polarization voltage), and the voltage needed to charge the cell is V+Vp.

17 
 

Figure 6. Polarization during discharge of a lithium ion battery.
Besides polarization, the conductivity of polymer lithium salt complexes tends to
reach a maximum and then decrease as the lithium salt concentration increases.
At high salt concentrations, triple or higher ionic aggregates form, decreasing the
number of charge carriers, conductivity, and cell efficiency. Additives that
preferentially interact with anions have been explored to increase the lithium ion
transference number in electrolytes. Alternatively, the anions can be immobilized
on polymers or on particle surfaces.
1.5 . Electrolytes with high tLi+ in literature
McBreen
aza-ether

13,14

and

coworkers

synthesized

anion

receptors

based

and fluorinated boron compounds (Scheme 2).

15

These

compounds were used as additives in aprotic polar solvents, THF for aza-ether
18 
 

on

F 3C
R

R

N

N

N

N

CF3

R

R

F3C

R=CF3SO2

O

O
B

CF3
O

F 3C

CF3

13, 14

Scheme 2. Typical aza-ether and boron derivatives in McBreen’s work.

derivatives, 1,2-dimethoxyethane for boron derivatives. While the receptors
successfully complexed bromide and chloride anions, the corresponding lithium
salts (LiBr, LiCl) are not typically used in lithium ion batteries.

Scheme 3. A typical calixarene derivatives in Wieczorek’s work.
Wieczorek and coworkers synthesized calixarene

16,17

16

and calixpyrrole

18,19

derivatives for use as anion-binding additives in PEO-lithium salt electrolytes. The
lithium transference number, measured by the steady-state technique, increased
from 0.1 to 0.6, depending on the concentration of the additives and the
temperature.

19 
 

Scheme 4. Boroxine containing polymer in Mehta’s work.

21

In conventional lithium salts, such as Li(CF3SO2)2N, LiCF3SO3, LiBF4, LiClO4,
LiBr, LiCl, LiPF6, the anions are intrinsicly Lewis bases. Mehta et al. constructed a
PEO gel with boroxine rings as a cross-linker.

20,21,22

The Lewis acidic boroxine

rings interact with anions and suppress their mobility. Lithium transference
numbers in these systems range from 0.6-0.9, compared to 0.2-0.3 for normal
PEO-lithium salt systems. The increase in the lithium transference number in
Mehta’s system is high, considering the simplicity of its synthesis. The maximum
-5

room temperature conductivity was ~10

S/cm.

Using the concept of trapping anions, Angell and co-workers synthesized
“base-in-chain” and “acid-in-chain” polymers. Incorporating sulfonylimide groups
into the polymer backbone gave a base-in-chain polymer. AlCl3 was added to
reduce the dissociation energy between the imide anion and lithium cation,
-4

resulting in a high ambient conductivity (1.6  10

S/cm). The conductivity was

thought to be mainly cationic, but transference number measurements were not
reported.

23

In addition, free AlCl3 caused side reactions in the cells. For the

acid-in-chain case, an open chain boron was incorporated into a PEO backbone.
20 
 

A conventional lithium salt was added and the anion was trapped by the acidic
boron. The properties of this system depended on the added lithium salt; for
-5

LiTFSI, the ambient conductivity was 7.6  10

S/cm.

24

The boron structure in

the polymer also was converted to anionic sites by reaction with phenyllithium,

25

leaving Li+ as the only mobile ion species and fulfilling the single-ion concept.
Scrosati promoted the use of composite polymer electrolytes, formed by
adding ceramic powders to mixture of PEO and lithium salt.

26

It was claimed that

the ceramic dispersion improved the interfacial stability between lithium metal and
the polymer electrolyte,

27

and that the ionic conductivity and transference number

were enhanced. However, Shriver and coworkers

28

studied lithium salt/PEO/SiO2,

Al2O3 nanoparticle systems, and their data showed that most important effect of
the particles was a reduction of the PEO crystallinity, but there was no direct
evidence for enhancement of conductivity in the amorphous phase.
Another way to get single cation conductors is to attach anions to polymers
through covalent bonds, leaving the lithium cation as the only mobile ion. As long
as all anions are tethered to the polymer, the lithium transference number should
be unity.

21 
 

Scheme 5. Two prepolymer structures (top) and the boron cross-linker (bottom)
29,30,31

used in Kerr’s work.

Kerr et al. constructed two cross-linked networks that exhibited single ion
conduction, where a lithium borate structure served as the lithium source and the
cross-linker (Scheme 5). In one case, a polyacrylate/polymethacrylate backbone
with a mixture of methoxy and allyl terminated PEO side chain structure was used.
-7

The highest ambient temperature conductivity was 3.510

S/cm and the

single-ion conducting behavior was confirmed in a symmetrical lithium cell cycling
test.

29

In another example, PEO and trimethylene oxide (TMO) chains were

attached to a poly(epoxide ether) backbone.

30,31

Conductivities were ~10

-6

S/cm,

and Li/Li cell cycling tests showed no polarization.
Mayes and coworkers synthesized copolymers from lithium methacrylate
32,33

(LMA) and polyethylene glycol methyl ether methacrylate (PEGMA).

Both

block copolymerization and random copolymerization were tried in their work. In
22 
 

block copolymer, carboxylate anions phase separated from the ion-conducting
PEGMA block, yielding room temperature conductivities of 10

-7

S/cm; in random

copolymer, ionic conductivities were one to two orders of magnitude lower. In
another example from the same group, poly (dimethyl siloxane) methacrylate
(PDMSMA) was incorporated into the copolymer,

34

When BF3 was added to

improve dissociation between the lithium and carboxylate ion, the room
-6

temperature conductivity was 710

S/cm at a O/Li ratio of 32. A test cell was

constructed from a lithium anode, a vanadium oxide cathode, and the electrolyte.
Lithium-ion intercalation into the cathode was hindered by the immobile anion in
the electrolyte, causing an unexpectedly high polarization.

35, 36

39

Scheme 6. Polysiloxane and side functionalities in Shriver’s work.

Shriver and coworkers used Pt-catalyzed hydrosilylation reactions to
attach trifluoromethylsulfonamide, perfluoroether sulfonate, and polyethylene
glycol methyl ether side chains to polysiloxane backbones. Because of the low Tg
of polysiloxanes, these comb polymers were expected to have a high conductivity.
23 
 

Polysiloxanes with a mixture of polyethylene glycol methyl ether and
trifluoromethylsulfonamide groups had a room temperature conductivity of
-6

1.210

S/cm.

37,38

Replacing the trifluoromethyl sulfonamide group with
-6

perfluoroether sulfonate slightly increased the conductivity to 2.510
an O/Li ratio of 33:1.

S/cm with

39

Lithium siloxyaluminate polymers containing ethylene oxide side chains
was synthesized by Rawsky and coworkers.

40

Conductivities of ~10

-5

S/cm were

obtained, with a lithium transference number of 0.71.

Scheme 7. Monomers used in Endo’s work.

43,44

Ohno and co-workers attached anion groups onto the chain ends of PEO
-7

oligomers. Ambient conductivities were in the range of 10 S/cm for carboxylate
derivatives,

41

-7

-8

10 -10

-6

S/cm for benzenesulfonate derivatives, and 3  10 S/cm

for sulfonamide derivatives.

42

Endo and coworkers formed single ion conductors

(Scheme 8) by using PEO diamines to ring-open epoxide monomers.

43,44
-5

best ambient conductivity for these self-standing rubbery films was 3.410
-6

with the rest in the range of 10

The
S/cm,

S/cm.

A variety of particles have been tested for their ability to improve Li

+

transference numbers. Kurian et al. mixed lithium-exchanged montmorillonite clay
24 
 

-8

with poly [(oxyethylene)8 methacrylate] and obtained low conductivities (10
S/cm).

45

Singhal et al. dispersed lithium-exchanged hectorite in a mixture of

ethylene carbonate (EC) and polyethylene glycol dimethyl ether (Mw = 250), and
-4

reported ambient conductivities of ~10

S/cm.

46

Riley combined lithium hextorite

with carbonate solvents. The resulting mixture formed a physical gel with
-3

-4

conductivities in the range of 10 -10
0.6-0.8.

47

S/cm and Li+ transference numbers of

To inhibit anion mobility, Fedkiw and coworkers attached sulfonate

groups on silica nanoparticles and dispersed the particles in PEO oligomers.
50

48,49,

+

While the Li transference number was encouraging, the room temperature
-6

conductivity was ~10

S/cm. Fedkiw and coworkers used free radical

polymerization to modify the silica particles, which may have resulted in low
polymer content and low lithium content.
Archer et al. grafted PEG chains to SiO2 (8 and 18 nm) and TiO2
nanoparticles (15 nm). LiTFSI was added to form a 1 M electrolyte in the organic
phase. The conductivity was insensitive to the type of metal oxide or the
nanoparticle size, but it increased with the molecular weight of the PEG chain.
The highest room temperature conductivity was ~10

-4

S/cm with PEG chains with

ranging from 595-725 g/mol. Most of the PEGs used in this work were liquids, but
51

after grafting, the particle surface was an amorphous soild.

With the development of controlled radical polymerization methods, such
as atom transfer radical polymerization (ATRP), reversible addition-fragmentation
25 
 

chain transfer polymerization (RAFT), and nitroxide meditated polymerization
(NMP), surface polymerization became a practical method for modifying surfaces.
A broad range of organic/inorganic hybrid materials have been synthesized that
are comprised of an inorganic core with a polymer shell. Once initiators are bound
to the surfaces of an inorganic particles, a dense layer polymer film can grown
from the surface, termed the “grafting from” approach.

52-55

Most work in this area

have used ATRP, but other controlled radical methods also can be used.
In order to improve proton conductivity in membranes, Yameen and coworkers
coated the membrane pores with ATRP initiators, and then grew brushes from a
mixture of oligo(ethylene glycol) methacrylate and 2-sulfoethyl methacrylate.
Including the hygroscopic oligo(ethylene glycol) chains was claimed to cause a
5

10 increase of the proton conductivity.

56

A previous group member, Dr. Yuan grafted sulfonic acid containing polymer on
7 nm silica particles, she mixed the hybrid particles with polyvinylidene fluoride
(PVDF) to form composite proton exchange membranes (PEM). This composite
PEM with 40 and 50 wt% particle contents gave proton conductivities comparable
or even higher than Nafion 117.

57

We investigated several nanoparticle systems where lithium counterions are
immobilized on polymers, the polymers are further attached to nanoparticles.
Electrolytes were prepared from the purified particles and low-molecular weight
(~500 g/mol) polyethylene glycol dimethyl ether (PEG -DME500). Initiators were
26 
 

attached to 10 nm silica nanoparticles, and using Atom Transfer Radical
Polymerization (ATRP), a controlled method useful for growing polymers from the
surfaces. Four different structures were used in this work, aiming to get a high
lithium conductivity.
2. Results and Discussion
Anchoring anions on nanoparticles should increase the Li

+

transference

number to near unity since anions bound to particles have reduced mobilities.
Previously, we and others examined anions attached to the surface of fumed silica
2

(AEROSIL 200 from Degussa, surface area of 200 m /g) and spherical
silicas.

58,59

Apparently, decorating particle surfaces with a monolayer of anions

provides too few anions to support high conductivity in a bulk electrolyte (O:Li
ratio = 120). One strategy for increasing conductivity is to increase the available
surface area by using smaller nanoparticles or higher surface-area particles
2

(roughened). Commercially available particles have surface areas of 302 m /g
(for Snowtex-S of Nissan Chemical). Assuming all of the surface area is
accessible, the O/Li ratios would improve to ~ 80.
An alternative approach to increasing anion concentrations is to move from
monolayer coverage of particles to polyelectrolytes anchored to particle surface
(Scheme 9). In principle, the carrier concentration can be increase arbitrarily and
approach the carrier concentration of the bulk polyelectrolyte. This can be
accomplished by growing polymers from particle surfaces, especially by
27 
 

controlled polymerization methods, such as ATRP. The research described in this
dissertation focuses on nanoparticle systems where lithium counterions are
immobilized on polyelectrolytes grown from nanoparticles.

Scheme 8. Synthetic approach to polyelectrolyte-decorated nanoparticles.
The synthetic route to polyelectrolyte-decorated nanoparticles is shown in
Scheme 8. The nanoparticles were Snowtex-S (Nissan Chemical), a colloidal
silica dispersed in pH 9-10 water, with an average particle size of 7-10 nm. The
TEM image of the as-received particles is shown in Figure 7. Each step in the
synthesis was confirmed by IR spectroscopy.

As shown in Figure 8, the IR

spectrum of Snowtex-S particles shows the characteristic bands 800-900

28 
 

-1

cm-1associated with to the Si-OH, and a broad absorption band at 1150 cm .

Figure 7. TEM image of Snotex-S silica nanoparticle (samples were dispersed in
water and dropped on copper grid, and dried before taking TEM).
In addition, C-H stretching bands at 2800-3000 cm
cetyltrimethylammonium bromide

42

-1

associated with

(CTAB), a surfactant added to stabilize the

nanoparticles in non-aqueous solvents.

After anchoring the ATRP initiator

(11-(2-bromo-2-methyl)propionyloxy)undecyltrichlorosilane to the nanoparticle
surface, a new band appeared at 1700 cm

-1

(C=O) as the initiator displaced the

CTAB.
2.1.

Particles modified with lithium 4-styrenesulfonate

For the preparation of poly(lithium styrene sulfonate) brushes on nanoparticles,
we followed the procedure of Yuan.

52

ATRP was used to grow poly(sodium

styrene sulfonate on particle surfaces at room temperature in a mixture of
29 
 

methanol and water (1:3), using Cu(bpy)2 as the catalyst. After polymerization,
the particles were purified by washing with mili-Q water (50 mL x 10). The poly
(sodium

styrene

sulfonate)

modified

particles

were

converted

to

the

+

corresponding Li salt by treating the particles with aqueous LiCl (100 equivalents)
for four days at room temperature (Scheme 10). The particles were recovered by
centrifugation, washed 10 times with Mili-Q water, and dried at 90 °C under
vacuum overnight.

HO OH
Cl
OH
HO
+ Cl Si
HO SiO2 OH
OH
HO
Cl
HO OH

=

Cl
O Si
Cl

O

Br
O

O

Toluene
65 °C

SiO2

H2O/CH3OH
CuBr
Bipyridine

Br
O

SO3Na
r.t.

n
SiO2
SO3Na

Scheme 9. Immobilization of initiator on silica particle and the following surface
ATRP with sodium 4-styrenesulfonate (SSNa).

30 
 

Scheme 10. Lithiation of silica particle grafted by poly(sodium 4-styrenesulfonate)
(Si-PSSNa).
The IR spectrum of the poly(lithium styrene sulfonate) brushes showed strong
-1

bands associated with the sulfonate group (S=O stretching, 980-1225 cm , S-O
-1

stretching, 690-870 cm ). Overtones and combination bands at 1800-2000 cm
and the out-of-plane CH bending at 817 cm

-1

-1

are characteristic of the

para-substituted benzene rings, expected for the styrene moiety, confirming the
successful growth of the polymer from the surface.

Figure 8. FTIR spectra of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
31 
 

grafted silica particles; (c) Si-PSSLi, poly (lithium 4-styrene sulfonate) grafted
silica particles.

Figure 9. Thermogravimetric Analysis (TGA) curves of (a) Snowtex-S bare
particles; (b) initiator grafted silica particles; (c) Poly(sodium 4-styrenesulfonate)
grafted silica particles; (d) Poly(lithium 4-styrenesulfonate) grafted silica particles.
Figure 9 shows the Thermogravimetric Analysis (TGA) data of sulfonate styrene
monomer system up to 900 oC in air. Under oxygen at this high temperature, all
the organic component were lost, and the weight loss here gave information about
organic content on silica particle surface. All the particle samples were dired at 90
o

C under vacuum line before characterization. The silica particles had a weight

loss of 15 wt% at 800 °C. The mass loss includes the CTAB surfactant, adsorbed
32 
 

water, and water formed by condensation of surface bound silanols. Accroding to
the information provided by Nissan Chemical Co., the surface area of Snowtex-S
2

sample is 302 m /g, cross section of the ammonium bromide group on CTAB is
2

0.8 nm . On one extreme case, to estimate the minimum residual CTAB on
particle surface, all the CTAB would be washed out, which give a 0 wt% weigt loss;
on the other extreme case, to estimate the maximum residual CTAB on particle
surface, there would be one whole layer of CTAB molecule on the particle surface,
based on the surface area of the particle and cross section of the ammonium
group, it gives 16.5 wt% weight loss. This TGA value of 15 wt% fall in between 0 ~
16.5 wt%. For calculation of the initiator-modified particles, assuming 0.6
nm2/initiator chain, theoretical weight loss of the organic component should be
around 15.1 wt%.

60

Initiator grafted silica particles showed weight loss of 27%,

the residual water might add up the weight loss of the sample. The high weight
loss of sample c (64%) and d (72%) proved that surface polymerization was
successfully done. For the TGA data of Na and Li polyelectrolytes, the residue at
900 °C should be comprised of the silica particle and Li and Na oxides or salts. As
expected, PSSLi had the largest weight loss, as lithium has the lower atomic
weight.

33 
 

Scheme 11. Quantitative determination of metal content.
Quantitative analysis of the metal content in the TGA residue by Inductive
Coupled Plasma (ICP) optical emission spectrometry provides an estimate of the
number of anions bound to the nanoparticles. The lithium should be in the form of
Li2O or Li2CO3; both are water soluble. For the ICP analysis, the TGA residue
was stirred in Mili-Q water for 12 h, and filtered prior to the analysis. ICP analysis
of the TGA residue from a 6.6 mg Si-PSSLi sample gave 0.16 mg lithium, which
-2

corresponds to 2.4  10

g Li/g sample. This means more than 95% of the Na

was exchanged with Li, this relatively high lithium content enables us to get a high
lithium content concentration in the electrolyte (as high as a O/Li ratio of 32:1).

34 
 

Figure 10. TEM image of Si-PSSLi (samples were dispersed in water and
dropped on copper grid, dried before taking TEM).
Figure 10 shows TEM image of Si-PSSLi dispersed in water. The larger size of
the polymer grafted particles is 15-20 nm, compared to the 7-10 nm of bare
particles (Figure 7), consistent with successful growth of polymer from the particle
surface. Electrolytes were prepared by mixing polymer grafted silica particles and
poly(ethylene oxide) dimethyl ether oligomer (Mw=500) (PEGDME-500) at various
O/Li ratios were made. Vigorous stirring for 24 hours resulted in homogeneous
electrolytes.
O/Li ratio

32

64

96

128 192 256

Particle content (wt%) 17.2 9.4 6.5

4.9

3.3

2.5

Table 6. Particle weight content for electrolytes prepared from Si-PSSLi (2.4 
-2

10

g Li/g nanoparticles) dispersed in PEGDME-500.
35 

 

Figure 11. Temperature dependent conductivity for Si-PSSLi/PEGDME-500 at
various O/Li ratios.
Figure 11 shows the temperature-dependent conductivity measured by
impedence spectroscopy for electrolytes with different O/Li ratios. As the
temperature increased, the conductivity increased, as expected for increased
chain mobility for the PEGDME-500 host. The data are roughly linear, consistent
with themally activated transport and the VTF equation. The conductivity
increased with decreased O/Li ratios (more carriers), but at the lowest ratios
(highest particle content), 32, the conductivity decreased. The trends are more
clear in Figure 12, where the room temperature conductivity is plotted as a
36 
 

function of the O/Li ratio. Similar trends are see in salt/PEO electrolytes (see
Figure 5), which are usually interpreted as the formation of less mobile ionic
aggregates or an increase in the Tg of the elctrolyte. Here the anions are tethered
to the polymer chains and the effects occur at higher O:Li ratios, suggesting that
either an increase in the elctrolyte Tg or a decline in the volume fraction of the
conductive phase is responsible. Overall, the room temperature conductivity of
-7

the electrolytes is around 1  10

S/cm.

Figure 12. Room temperature conductivity vs. O/Li ratio for Si-PSSLi/
PEGDME-500 electrolytes.
2.2.

Particles modified with lithium 4-styrenesulfonate and PEGMA

We reasoned that the ion mobility may be hindered within the dense layer of
lithium ions and aryl sulfonates coating the nanoparticles. To test that hypothesis

37 
 

and hopefully increase the conductivity, we copolymerized PEGMA (Mn=1100)
with sodium 4-styrenesulfonate from the particle surface. Copolymerization of
PEGMA with sodium 4-styrenesulfonate would introduce PEG chains on the
particles that might facilitate transport of lithium ion.

In addddition, adding PEG

chains to the particles might improve the interface between the particles and the
PEGDME-500 host.

Scheme 12. Copolymerization of SSNa and PEGMA monomers on silica particle.
The conditions for the surface-initiated copolymerization and workup procedure
were

similar

to

those

used

for

the

homopolymerization

of

sodium

4-styrenesulfonate (Scheme 12). The mole ratio of sodium 4-styrenesulfonate to
PEGMA monomers was 3:1, which corresponds to a polymer layer with an O/Li
ratio of 8:1 (two oxygen atoms of the ester group not included).
Since TGA was performed up to 900

o

C, all the organic content would

contribute to the weight loss, which provided information about amount of both
monomers; ICP gave information about sodium content, which was only from the
PSSNa structure. Combining TGA and ICP data (Scheme 13) gave ratio of this
two monomers on particle surface, which was very close to the targeted monomer
38 
 

ratio (3.45:1).

Scheme 13. Characterization for the copolymer content in Si-PSSNa-PEGMA.

Scheme 14. Lithiation of Si-PSSNa-PEGMA.
Lithiation of this copolymer grafted particles was similar as previous. Excess of
lithium chloride aqueous solution was used to change it to lithium electrolyte
(Scheme 14). ICP data showed that 87.6% sodium content was converted to
lithium, and lithium content was 1.6 x 10

-2

g Li/g sample. Compared to Si-PSSLi

sample, lithium content in this copolymer version was relatively lower, more
particles need to be mixed to get the same O/Li ratio.

39 
 

Figure 13. FTIR spectra of (a) Snowtex-S nanoparticles; (b) initiator grafted silica
particles; (c) Co-poly(4-styrenesulfonate)-PEGMA grafted silica particles.

Compared to initiator grafted silica particles from FTIR spectra (Figure 13), the
-1

multiple bands at 980-1225 cm

from S=O stretching and peaks at 690-870 cm

-1

from S-O indicate that polymer was successfully grown onto the particle surface.

40 
 

Figure 14. TGA curves of (a) Snowtex-S bare particles; (b) initiator grafted silica
particles; (c) Co-Poly(sodium 4-styrenesulfonate)-PEGMA grafted silica particles;
(d) Co-Poly(lithium 4-styrenesulfonate)-PEGMA grafted silica particles.

Figure 14 shows TGA curves of copolymer grafted silica particles. The high
weight loss of sample c (68%) and d (82%) proved that surface polymerization
was successfully achieved. The relative bigger weight loss compared to pure
sulfonate monomer grafted system was due to the large side chain structure of
PEGMA repeating unit.

41 
 

Figure 15. TEM image of Si-PSSLi-PEGMA (samples were dispersed in water
and dropped on copper grid, dried before taking TEM).

Figure 15 shows TEM image of Si-PSSLi-PEGMA. The average size of the
polymer grafted particles increased to 15 nm. The particles were mixed with
PEGDME-500 to measure the conductivity at different O/Li ratios.
O/Li ratio

32

64

Particle content (wt%) 24.2 13.7

128 192 256
7.4

5.0

3.8
-2

Table 7. Particle weight fraction for Si-PSSLi-PEGMA (1.56 x 10
dispersed in PEGDME-500.

42 
 

g Li/g sample)

Figure

16.

Temperature

dependent

conductivity

of

Si-PSSLi-PEGMA/

PEGDME-500 at various O/Li ratios.

Figure 16 showed that for all samples at different O/Li ratio, conductivity
increases with increase of the temperature. However, shapes of curves for
different samples seem quite “noisy” for this system here.

43 
 

Figure 17. Room temperature conductivity VS. O/Li ratio for Si-PSSLi-PEGMA
/PEGDME-500.
Room temperature conductivity reached a peak value of 4.3 x 10

-7

S/cm at O/Li

of 64 (Figure 17). Further increase of the particle content only resulted in
decrease of the conductivity. Compared to the homopolymer modified particles
Si-PSSLi, which has a peak value of 2.9 x 10

-7

S/cm, conductivity of copolymer

modified particles is relatively increased.
2.3.

LiTFSI analogue polymer modified particles

Transference number and conductivity are the most important parameters for
electrolyte material in lithium ion battery. In our study, the anion group was
anchored onto the polymer backbone; to further immobilize the anion, the
polyelectrolyte was anchored on surface of silica nanoparticles. This architecture
44 
 

could ensure that most of the ionic conductivity was contributed from lithium
cation. In this case, how to improve conductivity became our major concern.
According to the equation

σ (T)=n(number of carriers)*q(charge of the carrier)* (mobility)
At a specific temperature, ionic conductivity is directly related to number of
carriers, charge of the carrier and mobility of the carrier. We used PEGDME-500
oligomer as a model compound to study the properties of this architecture, this
oligomer dominated mobility of the lithium cation, which is a factor we could not
change; charge of lithium cation is 1.6*e

-19

Coulombic. We decided to focus on

the first factor and try to increase number of carriers. In most of the cases, cation
and anion formed compact ion pair, this hindered lithium cation mobility. Ue and
coworkers'

61

study showed dissociation degree of common lithium salt followed

this order: Li[N(SO2CF3)2], LiAsF6 > LiPF6 > LiClO4 > LiBF4 > LiC4F9SO3 >
LiCF3SO3.
Because lithium bis(trifluoromethanesulphonyl)imide (LiTFSI) is well known for
good conductivity and high dissociation degree, we proposed a new monomer to
mimic its structure (Scheme 15). The methacrylate part of this structure makes it
easily radical polymerized; a mono-trifluoromethane sulfonyl structure was
retained to ensure a good dissociation degree with lithium cation.

45 
 

Scheme 15. Trifluoromethane sulfonic aminoethylmethacrylate (TFMA) monomer
inspired by LiTFSI structure.
Once polymerized, post-modification with n-butyl lithium gave a lithium
electrolyte. Admittedly, conductivity data of a mono-trifluoromethane sulfonyl
structure would not be as good as its bis-version, but this was a necessary
sacrifice to get a polymerizable structure.

HO

(1) Hydroquinone, 95 °C,

SOCl2

NH3Cl

NH2 ethanol, 0 °C HO

Cl
(2)115 °C, 2 h
(3)65 °C, overnight

O
O

O

NH3Cl

(CF3SO2)2O
Et3N
Chloroform
-45 °C

0 °C

O

O

O
HN S
O

F
F
F

Scheme 16. Synthesis of trifluoromethane sulfonic aminoethylmethacrylate
(TFMA).
Ethanolamine was converted to amino salt to protect amine group. The
following chemistry was done in the molten state without any solvent;
hydroquinone was used as a radical scavenger. The methacrylate organic salt
46 
 

intermediate was found only soluble in chloroform (Scheme 16). Trifluoromethane
o

o

sulfonyl group was normally installed at -78 C, -45 C was chosen as reaction
temperature to make sure the solvent was maintained as liquid state (m.p. of
o

Chloroform -63 C ).

Scheme 17. ATRP of trifluoromethane sulfonic aminoethylmethacrylate (TFMA)
on silica particles.
Surface ATRP on silica particle using bipyridine as ligand at r.t. gave 45%
polymer content. To get a thicker polymer layer, N, N, N’, N’, N’’-Pentamethyldi-ethylenetriamine (PMDETA) was used as ligand, reaction temperature was
o

increased to 90 C. As a result, product with 62% polymer content was obtained
(Scheme 17).
47 
 

n
O

O

SiO2

n-BuLi Toluene
r.t. overnight

n
O

SiO2

HN O
S
O CF3

O
LiN O
S
O CF3

Si-TFMA

Si-TFMA-Li

Scheme 18. Lithiation of Si-TFMA to Si-TFMA-Li.
Lithiation of Si-TFMA was done using n-butyl lithium in dry toluene (Scheme 18).
Overnight reaction at room temperature gave lithiation efficiency of 78%, and the
lithium content in Si-TFMA-Li was 2 x 10

-2

g Li/g sample.

Figure 18. FTIR spectra of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
grafted silica particles; (c) Si-TFMA-Li, lithiated poly(trifluoromethane sulfonic
aminoethylmethacrylate) grafted silica particles.
48 
 

Figure 18 shows FTIR spectra of the PTFMA grafted particles at different
stages. Further development of the peak at 980-1300 cm

-1

and split of the peak

from FTIR spectra corresponds to S=O stretching, the sharp peak at 680-780
cm

-1

was assigned to CF3 stretching, both proved the polymer content on particle

surface.

Figure 19. FTIR spectra of (a) Si-TFMA, poly(trifluoromethane sulfonic
aminoethylmethacrylate) grafted silica particles; (b) Si-TFMA-Li, lithiated
poly(trifluoromethane sulfonic aminoethylmethacrylate) grafted silica particles.
A potential problem with the lithiation of Si-TFMA was that the ester group might
be susceptible to the strong base (n-BuLi). This reaction was characterized with
ICP, which only gave the lithium content of the final product. However, lithium
49 
 

cation could be on the imide group; or if the ester got hydrolyzed, lithium cation
could also stay on the carboxylic group. To prove that final product Si-TFMA-Li
maintained the original structure, FTIR spectra before and after lithiation was
shown in Figure 18. If the polymer structure survived lithiation chemistry, ideally
the two spectra would be the same, but there are subtle difference between the
two spectra. As another tentative evidence, for the samples from different batches
with the same weight loss in TGA, lithium content is repeatable after lithiation.
However, none of these are conclusive.

Figure 20. TGA curves of (a) Snowtex-S silica particles; (b) Si-Initiator,

initiator

grafted

sulfonic

silica

particles;

(c)

Si-TFMA,

poly(trifluoromethane

aminoethylmethacrylate) grafted silica particles; (d) Si-TFMA-Li, lithiated
poly(trifluoromethane sulfonic aminoethyl -methacrylate) grafted silica particles.
50 
 

Figure 20 shows TGA curves of TFMA monomer grafted silica particles. The
large weight loss of sample c (62%) and d (57%) proved that surface
polymerization was successfully conducted. After lithiation, the lithium content
was retained as oxide or salt, this made the weight loss of sample d smaller than
sample c.
O/Li ratio

16

32

64

96

128

Partcile content (wt%) 33.2 19.9 11.1 7.7

5.8

Table 8. Particle weight content for electrolytes prepared from Si-TFMA-Li (2.0 x
-2

10

g Li/g sample) dispersed in PEGDME-500

Figure 21. Temperature dependent conductivity of Si-TFMA-Li/PEGDME-500 at
various O/Li ratios.
51 
 

Figure 22. Room temperature conductivity vs. O/Li for Si-TFMA-Li/PEGDME-500.
For samples at different O/Li ratios, conductivity data smoothly increases with
increase of temperature. For sample with O/Li of 32, room temperature value
gives the highest conductivity of all samples. The room temperature conductivity
VS. O/Li ratio plot shows an increase in conductivity with the increase of O/Li ratio.
After reaching a peak value of 10

-6

S/cm at an O/Li of 32, the conductivity

decreases. As the particle content increases, the PEGDME-500 and particle
mixture becomes inhomogeneous, particle could not get well dispersed, this only
results in worse conductivity. Minimizing the volume fraction of the silica cores, i.e.,
increasing the molecular weight of the polyelectrolytes attached to the particles is
the key issue to improve the physical attributes of the electrolyte. An alternative
approach is to increase the number of lithium carriers in each monomer unit.
52 
 

2.4.

Phosphonic acid modified particles

To improve the conductivity, we used a styryl phosphonate which is a precusor
to polymers with phosphonic acid as the repeating unit. Once lithiated, each
repeating unit would have two lithium cations, and increase the local lithium cation
density on the particles. To avoid the potential influence of the phosphonic acid on
the ATRP initiating system, we used a phosphonate ester as the monomer, the
ester group was hydrolyzed after polymerizing the styryl phosphonate on the
particle surface.

Scheme 19. Synthesis of diethyl 4-vinylbenzylphosphonate (DEVBP).
We used a modified version of Chidsey’s protocol

62

for the synthesis of diethyl

4-vinylbenzylphosphonate (DEVBP). Our initial attempts suffered from low yields.
Since the benzyl iodide intermediate was light sensitive, we decided to add the
triphenyl phosphite without isolating the 4-iodomethyl styrene intermediate.
From this one-pot variation, we isolated the desired product in 55 % yield
(Scheme 19).

53 
 

Scheme 20. ATRP of diethyl 4-vinylbenzylphosphonate (DEVBP) on particle
surface.
ATRP of DEVBP on silica particle surface used PMDETA as ligand with a metal
ligand ratio of 1:6, the reaction normally continues for 24 h before work-up
(Scheme 20). This polymerization chemistry was done 20 times or so to get
enough material for characterization. Only the first trial gave a polymer content of
more than 70 wt%, samples typically have polymer content of 50 wt%, these
sample were used for lithiation and conductivity measurement.

54 
 

Scheme 21. Lithiation procedure for poly(diethyl 4-vinylbenzylphosphonate)
grafted silica particle (Si-PDEVBP).
The monomer, diethyl 4-vinylbenzylphosphonate, was used as a model
compound to establish suitable conditions for the hydrolysis of the phosphonate
ester. Refluxing in 4 M HCl or 2 M NaOH, failed to hydrolyzed product. In a
two-step reaction, the phosphonate was treated with TMSCl, followed by
methanolysis. However, the procedure of Mullen,

63

using TMSBr followed by

methanolysis, successfully converted the phosphonate ester to the acid (Scheme
21). The lithium content in Si-PPLi was 3.5 x 10

-2

g Li/g sample. With two lithium

atoms in one repeating unit, lithium content increases. With this relatively high
lithium content, we could get a high lithium concentration with less particle content,
this could potentially increase the conductivity of the electrolytes.

55 
 

Figure 23. TGA curves of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
grafted silica particle; (c) Si-PDEVBP, silica particle grafted by poly(diethyl
4-vinylbenzylphosphonate); (d) Si-PPLi, poly(lithium 4-vinylbenzylphosphonate).

Figure 23 shows TGA curves of silica particles at different stages. Curve c and
d proved successful ATRP of diethyl 4-vinylbenzylphosphonate monomer on
particle surface.

56 
 

Figure 24. FTIR spectra of (a) Snowtex-S silica particles; (b) Si-Initiator, initiator
grafted silica particles; (c) Si-PPLi, silica particle grafted by poly(lithium
4-vinylbenzylphosphonate).
In curve c Figure 24, the very strong peak at 960-1300cm
P=O stretching. The out-of-plane CH bending at 817 cm

-1

-1

was assigned to

is characteristic of the

para-substituted benzene rings, expected for the styrene moiety, confirming the
successful growth of the polymer from the surface.
O/Li ratio

32

64

96

Particle content (wt%) 12.6 6.7 4.6

128 192 256
3.5

2.3

1.8

Table 9. Particle weight content for electrolytes prepared from Si-PPLi (3.45 x
-2

10 g Li/g sample) dispersed in PEGDME-500.

57 
 

Figure 25. Temperature dependent conductivity for Si-PPLi/PEGDME-500 at
various O/Li ratios.
Figure 25 shows conductivity change with the increase of temperature for
Si-PPLi mixed with PEGDME-500, conductivity increases with temperature for all
samples. Samples with six different O/Li ratios were characterized, particle
content reaches 13 wt% at O/Li ratio of 32. In different sets of samples,
conductivity at O/Li 32 were lower than those at O/Li 64, further increase of
particle content was not tried.

58 
 

Figure 26. Room temperature conductivity vs. O/Li for Si-PPLi/PEGDME-500.
Room teperature conductivity reaches peak value of 2.4 x 10

-7

S/cm at O/Li of

64 (Figure 26). Conductivity of Si-PPLi is not evidently improved, probably
because the low degree of dissociation between lithium cation and the
phosphonate group. The low dissociation degree between two counterions
sacrificed increase of lithium content, free moving lithium cations in the materials
were not increased, conductivity was not improved.
In this project, we aimed at developing single lithium ion electrolytes based on
polymer modified silica nanoparticles, conductivity data show that different
polymer structure gave different properties, a TFSI analogue monomer gave the
best conductivity. However, at this point, there are several measurements we
need to do in the near future:
59 
 

1) People have been using ATRP chemistry to modify silica nanoparticles,
characterization of the polymer content on particle surface was done using
hydrofluoric acid to etch the particle, extracting the polymer content to
organic (toluene) phase. This polymer solution could be subject to GPC to
get related data. Kinetics of surface ATRP could be done using the same
protocol. However, most of the previous work used styrene or methyl
methacrylate as monomer, in their case the polymer has a good distribution
in the organic phase. In our case, the polymer electrolytes are quite soluble
in water solution, it is difficult to extract the polymer content to organic
phase, NMR spectra from the organic phase were quite messy at this
point;
2) To prove the single-ion concept, lithium transference number will be
measured. The electrolyte in this work is composed of PEGDME-500 and
polymer grafted particle. PEO is basicly non-conductive, anions were
covalently bonded on particle surface, lithium cation is the only conductive
ion in the materials, the lithium transference number is supposed to be
close to unity.
3. Conclusion
Four different monomer structures were grown on silica nanoparticles using
ATRP. The first monomer, sodium 4-styrenesulfonate was a commonly used
precursor for lithium single ion electrolyte; by intramolecularly incorporating a
60 
 

PEO containing monomer, it was aimed to improve lithium cation conducting
behavior and interfacial property between particles and PEGDME-500; the third
monomer featured a trifuoromethane sulfonic imide strucute, which could give
higher dissociation degree between anion and lithium cation; in the fourth
monomer, each phosphonate group contained two lithium cations, this was
supposed to increase the lithium content. In each case, the polymer content on
particles was more that 50 wt%. These particles were post-grafted to lithium
conductor, and mixed with PEGDME-500 oligomer. This protocol gave high
polymer content and relatively high lithiation efficiency, which resulted in a high
lithium content. Room temperature conductivity of these materials were normally
-7

in the range of 10

-6

to 10

S/cm.

4. Experimental Section
4.1.

Materials

Lithium hydroxide monohydrate (Aldrich, 98%), cetyltrimethylammonium
bromide (CTAB, Aldrich), 4-styrenesulfonic acid sodium salt hydrate (Aldrich),
trifluoromethanesulfonic anhydride (Aldrich, 99%), bromotrimethylsilane (Alfa
Aesar, 97%), butyllithium, 2.5 M solution in hexane (Aldrich), 4-vinylbenzyl
chloride (Aldrich, 90%), triethyl phosphite (Aldrich, 98%), ethanolamine (Aldrich,
99%), thionyl chloride (CCI, 99.7%), methacryloyl chloride (Aldrich, 97%),
triethylamine (J. T. Baker), hydroquinone (J. T. Baker Chemical),l chloride (J. T.
Baker Chemical), sodium thiosulfate (CCI), Sodium Iodide (Jade Scientific), N, N,
61 
 

N’, N’, N’’-pentamethyldiethylenetriamine (PMDETA, Aldrich, 99%) were used as
received. Cu(I)Br (Aldrich,99.999%) was purified using saturated aqueous NaBr
solution. Bipyridine (Aldrich, 99%) was recrystallized from hexane and sublimed.
Poly(ethylene glycol) methyl ether methacrylate (Average Mn 1,100, Aldrich) was
passed through a basic column before use. Snowtex-S (7-10 nm) were received
as a gift from Nissan Chemical.

4.2.

Characterization

1

H NMR and

13

C NMR spectra were collected using a Varian UnityPlus-500

spectrometer in CDCl3 with the residual proton signals from the solvent as the
chemical shift standard. Thermogravimetric analysis (TGA) were carried out in air
on Perkin-Elmer TGA 7 instruments at heating rate of 10 oC/min. Sample were
held at 120 oC for 30min before starting the heating process. FTIR spectra were
collected in a Mattson Galaxy 300 spectrometer. FTIR samples were made by
mixing with KBr, grounded and pressed to pellets. A JEOL-100CX transmittance
electron microscope was used to detect the microstructure of the particles.
Particles were dispersed in Mili-Q water using ultrasonication at low concentration.
A drop of this suspension was put onto a carbon-coated copper mesh grid.
Excess of water was absorbed carefully by filter paper and then air dried.
Conductivity data were collected from an HP 4192A LF Impedance Analyzer
scanning from 5 Hz to 13 MHz with an applied voltage of 10 mV. Sample cell uses
62 
 

two steel disks as symmetrical electrodes, containing sample of 0.61 cm in radii
and 0.0175 cm in thickness. Polymer grafted particle were finely ground using a
pestle and mortar then transferred into glovebox. PEGDME-500 was passed
through a basic column and diluted with ether. The solution was stirred with
molecular sieves for 24 h before removing solvent on rotovac and put onto
vacuum line at 50 oC for 24 h. Purified PEGDME-500 was transferred into
glovebox. For different O/Li particles were mixed with 0.2 g PEGDME-500 and
stirred in 8ml vial for 24 h to make it well dispersed. Data were taken at r.t., 30, 35,
40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90 oC. Samples were equilibrated at each
temperature for at least 10 minutes before measurement. Sample cell was
constructed of stainless steel disks separated by a Teflon collar.
Lithium quantitative analysis was done using Varian 710-ES ICP Optical Emission
Spectrometer under argon atmosphere. Six standard solutions (LiCl concentration
of 0, 0.05, 0.5, 1, 2, 4.5 ppm, 2 wt% HNO3 for each solution) were used to make
working curve for each measurement. Peaks at wavelength of 460.289 nm,
670.783 nm, 610.365 nm were all used for data collection, results showed that
data from the three different emission peaks were consistent. Following are data
from wavelength of 670.783 nm, calibration solution concentration/ppm
(instrument peak intensity):
0 (22457), 0.05 (187440.0), 0.5 (1958228.6), 1 (3490400.0), 2 (8897000.0), 4.5

63 
 

(15000000.0), Si-PSSLi (6704000.0) mass of sample 6.583 mg, 2.40 x 10

-2

g Li/g

sample.
0 (22457), 0.05 (187440.0), 0.5 (1958228.6), 1 (3490400.0), 2 (8897000.0), 4.5
(15000000.0), Si-PSSLi-PEGMA (5581714.3) mass of sample 8.480 mg, 1.56 x
-2

10

g Li/g sample.

0 (18606.4), 0.05 (75106.4), 0.5 (368117.0), 1 (548085.1), 2 (1106914.9), 4.5
(2336170.2), Si-TFMA-Li (6123404.3), mass of sample 7.953 mg, 2.00 x 10

-2

g

Li/g sample.
0 (22542.9), 0.05 (182800.0), 0.5 (1954857.1), 1 (3460800.0), 2 (8737000.0), 4.5
(14800000.0), Si-PPLi (5407978.7), mass of sample 3.777 mg, 3.45 x 10

-2

g Li/g

sample.
4.3.

Synthesis

ATRP of sodium 4-styrene sulfonate with Snotex-S particle: In a 50 mL
Schlenk flask, 1.72 g 4-styrene sulfonate (8.35 mmol), 56 mg bipyridine (0.32
mmol), 6 mL Mili-Q water and 2 mL methanol was mixed, which was charged with
a magnetic stir bar to stir to be a clear solution, 0.124 g initiator grafted silica
particle was added and the mixture was ultrasonicated for 15 mins. After 3 cycles
of Freeze-Pump-Thaw, the Schlenk flask was filled with nitrogen and 17.2 mg
CuBr was added. After another cycle of Freeze-Pump-Thaw, the reaction was
done under room temperature for 12 hours. To work up the reaction, Schlenk flask
was open to air, the mixture was transferred to 50 mL centrifuge tube and washed
64 
 

with MiliQ water twice, N,N,N’,N’-ethylenediaminetetraacetic acid (disodium salt
dehydrate, EDTA.2Na) saturated solution 3 times, Mili-Q water once. In each step
the polymer grafted silica particle was recovered using centrifuge. The product
was put onto vacuum line dried at 90 oC overnight.
Lithiation of Poly(4-styrene sulfonate) grafted silica particle: In a 250 mL
flask, 2.4 g lithium chloride was dissolved in 100 mL Mili-Q water. 0.32 g
poly(4-styrene sulfonate) grafted silica particle was added in the solution and
ultrasonicated 15 min before reaction at room temperature fro 48 h. Product was
collected by centrifuge and washed by Mili-Q water 3 times. It was dried under
vacuum at 90 oC overnight.
Synthesis

of

ethanolamine

hydrochloride and aminoethylmethacrylate

hydrochloride followed the procedure adapted from Forcada’s work with slight
modification. 10 mL ethanolamine was dissolved with 250 mL ethanol in an 500
mL Erlenmeyer flask on ice bath. Thionyl chloride was added dropwise using pipet
until no smoke generated. The suspension was stirred for another 10 minutes with
ice bath, the white crystals were collected by filtration and washed with hexane.
The product was air dried and 15.7 g product was obtained (yield 97%).
Synthesis of aminoethylmethacrylate hydrochloride: In a 250 mL 3-neck flask,
6.5 g ethanolamine (67 mmol) hydrochloride and 0.05 g hydroquinone (0.45 mmol)
was mixed. The mixture was heated up to 95 oC for 20 minutes with magnetic
stirring until the salt melted. 10 mL methacryloyl chloride (96 mmol) was added by
65 
 

syringe over a period of 20 mins. The temperature was increased to 115 oC for 2
hours, then decreased to 65 oC, 200 mL ethyl acetate was added while the
reaction mixture was still at molten state. White precipitant was produced and
collected by filtration. 7.85 g product was obtained (yield 72%).
Synthesis of trifluoromethane sulfonic aminoethylmethacrylate: In a 100 mL
flask, 0.57 g aminoethylmethacrylate hydrochloride (3.43 mmol), 0.6 g
triethylamine (5.94 mmol), and 20mL chloroform was mixed. The mixture was put
onto acetonitrile/dry ice bath (-45 oC). Using syringe to add 8 ml trifluoromethane
sulfonic anhydride (4.76 mmol). After addition of Tf2O was over, CH3CN/dry ice
bath was removed and the reaction was under room temperature for another 2
hours. 30 mL saturated NaHCO3 solution was added to the reaction mixture,
which was extracted by 3*30 mL dichloromethane. The organic layer was washed
with 2 x 50 mL 2 M HCl solution, and further wash with 2 x 30 mL brine. MgSO4
was used to dry the dichloromethane solution and condensed on rotovac. 0.6 g
product was obtained (yield 67%).
ATRP of trifluoromethane sulfonic aminoethylmethacrylate with Snotex-S
particle: In a Schlenk flask, 0.62 g 4-vinylbenzylphosphonate (2.36 mmol), 62.3
mg N,N,N’,N’,N’’- -Pentamethyldiethylenetriamine (PMDETA 0.36 mmol), 5 mL
DMF and 62 mg initiator grafted particle was mixed and ultrasonicated for 15 mins.
After 3 cycles of Freeze-Pump-Thaw, the Schlenk flask was filled with nitrogen
and 8.6 mg CuBr (0.06 mmol) was added. After another cycle of
66 
 

Freeze-Pump-Thaw, the reaction was done at 90 oC for 12 hours. To work up the
reaction, Schlenk flask was open to air, the mixture was transferred to 50 mL
centrifuge tube and washed with DMF twice, EDTA.2Na saturated solution 3 times,
MiliQ water once. In each step the polymer grafted silica particle was recovered
using centrifuge. The product was put onto vacuum line dried at 90 oC overnight.
Lithiation

of

poly(trifluoromethane

sulfonic

aminoethylmethacrylate)

grafted particles (Si-TFMA to Si-TFMA-Li) In a 200 mL round bottom flask, 1 g
Si-TFMA was dispersed in 150 mL dry toluene, 1 mL n-BuLi hexane solution (2.5
M) was added slowly in to the suspension. It was stirred overnight at r.t. and
product recovered by centrifuge. It was washed three times with toluene and three
times with acetone, dried under vacuum at 90 oC overnight.
Synthesis of trifluoromethane sulfonic octylamine In a 100 mL round bottom
flask, 2.58 g octylamine (0.02 mol), 3.03 g triethylamine (0.03 mol) and 30 mL dry
dichloromethane were added. The mixture was placed onto a -78 oC cold bath
under nitrogen, temperature of the cold bath was monitored using a low
temperature thermometer. 4 mL of trifluoromethane sulfonic anhydride (1.2 equiv.)
was slowly added over 15min. After addition of the anhydride was over, cold bath
was removed, the reaction was performed overnight under nitrogen atmosphere.
To work up the reaction, 30 mL saturated NaHCO3 solution was added to the
reaction mixture, which was extracted by 3 x 30 mL dichloromethane. The organic
layer was washed with 2 x 50 ml 2 M HCl solution, and further wash with 2 x 30
67 
 

mL brine. MgSO4 was used to dry the dichloromethane solution and condensed
on rotovac. To separate the mono TFSI amine from bi TFSI amine, the crude
product was added to THF solution of potassium t-butoxide (2 equiv.), the mixture
was stirred for 5 h and solvent removed on rotovac. Concentrated HCl was added
and the mixture was extracted with dichloromethane 3 x 50 mL. Organic layer was
washed with water and dried over MgSO4. The solvent was removed and dried
under vacuum line, giving the pure mono TFSI product.
Lithiation of trifluoromethane sulfonic octylamine A 100 mL round bottom
flask was flame dried before use, 1.3 g of trifluoromethane sulfonic octylamine (5
mmol) and 30 mL dry THF was added under nitrogen. 3 mL n-BuLi hexane
solution (2.5 M) was added slowly to the solution, It was stirred overnight at r.t.
and solvent was removed under rotovac and dried under vacuum line.
Synthesis of diethyl 4-vinylbenzylphosphonate followed procedure in a paper
from Chidsey and coworkers, with a slight modification.10.5 g NaI (70.4 mmol)
and 50 mL methyl ethyl ketone were mixed in an Aluminum foil wrapped 250 mL
round bottom flask, the suspension was purged with nitrogen for 30 mins. 2.5 mL
(14.1 mmol, 90%) vinyl benzyl chloride was added by syringe. It was reacted at
room temperature for 8 hours under nitrogen atmosphere. 2.4 g triethyl phosphite
was added, the flask was put onto 90 oC oil bath for additional 7 hours. Solvent
and ethyl iodide by-product were removed on rotovap. Crude product was
dissolved in 300 mL diethyl ether, which was washed with 3 x 150 mL DI water.
68 
 

The organic layer was dried by MgSO4. The ether solution was condensed and
purified by flash column (hexane/EtOAc=1:1) to give 1.97 g product (yield 55%).
1

H NMR (500 MHz, CDCl3, ppm): δ 4.46 (s, 2 H), 5.26 (d, 1 H), 5.75 (d, 1 H), 6.68

(d of d, 1 H), 7.34 (s, 4H).
ATRP of 4-vinylbenzylphosphonate with Snotex-S particle: In a Schlenk flask,
0.6

g

4-vinylbenzylphosphonate

(2.36

mmol),

62.3

mg

N,N,N’,N’,N’’-

-Pentamethyl-diethylenetriamine (PMDETA 0.36 mmol), 5 mL DMF and 62 mg
initiator grafted particle was mixed and ultrasonicated for 15 mins. After 3 cycles
of Freeze-Pump-Thaw, the Schlenk flask was filled with nitrogen and 8.6 mg CuBr
(0.06 mmol) was added. After another cycle of Freeze-Pump-Thaw, the reaction
was done at 90 oC for 12 h. To work up the reaction, Schlenk flask was open to air,
the mixture was transferred to 50 mL centrifuge tube and washed with DMF twice,
EDTA.2Na saturated solution 3 times, Mili-Q water once. In each step the polymer
grafted silica particle was recovered using centrifuge. The product was put onto
vacuum line dried at 90 oC overnight.
Lithiation

of

poly(4-vinylbenzylphosphonate)

grafted

silica

particle

(Si-PDEVBP to Si-PPLi) in a 500 mL round bottom flask, 1 g Si-PDEVBP was
dispersed in 200 mL dry dichloromethane by ultrasonication, 0.7 mL TMSBr (5.3
mmol) was injected into the solution. The solution was stirred at 50 oC for 24 h
before removing the solvent on rotovac. 300 mL MeOH was added into the flask
slowly, the suspension was ultrasonicated for another 10 min and put onto 50 oC
69 
 

oil bath for 24 h. Solvent was removed on rotovac, 10 g lithium hydroxide
monohydrate (0.24 mol) in 400ml Mili-Q water was added into the flask,
ultrasonicated for 10 mins, and reacted at r.t. for 12 h. Product was recovered by
centrifuge, washed ten times with large amount of Mili-Q water and dried at 90 oC
under vacuum overnight.

70 
 

 
 
 
 
 
 
 
 
 
 

REFERENCES
 
 

71 
 

REFERENCES

(1) Whittingham, M. S. Science, 1976, 192, 1126 -1127.
(2) Guyomard, D.; Tarascon, J-M. Adv. Mater. 1994, 6, 408.
(3) Ahmad, S. Ionics, 2009, 15, 309 -321.
(4) Wright, P. V. Br. Polymer J.,1975, 7, 319 -327.
(5) Armand, M. B.; Chabagno, J. M.; Duclot, M. Second International Meeting
on Solid Electrolytes, St Andrews, Scotland, 20-22 Sept., 1978, Extended
Abstract.
(6) Takahashi, Y.; Takadoro, H. Macromolecules, 1973, 6, 672-675.
(7) Ratner, M. A.; Shriver, D. F. Chem. Rev. 1988, 88, 109-124.
(8) Vogel, H. Phys. Z. 1921, 22, 645.
(9) Tamman, G.; Hesse, W. Z. Anorg. Allg. Chem. 1926, 156, 245-257.
(10) Fulcher, G. S. J. Am. Ceram. Soc. 1925, 8, 339-355.
(11) Dias, F. B.; Plomp, L.; Veldhuis, J. B. J. J. Power Sources, 2000, 88,
169-191.
(12) Evans, J.; Vincent, C. A.; Bruce, P. G. Polymer, 1987, 28, 2324-2328.
(13) Lee, H. S.; Yang, X. Q.; McBreen, J.; Choi, L. S.; Okamoto, Y. J.
Electrochem. Soc. 1996, 143, 3825-3829.
(14) Lee, H. S.; Yang, X. Q.; Xiang C. L.; McBreen, J.; Choi, L. S.; J.
Electrochem. Soc. 1998, 145, 2813-2818.
72 
 

(15) Lee, H. S.; Yang, X. Q.; Sun, X.; McBreen, J. J. Power Sources 2001,
97-98, 566-569.
(16) Blazejczyk, A.; Wieczorek, W.; Kovarsky, R.; Golodnitsky, D.; Peled, E.;
Scanlon, L. G.; Appetecchi, G. B.; Scrosati, B. J. Electrochem. Soc. 2004,
151, 1762-1766.
(17) Blazejczyk, A.; Szczupak, M.; Wieczorek, W.; Cmoch, P.; Appetecchi, G.
B.; Scrosati, B.; Kovarsky, R.; Golodnitsky, D.; Peled, E. Chem. Mater.
2005, 17, 1535-1547.
(18) Kalita, M.; Bukat, M.; Ciosek, M.; Siekierski, M.; Chung, S. H.; Rodriguez,
T.; Greenbaum, S. G.; Kovarsky, R.; Golodnitsky, D.; Peled, E.; Zane, D.;
Scrosati, B.; Wieczorek, W. Electrochim. Acta, 2005, 50, 3942-3948.
(19) Plewa, A.; Chylinski, F.; Kalita, M.; Bukat, M.; Parzuchowski, P.;
Borkowska, R.; Siekierski, M.; Zukowska, G. Z.; Wieczorek, W. J. Power
Sources, 2006, 159, 431-437.
(20) Mehta, M. A.; Fujinami, T.; Solid State Ionics 1998, 113-115, 187-192.
(21) Mehta, M. A.; Fujinami, T.; Inoue, T.; J. Power Sources, 1999, 81-82,
724-728.
(22) Mehta, M. A.; Fujinami, T.; Inoue, S.; Matsushita, K.; Miwa, T.; Inoue, T.;
Electrochim. Acta 2000, 45, 1175-1180.
(23) Zhang, S. S.; Chang, Z.; Xu, K.; Angell, C. A.; Electrochim. Acta 2000, 45,
1229-1236.
73 
 

(24) Sun, X. G.; Angell, C. A.; Electrochim. Acta 2001, 46, 1467-1473.
(25) Xu, W.; Sun, X. G.; Angell, C. A.; Eectrochim. Acta 2003, 48, 2255-2266.
(26) Croce, F.; Appetecchi, G. B.; Perci, L.; Scrosati, B. Nature 1998, 394,
456-458.
(27) Scrosati, B.; Croce, F.; Panero, S. J. Power Sources, 2001, 100, 93-100.
(28) Johansson, P.; Ratner, M. A.; Shriver, D. F. J. Phys. Chem. B 2001, 105,
9016-9021.
(29) Sun, X. G.; Reeder, C. L.; Kerr, J. B.; Macromolecules, 2004, 37,
2219-2227.
(30) Sun, X. G.; Kerr, J. B.; Macromolecules, 2004, 37, 5133-5135.
(31) Sun, X. G.; Kerr, J. B.; Macromolecules, 2006, 39, 362-372.
(32) Ryu, S-W.; Trapa, P. E.; Olugebefola, S. C.; G-Leon, J. A.;Sadoway, D. R.;
Mayes, A. M. J. Electrochem. Soc. 2005, 152, 158-163.
(33) Sadoway, D. R., Huang, B.; Trapa, P. E.; Soo, P. P.; Bannerjee, P.; Mayes,
A. M. J. Power Sources, 2001, 97-98, 621-623.
(34) Trapa, P. E.; Acar, M. H.; Sadoway, D. R.; Mayes, A. M. J. Electrochem.
Soc. 2005, 152, 2281-2284.
(35) Trapa, P. E.; Reyes, A. B.; Gupta, R. S. D.; Mayes, A. M.; Sadoway, D. R.
J. Electrochem. Soc. 2006, 153, 1098-1101.
(36) Kurian, M.; Galvin, M. E.; Trapa, P. E.; Sadoway, D. R.; Mayes, A. M.
Electrochim. Acta 2005, 50, 2125-2134.
74 
 

(37) Siska, D. P.; Shriver, D. F. Chem. Mater. 2001, 13, 4698-4700.
(38) Snyder, J. F.; Ratner, M. A.; Shriver, D. F. J. Electrochem. Soc. 2003, 150,
1090-1094.
(39) Snyder, J. F.; Hutchison, J. C.; Ratner, M. A.; Shriver, D. F. Chem. Mater.
2003, 15, 4223-4230.
(40) Fujinami, T.; Tokimune, A.; Mehta, M. A.; Shriver, D. F.; Rawsky, G. C.
Chem. Mater. 1997, 9, 2236-2239.
(41) Ito, K.; Ohno, H. Solid State Ionics 1995, 79, 300-305.
(42) Tominaga, Y.; Ohno, H. Solid State Ionics 1999, 124, 323-329.
(43) Endo, T.; Matsumoto, K. J. Polym. Sci. Part A: Polym. Chem. 2010, 48,
3113.
(44) Endo, T.; Matsumoto, K. Macromolecules, 2009, 42, 4580.
(45) Kurian, M.; Galvin, M. E.; Trapa, P. E.; Sadoway, D. R.; Mayes, A. M.
Electrochimica Acta, 2005, 50, 2125-2134.
(46) Singhal, R. G.; Capracotta, M. D.; Martin, J. D.; Khan, S. A.; Fedkiw, P. S.
J. Power Sources, 2004, 128, 247-255.
(47) Riley, M.; Fedkiw, P. S.; Kha, S. A. J. Electrochem. Soc. 2002, 149,
667-674.
(48) Zhang, X-W.; Khan, S. A.; Fedkiw, P. S. Electrochem. Solid-State Lett.
2004, 7, 361-364.
(49) Zhang, X-W.; Fedkiw, P. S. J. Electrochem. Soc. 2005, 152, 2413-2420.
75 
 

(50) Zhang, H.; Zhang, X-W.; Shiue, E.; Fedkiw, P. S. J. Power Sources, 2008,
177, 561-565.
(51) Nugent, J. L.; Moganty, S. S.; Archer, L. A. Adv. Mater. Early View, 2010,
DOI: 10.1002/adma.201000898.
(52) Werne, T.; Patten, T. E. J. Am. Chem. Soc 2001, 123, 7497-7505.
(53) Bombalski, L.; Min, K.; Dong, H.; Tang, C.; Matyjaszewski, K.
Macromolecules, 2007, 40, 7429-7432.
(54) Sunday, D.; C-Medina, S.; Green, D. L. Macromolecules, 2010, 43,
4871-4878.
(55) Jiang, X.; Zhong, G.; Horton, J. M.; Jin, N.; Zhu, L.; Zhao, B.
Macromolecules, 2010, 43, 5387-5395.
(56) Yameen, B.; Kaltbeitzel, A.; Langer, A.; Muller, F.; Goesele, U.; Knoll, W.;
Azzaroni, O. Angew. Chem. Int. Ed. 2009, 48, 3124-3128.
(57) Qin, Y; 2009, PhD Dissertation, MSU.
(58) Zhang, X-W.; Fedkiw, P. S. J. Electrochem. Soc. 2005, 152, 2413-2420.
(59) Asfour, F. H.; 2004, PhD Dissertation, MSU.
(60) Sander, L. C.; Wise, S. A., J. Chromatogr. 1984, 316, (DEC), 163-181.
(61) Ue, M. J. Electrochem. Soc. 1994, 141, 3336-3342.
(62) Dudek, S. P.; Skes, H. D.; Chidsey, C. E. D. J. Am. Chem. Soc. 2001, 123,
8033-8038.
(63) Markova, D.; Kumar, A.; Klapper, M.; Mullen, K.; Polymer, 2009, 50,
76 
 

3411-3421.

77