Composite electrolytes comprised of poly(ethylene oxide) and silica nanoparticles with grafted poly(ethylene oxide)-containing polymers

Article abstract of DOI:10.1039/c4ra07262f

We designed, synthesized and characterized several novel hybrid inorganic/organic nanocomposite electrolytes that consist of poly(ethylene oxide) (PEO) based polymer grafted from silica nanoparticles. Poly(ethylene glycol)methyl ether methacrylate (PEGMA) was tailored on the silica surface through atom transfer radical polymerization (ATRP). A series of silica-polymers were synthesized with different lengths of PEO side chains. Electrolytes were prepared from the functionalized particles and low-molecular weight polyethylene glycol dimethyl ether (PEGDME) with the addition of LiI. Upon the introduction of particles, electrolytes became viscous and gel-like. With the increase of PEO side chains, the viscosity of the electrolytes increased dramatically, among which, silica-poly(PEGMA-1100) became solid-state. The room temperature conductivities of the hybrid silica-polymer electrolytes are in the range of 6 × 10^-5to 1.2 × 10^-4S cm^-1. Silica-poly(PEGMA-475) and silica-poly(PEGMA-1100), with higher viscosity, exhibited better ionic conductivity. Surface-initiated copolymerization was also conducted to optimize the electrochemical performance of polymer coated silica nanoparticles. This journal is

Full text of DOI:10.1039/c4ra07262f

View Article Online
View Journal
RSC Advances
This article can be cited before page numbers have been issued, to do this please use: Z. Jia, W. Yuan, H.
Zhao, H. hu and G. L. Baker, RSC Adv., 2014, DOI: 10.1039/C4RA07262F.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. This Accepted Manuscript will be replaced by the edited,
formatted and paginated article as soon as this is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/advances

Page 1 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
This work reported an extensive study of poly(ethylene oxide)-containing polymers grafted silica
nanoparticles and their performance as polymer composite electrolytes

RSC Advances
Page 2 of 39
View Article Online
DOI: 10.1039/C4RA07262F
Composite Electrolytes Comprised of Poly(Ethylene Oxide) and Silica Nanoparticles
with Grafted Poly(Ethylene Oxide)-Containing Polymers
Zhe Jia, Wen Yuan, Hui Zhao, Heyi Hu, Gregory L. Baker§
Department of Chemistry, Michigan State University, East Lansing, Michigan 48824,
United States
Correspondence to: Zhe Jia (jiazhe@msu.edu)
§ Prof. Gregory L. Baker passed away on October 18, 2012.We dedicate this work as a
memorial to him
1

Page 3 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
Abstract
We designed, synthesized and characterized several novel hybrid inorganic/organic
nanocomposite electrolytes that consist of poly(ethylene oxide) (PEO) based polymer
grafted from silica nanoparticles. Poly(ethylene glycol) methyl ether methacrylate
(PEGMA) was tailored on the silica surface through atom transfer radical polymerization
(ATRP). A series of silica-polymers were synthesized with different length of PEO side
chains. Electrolytes were prepared from the functionalized particles and low-molecular
weight polyethylene glycol dimethyl ether (PEGDME) with the addition of LiI. Upon the
introduction of particles, electrolytes became viscous and gel-like. With the increase of
PEO side chains, viscosity of the electrolytes increased dramatically, among which,
silica-poly(PEGMA-1100) became solid-state. The room temperature conductivity of
hybrid silica-polymer electrolytes are in the range of 6×10-5 to 1.2×10-4 S/cm. Silica-
poly(PEGMA-475) and silica-poly(PEGMA-1100), with higher viscosity, represented
better ionic conductivity. Surface-initiated copolymerization was also conducted to
optimize the electrochemical performance of polymer coated silica nanoparticles.
Keywords
Composite electrolyte; surface-initiated ATRP; poly(ethylene glycol) methyl ether
methacrylate; conductivity
2

RSC Advances
Page 4 of 39
View Article Online
DOI: 10.1039/C4RA07262F
1
Introduction
The development of sustainable energy-harvesting devices is perhaps the most
crucial technological challenge for mankind,1, 2 and dye-sensitized solar cells (DSSCs)3-5
and rechargeable lithium ion batteries (LIBs)6-8 represent important advances in this area.
Electrolytes that carry charge between the electrodes of these systems strongly affect
device construction and efficiency.9 Of particular relevance to this work, difficulties in
sealing liquid electrolytes in batteries or solar cells and safety issues raised during device
operation are driving the development of solid-state electrolytes.10 One approach for
creating highly conductive solid electrolytes employs organic/inorganic nanocomposite
materials, which often combine the advantages of both organic and inorganic
components.11, 12 These nanocomposites usually have core-shell architectures comprised
of organic polymer shells and inorganic nanoscale cores, to combine the flexibility, and
processability of organic polymers with the rigidity and thermal stability of the inorganic
component.13
Among the candidates for replacing conventional liquid electrolytes,
poly(ethylene oxide) (PEO−) based polymers are particularly attractive for LIBs,14
DSSCs,15 fuel cells16 and other solid-state electrochemical devices because these
polymers are relatively stable and dissolve high levels of salts due to their high polarity.17
In 1973, Wright first studied the ionic conductivity of PEO-alkali metal salt complexes.18
Since that time, many groups worked to develop PEO and its derivative polymers as
stable and highly conducting solid polymer electrolytes. However, the ion conduction
only occurs in the amorphous PEO phase, where conductivity is two to three orders of
magnitude higher than in the crystalline phase,19 so the ionic conductivities of common
3

Page 5 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
PEO-salt systems are not satisfactory because of PEO crystallization.20 Improvements in
conductivity require the introduction of disorder to prevent crystallization,21 and
approaches to this end including blending polymers, cross-linking, copolymerization, and
plasticizer addition.17, 22
With the development of controlled living radical polymerization, including atom
transfer radical polymerization (ATRP),23-26 reversible addition-fragmentation chain
transfer radical polymerization (RAFT)27-30 and nitroxide-mediated polymerization
(NMP)31-34, the scope of surface modification with polymers increased dramatically.
Researchers have synthesized numerous inorganic/organic composite materials with
tailored polymer shells coated on inorganic cores.35 After anchoring an initiator to the
target surface, initiation of polymerization leads to a dense layer of immobilized polymer
via a “grafting from” approach.36-38 Our lab invoked controlled living radical
polymerization to grow customized polymers from silica39 and gold.40
The research described here examines the hypothesis that utilizing controlled
polymerization of optimized PEO-based polymers from silica nanoparticles and
incorporation of these particles in pure PEO with metal salts will lead to solid or highly
viscous materials with high conductivity. The nanoparticle should enhance mechanical
strength and may inhibit PEO crystallization to increase conductivity. Grafted
copolymers with a significant fraction of short PEG side chains should also lead to a
deceased crystallization and increased conductivity.
Scheme 1 shows the synthesis of modified electrolytes and their incorporation in
PEO. We synthesized several hybrid inorganic/organic electrolytes by first polymerizing
poly(ethylene glycol) methyl ether methacrylate (PEGMA) from silica nanoparticles via
4

RSC Advances
Page 6 of 39
View Article Online
DOI: 10.1039/C4RA07262F
surface-initiated ATRP. The hydroxyl groups on the surface of the silica particles afford
sites for initiator attachment and subsequent PEGMA growth, and grafting of both
homopolymers and copolymers from particle surfaces occurs with controlled growth rates
and thicknesses. Corresponding free polymers were synthesized in solution via ATRP to
monitor macromolecule behavior.
Scheme 1
Surface-initiated polymerization from a silica nanoparticle and subsequent
formation of a composite electrolyte
One of the target applications of these nanocomposite PEGMA electrolytes is
−
DSSCs, so we selected LiI as the lithium source during electrolyte preparation. The I−/I3
redox couple is the most common system employed to regenerate dyes in DSSCs. Thus,
we prepared electrolytes from the functionalized particles and low-molecular weight
polyethylene glycol dimethyl ether (PEGDME) with the addition of LiI and I2. Upon the
introduction of particles in PEGDME, electrolytes formed gels, and with the increasing
5

Page 7 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
length of PEO side chains in grafted polymers, the viscosity of the electrolytes increased
dramatically. Some electrolytes solidified. Surprisingly, electrolytes containing longer
PEO side chains show higher ionic conductivity than electrolytes with shorter side chains
on the silica coated particles, despite much lower viscosity with shorter side chains.
Surface-initiated copolymerization allows optimization of the ion conductivity of the
composite electrolytes via prevention of crystallization. Conductivity values for hybrid
silica-copolymer electrolytes ranged from 1.1×10-4 to 1.5×10-4 S/cm, an order-of-
magnitude improvement compared to the homopolymer-coated silica particle electrolytes,
and very close to the conductivity of free poly(PEGMA) electrolytes, which have much
lower viscosity. The high viscosity of the hybrid electrolytes is attractive for their
encapsulation.
2
Experimental
2.1 Materials
Lithium iodide (crystalline powder, 99.9%), cetyltrimethylammonium bromide
(CTAB), copper (I) bromide (CuBr, 99.999%), ethyl 2-bromoisobutyrate (EBiB, 98%),
methacryloyl chloride (97%), 10-undecen-1-ol (98%), 2-bromoisobutyryl bromide (98%),
trichlorosilane (99%), tri(ethylene glycol) monomethyl ether (96%), and di(ethylene
glycol) monomethyl ether (96%) were obtained from Aldrich and used as received. N, N,
N’, N’-ethylenediaminetetraacetic acid (disodium salt dehydrate, EDTA·2Na) was
purchased from Spectrum. Triethylamine and hydroquinone were obtained from J. T.
Baker Chemical and used as received. Iodine (Aldrich, 99.99%) and 2, 2-bipyridine (bpy,
Aldrich, 99%) were sublimed prior to use. Poly(ethylene glycol) methyl ether
6

RSC Advances
Page 8 of 39
View Article Online
DOI: 10.1039/C4RA07262F
methacrylates (Aldrich, average Mn=300, 475, or 1100) were passed through an activated
basic alumina column before use. Snowtex-XS (7-10 nm) was a gift from Nissan
Chemical, and the colloidal silica was received as an aqueous dispersion in pH 9-10
water. All other chemicals and solvents were ACS reagent grade and used as received
from commercial suppliers without further purification unless otherwise specified.
2.2 Instruments
1H NMR spectra were obtained using Varian Inova 500 MHz instruments.
Thermogravimetric analyses (TGA) were performed in air using a Perkin-Elmer TGA 7
Instrument, with a heating rate of 10 °C/min; samples were heated up to 850 °C after
holding at 120 °C for 30 min prior to the temperature ramp. FT-IR spectra were collected
with a Mattson Galaxy 300 spectrometer through KBr pellets. Differential scanning
calorimetry (DSC) was carried out using a TA DSC Q100 instrument with a
heating/cooling/heating cycle of 10 °C/min under nitrogen. Transmission electron
microscopy (TEM) images were obtained with a high resolution JEOL100 CXII
instrument with both digital and film image recording capabilities. Samples were spotted
on copper grids coated with ultrathin polymer films (PELCO® TEM grid 01822).
Electrical impedance spectroscopy (EIS) analysis was done with a Hewlett-Packard
4192A LF Impedance Analyzer to obtain temperature-dependent ionic conductivity data.
Electrolytes were pipetted into a cell and resistance measurements were taken when
voltage was passed through the cell at 5 ºC intervals from 30-90 ºC. The sample cell uses
two steel disks as symmetrical electrodes separated by a sample 0.6 cm in radius and 0.02
cm in thickness. The EIS testing apparatus were placed inside a N2-filled glove box and
all measurements were conducted inside the glove box to avoid moisture.
7

Page 9 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
2.3 Synthesis and characterization of initiator-coated silica nanoparticles (silica-
initiator)
Preparation of silica-initiator nanoparticles. Particle preparation followed a
literature procedure.39 1 g of cetyltrimethylammonium bromide (CTAB) was added into
20 mL of Snowtex-XS, and the mixture was stirred at room temperature for 30 min,
during which the particles aggregated and precipitated from solution. Silica particles were
collected by centrifugation and were then redispersed in 100 mL deionized water to
remove CTAB residues. Particles were recollected by centrifugation, and the
collecting/redispersing process was repeated 20 times to remove CTAB as completely as
possible. Finally, silica particles were dried under high vacuum≤ (10 mTorr) at 70 °C
overnight.
(11-(2-bromo-2 methyl) propionyloxy)undecyl-trichlorosilane was synthesized
via a two-step literature procedure.41 10-undecen-1-ol (1.71g, 10mmol) and pyridine
(0.96 mL, 12 mmol) were dissolved in 10 ml of distilled THF, followed by the addition
of 2-bromoisobutyryl bromide (1.24 mL, 10 mmol) drop by drop over a period of 10
minutes. The mixture was stirred at room temperature under N2 atmosphere overnight.
Following dilution with 20 mL hexane, and washing with 3N HCl three times and
distilled water three times, the organic phase obtained was dried over Na2SO4 and filtered
under reduced pressure. Finally, solvent was removed and a colorless oily liquid product,
1
10-undecen-1-yl 2-bromo-2-methylpropionate, was obtained (2.63 g, 82.5% yield). H
NMR (500 MHz, CDCl3): δ 1.22 – 1.41 (m, 12H), 1,60 – 1.70 (m, 2H), 1.92 (s, 6H), 1.97
– 2.06 (m, 2H), 4.15 (t, 2H), 4.86 – 5.02 (m, 2H), 5.85 – 5,72(m, 1H).
8

RSC Advances
Page 10 of 39
View Article Online
DOI: 10.1039/C4RA07262F
10-undecen-1-yl 2-bromo-2-methylpropionate (2 g, 6.27 mmol) and
trichlorosilane (5 mL, 50.7 mmol) were added into a flame-dried flask, followed by the
addition of Karstedt’s catalyst (10 μL). The reaction took place at room temperature
overnight under N2, and, the mixture was passed through a flash column to remove
catalyst. Excess trichlorosilane was removed by vacuum distillation (60 °C, 100 mTorr),
to give a colorless oil, (11-(2-Bromo-2-methyl)propionyloxy) undecyltrichlorosilane
1
(1.81 g, 72% yield). H NMR (500 MHz, CDCl3): δ 1.21 – 1.41 (t, 2H), 1,46 – 1.75 (m,
4H), 1.92 (s, 6H), 4.15 (t, 2H).
Freshly prepared surface initiator, (11-(2-bromo-2 methyl) propionyloxy)undecyl-
trichlorosilane, was dissolved in dry toluene, and stirred with silica nanoparticles under
N2 at 60 – 70 °C overnight. Initiator was in excess to ensure effective anchoring onto the
surface of the silica nanoparticles. The initiator-coated particles were collected by
centrifugation, dispersed into toluene, washed with pentane, and then recollected by
centrifugation. The redispersion /wash/centrifugation/recollecting cycle was repeated at
least 10 times to remove any free initiator from the solution or particle surface. The
initiator-coated nanoparticles were finally dried under vacuum≤ (10 mTorr) at 70 °C
overnight.
Calculation of initiator content grafted on silica-initiator nanoparticles. After
drying, TGA was conducted to determine the amount of initiator attached to the particles.
TGA was performed under a dry air atmosphere, holding the temperature at 120 °C for 30
min before starting the temperature ramp at 10 °C/min up to 850 °C. The TGA data
showed 23.7% weight loss, implying that 1 mg of initiator-coated nanoparticles contained
0.237 mg initiator. The molar mass of the initiator is 342.54 g/mol, so there is 6.9 × 10-7
9

Page 11 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
mole of initiator per mg of silica-initiator. Thus, when surface-initiated ATRP was
conducted with 62 mg of initiator coated particles, the solution contained 0.042 mmol of
initiator. Therefore, for each surface-initiated ATRP reaction, the monomer (2 mmol) to
initiator ratio was roughly 47:1.
2.4 Surface-initiated ATRP of silica-poly(PEGMA) homopolymers
Synthesis of tri(ethylene glycol) methyl ether methacrylate.42 Tri(ethylene
glycol) monomethyl ether (13.7 g) and 20 mL of triethylamine (0.14 mol) were dissolved
in 50 mL of diethyl ether in a three-neck round-bottom flask and placed in an ice bath. To
this flask a solution containing 6.87 mL of methacryloyl chloride (0.07 mol) in 15 mL of
ethyl ether was added dropwise over 15 minutes. The reaction was allowed to proceed
overnight at room temperature under N2. The mixture was filtered under reduced pressure
to remove the precipitate, and the residue was purified by silica gel column
chromatography (hexanes/ethyl acetate 1/2, v/v). After removal of the solvent, a small
amount (ca. 3 mg) of hydroquinone was added to the product as a polymerization
inhibitor. The product was further purified by vacuum distillation at 120 °C, 100 mTorr
1
to obtain a colorless liquid (11.29 g, 70% yield). H NMR (500 MHz, CDCl3): δ 1.92 (s,
3H), 3.37 (t, 3H), 3.64 (m, 10H), 4.28 (t, 2H), 5.55 (s, 1H), 6.11 (s, 1H).
Synthesis of di(ethylene glycol) methyl ether methacrylate.42 This synthesis
followed the same protocol described with the same molar ratios of reactants to obtain a
1
colorless liquid (78% yield). H NMR (500 MHz, CDCl3): δ 1.93 (s, 3H), 3.36 (t, 3H),
3.64 (m, 8H), 4.28 (t, 2H), 5.55 (t, 1H), 6.11 (q, 1H).
Surface-initiated ATRP of tri(ethylene glycol) methyl ether methacrylate
(Silica-poly(PEGMA-232)).11 The monomer was first dissolved in methanol and passed
10

RSC Advances
Page 12 of 39
View Article Online
DOI: 10.1039/C4RA07262F
through an activated basic alumina column to remove inhibitor. After removing the
methanol, 0.464 g of tri(ethylene glycol) methyl ether methacrylate (2 mmol), 28 mg of 2,
2-bipyridine (bpy, 0.18 mmol), and 62 mg of initiator grafted-particles (ca. 0.042 mmol
of initiators) were dissolved or dispersed in 1 mL of deionized water and 3 mL of
methanol in a Schlenk flask. The mixture was ultrasonicated for 15-20 minutes. After 3
freeze-pump-thaw cycles, the Schlenk flask was filled with nitrogen, and 8 mg of CuBr
(0.06 mmol) was added to the mixture, which became dark brown. After another freeze-
pump-thaw cycle, the reaction was allowed to take place at room temperature for 12
hours. To quench the reaction, the Schlenk flask was opened and exposed to air, and the
solution turned blue. The mixture was transferred to a 50-mL centrifuge tube and washed
with an EDTA•2Na-saturated solution several times, until no blue color was observed.
Subsequent washes employed deionized water 10 times and acetone 3 times. In each step
the polymer-grafted silica particle was recollected by centrifugation. The product was
dried under vacuum (≤ 10 mTorr) at 70 °C overnight. After drying, TGA was performed
to determine the extent of polymer grafting on the particles.
Surface-initiated ATRP of silica-poly(PEGMA-188), silica-poly(PEGMA-300),
silica-poly(PEGMA-475), and silica-poly(PEGMA-1100) followed the same protocol
with 2 mmol of monomer in the polymerization solution in all cases.
2.5 Surface-initiated ATRP of silica-copolymer
Surface-initiated ATRP of Silica-Copolymer-1 (Silica-poly(PEGMA-1100-co-
PEGMA-232), 1:1). Two monomers for copolymerization, PEGMA-1100 and PEGMA-
232, were dissolved separately in methanol and passed through an activated basic
alumina to remove inhibitor. Copolymerization of PEGMA-1100 and PEGMA-232 at a
11

Page 13 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
1:1 monomer ratio was carried out via surface-initiated ATRP. Initially, 1.1 g of
poly(ethylene glycol) methyl ether methacrylate (Mn=1100, 1 mmol, PEGMA-1100),
0.232 g of tri(ethylene glycol) methyl ether methacrylate (1 mmol, PEGMA-232), 28 mg
of 2, 2-bipyridine (bpy, 0.18 mmol), and 62 mg of initiator-modifies particles (ca. 0.042
mmol) were dissolved or dispersed in 1 mL of deionized water and 3 mL of methanol in a
Schlenk flask. The mixture was ultrasonicated for 15-20 minutes. After 3 freeze-pump-
thaw cycles the Schlenk flask was filled with N2, and 8 mg of CuBr (0.06 mmol) was
added to the solution, which became dark brown. After another freeze-pump-thaw cycle,
the reaction was allowed to proceed at room temperature for 12 hours. To quench the
reaction, the Schlenk flask was exposed to air, and the solution turned blue. The mixture
was transferred to a 50-mL centrifuge tube and washed with an EDTA•2Na saturated
solution 3 times until no blue color was observed, deionized water 8-12 times, and
acetone twice. In each step the polymer-grafted silica particle was recollected via
centrifugation. The product was dried in vacuum at 60 °C overnight. After drying, TGA
was performed to determine the extent of polymer grafting.
Surface-initiated ATRP of silica-copolymer-2 (silica-poly(PEGMA-1100-co-
PEGMA-300), 1:1), silica-copolymer-3 (silica-poly(PEGMA-1100-co-PEGMA-475),
1:1), silica-copolymer-4 (silica-poly(PEGMA-300-co-PEGMA-475), 1:1), silica-
copolymer-5 (silica-poly(PEGMA-1100-co-PEGMA-232), 3:7), silica-copolymer-6
(silica-poly(PEGMA-1100-co-PEGMA-300),
3:7),
silica-copolymer-7
(silica-
poly(PEGMA-1100-co-PEGMA-475), 3:7), silica-copolymer-8 (silica-poly(PEGMA-
300-co-PEGMA-475), 3:7), silica-copolymer-9 (silica-poly(PEGMA-300-co-PEGMA-
475), 7:3) were conducted following the same protocol with the specified monomer ratios.
12

RSC Advances
Page 14 of 39
View Article Online
DOI: 10.1039/C4RA07262F
In all cased the total amount of monomer in solution was 1 mmol. Fig. 2 and Table 2
describe TGA profiles and weight loss.
2.6 ATRP of free poly(PEGMA) homopolymers and copolymers in solution
ATRP of free tri(ethylene glycol) methyl ether methacrylate (poly(PEGMA-
232)) in solution. The monomer was dissolved in methanol and passed through an
activated basic alumina column to remove inhibitor. After removing solvent, 1160 mg of
tri(ethylene glycol) methyl ether methacrylate (5 mmol), 19.5 mg of ethyl 2-
bromoisobutyrate (0.1 mmol, EBiB), and 62.5 mg of 2, 2-bipyridine (bpy, 0.4 mmol)
were dissolved in 1 mL of deionized water and 3 mL of methanol in a Schlenk flask. The
mixture was ultrasonicated for 15-20 minutes. After 3 freeze-pump-thaw cycles, the
Schlenk flask was filled with nitrogen, 28.7 mg of CuBr (0.2 mmol) was added, and the
solution immediately turned brown. After another freeze-pump-thaw cycle, the reaction
was allowed to proceed at room temperature for 12 hours. To quench the reaction, the
Schlenk flask was exposed to air, after which the solution turned blue. The mixture was
diluted in deionized water and passed through an activated basic alumina column to
remove copper catalyst. After removing solvent, the polymer was dried under vacuum (≤
1
10 mTorr) for 24 hours at 70 °C. Since no vinyl proton signal was observed in the H
NMR spectrum, indicating all monomers were consumed during polymerization, the
reaction conversion should be ≥ 99%. The polymer (1.037 g, 89% yield) obtained was a
colorless, highly viscous liquid.
ATRP of poly(PEGMA-188), poly(PEGMA-300), poly(PEGMA-475), and
poly(PEGMA-1100) free polymers followed the same protocol using the same number
of moles of monomer in each case.
13

Page 15 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
ATRP of free copolymer-1 (poly(PEGMA-1100-co-PEGMA-232), 1:1),
copolymer-2 (poly(PEGMA-1100-co-PEGMA-300), 1:1), copolymer-3 (poly(PEGMA-
1100-co-PEGMA-475), 1:1), copolymer-4 (poly(PEGMA-300-co-PEGMA-475), 1:1),
copolymer-5 (poly(PEGMA-1100-co-PEGMA-232), 3:7), copolymer-6 (poly(PEGMA-
1100-co-PEGMA-300), 3:7), copolymer-7 (poly(PEGMA-1100-co-PEGMA-475), 3:7),
copolymer-8 (poly(PEGMA-1100-co-PEGMA-475), 3:7), copolymer-9 (poly(PEGMA-
300-co-PEGMA-475), 7:3) were conducted following the same protocol with different
copolymerization ratio calculated. The total amount of monomer employed in all cases
was 1 mmol.
2.7 Preparation of hybrid silica/polymer and free polymer electrolytes
Hybrid silica-poly(PEGMA) nanocomposite electrolyte preparation. Hybrid
electrolytes were prepared by dissolving LiI and I2 (10:1 mole ratio) in the host
poly(ethylene glycol) dimethyl ether (PEGDME-500, Mn = 500) electrolyte, and then
dispersing PEGMA/silica nanoparticles in this PEGDME. The mole ratio of PEGDME-
500/LiI/I2 was fixed at 100:10:1. Since the number of ethylene oxide (EO) repeating units
in PEGDME-500 is around 10, the EO to Li ratio is 100:1 in the PEGDME-500/LiI/I2
mixture. Specifically, 78 mg of I2 (0.3 mmol) and 403.7 mg of LiI (3 mmol) were mixed
into 15 g of PEGDME-500 (30 mmol) in a vial. The vial was sealed with parafilm,
covered with aluminum foil, and the electrolyte was stirred overnight until the salt
species dissolved. A clear, dark orange/brown liquid resulted. In this electrolyte, every
258 mg of liquid contained 5 mmol PEO (50 mmol EO), 0.5 mmol LiI, and 0.005 mmol
I2 (PEGDME-500/LiI/I2 = 100/10/1). When dispersing nanoparticles in this electrolyte, to
guarantee the same EO/Li ratio for all hybrid electrolytes, the mole ratio of PEGMA EO
14

RSC Advances
Page 16 of 39
View Article Online
DOI: 10.1039/C4RA07262F
to the previously added lithium salt was fixed at 10:1 during each particle addition.
Therefore, when including ethylene oxide units in both PEGDME-500 and silica-
poly(PEGMA), the total EO/Li ratio was 110:1 for all hybrid electrolytes. The amount of
poly(PEGMA)-coated silica to add to the electrolyte was calculated according to TGA
data and molecular weight (for the copolymers, we assumed that the ratio of monomers
incorporated into the polymer was the same as that in solution). Therefore, to make
hybrid electrolyte, in every 258 mg of freshly prepared PEGDME-500/LiI/I2 electrolyte,
the following amount of hybrid particles were added (0.5 mmol EO from silica-
poly(PEGMA) were added in all cases): 57.1 mg silica-poly(PEGMA-188), 43.1 mg
silica-poly(PEGMA-232), 33.1 mg silica-poly(PEGMA-300), 29.4 mg silica-
poly(PEGMA-475), 26.5 mg silica-poly(PEGMA-1100), 31.3 mg silica-poly(PEGMA-
1100-co-PEGMA-232) (1:1), 28.6 mg silica-poly(PEGMA-1100-co-PEGMA-300) (1:1),
29.9 mg silica-poly(PEGMA-1100-co-PEGMA-475) (1:1), 31.4 mg silica-poly(PEGMA-
300-co-PEGMA-475), 32.4 mg silica-poly(PEGMA-1100-co-PEGMA-232) (3:7), 29.1
mg silica-poly(PEGMA-1100-co-PEGMA-300) (3:7), 30.5 mg silica-poly(PEGMA-
1100-co-PEGMA-475) (3:7), 30.3 mg silica-poly(PEGMA-300-co-PEGMA-475) (3:7),
31.5 mg silica-poly(PEGMA-300-co-PEGMA-475) (7:3). Electrolytes containing silica-
poly(PEGMA-1100) were particularly viscous and solidified easily. A Wig-L-Bug was
used to homogenize these particles in the PEGDME-500/LiI/I2 electrolyte, while the
other samples were stirred for 24 hours with a magnetic stirring bar in a parafilm-sealed,
aluminum-covered vial. For a control experiment, a set of electrolytes containing
PEGDME-500/LiI/I2 with no particles but having the same EO/Li ratio was also prepared
15

Page 17 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
by adding 25 mg PEGDME-500 into 258 mg of previously prepared PEGDME-500/LiI/I2
electrolyte.
Poly(PEGMA) free polymer electrolyte preparation. The mole ratio of
PEGDME-500/LiI/I2 was again fixed at 100:10:1, with EO:Li = 100:1 in the mixture.
While dispersing free polymers into the system, the mole ratio of PEGMA EO to lithium
salt was fixed at 10:1. Therefore, when including EO units in both PEGDME-500 and
free poly(PEGMA), the total EO/Li ratio was fixed at 110:1 in all electrolytes. The
amount of free poly(PEGMA) was calculated according to molecular weight. For every
258 mg of PEGDME-500/LiI/I2 electrolyte, the following amounts of free polymers were
added (0.5 mmol of EO in poly(PEGMA) in all cases): 38.8 mg poly(PEGMA-232), 29.9
mg poly(PEGMA-300), 26.5 mg poly(PEGMA-475), 23.9 mg poly(PEGMA-1100), 25.6
mg poly(PEGMA-1100-co-PEGMA-232) (1:1), 25.0 mg poly(PEGMA-1100-co-
PEGMA-300) (1:1), 24.6 mg poly(PEGMA-1100-co-PEGMA-475) (1:1), 27.7 mg
poly(PEGMA-300-co-PEGMA-475) (1:1), 27.4 mg poly(PEGMA-1100-co-PEGMA-232)
(3:7), 26.0 mg poly(PEGMA-1100-co-PEGMA-300) (3:7), 25.1 mg poly(PEGMA-1100-
co-PEGMA-475) (3:7), 27.0 mg poly(PEGMA-300-co-PEGMA-475) (3:7), 28.4 mg
poly(PEGMA-300-co-PEGMA-475) (7:3). All these electrolytes were dark brown,
viscous liquids.
2.8 Ionic conductivity measurements
Temperature-dependent ionic conductivity measurements were determined from
electrical impedance analysis (EIS) using a homemade cell. The data were obtained with
a Hewlett-Packard 4192A LF Impedance Analyzer over a frequency range of 5 Hz to 13
MHz with an applied oscillating voltage of 10 mV. The sample cell uses two steel disks
16

RSC Advances
Page 18 of 39
View Article Online
DOI: 10.1039/C4RA07262F
as symmetrical electrodes separated by a 0.02 cm-thick sample in a Teflon collar. The
sample radius is 0.6 cm. All electrolytes were pipetted into the cell, and impedance
measurements were obtained at 5 ºC intervals from 25 – 90 ºC. The sample was
equilibrated at the pre-determined temperature for 15 minutes before each measurement.
Ionic conductivity was calculated:
σ = 1/ρ = l/(R·A)
where l (cm) refers to the measured distance and A (cm2) is the cross-sectional area. To
avoid moisture, all samples were dried under vacuum (≤ 10 mTorr) at 70 °C for at least
48 hours. All conductivity measurements were conducted in the glove box.
3
Results and discussion
3.1 Synthesis of silica-poly(PEGMA) nanoparticles
The first step in growing polymers from silica nanoparticles is attachment of an
ATRP initiator to the surface via silanization (Scheme 2). Synthesis of the silane-
containing
surface
initiator,
(11-(2-bromo-2-methyl)
propionyloxy)undecyl-
trichlorosilane, proceeded in two steps according to a literature procedure.41 The initiator
was then anchored onto silica surface via silane chemistry. Subsequent surface-initiated
ATRP (Scheme 3) of poly(ethylene glycol) methyl ether methacrylate (PEGMA) (Table
1) gave the modified particles. Variation of the length of the PEG chain in PEGMA
should allow tailoring of particle properties.
17

Page 19 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
Scheme 2
Synthesis of a silane-containing ATRP initiator, and attachment of this
initiator to silica nanoparticles
Scheme 3
Surface-initiated ATRP of PEGMA homopolymer from silica-initiator
nanoparticles to form silica-poly(PEGMA)
18

RSC Advances
Page 20 of 39
View Article Online
DOI: 10.1039/C4RA07262F
Table 1
Silica-poly(PEGMA) synthesized in this work
Silica-Polymer
Silica-poly(PEGMA-188)
Silica-poly(PEGMA-232)
Silica-poly(PEGMA-300)
Silica-poly(PEGMA-475)
Monomer
xa
2
PEGMA-188, Mn=188
PEGMA-232, Mn=232
PEGMA-300, Mn=300
3
6
PEGMA-475, Mn=475 10
Silica-poly(PEGMA-1100) PEGMA-1100, Mn=1100 25
a
x refers to an estimation of the number of ethylene oxide repeating units in the
monomer, as derived from the average molecular weight given by the manufacturers.
FT-IR spectroscopy confirms the polymer grafting. The FT-IR spectrum (Fig. 1)
of bare silica nanoparticles shows a characteristic broad absorption band associated with
Si-O-Si stretching (1100 cm-1), and a sharp band at around 800-770 cm-1 corresponding
to Si-OH stretching. The C-H stretching band at 3000-2800 cm-1 likely stems from
residual CTAB. The abundant broad band at 3600-3200 cm-1 (Fig. 1a) is due to the
hydroxyl groups on the silica surface, which are partly consumed after reaction with the
initiator (11-(2-bromo-2-methyl)propionyloxy) undecyltrichlorosilane. After surface-
initiated ATRP, the ester carbonyl stretch of the methacrylate appears between 1750 and
1700 cm-1. Another characteristic peak associated with the C-O-C stretching of PEO is
present between 1300 and 1100 cm-1. This absorbance along with C-H stretching modes
(3000 – 2800 cm-1) become less intense with increasing length of the PEO side chain.
19

Page 21 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
Fig. 1 FT-IR spectra of (a) bare silica particles, (b) silica-initiator nanoparticles, (c)
silica-poly(PEGMA-1100) nanoparticles, (d) silica-poly(PEGMA-475) nanoparticles, (e)
silica-poly(PEGMA-300) nanoparticles, (f) silica-poly(PEGMA-232) nanoparticles, and
(g) silica-poly(PEGMA-188) nanoparticles
The polymer grafted nanoparticles are also confirmed through TEM (Fig. S1).
Bare silica particles precipitated by CTAB show diameters around 16 nm, which is about
three times the diameter given by the manufacture. This may reflect some particle
aggregation during the extensive wash/centrifugation/redispersion process to remove
excess surfactant. After initiator attachment, aggregation became more severe even with
ultrasonication during sample preparation. TEM clearly shows the aggregation (Fig. S1b).
20

RSC Advances
Page 22 of 39
View Article Online
DOI: 10.1039/C4RA07262F
This is presumably due to the long PEO-containing polymer side chains were tangle
together make them hard to isolate into well-dispersed single particle. Moreover, Fig. S1c
shows that after surface-initiated polymerization, silica-poly(PEGMA) particles can
aggregate into a 100-nm particles. The dark spots in Fig. S1c are the silica cores, while
the abundant gray area between these cores is the poly(PEGMA) shells. The TEM image
suggests significant polymer grafting from the silica surface, and thermogravimetric
analysis was performed to quantitatively determine the amount of organic content.
3.2 Thermal properties of silica-poly(PEGMA) nanocomposites
Thermogravimetric analysis (TGA) allows estimation of the extent of surface-
initiated polymerization. TGA occurred in air with temperatures ramped to 850 °C.
Under the presence of oxygen at this high temperature, the organic portion (i.e. polymer
and initiator) of the modified particles completely decompose and vaporizes with air flow,
to leave behind an inorganic silica core. Therefore, weight loss quantitatively reveals the
amount of polymer grafted on silica particles. Table 2 and experimental section examine
the extent of polymer growth in more detail.
21

Page 23 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
Fig. 2 TGA data for silica-initiator and silica-poly(PEGMA) nanoparticles
Table 2
DSC and TGA result for silica-poly(PEGMA)
Tg (°C)a Tm (°C)a Weight loss b Maximum weight lossc
Silica-poly(PEGMA)
Silica-poly(PEGMA-188)
Silica-poly(PEGMA-232)
Silica-poly(PEGMA-300)
Silica-poly(PEGMA-475)
-49.4
-61.6
-62.4
-70.5
n/a
n/a
n/a
82.2%
89.2%
90.8%
89.9%
90.3%
88.2%
90.2%
92.2%
95.7%
97.7%
n/a
-7.3
35.7
Silica-poly(PEGMA-1100)
a
b
Tg and Tm of silica-polymer nanoparticles were determined from DSC data. % of
c
weight loss was obtained from TGA. Materials were heated up to 850°C under air.
22

RSC Advances
Page 24 of 39
View Article Online
DOI: 10.1039/C4RA07262F
Maximum weight loss is the theoretical weight loss calculated from monomer to initiator
ratio during polymerization. The calculation details are in experimental section.
The Tg of pure PEO ranges from -60 to -30 °C, depending on the number of
repeating ethylene oxide (EO) units. Additionally, when the number of EO repeating
units is > 8, PEO tends to crystallize,43 choosing its most stable conformation—the 72
helix.44 Upon crystallization, the polymer chain becomes rigid, and hence the
conductivity of PEO-based electrolyte decreases. DSC data (Fig. 3) show that with
shorter PEO side chains (e.g. silica-poly(PEGMA-188), silica-poly(PEGMA-232), and
silica-poly(PEGMA-300)), no crystallization occurs so a melting temperature is not
observed over the operation range. However, for silica-poly(PEGMA-475) and silica-
poly(PEGMA-1100), especially the latter, a sharp and strong endothermic peak due to
crystallization appears in the DSC curve. Fig. 3 and Table 2 show that the value of Tg
decreases with the length of the PEO side chain, and the Tg of silica-poly(PEGMA-1100)
is below the range of our instrument (≥ −85 °C) . Below Tg, the polymer is a glass where
chains show regions of three-dimensional order, but after passing through glass transition,
the polymer softens and becomes rubber like. The Tg correlates with the flexibility of the
polymer chains, and greater flexibility should enhance ion conductivity.
23

Page 25 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
Fig. 3 DSC results of (a) silica-poly(PEGMA-188); (b) silica-poly(PEGMA-232); (c)
silica-poly(PEGMA-300); (d) silica-poly(PEGMA-475) and (e) silica-poly(PEGMA-
1100)
3.3 Ionic conductivity of electrolytes containing PEO mixed with silica-
poly(PEGMA) nanoparticles
Hybrid electrolytes were prepared by dissolving LiI and I2 (10:1 mole ratio) in
poly(ethylene glycol) dimethyl ether (PEGDME-500, Mn = 500), with the mole ratio of
PEGDME-500/LiI/I2 fixed at 100:10:1 (EO to Li ratio of ~ 100:1, PEGDME-500
contains around 10 EO units), and subsequently dispersing the silica-poly(PEGMA)
nanoparticles into the mixture. The amount of coated nanoparticles added was calculated
from TGA data (see experimental section for details) to achieve a total EO:Li ratio in the
24

RSC Advances
Page 26 of 39
View Article Online
DOI: 10.1039/C4RA07262F
electrolyte (including EO from PEGDME and silica-poly(PEGMA) of approximately
110:1. Control experiments employed a set of electrolytes consisting of PEGDME-
500/LiI/I2 with no particles but an EO/Li ratio of 110:1. Fig. 4 shows that vigorous
stirring of functionalized particles in the PEGDME/LiI/I2 mixture for at least 24 hours
resulted in homogeneous electrolytes. For very viscous samples, such as silica-
poly(PEGMA-1100), agitation with a Wig-L-Bug facilitated formation of a homogeneous
hybrid electrolyte. Upon the introduction of particles, the electrolyte gelled, and with an
increasing length of PEO side chains, the viscosity of the electrolyte increased
dramatically. Electrolytes containing silica-poly(PEGMA-475) and silica-poly(PEGMA-
1100) did not flow.
Fig. 4 Photographs showing electrolytes formed by dispersing silica-poly(PEGMA) in
PEGDME containing LiI and I2
Using electrochemical impedance spectroscopy (EIS), we determined the
conductivity of the hybrid polymer electrolytes. Fig. 5 shows conductivity values at
temperature ranging from 20 °C to 90 °C with 5 °C intervals, plotted versus 1000/T. The
25

Page 27 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
slight convex curvatures of hybrid silica-poly(PEGMA) electrolytes plots indicates a
typical VFT model of ion conduction mechanism.45 As expected, the ionic conductivity
increased with temperature because of enhanced chain mobility. Room-temperature
conductivities of all electrolytes containing silica-poly(PEGMA) range from 6×10-5 to
1.2×10-4 S/cm. Despite its solidification, the electrolyte with silica-poly(PEGMA-1100)
shows higher ionic conductivity than other electrolytes with modified silica particles. As
discussed previously, three parameters control ionic conductivity: the charge carrier (ion)
concentration (n), ion charge (q), and ion mobility (μ). The amount of silica-
poly(PEGMA) particles added into PEGDME/LiI/I2 gave the same EO/Li ratio in all
electrolytes, and the silica cores occupy only a small part of the volume because over 90%
of the nanoparticle mass is polymer. Hence, the charge carrier concentration for all
electrolyte samples should be similar. Because ion transport in the polymer matrix
depends on the segmental movement of the polymer chain in the amorphous phase, and
lithium ion coordinates to 4 – 8 oxygen atoms, the relatively high conductivity of silica-
poly(PEGMA-1100) has two possible explanations: a lower Tg (i.e. higher chain
mobility); or lithium coordination with more oxygen atoms.46 To further study the Li—
oxygen coordination, the equivalent weight (EW) value of EO functional group in silica-
poly(PEGMA) particles were calculated from TGA data (Table 2). The calculation was
shown in Table S1. As concluded from the table, the EW values of EO functional group
in silica-poly(PEGMA) nanoparticles increase with the increase of the PEO side chain.
Therefore, with shorter PEO side chains, lithium coordination to fewer oxygen atoms
might allow formation of Li cation, which will decrease ionic conductivity.
26

RSC Advances
Page 28 of 39
View Article Online
DOI: 10.1039/C4RA07262F
Fig. 5 Temperature dependent ionic conductivities of electrolytes formed by dispersing
silica-poly(PEGMA) in PEGDME containing LiI and I2 (The figure also shows data for
PEGDME/LiI/I2 without the modified particles.)
3.4 Copolymerization to form silica-poly(PEGMA) nanocomposites
Since ion transport only takes place in the amorphous phase, crystallization of
long PEO side chains should decrease conductivity.21 Even though silica-poly(PEGMA-
475) and silica-poly(PEGMA-1100) show relatively high ionic conductivity, partial
crystallization of PEO side chains might hinder the ion transport in the polymer matrix.
Crystallization should be especially detrimental for silica-poly(PEGMA-1100), which has
a Tm value around 35 °C. Consequently, decreasing the degree of crystallinity and thus
27

Page 29 of 39
RSC Advances
View Article Online
DOI: 10.1039/C4RA07262F
increasing the ionic mobility, by introducing disorder into the poly(PEGMA-1100)
structure should increase conductivity. Methods to decrease crystallinity include blending
different polymers, copolymerization, forming cross-linked networks, or addition of
plasticizer.22 Among these methods, copolymerization is particularly attractive for
chemically altering the chain tacticity, thereby inhibiting crystallization.21 Hence, we
designed a copolymer with two PEGMA monomers, one with a short PEO side chain to
prevent crystallization and one with a longer PEO side chain to complex lithium cation
and promote ion transport. As Scheme 4 illustrates, the surface-initiated polymerization
strategy is similar to that for coating nanoparticles with PEGMA homopolymers. We
synthesized various silica-copolymer nanocomposite materials with different monomer
ratios.
Scheme 4
ATRP from silica nanoparticles to form PEGMA copolymers
28