"Clicked" fluoropolymer elastomers as robust materials for potential microfluidic device applications

Article abstract of DOI:10.1039/c1jm14131g

We report the design and synthesis of a new perfluoropolyether-based material, which has liquid-like viscosity and can be cured into a tough, highly durable elastomer when "clicked" with selected tri-pod organic small molecules. This highly fluorinated elastomer exhibits remarkable resistance to a variety of organic solvents, water, heat and even harsh acidic and basic conditions. Whereas PDMS-based microfluidic devices are commonly used for aqueous based applications, their limited chemical resistance and high swellability in many common organic solvents make it unfeasible for microfluidic applications involving organic solvents and/or harsh conditions. With excellent chemical resistance and low swellability, our newly synthesized fluoro-elastomers will hopefully provide an alternative material for organic based microfluidic devices. Furthermore, the alkyne-azide "click" chemistry employed in curing not only provides high efficiency of synthesis and ease of device fabrication, but, more importantly, produces 1,2,3-triazole linkages that are very stable against harsh acidic or basic conditions. This work has great potential to expand microfluidics to a series of novel applications especially in organic and medicinal chemistry. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c1jm14131g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 1100
www.rsc.org/materials
PAPER
‘‘Clicked’’ fluoropolymer elastomers as robust materials for potential
microfluidic device applications†
Ying-Wei Yang,a Jens Hentschel,a Yi-Chun Chen,b Mark Lazari,cd Hanxiang Zeng,a R. Michael van Dambcd
a
*
and Zhibin Guan
Received 24th August 2011, Accepted 24th October 2011
DOI: 10.1039/c1jm14131g
We report the design and synthesis of a new perfluoropolyether-based material, which has liquid-like
viscosity and can be cured into a tough, highly durable elastomer when ‘‘clicked’’ with selected tri-pod
organic small molecules. This highly fluorinated elastomer exhibits remarkable resistance to a variety of
organic solvents, water, heat and even harsh acidic and basic conditions. Whereas PDMS-based
microfluidic devices are commonly used for aqueous based applications, their limited chemical
resistance and high swellability in many common organic solvents make it unfeasible for microfluidic
applications involving organic solvents and/or harsh conditions. With excellent chemical resistance and
low swellability, our newly synthesized fluoro-elastomers will hopefully provide an alternative material
for organic based microfluidic devices. Furthermore, the alkyne–azide ‘‘click’’ chemistry employed in
curing not only provides high efficiency of synthesis and ease of device fabrication, but, more
importantly, produces 1,2,3-triazole linkages that are very stable against harsh acidic or basic
conditions. This work has great potential to expand microfluidics to a series of novel applications
especially in organic and medicinal chemistry.
limitations of hard materials. In particular, mechanical valves are
1. Introduction
difficult and expensive to fabricate in silicon/glass-based micro-
fluidic systems,4 and devices of high complexity have been almost
impossible due to the large size (several millimetres) of these
valves.4,10 The size can be reduced by using alternative, passive
valve approaches, but their operation depends greatly on the
physical and chemical properties of the fluid.11 In contrast, the
fabrication and manipulation of microfluidic devices containing
valves, pumps and mixers in elastomers is much easier than that
in rigid materials.1,3,4,12–15 The employment of soft elastomeric
materials has allowed microfluidics to explode rapidly into
a modern technology that has found many applications in ink-jet
printing, nanoscale chemical reactions, drug screening, micro-
analysis related to molecular biology, optics and fuel cells.1–8,16–20
Poly(dimethylsiloxane) (PDMS), which is optically clear, inert,
non-toxic and non-flammable, has become the most common
elastomeric material in fabricating microfluidic devices for
applications especially in biotechnology.2–8,12–17 PDMS possesses
many attractive properties for microfluidics, such as viscoelastic
mechanical properties, low Young’s modulus, biocompatibility,
low surface energy and outstanding gas permeability.4,12,14
Despite these advantages, PDMS is limited to applications
mostly involving aqueous solutions. Glass and silicon micro-
fluidic devices are still preferable to PDMS devices in many areas
of microfluidic chemical synthesis and analysis, where acids,
bases, and organic solvents are frequently used. This stems from
the serious drawback that PDMS can dissolve or swell in many
The precise control and manipulation of small (10ꢀ9 to 10ꢀ18
litres) amounts of fluids in channels with dimensions of tens of
micrometres—microfluidics—has emerged as a distinct new field
in the beginning of the 1990s and is increasingly being employed
to scale down and automate laboratory procedures in the fields
of chemistry and biotechnology.1–4 The small dimensions of
microchannels can reduce reagent consumption and waste
production, improve the speed and accuracy of chemical or
biochemical reactions, and also allow the replication of identical
reactions or screening assays on a single microfluidic chip to
harness parallelism and increased throughput.5–8
Benefiting from the great success of photolithography and
related technologies in silicon microelectronics, most of the early
work in microfluidic systems employed silicon and glass,3,4,9
which with time were largely replaced by soft materials due to the
aDepartment of Chemistry, University of California, 1102 Natural Sciences
2, Irvine, CA, 92697-2025, USA. E-mail: zguan@uci.edu
bCrump Institute for Molecular Imaging, University of California, Los
Angeles, CA, 90095, USA
cDepartment of Molecular and Medical Pharmacology, University of
California, Los Angeles, CA, 90095, USA
dBiomedical Engineering Interdepartmental Program, University of
California, Los Angeles, CA, 90095, USA
† Electronic supplementary information (ESI) available: TGA, Instron
tensile tests, rheometric measurement and DMA. See DOI:
10.1039/c1jm14131g
1100 | J. Mater. Chem., 2012, 22, 1100–1106
This journal is ª The Royal Society of Chemistry 2012

View Article Online
such solvents,21 which makes it impossible for organic solvents
and harsh aqueous solutions to flow inside the micro-scale
channels. Therefore, these applications will need to revert to the
early generation systems using rigid inert materials, such as glass,
silicon, ceramics and metals, which are limited by the disad-
vantages mentioned earlier.21
substrates. We believe that our new elastomer technology will
open a door to the fabrication of microfluidic devices for many
organic-based applications.
2. Experimental section
2.1 General
To address this issue, DeSimone and Quake research groups
reported together an organic solvent-compatible microfluidic
device based on a photocurable perfluoropolyether (PFPE)
elastomer.22 This highly fluorinated elastomer exhibits a number
of attractive properties, including low surface energy, high elas-
ticity, high gas permeability, low toxicity, high chemical resis-
tance,22,23 and low swellability in common organic solvents.22
This important work demonstrates the potential of extending the
use of microfluidic devices to a wide variety of new chemical
applications and other domains. Nevertheless, the ester and
carbamate linkages involved in this elastomer system are
susceptible to cleavage under basic conditions, limiting their
general application for microfluidic devices used for chemical
reactions. To address this critical issue, herein we report a new
design of PFPE-based elastomers cured by efficient ‘‘click
chemistry’’ that forms chemically stable 1,2,3-triazole linkages.
A critical design principle in our system is to avoid any weak
linkages that are susceptible to degradation under harsh condi-
tions, for example, in the presence of strong bases or nucleo-
philes. With this in mind, chemically very stable alkyl ether
linkages were employed in the functionalization of both the
PFPE polymer (1) and the cross-linker (2) (Chart 1). Further-
more, copper(I)-catalyzed alkyne–azide cyclization reaction, the
so-called ‘‘click reaction’’,24 was chosen for curing the PFPE
elastomer because this reaction is highly efficient and the
resulting 1,4-disubstituted 1,2,3-triazole linkages are chemically
very stable. Due to many advantageous features, ‘‘click chem-
istry’’ has been extensively employed in various materials
synthesis and applications.25 Herein, we successfully employed
a robust, efficient, orthogonal ‘‘click reaction’’ between terminal-
alkyne functionalized PFPE (1) and tri-azide (2) to form
a transparent elastomer suitable for microfluidic device fabrica-
tion (Chart 1). The focus of this manuscript is to report our
material synthesis, characterization of their basic materials
properties critical for microfluidic devices, and investigation of
experimental conditions suitable for fabricating the new material
for microfluidic device applications. Compared to previous
elastomeric materials used for microfluidic applications, our new
elastomer is more robust with excellent resistance to organic
solvents, heat and base, and shows strong adhesion to glass
1H NMR, 13C NMR and 19F NMR spectra were recorded at 500
MHz, 125 MHz and 376 MHz, respectively on a Bruker Avance-
500 Spectrometer or a Bruker ARX-400 Spectrometer. All NMR
spectra are reported as d in parts per million (ppm). 1H NMR and
13C NMR spectra are reported relative to residual solvent peaks.
Coupling constants are reported in Hz. Electrospray mass
spectrometric analysis (ESI-MS) was performed on a Micromass
LCT or a Micromass Autospec Spectrometer. IR spectra were
collected on a Perkin Elmer 1600 Series FT-IR spectrometer.
All reagents were used as received from commercial suppliers
unless otherwise noted. Anhydrous solvents were passed through
a column of activated alumina (type A2, size 12 ꢁ 32, purify)
under nitrogen pressure and sparged with nitrogen before use.
Deuterated solvents were purchased from Cambridge Isotope
ꢀ
Laboratories. CD2Cl2 was placed over activated 4 A molecular
sieves under inert atmosphere. Flash column chromatography
was performed using forced flow on EM Science 230–400 mesh
silica gel.
2.2 Synthesis of starting materials
2.2.1 PFPE-dialkyne ether (1). Poly(tetrafluoroethylene
oxide-co-difluoromethylene oxide)-a,u-diol (Fomblin ZDOL,
Aldrich, Mw 3800 Mꢀ1, 3.46 g, 0.91 mmol) was dissolved in
a mixed solvent of anhydrous tert-butanol (3.5 mL) and 1,1,2-
trichlorotrifluoroethane (3.5 mL) containing potassium tert-
butoxide (270 mg, 2.4 mmol). This solution was heated to 35–40
ꢂC under stirring. Propargyl bromide (80 wt% solution in
toluene, Aldrich, 542 mg, 0.4 mL, 3.6 mmol) was added to the
reaction mixture dropwise. The reaction mixture was heated
under vigorous stirring at 35–40 ꢂC for 4 days. The solution was
then cooled, filtered through Celite over a medium frit and the
solvent was evaporated in vacuo to give a crude product, which
was further purified by filtering through a 0.2 mm poly-
ethersulfone filter to yield a clear, colorless, viscous oil (yield:
95%). 1H NMR (1,1,2-trichlorotrifluoroethane/Me2(CO)-d6
4 : 1): d 4.34 (s, 4H, OCH2C^CH), 4.01 (q, 4H, CF2CH2O), 2.74
(s, 2H, OCH2C^CH); 19F NMR (1,1,2-trichlorotrifluoroethane/
Me2(CO)-d6 4 : 1): d ꢀ52.9, ꢀ54.3, ꢀ54.5, ꢀ56.3, ꢀ78.4, ꢀ80.5,
ꢀ89.6, ꢀ89.9, ꢀ91.5, ꢀ126.8, ꢀ130.4; 13C{1H} NMR (CDCl3):
d 126.3, 126.0, 123.9, 123.6, 121.5, 121.2, 117.9, 116.8, 116.5,
116.1, 114.3, 114.0, 113.7, 75.5, 75.4, 63.3, 58.6.
2.2.2 3-Azidopropan-1-ol. 3-Bromo-1-propanol (10 g, 6.3
mL, 72 mmol) and sodium azide (7.8 g, 120 mmol) were dissolved
in a mixture of acetone (120 mL) and water (20 mL) and the
resulting solution was heated to reflux overnight. After removing
acetone under reduced pressure, 80 mL of water were added and
the mixture was extracted with diethyl ether (4 ꢁ 80 mL). The
organic layers collected were dried over MgSO4 and the solvent
was removed under reduced pressure to give a product as
colorless oil (yield: 5.82 g, 80%). 1H NMR (CDCl3): d 3.77 (q, 2H,
Chart 1 Structural representation of the materials used in the current
study and the final gel material produced by alkyne–azide ‘‘click’’
chemistry.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 1100–1106 | 1101

View Article Online
J ¼ 5.8 Hz, CH2–OH), 3.47 (t, 2H, J ¼ 6.5 Hz, CH2–N3), 1.85
(quint, 2H, J ¼ 6.0 Hz, CH2CH2CH2), 1.53 (t, 1H, J ¼ 5.0 Hz,
CH2OH).
removed, which are suitable for solvent-resistant and bonding
strength tests.
2.4 Characterization of properties of the elastomeric materials
2.2.3 1,3,5-Tris((3-azidopropoxy)methyl)-2,4,6-trimethyl-
benzene (2). 1,3,5-Tris(bromomethyl)-2,4,6-trimethylbenzene
(5.8 g, 14.4 mmol) and sodium hydride (60% suspension in
mineral oil, 3.4 g, 86.4 mmol) were placed in a 300 mL round
bottom flask. Under a nitrogen atmosphere, THF (dry, 150 mL)
was added at room temperature. At 0 ꢂC, a solution of 3-azido-1-
propanol (5.82 g, 57.6 mmol) in 15 mL THF was added drop-
wise. After complete addition, the mixture was allowed to warm
to room temperature and was stirred for further 8 hours.
Aqueous ammonium chloride solution (100 mL, 1 M) was then
added in order to remove any residual hydride, which was fol-
lowed by extractions with diethyl ether (4 ꢁ 120 mL). The
organics were dried over MgSO4. After removal of the solvent
under vacuum, the product was isolated as colorless oil via flash
column chromatography (1 : 5 EtOAc/Hexanes) (yield: 5.95 g,
2.4.1 Chemical stability and solvent resistance. For the
chemical stability and solvent resistance study, the general
procedure was as follows. The elastomeric cylinder material was
immersed in a sealed container of solvent or solution. The dry
weight was taken after a period of time with or without heating.
The color change of the material and material properties, such as
elasticity and brittleness, as well as appearance of the solution
were checked by naked eyes.
2.4.2 Surface bonding. Two bonding methods were compared
for the bonding strength test. (1) One cylindrical sample of the
material was attached to a second cylinder with a gas-flow hole
punched through the middle. Pressure was applied to make a tight
contact between surfaces, and finally the assembly was baked at
50 ꢂC to form covalent bonding (Fig. 1b–d). (2) Freshly prepared
polymer syrup was poured on a glass slide, and made into a thin
film by spreading or spin-coating to act as a ‘‘glue’’ layer. A gas-
flow hole was punched through a cylinder of the polymer, which
was attached to the glass slide coated with glue, and baked at 50
ꢂC to form strong covalent bonding. The interfacial bonding
strength was tested by applying increasing pressure of nitrogen to
the gas-flow hole and evaluating the highest pressure that could
be applied in a cyclic on/off fashion for over 2 minutes without
bond failure. Pulsing the pressure is a more stringent test (i.e.
leads to earlier failure) than continuous pressure.
1
89.0%). H NMR (CDCl3): d 4.55 (s, 6H, Ph(CH2O–)3), 3.60 (t,
6H, J ¼ 6.0 Hz, (OCH2)3), 3.40 (t, 6H, J ¼ 6.5 Hz, (CH2N3)3),
2.42 (s, 9H, Ph(CH3)3), 1.87 (quint, 6H, (CH2CH2CH2)3); 13C
{1H} NMR (CDCl3): d 138.3, 132.9, 68.0, 67.1, 48.8, 29.6, 15.9.
ESI-MS (CH2Cl2/MeOH), m/z 482.26 [M + Na]+ (found); 482.26
[M + Na]+ (calc).
2.2.4 1,4-Tris((3-azidopropoxy)methyl)-benzene (2a). a,a0-
Dibromo-p-xylene (1.32 g, 5 mmol) and sodium hydride (60%
suspension in mineral oil, 0.52 g, 13 mmol) were placed in a 250
mL round bottom flask. Under a nitrogen atmosphere, THF
(dry, 50 mL) was added to the reaction mixture at room
ꢂ
temperature. At 0 C, a solution of 3-azido-1-propanol (1.32 g,
2.4.3 Thermogravimetric analysis (TGA). TGA was per-
formed on a TA Instruments Q500. Samples (ꢃ5 mg) were
equilibrated at 105 ꢂC before ramping up to 800 ꢂC at a rate of 20
ꢂC minꢀ1. The experiments were carried out in air.
13 mmol) in 5 mL THF was added dropwise. After complete
addition, the mixture was allowed to warm to room temperature
and was stirred for further 8 hours. Aqueous ammonium chlo-
ride solution (30 mL, 1 M) was then added in order to remove
residual hydride, which was followed by extractions with diethyl
ether (4 ꢁ 30 mL). The organics were dried over MgSO4. After
removal of the solvent under vacuum, the product was isolated as
colorless oil via flash column chromatography (1 : 2 CH2Cl2/
Hexanes) (yield: 0.95 g, 62.5%). 1H NMR (CDCl3): d 7.33 (s, 4H,
aryl), 4.52 (s, 4H, Ph(CH2O–)2), 3.56 (t, 4H, J ¼ 6.0 Hz,
(OCH2)2), 3.43 (t, 4H, J ¼ 6.5 Hz, (CH2N3)2), 1.89 (quint, 4H,
(CH2CH2CH2)2); 13C{1H} NMR (CDCl3): d 137.9, 128.0, 73.1,
67.1, 48.7, 29.5; ESI-MS m/z 327.15 [M + Na]+ (found); 327.15
[M + Na]+ (calc).
2.4.4 Mechanical characterization. Films for mechanical
testing were prepared in the similar manner as described in the
Synthesis section. The reaction solutions were cast into Teflon
molds, dried for 4 hours at room temperature and 16 hours at 40
ꢂC under vacuum.
Instron tensile tests. Standard stress/strain experiments were
performed using an Instron 3365 machine. The specimens were
stamped out from solution cast films and extended at 50 mm
minꢀ1 at room temperature. The measurement was repeated
three times. Young’s modulus (E) was determined from the
initial stress–strain slope.
2.3 Synthesis of elastomers
To a 10 mL vial were added tri-azide 2 (38.3 mg), PFPE-dialkyne
ether 1 (498.9 mg) in 0.55 mL of the mixed 1,1,2-trichlorotri-
fluoroethane and trifluorotoluene (v/v 6 : 1). Copper wire pieces
or copper turnings (1 gram) were added to the reaction solution.
The mixture was stirred for ca. 15 hours until the solution
became viscous enough for casting. Then, the reaction mixture
was transferred into a PDMS mold containing cylindrical cavi-
ties (approx. 5 mm diameter, 5 mm deep), and allowed to air-dry
after 4 hours. Curing was performed in an oven at 50 ꢂC for four
hours. The PDMS mold was destroyed and cylinders of the gel
Dynamic mechanical analysis (DMA). DMA was carried out
on a TA Instruments Q800 using a standard temperature sweep
experiment between ꢀ80 and 80 ꢂC. Samples were cut into 6 mm
ꢁ 3 mm ꢁ 0.5 mm (length ꢁ width ꢁ thickness) rectangles. The
initial force was 10 mN, and the heating rate was 3 C minꢀ1 at
ꢂ
a constant frequency of 3.2 Hz. The glass transition temperature
(Tg) was determined at the peak maximum of the tan d signal.
Rheometry. The storage modulus (G0) and loss modulus (G00)
of the final cross-linked polymer material were determined at
1102 | J. Mater. Chem., 2012, 22, 1100–1106
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 1 (a) Casting of syrup in PDMS mold to make polymer cylinders for testing. (b) Fabrication of bond strength testing devices from molded
cylinders of polymerized syrup. (c) Photograph of the finished structure. (d) Pressurized nitrogen is supplied via the gas-flow hole to the bonding
interface.
25 ꢂC and at different frequencies (0.01–100 Hz) in standard
frequency sweep measurements (0.1% strain, 5 N normal force)
using an AR G2 Rheometer from TA Instruments (12 mm steel
plate).
hydroxyl into alkyne for the PFPE diol. In the 19F NMR spec-
trum of the final product 1, the signals from ꢀ52.9 to ꢀ56.3 ppm
and ꢀ89.6 to ꢀ91.5 correspond to the backbone repeat units of
PFPE. The end group region extends from ꢀ78.4 to ꢀ80.5 ppm.
The amount of ‘‘free’’ (not etherified) –CF2CH2OH groups on
the crude oligomer can be evaluated by comparing the integral of
signals at ꢀ81.4 (and ꢀ83.5) ppm with those at ꢀ78.4 (and
ꢀ80.5) ppm (etherified). The free alcohol end is less than 5%,
which is sufficient for device fabrication. Further evidence of this
substitution was the appearance of a strong peak at 3325 cmꢀ1 in
FT-IR spectroscopy, corresponding to the –C^C–H stretching.
Many practical difficulties were met in the cross-linking
polymerization reaction shown in Scheme 1, mainly due to the
poor miscibility of the product in the solvent mixture. Under an
optimized reaction condition, we successfully ‘‘clicked’’ poly-
merized PFPE-dialkyne 1 and tri-azide 2 to form a crosslinked
network. The FT-IR spectrum of the polymerization product
shows the appearance of a new peak at 1631 cmꢀ1 corresponding
to the newly formed triazole ring as well as the disappearance of
the peak at 2102 cmꢀ1 corresponding to the azide group. A
control experiment using a di-azide 2a, 1,4-tris((3-azidopropoxy)
methyl)-benzene, instead of tri-azide 2, for the click reaction did
not produce any gel material, indicating no crosslinking of the di-
functional monomers.
3. Results and discussion
3.1 Material synthesis
As shown in Scheme 1, the ZDOL potassium dialkoxide solution
was prepared directly by simply dissolving it in an equivolu-
metric mixture of anhydrous tert-butanol and 1,1,2-tri-
chlorotrifluoroethane (final concentration 0.5
g
mLꢀ1
)
containing a slightly excess amount of potassium tert-butoxide.
Given the significantly more acidic terminal alcohol on ZDOL
PFPE diol (pKa z 12.5 in H2O) as compared to tert-butanol
(pKa z 17 in H2O), the PFPE diol should be quantitatively
converted to its potassium dialkoxide. The temperature was
raised to 35–40 ꢂC, and an excess amount of propargyl bromide
in toluene (3–4 mol per mol of ZDOL) was added dropwise.
Etherification at both termini of PFPE diol via SN2 substitution
resulted in the formation of PFPE-dialkyne, 1. 19F NMR and 1H
NMR analyses of the product showed a conversion of the
etherification reaction to be more than 95%. The use of a mixed
solvent of anhydrous tert-butanol and 1,1,2-trichlorotrifluoro-
ethane (Freon 113) is necessary because of the difficulty of dis-
solving ZDOL of this molecular weight (Mn around 3800) in
pure tert-butanol. Etherification of ZDOL with propargyl
bromide was highly efficient even at low temperature owing to
3.2 Investigation of material fabrication conditions
Initially, the fabrication condition with our new material was
investigated without using any copper catalyst. In classical
Huisgen reaction, 1,3-dipolar cycloaddition between a terminal
alkyne and an organic azide occurs at elevated temperature
1
the high reactivity of propargyl halides. The H NMR and 13C
NMR spectra confirmed the quantitative conversion of terminal
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 1100–1106 | 1103

View Article Online
Scheme 1 Synthesis of starting materials and the final elastomer.
without any catalyst.25 Therefore, PFPE-dialkyne 1 and tri-azide
2 were dissolved in a minimum amount of mixed Freon 113 and
trifluorotoluene (1 : 1) in a round bottom flask and the resulting
homogeneous solution was then heated at reflux under stirring
overnight until the solution became much more viscous. The
reason for employing a mixed solvent is to avoid phase separa-
tion. Then the reaction mixture was transferred to a small vial,
which was heated slowly from 80 ꢂC to 130 ꢂC to prevent
porosity due to fast evaporation of solvent. No phase separation
was observed during the reaction. After several hours, the gel
formed after all the solvents were removed. This gel material is
suitable for further solvent-resistant tests.
device fabrication. The presence of solvents can interfere with
microfluidic device fabrication by the formation of solvent vapor
bubbles that disrupt the contact between PFPE surfaces during
the bonding process. However, bulk polymerization without any
solvent did not proceed well because the two starting materials,
i.e., PFPE-dialkyne (1) and tri-azide (2), are immiscible. There-
fore, the optimal condition is to use a minimal amount of Freon
113/trifluorotoluene mixed solvents to dissolve both starting
materials, and then stir the reaction mixture overnight in the
presence of an appropriate amount of copper metal as the
catalyst (see the Experimental part for the detailed reaction
conditions). It is critical to reach an appropriate polymerization
degree, as indicated by the viscosity of the polymerization solu-
tion, before casting of the material into the mold. Once the
polymerization formed a syrup with suitable viscosity for casting,
the reaction mixture was cast into a mold and/or spin-coated
onto a substrate surface. With slow evaporation of solvents and
further curing, the syrup will eventually form a robust trans-
parent elastomer. The fabricated samples were finally cured in an
oven at 50 ꢂC to allow the complete polymerization of azide and
alkyne functional groups (Fig. 1a).
However, this procedure for casting the gel is complicated by
the need for a complex temperature profile that depends on
somewhat subjective evaluations of viscosity. The ideal reaction
condition would be simply mixing the starting materials together
and then casting them into a mold.
Toward this goal, the first optimization of gel fabrication is to
remove this complex heating profile. Without heating, the click
reaction is significantly slowed down, which requires the use of
an appropriate catalyst to accelerate the click reaction. A
preliminary screening of various copper(I) salts as potential
catalysts did not offer us any good results, presumably due to the
insolubility of those salts in the reaction mixture. Furthermore,
removal of the catalyst after the reaction is also an issue. On the
other hand, the use of copper metal, copper turnings and copper
wires at room temperature resulted in significant acceleration of
the reaction. The removal of copper metal is extremely easy
because it settles at the bottom of the reaction container prior to
casting.
3.3 Thermal stability
As reported by Whitesides et al., PDMS-based microfluidic
devices have limited compatibility with organic solvents.1,21,23 In
comparison, the PFPE-based elastomers prepared in this study
showed a number of improved features in terms of the thermal
stability, acid/base resistance and solvent resistance. The initial
thermal stability test shows that the gel is very stable with no
weight or color change after heating at 50 ꢂC for 5 hours. More
quantitative thermal stability analysis was conducted by ther-
mogravimetric analysis (TGA) (Fig. S1 in the ESI†), which
In further optimization of material fabrication conditions, we
aimed at minimizing the amount of solvents used in the reaction,
especially trifluorotoluene, which is difficult to remove during the
1104 | J. Mater. Chem., 2012, 22, 1100–1106
This journal is ª The Royal Society of Chemistry 2012

View Article Online
shows that the material has no appreciable thermal decomposi-
ꢂ
1). Finally, we tested the materials in a strong basic solution,
namely 1 or 3 N NaOH aqueous solution, at various tempera-
tures and times. The two types of PFPE gels showed significantly
different stability to strong basic solutions. Under all NaOH test
conditions (entry 11–15, Table 1), the ‘‘click’’ PFPE gel showed
no or only slight color change, with no suspension or change in
flexibility being observed. In sharp contrast, the previously
reported PFPE gel showed substantial degradation under 1 or 3
N NaOH aqueous solution for prolonged heating (entries 14 and
15, Table 1), as evidenced by the turbidity and color change of
the solution. After the base treatment, the material also became
noticeably more flaky and weaker. These tests confirm that the
‘‘click’’ PFPE material shows excellent resistance to heat, organic
solvents, and acidic and basic solutions. We attribute the excel-
lent stability to the chemical design of the ‘‘click’’ PFPE gel: there
is no weak chemical linkage in the polymer that is susceptible to
chemical degradation. Both the ether linkage and the 1,2,3-tri-
azole ring formed from ‘‘click’’ reaction are stable. These new
design features, in combination with the inherent chemical
stability and solvent resistance for PFPE, make our ‘‘click’’
PFPE material an excellent candidate for microfluidic device
fabrication for enhanced solvent resistance and chemical/thermal
stability.
tion at temperatures below 350 C.
3.4 Solvent and chemical resistance
PDMS-based microfluidic devices have very limited compati-
bility with organic solvents. Here, we choose some commonly
used organic solvents available in most organic labs to test the
solvent resistance of our new material. The initial swelling test in
DMSO at room temperature with 7 days immersion shows
a weight change of less than 4%, suggesting excellent compati-
bility with this solvent. Screening of a series of commonly used
organic solvents, such as DCM, MeCN, DMSO, DMF, H2O,
Freon 113, toluene, and trifluorotoluene, further validated that
the ‘‘click’’ PFPE gel material is highly compatible with a variety
of organic solvents (entries 1–4 in Table 1). For comparison, the
PFPE gels reported previously22a show higher swellability,
especially in fluorinated solvents such as trifluorotoluene (entry 2
in Table 1). In further tests, the ‘‘click’’ PFPE material exhibited
an excellent chemical stability under harsh basic or acidic
conditions even at elevated temperatures. Under strong acidic
conditions (3 N HCl or 5 M TFA) at 120 ꢂC, the material shows
very small changes (entries 5 and 6 in Table 1). The test of
material stability under basic conditions is particularly important
because the previously reported PFPE material contains ester
and carbamate linkages,22a which are presumably sensitive to
basic conditions. We first tested the stability in 1 M K2CO3
aqueous solution at various temperatures and times. The ‘‘click’’
PFPE gel showed a negligible change under all the conditions.
However, the previously reported PFPE showed a significant
amount of suspension present in the solution after being heated
at 80 ꢂC in 1 M K2CO3 aqueous solution for 90 h (entry 9, Table
3.5 Mechanical properties
The mechanical properties of the PFPE elastomer prepared by
click reaction were investigated by several methods. Instron
tensile tests revealed a typical rubber-like stress–strain behavior
(Fig. S2 in the ESI†), with a strain at break about 175% at a stress
of ca. 2.1 MPa. The Young’s modulus (E) measured from the
initial slope on the stress–strain curve was 3.5 MPa, which is
Table 1 Solvent resistance and thermal stability tests on the ‘‘click’’ PFPE elastomer compared to the DeSimone/Quake PFPE elastomerb
‘‘Clicked’’ PFPE
Weight
DeSimone/Quake PFPE22a
Weight
change
Appearance
change
Entry
Conditions
Temp./ꢂC
25
Time/h
14
change
Appearance change
In pure organic solvents
Pure solventsa
1
<ꢄ3%
No
<6%
Slightly cloudy for
DMSO and toluene
2
3
4
Trifluoro-toluene
MeCN
DMSO
25
120
130
14
0.33
0.50
<ꢄ3%
+1.9%
+2.9%
No
No
Darker
+13%
+1.9%
+2.6%
No
No
No
Under acidic conditions
5
6
3 N HCl
5 M TFA
120
120
0.33
0.33
<1%
ꢀ3.9%
No
No
<1%
+1.4%
Slightly darker
Slightly darker
Under basic conditions
7
8
9
10
1 M K2CO3
1 M K2CO3
1 M K2CO3
1 M K2CO3
25
80
80
120
10
10
90
<1%
<1%
<1%
<1%
No
No
<1%
<1%
<1%
<1%
No
No
Slightly yellow, no suspension
No
Significant suspension in solution
No
0.33
Under strong basic conditions
11
12
13
1 M NaOH
1 M NaOH
1 M NaOH
25
80
80
10
10
90
<1%
<1%
<1%
No
No
<1%
<1%
<1%
No
No
Slightly yellow; no suspension;
no change in flexibility
Slightly yellow; no suspension;
no change in flexibility
Slightly yellow; no suspension;
no change in flexibility
Significant suspension in solution;
slight flexibility change
Darker, turbid solution; noticeably
flaky and easy to break
Darker, very turbid solution; weak
and flaky, easy to break
14
15
3 N NaOH
1 N NaOH
80
90
90
<1%
+1.62%
120
ꢀ3.41%
ꢀ6.51%
a
CH2Cl2, MeCN, DMSO, DMF, H2O, and PhMe. b There is an uncertainty of weight measurement due to the potential loss of flaky material.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 1100–1106 | 1105

View Article Online
Table 2 Mechanical properties of the PFPE elastomer
candidate for microfluidic applications involving organic
solvents and/or harsh conditions. Further studies including
microfluidic device fabrication using the ‘‘click’’ PFPE elastomer
are currently under investigation.
a
a
/
G0 /
G00
Strength at
Strain at
MPa MPa Eb/MPa
breakb/MPa breakb (%) Tg /ꢂC
c
Gel 1.04
0.10
3.5 ꢄ 0.2 2.1 ꢄ 0.3
175 ꢄ 30
ꢀ10
a
b
Measured at 25 ꢂC and 1 Hz. Estimated from 3 stress–strain tensile
Acknowledgements
experiments. c Determined by DMA.
We thank the Phelps Family Foundation for the generous
financial support. Z.G. acknowledges a Humboldt Bessel Award.
We also thank Dirk Williams and Darin Williams for assistance
in making molds to cast PFPE samples.
similar to previously reported tensile moduli of cured PFPE and
PDMS materials (3.9 MPa and 2.4 MPa, respectively).22a
Rheometric measurement confirms the PFPE gel as a robust
elastomer with a storage modulus of ꢃ1 MPa (Fig. S3 in the
ESI†). Finally dynamic mechanical analysis (DMA) reveals that
the PFPE elastomer has a glass transition temperature below
room temperature (ꢀ10 ꢂC), as shown in Fig. S4 in the ESI†. The
mechanical properties of the PFPE elastomer are summarized in
Table 2.
Notes and references
1 G. M. Whitesides, Nature, 2006, 442, 368.
2 A. J. de Mello, Nature, 2006, 442, 394.
3 J. Quellette, Ind. Phys., 2003, 9, 14.
4 S. R. Quake and A. Scherer, Science, 2000, 290, 1536.
5 J. W. Hong, V. Studer, G. Hang, W. F. Anderson and S. R. Quake,
Nat. Biotechnol., 2004, 22, 435.
6 S. H. DeWitt, Curr. Opin. Chem. Biol., 1999, 3, 350.
7 C. J. Cullen, R. C. R. Wootton and A. J. de Mello, Curr. Opin. Drug
Discovery Dev., 2004, 7, 798.
3.6 Interfacial bonding strength
8 P. Watts and S. J. Haswell, Curr. Opin. Chem. Biol., 2003, 7, 380.
9 K. E. Peterson, Proc. IEEE, 1982, 70, 420.
10 P. K. Yuen, L. J. Kricka and P. Wilding, J. Micromech. Microeng.,
2000, 10, 401.
11 N. L. Jeon, D. T. Chiu, C. J. Wargo, H. Wu, I. S. Choi, J. R. Anderson
and G. M. Whitesides, Biomed. Microdevices, 2002, 4, 117.
12 M. A. Unger, H. P. Chou, T. Thorsen, A. Scherer and S. R. Quake,
Science, 2000, 288, 113.
13 T. Thorsen, S. J. Maerkl and S. R. Quake, Science, 2002, 298, 580.
14 J. C. McDonald and G. M. Whitesides, Acc. Chem. Res., 2002, 35,
491.
Valve actuation in elastomeric microfluidic devices needs
a strong bond between two different layers of the material. One
advantage of our material is that there are still active alkyne and
azide functional groups left un-reacted on the surface. In prin-
ciple, it should be possible for the material to react further at
interfaces between layers with or without the use of fresh mate-
rial as glue. Here, two different bonding methods were employed
for the bonding strength test (see the Experimental part).
Preliminary tests of bonding strength show that the simple
devices made by both methods can hold pressure at the bonding
interface. The second method gave a much stronger layer-to-
layer bonding, capable of withstanding nitrogen pressure up to
about 80 psi [550 kPa]. This is comparable to the best bond
strengths achieved in PDMS devices26 and is significantly greater
than previously reported for PFPE materials.26 The ability to
sustain these pressures at the bonding interface suggests the
possibility to fabricate multi-layer microfluidic devices with
integrated microvalves.
15 A. Groisman, M. Enzelberger and S. R. Quake, Science, 2003, 300,
955.
16 J. Liu, C. Hansen and S. R. Quake, Anal. Chem., 2003, 75, 4718.
17 C.-C. Lee, G. Sui, A. Elizarov, C. J. Shu, Y.-S. Shin, A. N. Doole,
D. Stout, O. N. Witte, H. C. Kolb, N. Satyamurthy, J. R. Heath,
M. E. Phelps, S. R. Quake and H.-R. Tseng, Science, 2005, 310, 1793.
18 J. W. Hong and S. R. Quake, Nat. Biotechnol., 2003, 21, 1179.
19 C. Hansen and S. R. Quake, Curr. Opin. Struct. Biol., 2003, 13, 538.
20 D. Psaltis, S. R. Quake and C. H. Yang, Nature, 2006, 442, 381.
21 J. N. Lee, C. Park and G. M. Whitesides, Anal. Chem., 2003, 75, 6544.
22 (a) J. P. Rolland, R. M. van Dam, D. A. Schorzman, S. R. Quake and
J. M. DeSimone, J. Am. Chem. Soc., 2004, 126, 2322; (b)
J. P. Rolland, R. M. van Dam, D. A. Schorzman, S. R. Quake and
J. M. DeSimone, J. Am. Chem. Soc., 2004, 126, 8349; (c) Y. Huang,
P. Castrataro, C.-C. Lee and S. R. Quake, Lab Chip, 2007, 7, 24.
23 (a) J. Scheirs, Modern Fluoropolymers, John Wiley & Sons, Ltd., New
York, 1997, p. 435; (b) G. Maltezos, E. Garcia, G. Hanrahan,
F. A. Gomez, S. Vyawhare, R. M. van Dam, Y. Chen and
A. Scherer, Lab Chip, 2007, 7, 1209.
4. Conclusions
In summary, we designed and synthesized a new type of robust
elastomer by ‘‘click’’ polymerization of perfluoropolyethers
(PFPEs) with tri-azide terminated organic small molecules. The
alkyne–azide ‘‘click’’ chemistry employed in curing not only
provides high efficiency of synthesis and ease of device fabrica-
tion, but, more importantly, produces 1,2,3-triazole linkages that
are very stable against harsh acidic or basic conditions. These
new design features, in combination with the inherent chemical
stability and solvent resistance for PFPE, render the ‘‘click’’
PFPE material remarkable resistance to a variety of organic
solvents, heat and even harsh acidic and especially basic condi-
tions. In addition, the material also shows strong adhesion to
glass and itself, an important property for device fabrication.
This new material overcomes a number of disadvantages of the
current materials used in microfluidic device fabrication,
including PDMS and the previously reported PFPE gels. We
envision that our ‘‘click’’ PFPE elastomer will be an excellent
24 (a) C. W. Tornoe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057; (b) V. V. Rostovtsev, L. G. Green, V. V. Fokin and
K. B. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596; (c) Y. Liu,
D. D. Diaz, A. A. Accurso, K. B. Sharpless, V. V. Fokin and
M. G. Finn, J. Polym. Sci., Part A: Polym. Chem., 2007, 45, 5182.
25 (a) H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int.
€
Ed., 2001, 40, 2004; (b) R. Breinbauer and M. Kohn, ChemBioChem,
2003, 4, 1147; (c) W. R. Dichtel, O. S. Miljanic, J. M. Spruell,
J. R. Heath and J. F. Stoddart, J. Am. Chem. Soc., 2006, 128,
10388; (d) S. Angelos, Y.-W. Yang, K. Patel, J. F. Stoddart and
J. I. Zink, Angew. Chem., Int. Ed., 2008, 47, 2222; (e) P. Wu,
A. K. Feldman, A. K. Nugent, C. J. Hawker, A. Scheel, B. Voit,
ꢁ
J. Pyun, J. M. Frechet, K. B. Sharpless and V. V. Fokin, Angew.
Chem., Int. Ed., 2004, 43, 3928.
26 Maximum operating pressure in DeSimone and Quake’s paper (see
ref. 22) was 25 psi. 40 psi was reported in PFPE DNA synthesizer.
For PDMS, ‘‘high’’ bond strength reported as 600 kPa ¼ 87 psi,
please refer to: M. A. Eddings, M. A. Johnson and B. K. Gale, J.
Micromech. Microeng., 2008, 18, 067001.
1106 | J. Mater. Chem., 2012, 22, 1100–1106
This journal is ª The Royal Society of Chemistry 2012