PDMS–Fluorous Polyoxetane–PDMS Triblock Hybrid Elastomers

Article pubs.acs.org/Macromolecules

PDMS−Fluorous Polyoxetane−PDMS Triblock Hybrid Elastomers: Tough and Transparent with Novel Bulk Morphologies Souvik Chakrabarty,† Mark Nisenholt,‡ and Kenneth J. Wynne*,† †

Department of Chemical and Life Science Engineering, School of Engineering, Virginia Commonwealth University, 601 West Main Street, Richmond, Virginia 23284-3028, United States ‡ Department of Visual Arts, Lakehead University, 955 Oliver Rd., ThunderBay, Ontario, Canada P7B 5E1 S Supporting Information *

ABSTRACT: PDMS-3F-1.1-PDMS and PDMS-3F-4.5PDMS triblock hybrid elastomers are investigated, wherein (1) 3F-1.1 and 3F-4.5 are poly(3-methyl-3-trifluoroethoxymethyl)-1,3-propylene oxide with Mn = 1.1 or 4.5 kDa, (2) segments are linked by urethane/urea forming reactions, and (3) the intermediate PDMS-3F-PDMS aminopropyl end segments are end-capped with isocyanatopropyltriethoxysilane. After condensation cure, PDMS-3F-PDMS triblock hybrids (A-1.1 and A-4.5) form robust elastomers. In a second set, bis(triethoxysilylethane), BTESE, was incorporated to probe effects of increased siliceous domain content (B-1.1, B-4.5). All compositions are optically transparent due to nearly identical refractive indexes for 3F and PDMS segments. TM-AFM images for A-4.5, A-1.1, and B-4.5 fracture surfaces reveal microscale bulk phase separation. The A-4.5 triblock hybrid shows a particularly interesting morphology comprised of 2−3 μm ovaloids (low modulus) surrounded by a higher modulus matrix. A model is proposed for this microscale morphology based on the relative rates of physical network formation (PN, H-bonding) and chemical network formation (CN, SiO1.5) during coating deposition Despite low hard segment weight percents (2.6−3.5) the hybrid triblocks have moderate toughness with strain at break ranging from 260 to 492%. Triblock hybrid elastomer B-1.1 has the highest −SiO1.5 wt % (mostly from BTESE) and lowest 3F wt % (3F1.1). No sign of microscale phase separation is observed by TM-AFM imaging, and a separate Tg for the 3F segment is not detected by DMA; these findings are ascribed to network constrained phase separation of that results in 3F being incorporated in an “interphase”. The absence of a separate Tg for 3F leads to a gradual decrease in storage modulus (8 to 1.4 MPa) from −90 to 150 °C. In contrast to the complex bulk morphology, TM-AFM imaging shows the hybrid surfaces are devoid of microstructural features attributable to phase separation. Based on contact angle measurements and XPS analysis, the outermost surface for all PDMS-3F-PDMS hybrid triblocks elastomers is dominated by PDMS.

■

side chain confers oleophobicity.23−25 Polysiloxanes with higher F(CF2)n-fluorinated content are also oleophobic, but advantages for introduction of higher fluorine content must be weighed against higher cost.23,26 Alternatives continue to be explored including the introduction of oleophobic perfluoropolyether moieties for increased resistance to organics in applications such as microfluidics.2,27 Introducing CnF fluorinated groups (n = perfluoro C atoms) at low wt %, fluorous surface modifiers can introduce oleophobicity. Simply blocking diffusion of cyclics and other low molar mass components is important to micro- and nanomolding28 and biomedical applications.7 Copolyacrylates having C8F and silicone side chains ( B-4.5. The scale of phase separation for B-1.1 is below resolution limits (∼20 nm). Microscale phase separation was unexpected as the hybrid triblock plaques are optically transparent. Optical transparency for conventional polyurethanes is attributed to nanoscale hard segment phase separation and negligible visible light scattering. As noted above, the nearly identical refractive indexes for PDMS and 3F segments, which constitute >90 wt % of the hybrid compositions (Table 1), account for optical transparency (Figure 2). At the micrometer scale, we propose that the darker ovaloids are due to the lower stiffness of largely physically phase separated domains, whereas the lighter color of the surrounding domain is evidence for the dominance of a siliceous network formed by hydrolysis/condensation cure. Scheme 3 shows an illustration of physical association of urea/urethane segments (PN) and Si−O−Si network formation by the SiO1.5 domain (CN).

(3.6 wt %) compared to A-1.1 (2.6 wt %) and A-4.5 (1.8 wt %). Control C2-4.5 contains 10 wt % HS by the usual convention of including HMDI. Allowing for the higher urethane and urea content of the control, the position and relative intensity of the 1630 cm−1 carbonyl peak for A-1.1 and A-4.5 are in good agreement. The surprisingly weak, broad absorptions for the hybrids at ∼1710 cm−1 nominally correspond to “free” urethane/urea C O.55 By this measure, the concentration of “free” CO is very low. Siliceous weight percents based on −SiO1.5 are listed in Table 2. The siliceous domain formed by hydrolysis/ condensation cure for all hybrids is likely fringed with −Si− OH. The absence or weakness of the high-frequency absorption at 1710 cm−1 is attributed to carbonyl hydrogen bonding to Si− OH depicted in Figure 5B. This notion is supported by the virtual absence of the 1710 cm−1 absorption for B-1.1 and B-4.5, which have higher siliceous content. A similar H-bonding model was proposed by Saegusa and Chujo to account for a shift of the carbonyl stretching band of polyoxazoline (POZO) from 1637 to 1621 cm−1 in POZO−SiO2 hybrids.34,35 Thus, it seems likely that the 1630 cm−1 carbonyl peak for the triblocks hybrids is due to a combination of U−Ur and −Si−OH hydrogen bonding depicted in Figure 5. Control C1 does not contain HMDI. Hence, the synergistic hydrogen bonding of urea/urethane groups held in close proximity (Figure 5A) is not possible. As a consequence, the low-frequency absorption for C1 is weak (Figure 5). There is a significant shift in this absorption to higher frequency (1640 cm−1), which is another indicator of weak H-bonding. Finally, there is no clear indication of a discrete absorption for carbonyl hydrogen bonded to Si−OH in C1 or any of the hybrids. Apparently, a broad absorption envelope exists likely due to a variety of carbonyl interactions. Atomic Force Microscopy. Reactive triblocks III were cast and cured into plaques. After immersion in liquid nitrogen, samples retained toughness and were fractured with some difficulty. Examination with the AFM optical microscope revealed areas suitable for TM-AFM imaging. Light tapping (rsp, 0.95) introduces fewer artifacts (such as scan lines) when surfaces are relatively uneven. Phase and 2D height images (10 × 10 μm) for fracture surfaces are shown in Figure 6. Root-meansquare roughness is lower for hybrids with augmented siliceous domains (10 wt % BTESE), suggesting a cleaner fracture. PDMS-3F-PDMS Hybrid Elastomers. The AFM phase image for the A-4.5 fracture surface is striking, with microscale ovaloids and other irregularly shaped features (Figure 6). The ovaloids have a dark appearance signaling an overall lower modulus.56 Surrounding the ovaloid domains is a lighter (higher modulus) matrix. The degree to which micrometer scale ovaloid domains

Scheme 3. Kinetics of Physical vs Chemical Network Formation during the Coating Processa

a

Notation: PN = physical network via hard block hydrogen bonding; CN = covalently bonded network formed by Si−O−Si bonds; ES = elastomeric solid.

The ovaloids on the fracture surface must arise from the fracture of three-dimensional microscale domains. An idealized model for ordered ovaloid regions in A-4.5 is shown in Figure 7 based on the notion that PDMS/3F/HS phase separation forms a physical network rapidly during solvent evaporation. The physical network formed via hydrogen bonding (Figure 5A and

Figure 6. TM-AFM phase and 2D height images for triblock hybrid and C-2−4.5 fracture surfaces; rsp = 0.95, scan size = 10 μm × 10 μm, z = 60°; hard block wt % shown after sample designation in phase images; Rq (nm) noted in 2D height images.

Figure 7. Model depicting a volume element for an A-4.5 PN microdomain. 7905

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

phase separation leading to nanoscale hard segment domains that are too small for detection ( A-1.1 (2.2) > B-4.5 (5.4) ≫ B-1.1 (5.8) (Table 3). This correlation is consistent with the rate of CN formation via hydrolysis/ condensation of −Si(OEt)3 moieties being dependent on −Si(OEt)3 concentration. Following this reasoning, AFM images suggest Si−O−Si network formation interferes with PN formation of ovaloid and related structures. Neither microscale nor nanoscale features are resolvable for the B-1.1 fracture surface (Figure 6), which is attributed to relatively high initial −Si(OEt)3 concentration (III and BTESE). The competition between physical network formation (PN) and covalent network formation by hydrolysis/condensation cure is represented in Scheme 3. If the rate for PN formation, kPN, is greater than kCN for hydrolysis/condensation, physical network formation will occur first during coating development (Figure 6, A-4.5). Conversely, if kCN is rapid due to high −Si(OEt)3 concentration morphological features are not apparent by TM-AFM (Figure 6, B-1.1). A low Mn for 3F-1.1 and low wt % also contributes to the absence of microscale morphological features (vide infra). C2-4.5. This hybrid urea−urethane elastomer is comprised of a 3F segment and an HMDI urea−urethane HS (10 wt %) terminated with −Si(OC2H5)3 (Scheme 2). C2-4.5 has a high 3F soft segment content (86 wt %) compared to the PDMS-3FPDMS hybrid triblocks (9.2−25 wt %). The hard segment has close proximity to the −Si(OC2H5)3 end group. The complex phase image for C2-4.5 is comprised of irregular dark (lower modulus) regions surrounding elongated light (higher modulus) micrometer-scale features (Figure 6). An enlarged version of the C2-4.5 fracture image is provided in Figure S3. The competition between physical and chemical cross-linking is again thought to account for a complex morphology. As for the PDMS-3F-PDMS hybrids, the darker micrometer scale domains are assigned to PN regions while the lighter portions of the phase image are associated with the CN. For C2-4.5, hard and soft domains are comprised of the majority 3F soft segment, which accounts for optical transparency despite microscale phase separation. In comparison to the A-4.5 ovaloids, phase separation for C24.5 is irregular with ∼5× smaller microscale phase separation. Of note is the observation that soft domains have nanoscale “inclusions” of smaller high modulus regions, but higher modulus regions (yellow) generally do not have soft inclusions. This is accounted for by initial physical phase separation and later chemical cross-linking leading to “hard” −SiO1.5 inclusions in soft domains. On the other hand, CN regions constrain physical

Figure 8. Stress−strain curves for triblock elastomers C1 and C-4.5.

Modulus, ultimate strength, and elongation at break were determined for three specimens. Averages are reported in Table 3 along with standard deviations for elongation at break. Low moduli (∼2 MPa) reflect high soft block content and low glass transition temperatures for PDMS and 3F soft segments. The initial stress−strain curves from which tensile moduli are determined are similar for A-1.1, B-1.1, B-4.5, and C2-4.5 (2.1−2.2 MPa, Figure 8 and Table 3). The unique stress−strain curve for A-4.5 gives a somewhat higher initial modulus (2.3 MPa) that is attributed to clear microscale phase separation discussed above (TM-AFM, Figure 6). Figure 9 provides a correlation of HS wt % with strain at break. The order of strain to break (%) is A-4.5 (492) > A-1.1 (328) ≈

Figure 9. Abscissa: control and hybrid designations; left ordinate: average strain to break; right ordinate: wt % urea/urethane (C1) or hard block.

B-1.1 (315) > B-4.5 (260). Despite a lower HS content (3.1 wt %) than A-1.1 (4.5 wt %), A-4.5 has a significantly higher strain at break 492%. The strain at break for A-4.5 is exceeded only by control C2-4.5 (580%), which has 3.6 wt % urethane/urea but 10 wt % HS (Table 3). The better strain-to-break for A-4.5 is again attributed to superior microscale physical phase separation (TMAFM, Figure 6), strong hydrogen bonding (Figure 5A), high extensibility facilitated by minimized siliceous content, and perhaps higher 3F-4.5 molecular weight. Yilgor, McGrath, and Wilkes have shown that enhanced urea H-bonding results in good mechanicals and phase separation for polyurethanes and polyurethane−ureas.19,21,57 TDI(MDI)/ H2O/PTMO polyurethane−urea elastomers with 20−35 wt % HS have good mechanical properties with moduli (∼7.8−65 MPa) and strain at break (500−1200%).19 Considering A-4.5 has only 3 wt % HS, good mechanicals are attributed to microscale 7906

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

phase separation (Figure 6) favoring hydrogen bonding as noted above. Increased siliceous content decreases strain at break from 492% for A-4.5 to 260% for B-4.5. This is correlated with the weaker and broader H-bonded carbonyl absorption for B-4.5 compared to A-4.5. The low-frequency CO absorption for A4.5 indicates strong hydrogen bonding, while the weaker and broadened peak for B-4.5 reflects weaker CO hydrogen bonding to Si−OH (Figure 5B) and concomitant disruption of strong urea/urethane hydrogen bonding. The lower strain at break and weaker hydrogen bonding for B-4.5 is attributed to disruption of HS phase separation (decreased kPN, increased kCN; Scheme 3), network constrained phase separation, and the lowest HB wt % of this series (2.6). A-1.1 and B-1.1 have similar tensile mechanical properties. These compositions also have similar HS wt % (4.5 and 4.2, respectively) but different morphologies (Figure 6). Apparently, HS wt % is the strongest contributor to mechanicals and overrides network constraint on HS association for B-1.1. Without HMDI, control C1 has weak mechanicals with a very low modulus (0.9 MPa) and strain to break of 40%. Qualitatively, this observation correlates with negligible low frequency Hbonded carbonyl absorption (Figure 4). Phase Transitions and Morphology. Dynamic mechanical testing was used for determining elastomer phase transition temperatures. Compared to DSC, amplified sensitivity is obtained for the change from the glassy to rubbery state at Tg.60,61 Generally, Tg’s determined by DMA (1 Hz) are slightly higher than those from DSC.18 However, this point is moot as low temperature Tg’s could not be detected for hybrid triblocks by DSC (vide infra). Representative log E″ and tan δ for the each of the triblock hybrid elastomers are shown in Figure S4. For comparative purposes, stacked curves (arbitrary ordinate) are shown in Figures 10 (tan δ) and 11 (E′). Several grip techniques were tried

Figure 11. DMA: storage modulus (E′) versus temperature for triblock elastomers. Dashed line is for guiding the eye; ordinate, arbitrary units; E′ at selected temperatures in Table 3.

Storage modulus (E′) versus temperature for hybrid triblocks is shown in Figure 11. For A-1.1 and A-4.5 a 100-fold drop in storage modulus occurs from just above PDMS Tg (−90 °C, E′, 200−210 MPa) to 0 °C (E′, 1.7−2 MPa), which is above the 3F Tg. Although the phase image for A-1.1 microscale phase separation is less distinct than A-4.5, at small DMA deformations (∼0.05%) the physical network resulting from HS phase separation gives rise to a clear drop in storage modulus for A1.1 at 3F Tg. For B-4.5 intermediate behavior is observed. That is, the storage modulus at −90 °C is 90 MPa, about half that for A-4.5. However, a storage modulus similar to A-4.5 is found at 0 and 150 °C. Thus, network constrained phase separation, resulting in barely discernible micrometer-scale ovaloids in TM-AFM (Figure 6), has decreased the low-temperature modulus. However, the smaller scale of HS phase separation has not much affected higher temperature performance. A PDMS Tg is observed at −115 °C for B-1.1, but in contrast to A-1.1, A-4.5, and B-4.5, a 3F Tg is not seen (Figures 10 and 11, Figure S4). Above the PDMS Tg little change is seen in tan δ while a gradual decrease in the storage modulus (8 to 1.4 MPa) occurs from −90 to 150 °C. The retention of change in storage modulus E′ and loss peak for the PDMS domain with the absence of corresponding peaks for the 3F domain is noteworthy. Relevant prior work is briefly considered to place this observation in perspective. Conventionally, copolymers comprised of immiscible blocks form amorphous domains with separate Tgs.62,63,65 However, when the scale of phase separation is reduced to a few tens of nanometers, separate Tgs may not be observed. Baer reported layer-multiplied coextrusion of high-MW polycarbonate (PC, 62 kDa) and poly(methyl methacrylate) (PMMA, 132 kDa) sheets.66,67 When the interfacial separation of PC and PMMA layers was ∼10 nm (discernible by AFM), only a broad change in slope (Tg) was observed via DSC. When sheets of semicrystalline poly(ethylene oxide) (PEO, 200 kDa) and polystyrene (PS) were subjected to “forced assembly” by multiple coextrusions, the DMA β-relaxation peak (Tg) for PEO was greatly reduced in intensity, reaching a minimum at a PEO layer thickness of 120 nm.68 For perfluoropolyether (PFPE) and PEG blocks with very low molecular weights, photogenerated networks had microscale miscibility (optical transparency) but nanoscale phase separation (AFM).69 Separate Tg’s were not observed in DSC for combinations of low molecular weight PFPE and lowest weight fraction PEG, but two Tg’s and optical opacity were observed for other compositions and for networks derived from higher molecular weight components.

Figure 10. Dynamic mechanical analysis: tan delta for hybrid triblocks vs temperature; ordinate: arbitrary units.

for holding the samples, but DMA data are noisy (particularly for A-1.1 and B-1.1) at higher temperatures apparently because of the soft character of these elastomers. Tan δ versus temperature (Figure 10) show that compositions A-1.1, A-4.5, and B-4.5 have similar thermomechanical behavior, with loss peaks at −115 °C due to PDMS Tg and −35 °C for 3F block Tg. The clear separation of loss peaks for A-1.1, A-4.5, and B-4.5 is typical for phase separated soft blocks.62,63 A DMA study on PDMS−PTMO polyurethanes having 33−37 wt % MDI-BD hard segment showed separate Tgs for PDMS (−111 to −120 °C) and PTMO (−40 °C).64 7907

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

observed by TM-AFM for a base polyurethane modified with 2 wt % (3F-b-ME3n)-U, but for a base polyurethane modified with 2 wt % (3Fn-r-ME3n)-U, nanoscale phase separation could not be discerned by TM-AFM imaging. From the above, fluorous/aliphatic polymer blends and networks, even with relatively low molecular weights, can be phase separated at the nanoscale without observation of separate Tg’s for the respective domains. The absence of detectable nanoscale phase separation for B-1.1 (Figure 6) provides an example where reducing phase separation to a scale less than that detectable by TM-AFM (∼20 nm) results in the absence of a detectable Tg. The compositional contributors to this unique observation for B-1.1 include network constrained phase separation (kCN > kPN) driven by CN formation. The absence of a detectable Tg for one soft block domain (3F) with the retention of the phase transition for a second domain (PDMS, −115 °C) is understood considering the 8.9 wt % 3F-1.1, which is the lowest for the hybrid triblocks (Table 1) while the PDMS wt % is 80.6. Thus, 3F is apparently incorporated in an “interphase”.66,67 The volume fraction of PDMS in this “interphase” is unknown but the 80.6 wt % PDMS is high enough that the Tg is easily observed.

We have recently reported a mixed soft block polyurethane (U) that is described by the sequence −An−X−Bn−, where A and B are 3F and ME3 repeat units, respectively, and X is the hard block. Separate Tg’s for −An− and −Bn− soft blocks were observed by DSC at −47 °C (ME3) and −34 °C (3F).70 However, for a polyurethane having an −(An-b-Bn)−X− soft block (block soft block), namely (3Fn-b-ME3n), only one broad Tg at −38 °C was seen. An identical DSC result was observed for an analogous polyurethane with a (3Fn-r-ME3n) soft block.

Subsequently, contrasting nanomorphologies were observed for a base polyurethane modified with the above-described polyurethanes having −(An-b-Bn)−X− and −(An-r-Bn)−X− soft blocks. Nanoscale phase separation (25−100 nm domains) was

Figure 12. Phase and 3D height images for the triblock hybrids at rsp = 0.9, scan size = 10 μm × 10 μm, z (phase) = 60°, z (height) = 500 nm. The Rq values (nm) are shown in 3D height images. 7908

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

Network constrained phase separation (kCN > kPN) results in B-1.1 having a low storage modulus at −90 °C (8 MPa). The absence of a separate Tg for 3F leads to a gradual decrease in storage modulus (8 to 1.4 MPa) from −90 to 150 °C. Considering the 4.2 wt % HS, the relatively flat storage modulus from −90 to 200 °C is noteworthy (Figure 11). The constraint of HS physical cross-linking is tempered by higher HS wt % for B1.1 (4.2) compared to B-4.5 (2.6). The result is a higher strain to break for B-1.1 (Figures 8 and 9). MDSC (−70 to 200 °C) at a heating rate of 10 °C/min was used to assess hybrid elastomer thermal transitions (Figure S5). The Tg for 3F diol is −51 °C.42 The change in heat capacity for the PDMS-3F-PDMS elastomers at the 3F Tg was small (∼0.0009 W g−1 °C−1) and broadly spread so that identifying Tgs was not possible. Conventional DSC was tried, but 3F Tgs could not be discerned. A Tg (−40 °C, Figure S4) was found for the hybrid polyurethane C2-4.5, which has a much higher 3F wt % (86%) compared to the hybrid triblocks (Table 1). The limited results from DSC emphasize the importance of DMA in examining low-temperature transitions for these soft elastomers. The change in slope for A-1.1 (∼120 °C) and A-4.5 (∼150 °C) may be due to the HS Tg (Figure S4). The small change in heat capacity due to low wt % HS makes assignment uncertain. These higher temperature transitions are not observed for either B-1.1 or B-4.5. Surface Characterization. TM-AFM. PDMS-3F-PDMS Hybrids. The surface morphology of the hybrid elastomers was probed by TM-AFM at a set point ratio of 0.9, which is “soft” tapping. TM-AFM was also performed on controls C1 and C24.5. Figure 12 shows 10 × 10 μm phase and 3D images for A-1.1, A-4.5, B-1.1, and B-4.5 with a sectional analysis. Surfaces have varying degrees of microdepressions which are thought to be due to shrinkage from solvent evaporation and ethanol/water generation. A sectional analysis shows that depressions for A4.5, A-1.1, and B-4.5 have vertical dimensions kPN. Consistent with previous studies,72 there is no TM-AFM evidence for siliceous domain content at 8.3 wt % SiO1.5 (Table 2). In summary, the sequence for the scale of near surface phase separation is linear HMDI-BD(30)/P[3F-4.5] ≫ C2-4.5 > B-4.5 > A-1.1 ≈ A-4.5 > B-1.1. The notion of siliceous CN constrained phase separation explains this sequence of morphological observations. Atmospheric water availability at the coating surface that accelerates hydrolysis/condensation cure (kNC) compared to physical network formation (kPC) is a likely factor contributing to minimal near surface HD imaging. Network constrained phase separation is related to the results from a study of microphase-separated morphologies for poly(urethane urea) block copolymers by Runt.71 When solvent was removed slowly physical phase separation was enhanced. Conversely, rapid removal of solvent constrained phase separation. X-ray Photoelectron Spectroscopy. Atom percentages of F, O, C, N, and Si were obtained at a 90° takeoff angle (Table 4). All Table 4. XPS Atom Percent for the Hybrid Triblocks atom % (observed) designation

N

C

A-1.1 A-4.5 B-1.1 B-4.5

1.3 1.3 2.4 1.5

48.9 49.8 50.1 50.3

designation

N

C

O

1.6 1.3 1.6 1.3

50 58.3 53.3 54.4 53.3 54.4

25 16.7 22.6 21.4 22.6 21.4

PDMS 3F diol A-1.1 A-4.5 B-1.1 B-4.5

O

Si

22.6 27.2 22.5 26.4 22.4 23.5 22.6 24.6 atom % (calculated) Si

F

Si/O

1.6 0.95

1.19 1.17 1.04 1.09

F

Si/O

25 2.9 7.6 2.9 7.6

1 0 0.86 0.72 0.86 0.72

25 19.6 15.4 19.6 15.4

compositions have at. % Si approximating that for polydimethylsiloxane (25 at. % Si). A-1.1 (27.2) and A-4.5 (26.4) have somewhat higher at. % Si while B-1.1 (23.5) and B-4.5 (24.6) have less. Fluorine could not be detected in A-1.1 or A-4.5 but B-1.1 and B-4.5 have significant fluorine concentrations (Table 4). B-1.1 has the lowest 3F segment content of the hybrid triblocks (8.9 wt %, Table 1) but the highest XPS at. % F (1.6 at. %, Table 4). 7910

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

Figure 15. DCA force−distance curves (two cycles) for A-4.5: (a) isopropanol; (b) water.

interesting morphology comprised of 2−3 μm ovaloids (low modulus) surrounded by a higher modulus matrix (Figure 6). A model is proposed for this microscale morphology based on the relative rates of physical network formation (PN, H-bonding) and chemical network formation (CN, −SiO1.5) during coating deposition (Scheme 3). This model accounts for the prominence of a “softer” A-4.5 ovaloid morphology that is attributed to rapid PN formation followed by later formation of a “harder” CN formed by hydrolysis and condensation cure. The ovaloid morphology becomes less distinct with increasing −SiO1.5 content (A-4.5 < A-1.1 < B-4.5) as the increased rate of −Si−O−Si− network constrains PN formation. Despite low hard segment weight percents (2.6−3.5) the hybrid triblocks have moderate toughness with strain at break ranging from 260 to 492%. A-4.5, which has only 3.1 wt % hard segment, has the highest strain at break (492%). This is attributed to the phase-separated microscale morphology discussed above that synergistically favors PN formation with strong hydrogen bonding (ATR-IR, Figure 4). Dynamic mechanical analysis for A-1.1, A-4.5, and B-4.5 confirms phase separation with discrete PDMS and 3F glass transition temperatures. These hybrid elastomers retain rubbery behavior over a wide temperature range (−20 to 150 °C). Triblock hybrid elastomer B-1.1 has the highest −SiO1.5 wt % (mostly from BTESE) and lowest 3F wt % (3F-1.1). No sign of microscale phase separation is observed by TM-AFM fracture surface imaging (Figure 6). Dynamic mechanical analysis fails to detect the 3F-1.1 Tg (Figures 7 and 8) although a PDMS Tg is clear. Network constrained phase separation due to high −SiO1.5 content apparently “traps” the 3F-1.1 segment in an “interphase”.79 The absence of a separate Tg for 3F leads to a gradual decrease in storage modulus (8 to 1.4 MPa) from −90 to 150 °C. Considering the 4.2 wt % HS, the relatively flat storage modulus from −90 to 200 °C is noteworthy (Figure 11). In contrast to bulk morphology, triblock hybrid surfaces are devoid of microstructural features attributable to phase separation. Height variations are ascribed to solvent/ethanol evaporation. The absence of “ovaloid” or any other microscale phase separation is thought due to surface accessibility to atmospheric moisture. A relatively high water concentration at the outermost surface of ∼200 μm coatings favors hydrolysis/ condensation cure at the expense of PN formation (Scheme 3). Network constrained phase separation thus accounts for the absence of microscale morphology in a way paralleling that observed for the bulk morphology of B-1.1. Evidence for nanoscale phase separation driven by 3F immiscibility can be discerned by faint near surface features observed in phase images (Figure S7). A correlation with related compositions supports this hypothesis.

Figure 16. Sessile drop images and contact angles for water and hexadecane for A-4.5, condensation cured PDMS, and C2-4.5.

Figure 17. Model for hybrid triblock surfaces: red ● = −SiO1.5; blue ■ = urethane or urea.

and without BTESE are very similar (Table 3, Figures 15 and 16), suggesting PDMS character dominates the outermost surface.

■

CONCLUSION PDMS-3F-PDMS triblock hybrids are easily prepared by linking 3F diols and PDMS diamine segments with urethane/urea moieties followed by hydrolysis/condensation cure of −Si(OC2H5)3 end groups (Scheme 1). “Hybrid” follows the designation given by Saegusa and Chujo for materials with organic polymer and inorganic domains.35 For the triblock hybrids, a commercially available PDMS diamine 3 was used for the terminal blocks while the center block precursors were polyoxetane 3F-1.1 and 3F-4.5 diols.38 Augmenting the siliceous content with BTESE 5 brought to light interesting effects of competition between HS association and hydrolysis/condensation cure (Scheme 3). PDMS-3F-PDMS triblock hybrids form robust, optically transparent elastomers. Optical transparency is due to nearly identical refractive indexes for 3F and PDMS segments that constitute >90 wt % of the elastomers. TM-AFM images for A4.5, A-1.1, and B-4.5 fracture surfaces reveal microscale bulk phase separation. The A-4.5 triblock hybrid shows a particularly 7911

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

(18) Yilgor, I.; Eynur, T.; Bilgin, S.; Yilgor, E.; Wilkes, G. L. Polymer 2011, 52, 266−274. (19) Yilgor, E.; Isik, M.; Yilgor, I. Macromolecules 2010, 43, 8588− 8593. (20) Yilgor, I.; Eynur, T.; Yilgor, E.; Wilkes, G. L. Polymer 2009, 50, 4432−4437. (21) Yilgor, E.; Eynur, T.; Kosak, C.; Bilgin, S.; Yilgor, I.; Malay, O.; Menceloglu, Y.; Wilkes, G. L. Polymer 2011, 52, 4189−4198. (22) Zhang, W.; Zheng, Y.; Orsini, L.; Morelli, A.; Galli, G.; Chiellini, E.; Carpenter, E. E.; Wynne, K. J. Langmuir 2010, 26, 5848−5855. (23) Owen, M. J. Ind. Eng. Chem. Prod. Res. Dev. 1980, 19, 97−103. (24) Kobayashi, H.; Owen, M. J. Trends Polym. Sci. 1995, 3, 330−335. (25) Furukawa, Y.; Shin-ya, S.; Saito, M.; Narui, S.; Miyake, H. Polym. Adv. Technol. 2002, 13, 60−65. (26) Razzano, J. S. Momentive Performance Materials Inc., U.S. Patent 7361722, 2008. (27) Koshar, R. J. Minnesota Mining and Manufacturing Company, St. Paul, MN, 1986. (28) Merkel, T. J.; Herlihy, K. P.; Nunes, J.; Orgel, R. M.; Rolland, J. P.; DeSimone, J. M. Langmuir 2010, 26, 13086−13096. (29) Bertolucci, M.; Galli, G.; Chiellini, E.; Wynne, K. J. Macromolecules 2004, 37, 3666−3672. (30) Bullock, S.; Johnston, E. E.; Willson, T.; Gatenholm, P.; Wynne, K. J. J. Colloid Interface Sci. 1999, 210, 18−36. (31) Berglin, M.; Wynne, K. J.; Gatenholm, P. J. Colloid Interface Sci. 2003, 257, 383−391. (32) Harada, K.; Koizumi, A.; Saito, N.; Inoue, K.; Yoshinaga, T.; Date, C.; Fujii, S.; Hachiya, N.; Hirosawa, I.; Koda, S.; Kusaka, Y.; Murata, K.; Omae, K.; Shimbo, S.; Takenaka, K.; Takeshita, T.; Todoriki, H.; Wada, Y.; Watanabe, T.; Ikeda, M. Chemosphere 2007, 66, 293−301. (33) Guo, J.; Resnick, P.; Efimenko, K.; Genzer, J.; DeSimone, J. M. Ind. Eng. Chem. Res. 2008, 47, 502−508. (34) Saegusa, T.; Chujo, Y. J. Macromol. Sci., Chem. 1990, A27, 1603− 1612. (35) Chujo, Y.; Ihara, E.; Kure, S.; Saegusa, T. Macromolecules 1993, 26, 5681−5686. (36) Lai, X. J.; Li, X. R.; Wang, L.; Shen, Y. D. Polym. Bull. 2010, 65, 45−57. (37) Sardon, H.; Irusta, L.; Fernandez-Berridi, M. J.; Lansalot, M.; Bourgeat-Lami, E. Polymer 2010, 51, 5051−5057. (38) Zhang, W.; Henke, D.; Presnall, D.; Chakrabarty, S.; Wang, C.; Wynne, K. J. Macromol. Chem. Phys. 2012, DOI: 10.1002/ macp.201200053. (39) Skoog, D. A.; Holler, F. J.; Crouch, S. R. Principles of Instrumental Analysis, 6th ed.; Thomson Higher Education: 2007. (40) Wilhelmy, L. Ann. Phys. Chem. (Leipzig) 1863, 119, 177. (41) Varadaraj, R.; Bock, J.; Valint, P.; Zushma, S.; Brons, N. J. Phys. Chem. 1991, 95, 1679−1681. (42) Makal, U.; Uilk, J.; Kurt, P.; Cooke, R. S.; Wynne, K. J. Polymer 2005, 46, 2522−2530. (43) Wynne, K. J.; Ho, T.; Johnston, E. E.; Myers, S. A. Appl. Organomet. Chem. 1998, 12, 763−770. (44) Uilk, J.; Bullock, S.; Johnston, E.; Myers, S. A.; Merwin, L. H.; Wynne, K. J. Macromolecules 2000, 33, 8791−8801. (45) Ong, M. D.; Volksen, W.; Dubois, G.; Lee, V.; Brock, P. J.; Deline, V. R.; Miller, R. D.; Dauskardt, R. H. Adv. Mater. 2008, 20, 3159−3164. (46) Volksen, W.; Magbitang, T. P.; Miller, R. D.; Purushothaman, S.; Cohen, S. A.; Nakagawa, H.; Nobe, Y.; Kokubo, T.; Dubois, G. J. M. J. Electrochem. Soc. 2011, 158, G155−G161. (47) Volksen, W.; Miller, R. D.; Dubois, G. Chem. Rev. 2010, 110, 56− 110. (48) Burgmann, S.; Grosse, S.; Schroder, W.; Roggenkamp, J.; Jansen, S.; Graf, F.; Busen, M. Exp. Fluids 2009, 47, 865−881. (49) Smith, B. C. In Fundamentals of Fourier Transform Infrared Spectroscopy; Taylor and Francis Group, LLC: Boca Raton, FL, 2009; pp 132−134. (50) Sato, Y.; Ootsubo, M.; Yamamoto, G.; Van Lier, G.; Terrones, M.; Hashiguchi, S.; Kimura, H.; Okubo, A.; Motomiya, K.; Jeyadevan, B.; Hashida, T.; Tohji, K. ACS Nano 2008, 2, 348−356.

Based on contact angle measurements and XPS analysis, the outermost surface for PDMS-3F-PDMS hybrid triblocks is dominated by PDMS (Figure 15). Self-aggregation driven by 3F immiscibility with PDMS22,59 leads to surface depletion of 3F segments. The ready joining of soft segments by a straightforward addition reactions promises an entrée into many multiblock sytems. The interesting bulk and surface characteristics of PDMS-3F-PDMS hybrid elastomers described herein suggest further work such as varying segment sequence, investigating solvent effects and further physical and mechanical property investigations.

■

ASSOCIATED CONTENT

S Supporting Information *

Table S1 and Figures S1−S10. This material is available free of charge via the Internet at http://pubs.acs.org

■

AUTHOR INFORMATION

Corresponding Author

*Ph 001-804-828-9303, Fax +804-828-3846, e-mail kjwynne@ vcu.edu. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the National Science Foundation (DMR0802452 and DMR-1206259) for generous support of this research.

■

REFERENCES

(1) Opris, D. M.; Molberg, M.; Walder, C.; Ko, Y. S.; Fischer, B.; Nueesch, F. A. Adv. Funct. Mater. 2011, 21, 3531−3539. (2) Thanawala, S. K.; Chaudhury, M. K. Langmuir 2000, 16, 1256− 1260. (3) Okamoto, T.; Nakamura, S. Inst. Pure Appl. Phys. 2008, 521−526. (4) Botter, W.; Soares, R. F.; Galembeck, F. J. Adhes. Sci. Technol. 1992, 6, 791−805. (5) Hillborg, H.; Gedde, U. W. IEEE Trans. Dielectr. Electr. Insul. 1999, 6, 703−717. (6) Unger, M. A.; Chou, H. P.; Thorsen, T.; Scherer, A.; Quake, S. R. Science 2000, 288, 113−116. (7) Wynne, K. J.; Lambert, J. M. In Encyclopedia of Biomaterials and Biomedical Engineering; Wnek, G. W., Bowlin, G. B., Eds.; Marcel Dekker: New York, 2004; Vol. 1, pp 1348−1362. (8) Prasad, B. R.; Brook, M. A.; Smith, T.; Zhao, S. G. I.; Chen, Y.; Sheardown, H.; D’Souza, R.; Rochev, Y. Colloids Surf., B 2010, 78, 237− 242. (9) Bokobza, L. In Reinforcement of Elastomeric Networks by Fillers, Wiley-VCH Verlag Gmbh: Berlin, 2001; pp 243−260. (10) Bokobza, L.; Rapoport, O. J. Appl. Polym. Sci. 2002, 85, 2301− 2316. (11) Siloxane Polymers; Clarson, S. J., Semylen, J. A., Eds.; Prentice Hall: Upper Saddle River, NJ, 1993. (12) Burnside, S. D.; Giannelis, E. P. Chem. Mater. 1995, 7, 1597− 1600. (13) Garrido, L.; Mark, J. E.; Sun, C. C.; Ackerman, J. L.; Chang, C. Macromolecules 1991, 24, 4067−4072. (14) Stevenson, I.; David, L.; Gauthier, C.; Arambourg, L.; Davenas, J.; Vigier, G. Polymer 2001, 42, 9287−9292. (15) Winberg, P.; Eldrup, M.; Maurer, F. H. J. Polymer 2004, 45, 8253− 8264. (16) Mark, J. E. Curr. Opin. Solid State Mater. Sci. 1999, 4, 565−570. (17) Wen, J.; Mark, J. E. J. Mater. Sci. 1994, 29, 499−503. 7912

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913

Macromolecules

Article

(51) Shimomura, M.; Okumoto, H.; Kaito, A.; Ueno, K.; Shen, J. S.; Ito, K. Macromolecules 1998, 31, 7483−7487. (52) Seymour, R. W.; Estes, G. M.; Cooper, S. L. Macromolecules 1970, 3, 579−583. (53) Coleman, M. M.; Lee, K. H.; Skrovanek, D. J.; Painter, P. C. Macromolecules 1986, 19, 2149−2157. (54) Coleman, M. M.; Skrovanek, D. J.; Hu, J. B.; Painter, P. C. Macromolecules 1988, 21, 59−65. (55) MarcosFernandez, A.; Lozano, A. E.; Gonzalez, L.; Rodriguez, A. Macromolecules 1997, 30, 3584−3592. (56) Magonov, S. N.; Elings, V.; Whangbo, M. H. Surf. Sci. 1997, 375, L385−L391. (57) Sheth, J. P.; Aneja, A.; Wilkes, G. L.; Yilgor, E.; Atilla, G. E.; Yilgor, I.; Beyer, F. L. Polymer 2004, 45, 6919−6932. (58) Tyagi, D.; Yilgor, I.; McGrath, J. E.; Wilkes, G. L. Polymer 1984, 25, 1807−1816. (59) Lee, J. N.; Park, C.; Whitesides, G. M. Anal. Chem. 2003, 75, 6544−6554. (60) Uemura, S.; Takayanagi, M. J. Appl. Polym. Sci. 1966, 10, 113− 125. (61) Sperling, L. H. Introduction to Physical Polymer Science, 4th ed.; Wiley Interscience: Hoboken, NJ, 2006. (62) Feng, H. Q.; Feng, Z. L.; Yuan, H. Z.; Shen, L. F. Macromolecules 1994, 27, 7830−7834. (63) Chen, Z.; Gong, K. J. Appl. Polym. Sci. 2002, 84, 1499−1503. (64) Wang, L. F.; Ji, Q.; Glass, T. E.; Ward, T. C.; McGrath, J. E.; Muggli, M.; Burns, G.; Sorathia, U. Polymer 2000, 41, 5083−5093. (65) Sperling, L. H. In Applied Polymer Science; Craver, C., Carraher, C. J., Eds.; Elsevier Science: New York, 2000; pp 343−354. (66) Liu, R. Y. F.; Jin, Y.; Hiltner, A.; Baer, E. Macromol. Rapid Commun. 2003, 24, 943−948. (67) Liu, R. Y. F.; Ranade, A. P.; Wang, H. P.; Bernal-Lara, T. E.; Hiltner, A.; Baer, E. Macromolecules 2005, 38, 10721−10727. (68) Lai, C.; Ayyer, R.; Hiltner, A.; Baer, E. Polymer 2010, 51, 1820− 1829. (69) Hu, Z. K.; Chen, L.; Betts, D. E.; Pandya, A.; Hillmyer, M. A.; DeSimone, J. M. J. Am. Chem. Soc. 2008, 130, 14244−14252. (70) Zhang, W.; Fujiwara, T.; Taskent, H.; Zheng, Y.; Brunson, K.; Gamble, L.; Wynne, K. J. Macromol. Chem. Phys. 2012, in press. (71) Garrett, J. T.; Lin, J. S.; Runt, J. Macromolecules 2002, 35, 161− 168. (72) Inagi, S.; Ogoshi, T.; Miyake, J.; Bertolucci, M.; Fujiwara, T.; Galli, G.; Chiellini, E.; Chujo, Y.; Wynne, K. J. Chem. Mater. 2007, 19, 2141− 2143. (73) Chanda, M. Advanced Polymer Chemistry; Marcel Dekker: New York, 2000. (74) Sun, F.; Castner, D. G.; Mao, G.; Wang, W.; McKeown, P.; Grainger, D. W. J. Am. Chem. Soc. 1996, 118, 1856−1866. (75) Krishnan, S.; Paik, M. Y.; Ober, C. K.; Martinelli, E.; Galli, G.; Sohn, K. E.; Kramer, E. J.; Fischer, D. A. Macromolecules 2010, 43, 4733−4743. (76) Krishnan, S.; Kwark, Y. J.; Ober, C. K. Chem. Rec. 2004, 4, 315− 330. (77) Krishnan, S.; Ayothi, R.; Hexemer, A.; Finlay, J. A.; Sohn, K. E.; Perry, R.; Ober, C. K.; Kramer, E. J.; Callow, M. E.; Callow, J. A.; Fischer, D. A. Langmuir 2006, 22, 5075−5086. (78) Martinelli, E.; Menghetti, S.; Galli, G.; Glisenti, A.; Krishnan, S.; Paik, M. Y.; Ober, C. K.; Smilgies, D. M.; Fischer, D. A. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 267−284. (79) Liu, R. Y. F.; Bernal-Lara, T. E.; Hiltner, A.; Baer, E. Macromolecules 2005, 38, 4819−4827.

7913

dx.doi.org/10.1021/ma301447f | Macromolecules 2012, 45, 7900−7913