Article pubs.acs.org/Langmuir
Headgroup Effect on Silane Structures at Buried Polymer/Silane and Polymer/Polymer Interfaces and Their Relations to Adhesion Chi Zhang,† Nick E. Shephard,‡ Susan M. Rhodes,‡ and Zhan Chen*,† †
Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109, United States Dow Corning Corporation, Midland, Michigan 48686, United States
‡
S Supporting Information *
ABSTRACT: Sum frequency generation (SFG) vibrational spectroscopy was used to study the effect of silane headgroups on the molecular interactions that occur between poly(ethylene terephthalate) (PET) and various epoxy silanes at the PET/silane and PET/silicone interfaces. Three different silanes were investigated: (3-glycidoxypropyl) trimethoxysilane (γ-GPS), (3-glycidoxypropyl) methyl-dimethoxysilane (γGPMS), and (3-glycidoxypropyl) dimethyl-methoxysilane (γGPDMS). These silanes share the same backbone and epoxy end group but have different headgroups. SFG was used to examine the interfaces between PET and each of these silanes, as well as silanes mixed with methylvinylsiloxanol (MVS). We also examined the interfaces between PET and uncured or cured silicone with silanes or silane−MVS mixtures. Silanes with different headgroups were found to exhibit a variety of methoxy group interfacial segregation and ordering behaviors at various interfaces. The effect of MVS was also dependent upon silane headgroup choice, and the interfacial molecular structures of silane methoxy headgroups were found to differ at PET/silane and PET/silicone interfaces. Epoxy silanes have been widely used as adhesion promoters for polymer adhesives; therefore, the molecular structures probed using SFG were correlated to adhesion testing results to understand the molecular mechanisms of silicone-polymer adhesion. Our results demonstrated that silane methoxy headgroups play important roles in adhesion at the PET/silicone interfaces. The presence of MVS can change interfacial methoxy segregation and ordering, leading to different adhesion strengths.
■
INTRODUCTION Silicones, e.g., poly(dimethyl siloxane) (PDMS), have many superb characters such as good biocompatibility, high thermal stability, super hydrophobicity, and excellent fouling-release properties.1−4 Due to their unique properties, silicone elastomers are widely used as polymer adhesives in a variety of applications by the automotive, aviation, household, medical, electronic, and other industries. Ensuring the robust adhesion of silicone elastomers to polymers is therefore important to many applications. Platium catalyzed hydrosilylation provides a stable and predictable chemistry for cross-linking silicone polymer and is commonly used for mold making materials. For use in adhesives, this cure chemistry lacks the necessary condensation chemistry associated with interfacial covalent bonding on a variety of surfaces. Pretreatments such as corona or plasma treatments are usually used to alter polymer substrate surfaces and improve adhesion with PDMS materials, but these processes are expensive and time-consuming. An alternative way to enhance adhesion is to incorporate an adhesion promoter into PDMS to make it self-adhering when cured on a polymer surface. Organosilanes, such as alkoxysilane based coupling agents, have been developed as adhesion promoters to adhere silicone elastomers to organic polymers and inorganic materials.5−10 Even though it is well-known that © 2012 American Chemical Society
alkoxysilane adhesion promoters can enhance interfacial adhesion, the molecular-level mechanisms of how silane coupling agents enhance adhesion are not well understood. Interfacial properties, such as adhesion, are determined by interfacial structures. One way to study the structure of a buried interface is to break the interface and examine the two resulting surfaces. But this process may alter the original interfacial structure, especially for interfaces with good adhesion. Therefore there is a need for techniques that can directly study buried interfaces in situ. Sum frequency generation (SFG) vibrational spectroscopy has been developed into an effective analytical technique to study surfaces and buried interfaces in situ.11−31 SFG is a second order nonlinear optical technique that is surface sensitive due to its selection rules. Information on surface functional group identity, surface coverage, orientation distribution, and ordering can be deduced by analyzing polarized SFG spectra. SFG has been successfully applied to the study of interfaces involving polymer materials in situ.32−43 For example, molecular interactions such as interfacial hydroReceived: January 1, 2012 Revised: March 12, 2012 Published: March 16, 2012 6052
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
Article
Figure 1. Chemical structures of polymers and silanes employed in the study: (a) Poly(ethylene terephthalate) (PET); (b) PET with aliphatic chain deuterated (d4-PET); (c) Methylvinylsiloxanol (MVS); (d) (3-glycidoxypropyl) trimethoxysilane (γ-GPS); (e). (3-glycidoxypropyl) methyldimethoxysilane (γ-GPMS); (f) (3-glycidoxypropyl) dimethyl-methoxysilane (γ-GPDMS).
Figure 2. (a) Window geometry and (b) NCA geometry.
headgroups on interfacial epoxy silane behaviors and polymer adhesion.
gen bonding and the diffusion of silanes into polymer films have been investigated using SFG.44−48 Particularly, SFG has been applied to study interfacial behavior of epoxy silane (3-glycidoxypropyl) trimethoxysilane (γ-GPS) and its mixture with methylvinylsiloxanol (MVS).49,50 It has been reported that a mixture of γ-GPS and MVS effectively promotes adhesion of silicone elastomer to poly(ethylene terephthalate) (PET).5,6 SFG has been applied to examine interfaces between PET and γ-GPS or γ-GPS:MVS mixture, and interfaces between PET and uncured or cured silicone with 1.5% γ-GPS or a 3.0% γ-GPS:MVS blend mixed into the PDMS matrix.50 SFG has also been used to study interfacial behaviors of other silanes with different end groups at the polymer/silane and polymer/polymer interfaces to understand the silane end group effect.50 Among the silanes studied, only the γ-GPS:MVS mixture in silicone showed strong methoxy group segregation/ordering at PET/silicone interfaces and exhibited strong adhesion.50 This suggests that epoxy end groups affect interfacial behavior of silane headgroups and such headgroups may play important roles in adhesion. In this research, we therefore focused on the effect of silane headgroups on adhesion. We probed interfaces between PET and γ-GPS (with three methoxy groups), (3-glycidoxypropyl) methyl-dimethoxysilane (γ-GPMS, with two methoxy groups and one methyl group), and (3-glycidoxypropyl) dimethylmethoxysilane (γ-GPDMS, with one methoxy group and two methyl groups), and their mixtures with MVS. We also examined interfaces between PET and uncured or cured silicone with different silanes and their mixtures with MVS. SFG spectra detected at different interfaces were correlated to adhesion testing results. This research is a step toward developing a molecular understanding of the effect of
■
MATERIALS AND METHODS
Materials. Fused silica substrates (right angle prisms) were obtained from Altos Photonics, Inc. They were cleaned overnight by placing them in a concentrated sulfuric acid bath saturated with potassium dichromate at 60 °C. These substrates were then rinsed using deionized water and dried with nitrogen gas before polymer deposition. Poly(ethylene terephthalate) with deuterated ethylene glycol subunits (d4-PET, Mn = 72 000) was obtained from Polymer Source, Inc. and was used to avoid spectral confusion in the C−H stretching frequency region from silanes, MVS and PDMS. The d4PET sample was dissolved in dichlorophenol (Sigma Aldrich) to form a 2 wt % solution. The d4-PET polymer thin films were prepared by spin coating the solution at 2500 rpm on fused silica substrates. The spin coater was from Specialty Coating Systems (P-6000). Silane γGPDMS was ordered from Silar Laboratories. Silanes γ-GPS and γGPMS, MVS, and Sylgard 184 silicone elastomer kit were obtained from the Dow Corning Corporation. The Sylgard 184 silicone elastomer was prepared by mixing the base and curing agent in a 10:1 ratio. To incorporate silanes into PDMS, 1.5 wt % γ-GPS, 3.0 wt % 1:1 (w/w) γ-GPS/MVS mixture, 1.5 wt % γ-GPMS, 3.0 wt % 1:1 (w/w) γGPMS/MVS mixture, 1.5 wt % (γ-GPDMS), and 3.0 wt % 1:1 (w/w) γ-GPDMS/MVS mixture were mixed homogeneously using a vortex mixer (Vortex-Genie 2T, Scientific Industries Inc.) into the PDMS base before curing respectively. When MVS was added, a SiHfunctional PDMS (6−3570, Dow Corning) was added to maintain a 1.5:1 SiH/vinyl molar ratio. The 1:1 (w/w) silane/MVS mixtures were used because it was shown that the 1:1 (w/w) γ-GPS/MVS ratio can effectively enhance the adhesion between polymers and PDMS.5,6,47 The curing of the silicone samples was carried out in an oven at 140 °C for 120 min on d4-PET surfaces. The samples were then kept at room temperature for 24 h before use. Molecular formulas of the materials used in the experiments are shown in Figure 1. 6053
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
Article
SFG Experimental. The theory of SFG has been welldeveloped11−14,51−58 and the details of our SFG system have been reported in previous publications.38−50 Briefly, the visible and midinfrared (IR) input beams penetrate through a fused silica substrate and overlap at the interface spatially and temporally. At the sample stage, pulse energies of the visible and IR beams are 30 and 100 μJ, respectively. The reflected SFG signal is collected by a monochromator along with a photomultiplier tube (PMT). The monochromator can be tuned automatically while scanning the IR beam frequency. All SFG spectra in the experiment were collected using the ssp (s-polarized sum frequency output, s-polarized visible input, and p-polarized IR input) polarization combination. A window geometry (Figure 2a) was used to probe the polymer/silane and polymer/polymer interfaces previously.44−47,49,50 In this research, we used a near critical angle (NCA) geometry (Figure 2b) to study buried interfaces. The new geometry can provide stronger SFG signal intensity at the interfaces, and thus improve the signal-to-noise ratio. In the NCA reflection geometry, a right angle silica prism was used as the solid support for the PET thin film. The incident angles of the visible and mid-IR input beams used in the experiment are 57° and 54° relative to the surface normal, respectively. The calculated refractive angle of the signal at the interface, e.g., d4-PET/silicone interface is ∼81°, much larger than that in the window geometry (∼42°). Larger refractive angle provides stronger reflection, which increases the SFG signal intensity detected. It was found in our experiments that the overall SFG signal intensity collected using the NCA geometry is about 10 times stronger than that collected using the window geometry. For each experiment, at least three same samples were prepared and at least 10 SFG spectra were collected from each sample at different positions. The experimental results are reproducible. Adhesion Testing. The adhesion test in these experiments was based on ASTM D3163 Standard with some modifications using an Instron 5544 mechanical test machine. The shear test geometry used is
with a pulling speed of 1.3 mm/min while the shear strength was measured. SFG Spectral Fitting. SFG is a second order nonlinear optical process. The SFG signal intensity can be expressed as 2 ISFG = |χ(2) eff | IIR Ivis
(1)
Here IIR and Ivis are the intensities of the input IR beam and visible (2) is the effective second order nonlinear beam, respectively. χeff susceptibility, which can be written as (2) χ(2) eff = χ NR +
∑ q
Aq ω IR − ωq + i Γq
(2)
(2) is the nonresonant contribution from the sample. The Here χNR resonant contribution can be modeled as a sum of Lorentzian peaks with signal strength or amplitude Aq, frequency ωq, and line width Γq. All the SFG spectra presented in this paper were fitted according to eq 2 to obtain the resonant nonlinear susceptibility of the material, and the fitting parameters are shown in the Supporting Information. This study will focus on silane methoxy headgroups at interfaces, which produce symmetric C−H stretching signals around 2840 cm−1.
■
RESULTS AND DISCUSSION Interfaces between d4-PET and MVS or Silicone (without Silanes). As discussed above, to avoid overlapping signals from other compounds in the C−H stretching frequency range, deuterated d4-PET was used in the experiments. Therefore, no aliphatic SFG C−H stretching signal (usually 3000 cm−1) was detected.47 Here, we first studied the interfaces between d4PET and silicone elastomers before and after curing. In previous studies, no SFG signal could be observed from such interfaces.50 In this experiment, the signal enhancement provided by the new experimental geometry revealed two peaks at ∼2900 and 2965 cm−1 (from PDMS methyl symmetric and asymmetric stretches; Figure 4a,b). Furthermore, after spectral fitting we found that the symmetric/asymmetric stretching signal strength ratios are different before and after curing (0.9 before curing, 1.1 after curing). This indicates that the PDMS methyl group orientation changed at the interface during the curing process. An aromatic C−H stretching signal at ∼3085 cm−1 was also detected at the d4-PET/silicone interfaces, showing that the d4-PET phenyl groups are also ordered at these interfaces. We also collected the ssp SFG spectrum from the d4-PET/ MVS interface (Figure 4c). Two peaks at 2945 and 2970 cm−1
Figure 3. Adhesion test geometry for 180° shear test. shown in Figure 3. We cut a PET sheet (Ertalyte) into small pieces of the same size. The test pieces were sanded and cleaned using ethanol, rinsed extensively with deionized water, and dried before use. The contact surface area of the PET test pieces is 30.0(width) × 12.0(overlap length) mm2. Two PET pieces were attached using different silicone and silane (with or without MVS) mixtures and were cured at 140 °C for 120 min. Then the samples were kept at the room temperature for 24 h. The PDMS adhesive thickness is about 0.2 mm. The bonded PET pieces were pulled apart at the room temperature
Figure 4. SFG spectra collected from (a) the d4-PET/uncured silicone interface, (b) the d4-PET/cured silicone interface, and (c) the d4-PET/MVS interface. The dots are experimental data and the lines are fitted results. 6054
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
Article
Figure 5. SFG spectra collected from (a) the d4-PET/γ-GPS (Δ) and d4-PET/γ-GPS:MVS interfaces (○) using the window geometry, (b) the d4PET/γ-GPS (Δ) and d4-PET/γ-GPS:MVS interfaces (○) using the NCA geometry, (c) the d4-PET/γ-GPMS (Δ) and d4-PET/γ-GPMS:MVS interfaces (○) using the NCA geometry, (d) the d4-PET/γ-GPDMS (Δ) and d4-PET/γ-GPDMS:MVS interfaces (○) using the NCA geometry. For panels b−d, the dots are experimental data and the lines are fitted results.
were detected. The 2970 cm−1 signal is due to the asymmetric stretching mode of the Si-CH3 group in MVS. The 2945 cm−1 signal is unknown, which might be contributed by the Si−CH3 group or other C−H groups in MVS. Nevertheless, the detection of SFG signal here indicated that methyl groups (and perhaps other C−H groups) in MVS may be ordered at the d4PET/MVS interface. Interfaces between d4-PET and Silane or Silane−MVS Mixture. SFG spectroscopy was previously applied to study the interfaces between d4-PET in contact with a variety of silanes and their mixtures with MVS.47 When d4-PET was in contact with γ-GPS, a weak signal of the methoxy headgroups on γ-GPS was observed from the d4-PET/γ-GPS interface, while at the d4-PET/γ-GPS:MVS mixture interface, a much stronger methoxy C−H stretching SFG signal was observed.47 We first reproduced the SFG results at the d4-PET/γ-GPS and d4-PET/γ-GPS:MVS mixture interfaces using the window geometry (Figure 5a). We then collected SFG spectra from the same interfaces using the NCA geometry (Figure 5b). The spectra collected using different experimental geometries have similar features, but those collected using the NCA geometry exhibit stronger intensities, more spectral details, and better signal-to-noise ratios. In order to understand the effect of silane headgroups on interfacial silane behaviors, we also collected SFG spectra from the interfaces between d4-PET and silanes with a variety of different headgroups: γ-GPMS, γ-GPDMS, and their mixtures with MVS. All of these SFG spectra are shown in Figure 5. Furthermore, we fitted all of the spectra using eq 2. As mentioned above, since this study focused on the effect of headgroups upon adhesion, we will only discuss the SFG signal contributed by the headgroups (especially the methoxy peak at ∼2840 cm−1). The assignments for other C−H signals are complicated, and therefore other functional group signals are not discussed in detail here. Further information and possible peak assignments are presented in the Supporting Information. More detailed studies can be performed in the future using selectively isotope labeled silanes.
At the d4-PET/γ-GPS interface, the peak centered at ∼2840 cm−1 is assigned to the symmetric C−H stretch of the methoxy group. This indicates that at the interface, silane methoxy groups segregate with some order. Similar to our previous reports, at the d4-PET/γ-GPS:MVS 1:1 mixture interface, the intensity of the 2840 cm−1 peak increased greatly (Figures 5a and 5b). According to the fitting results in the Supporting Information, the signal strength increased from 1.2 to 3.2 from the methoxy groups at the interface after MVS was introduced into the system. This shows that MVS greatly increases the ordering of γ-GPS methoxy groups at the d4-PET/mixture interface. A methoxy peak was also detected from γ-GPMS in the spectrum collected from the d4-PET/γ-GPMS interface (Figure 5c). Whereas γ-GPS has three methoxy groups, a γ-GPMS molecule has only two methoxy groups plus a methyl group. The SFG signal of the methyl headgroup, which is connected to the Si atom of the silane, should be observed at ∼2910 cm−1 if they segregate with order at the interface. However, this peak is close to the methylene asymmetric stretching signal, and therefore the peak observed at about ∼2920 cm−1 cannot be definitely assigned to the Si−CH3 headgroup. The SFG signal intensity observed from the methoxy groups at the d4-PET/γGPMS interface (signal strength 1.4) is not very different from that observed at the d4-PET/γ-GPS interface (signal strength 1.2). In the SFG spectrum collected from the d4-PET/γGPMS:MVS 1:1 mixture interface, the methoxy C−H symmetric stretch at ∼2840 cm−1 can still be resolved. The spectral fitting results indicated that the signal strength grows from 1.4 to 1.7 after the introduction of MVS. This shows that the addition of MVS slightly enhanced the methoxy ordering of γ-GPMS at the interface. At the d4-PET/γ-GPDMS interface (Figure 5d), the SFG signal of the methoxy symmetric stretch at about 2840 cm−1 could also be detected. As discussed above, the observed peak at 2925 cm−1 may be contributed by the methylene asymmetric stretching and/or the methyl headgroup. Therefore, possibly 6055
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
Article
Figure 6. SFG spectra collected from the interfaces between d4-PET and uncured silicones mixed with (a) γ-GPS (Δ) and γ-GPS:MVS (○), (b) γGPMS (Δ) and γ-GPMS:MVS (○), and (c) γ-GPDMS (Δ) and γ-GPDMS:MVS (○). The dots are experimental data and the lines are fitted results.
Figure 7. SFG spectra collected from the interfaces between d4-PET and cured silicone with (a) γ-GPS (Δ) and γ-GPS:MVS (○), (b) γ-GPMS (Δ) and γ-GPMS:MVS (○), and (c) γ-GPDMS (Δ) and γ-GPDMS:MVS (○). The dots are experimental data and the lines are fitted results.
the methyl headgroup also exhibits some order at the d4-PET/ γ-GPDMS interface. At the d4-PET/γ-GPDMS:MVS 1:1 mixture interface, no SFG signal at ∼2840 cm−1 was detected, indicating that the interfacial methoxy group is disordered by the addition of MVS. This is different from the γ-GPS and γ-GPMS cases discussed above. The C−H symmetric stretching signal of the silane methyl headgroups at 2918 cm−1 could be resolved. Again, this peak could overlap with peaks from the methylene groups or even the Si-CH3 groups in MVS. The SFG results reported above demonstrate that silane methoxy headgroups exhibit some order at the d4-PET/γ-GPS, d4-PET/γ-GPMS, and d4-PET/γ-GPDMS interfaces. The addition of MVS to the first two silanes enhanced the interfacial ordering of the methoxy groups. In contrast, the addition of MVS disordered the methoxy groups at the d4PET/γ-GPDMS (with MVS) interface. Interfaces between d4-PET and Uncured Silicone Mixed with Silane or Silane−MVS Mixture. SFG spectra were collected from several d4-PET/uncured silicone interfaces, with PDMS silicone mixed with γ-GPS (1.5%), γ-GPS:MVS mixture (3.0%), γ-GPMS (1.5%), γ-GPMS:MVS mixture (3.0%), γ-GPDMS (1.5%), or γ-GPDMS:MVS mixture (3.0%) (Figure 6). Figure 6a shows the SFG spectra collected from the d4-PET/uncured silicone (mixed with γ-GPS with and without MVS) interfaces. In the absence of MVS, the ordering of the silane methoxy group at the interface could be inferred by the observed signal at ∼2840 cm−1. With MVS, the methoxy groups and the backbone still adopt some order at the interface. The strength of the methoxy signal at ∼2840 cm−1 increased only slightly after the addition of MVS (from 1.3 to 1.7). This is different from the situation where d4-PET was in contact with γ-GPS alone or with a γ-GPS:MVS mixture, in which the addition of MVS greatly enhanced the methoxy interfacial
order. The results indicate that uncured PDMS influences methoxy ordering at the d4-PET interface. At the d4-PET/uncured silicone (mixed with γ-GPMS, or with γ-GPMS:MVS) interface, the 2840 cm−1 peak was again observed, indicating ordering of the silane methoxy headgroups (Figure 6b). The fitting results indicate that the signal strength of this peak is very similar before and after the addition of MVS, at 1.0 and 1.2, respectively. This indicates that the interfacial segregation and/or interfacial orientation of the methoxy headgroups may only slightly change after the addition of MVS. Figure 6c shows the SFG spectra collected from the d4-PET/ uncured silicone (mixed with γ-GPDMS with and without MVS) interfaces. With only silane in silicone, weak methoxy symmetric stretching signals can be observed at ∼2840 cm−1. This shows that methoxy headgroups adopt some order at the interface. With MVS and silane in silicone, these signals disappear, suggesting that the methoxy groups are no longer ordered. Furthermore, in all of the samples that incorporate uncured silicone, signals at ∼3080 cm−1 provide evidence that phenyl groups in d4-PET adopt some order at the interface. This is different from the situations without the uncured silicone. At the d4-PET/uncured silicone interfaces for all the three silanes, the methoxy signal strengths are similar, at 1.3, 1.0, and 1.0, respectively. However, when mixed with MVS, the methoxy interfacial segregation and/or interfacial ordering increased slightly for γ-GPS, remained more or less the same for γ-GPMS, and decreased for γ-GPDMS. Interfaces between d4-PET and Cured Silicone Mixed with Silane or Silane−MVS Mixture. We cured all the samples investigated above in a convection oven for 120 min at 140 °C, and kept the samples at room temperature for at least 24 h. We then collected SFG spectra from the interfaces of all the cured samples. In the spectrum collected from the d4-PET/ cured silicone interface with 1.5% γ-GPS (Figure 7a), a peak at 6056
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
Article
∼2840 cm−1 was observed, indicating that the γ-GPS methoxy groups were segregated to and ordered at the polymer interface. The intensity of the γ-GPS methoxy peak was weaker after the addition of MVS, which was opposite to the relative intensities observed in the spectra collected when d4-PET in contact with silane or the uncured silicone (with or without MVS). Before and after the addition of MVS, the signal strengths are 1.4 and 1.1, respectively. Compared to the uncured silicone cases, the signal strength of the methoxy groups at the interface with γGPS is similar, at 1.3 and 1.4, respectively. However, for the γGPS with MVS, the signal decreased substantially from 1.7 to 1.1. As we will further discuss in the next section, the d4-PET/ cured silicone (with γ-GPS and MVS) interface has the best adhesion. The methoxy SFG signal decrease may be due to the high cure temperature of 140 °C, which may induce a chemical reaction or diffusion of the silane at the polymer interface. The loss of methoxy signal upon curing suggests that the methoxy headgroups must play a role in interfacial adhesion through chemical reactions or interfacial diffusion. This will be studied further in the future. SFG spectra were next collected from the d4-PET/cured silicone interfaces with silicone mixed with γ-GPMS (1.5%) and γ-GPMS:MVS mixture (3.0%), shown in Figure 7b. In both cases the peak at ∼2840 cm−1 was observed. The fitting results indicated that the methoxy signal strengths are about the same, at 0.8 and 0.7, respectively. As with γ-GPS, the fitted signal strength of the interfacial methoxy groups of γ-GMPS before and after curing the silicone is similar, at 1.0 and 0.8, respectively. After the addition of MVS, the signal strength decreased from 1.2 (before curing) to 0.7 (after curing), which suggests that adhesion strength may be related to interfacial chemical reactions or interfacial diffusion of the methoxy groups due to curing. SFG spectra were also collected from the d4-PET/cured silicone interfaces with silicone mixed with γGPDMS (1.5%) or γ-GPDMS:MVS mixture (3.0%). These spectra are displayed in Figure 7c. No methoxy signals were detected at either interface, which is different from the γ-GPS and γ-GPMS situations. Adhesion Testing. To correlate the SFG results on buried interfaces to polymer adhesion, adhesion strength was measured using shear testing (see the results reported in Figure 8). The forces to break the interfaces of the samples were measured in the experiment to compare the adhesion strengths at different interfaces. All samples were monitored to
exclude erroneous results due to defects (e.g., bubbling). The adhesion strength was calculated by dividing the measured breaking force by the contact area. Four samples were prepared for each adhesive mixture in every adhesion experiment, and the adhesion experiment was repeated several times. The shear testing results (Figure 8) show that pure silicone only has weak adhesion to PET ( γ-GPDMS (only one methoxy headgroup). Similarly, the addition of γ-GPS and MVS improves adhesion greater than the addition of γ-GPMS and MVS, which in turn improves the adhesion more than the addition of γ-GPDMS and MVS into silicone. This clearly shows that the methoxy headgroups play an important role in adhesion. The more methoxy groups a silane has, the stronger the adhesion. Also the mixing of MVS with silane greatly enhanced the adhesion between PET and PDMS silicone. Our adhesion testing results can be correlated to the SFG results quite well. The SFG data indicated that at the PET/ cured silicone interfaces, when γ-GPS (with or without MVS) and γ-GPMS (with or without MVS) were used, interfacial segregation and ordering of methoxy headgroups could be observed, leading to strong adhesion. When γ-GPDMS (with or without MVS) was used, no interfacial segregation and/or ordering of methoxy headgroups could be detected from the PET/cured silicone interfaces, and only weak adhesion was observed. Furthermore, the signal strength decreased substantially after the silicone was cured in the case of the two samples with the strongest adhesion (γ-GPS with MVS and γ-GPMS with MVS). Here, we believe that the signal strength decrease is related to interfacial chemical reactions and/or diffusion processes that could lead to stronger adhesion. For example, there may be a chemical reaction between the hydroxyl end groups and silane methoxy headgroups,59 which can be studied in the future by using samples of PET with different molecular weights. Interfacial diffusion will also be tested, using methods developed previously.43,46,49 The addition of the γ-GPS:MVS mixture to silicone leads to the strongest ordering of methoxy groups at the PET/uncured silicone interface, and the largest decrease in signal after the silicone curing. This suggests that this sample is subject to a large amount of interfacial chemical reaction and/or interfacial diffusion, which is correlated to observations of the strongest adhesion at the PET/silicone interface of the samples studied here.
Figure 8. Adhesion shear test results. Strength = force/contact area. 6057
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
■
■
SUMMARY
AUTHOR INFORMATION
Corresponding Author
This research focused on the effect of silane headgroups, especially, the methoxy headgroups on silane interfacial structures at the PET/silane, PET/silane:MVS mixture, PET/ uncured silicone (with silane or silane:MVS mixture), and PET/cured silicone (with silane or silane:MVS mixture) interfaces. We showed that silanes with different headgroups can behave differently at various interfaces. For γ-GPS, γGPMS, and γ-GPDMS, methoxy headgroups tend to order at the polymer/silane interfaces. MVS greatly increased the methoxy group ordering in γ-GPS at the interface, slightly affected the interfacial methoxy ordering in γ-GPMS, and disordered the interfacial methoxy groups in γ-GPDMS. Uncured silicone affected the methoxy interfacial segregation and/or order at the interfaces. At the d4-PET/uncured silicone interfaces for all the three silanes, the methoxy signal strengths are similar, at 1.3, 1.0, and 1.0, respectively. However, when mixing with MVS, the methoxy order increased slightly for γGPS, remained more or less the same for γ-GPMS, and decreased for γ-GPDMS. At the d4-PET/cured silicone interfaces, the SFG signal detected from the methoxy groups was stronger for the addition of γ-GPS alone to silicone compared to the addition of γ-GPS:MVS mixture to silicone, while it was similar for the addition of γ-GPMS alone and the addition of the γGPMS:MVS mixture. At the d4-PET/cured silicone interface, no methoxy SFG signal was detected when either adding γGPDMS alone or adding the γ-GPDMS:MVS mixture to silicone. A range of methoxy group interfacial segregation and ordering behaviors was observed, which was correlated to measurements of adhesion strengths. Strong adhesion was detected at the PET/cured PDMS interfaces when γ-GPS (with or without MVS) and γ-GPMS (with or without MVS) were used, because of the interfacial segregation and ordering of methoxy headgroups before and after curing. The methoxy SFG signal strength decreased substantially after the silicone was cured when using silicone with either γ-GPS with MVS or γ-GPMS with MVS, and these two samples had the highest adhesive strength of those studied. We therefore believe that stronger adhesion is related to (1) the interfacial segregation and ordering of methoxy groups at PET/uncured and cured silicone interfaces and (2) the methoxy SFG signal strength decrease during the curing process (which could be related to interfacial chemical reactions and/or diffusion). The second hypothesis will be further studied in the future. This study shows that SFG can be used to elucidate the interfacial segregation and ordering of molecules at buried interfaces in situ, which can be correlated to adhesion. This research also demonstrates that SFG is sensitive enough to detect the interfacial behavior of very small amount of silane molecules loaded into the polymer systems. We believe that SFG can be developed into a powerful, nondestructive tool to study the mechanisms of polymer adhesion. In the future, this will lead to the ability to relate interfacial structure to the adhesion strength of polymers.
■
Article
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research is supported by the University of Michigan, Dow Corning Corporation, and the Semiconductor Research Corporation (P10419).
■
REFERENCES
(1) Kinloch, A. J. Adhesion and adhesives: science and technology; Chapman and Hall: London, 1987. (2) Charles, H. Engineered Materials Handbook, Vol. 3, Adhesives and Sealants; ASM International: Materials Park, OH, 1990. (3) Yacobi, B. G.; Martin, S.; Davis, K.; Hudson, A.; Hubert, M. J. Appl. Phys. 2002, 91, 6227−6262. (4) Sun, Y. J.; Guo, S. L.; Walker, G. C.; Kavanagh, C. J.; Swain, G. W. Biofouling 2004, 20, 279−289. (5) Mine, K.; Nishio, M.; Sumimura, S. U.S. Patent 4,033,924, July 5, 1977. (6) Schulz, J. R. U.S. Patent 4,087,585, May 2, 1978. (7) Gray, T. E.; Lutz, M. A. U.S. Patent 5,595,826, Jan 21, 1997. (8) Suzuki, T.; Kasuya, A. J. Adhesion Sci. Technol. 1989, 3, 463−473. (9) Sathyanarayana, M. N.; Yaseen, M. Prog. Org. Coat. 1995, 26, 275−313. (10) Feresenbet, E.; Raghavan, D.; Holmes, G. A. J. Adhes. 2003, 79, 643−665. (11) Shen, Y. Nature 1989, 337, 519−525. (12) Shen, Y. R. The Principles of Nonlinear Optics; Wiley: New York, 1984. (13) Zhuang, X.; Miranda, P.; Kim, D.; Shen, Y. Phys. Rev. B 1999, 59, 12632−12640. (14) Chen, Z.; Shen, Y.; Somorjai, G. A. Annu. Rev. Phys. Chem. 2002, 53, 437−465. (15) Ye, S.; Morita, S.; Li, G. F.; Noda, H.; Tanaka, M.; Uosaki, K.; Osawa, M. Macromolecules 2003, 36, 5694−5703. (16) Baldelli, S. Acc. Chem. Res. 2008, 41, 421−431. (17) Kataoka, S.; Cremer, P. S. J. Am. Chem. Soc. 2006, 128, 5516− 5522. (18) Wang, J.; Chen, X.; Clarke, M. L.; Chen, Z. Proc. Natl. Acad. Soc. 2005, 102, 4978−4983. (19) Chen, C. Y.; Wang, J.; Even, M. A.; Chen, Z. Macromolecules 2002, 35, 8093−8097. (20) Yurdumakan, B.; Nanjundiah, K.; Dhinojwala, A. J. Phys. Chem. C 2007, 111, 960−965. (21) Shi, Q.; Ye, S.; Spanninga, S. A.; Su, Y. L.; Jiang, Z. Y.; Chen, Z. Soft Matter 2009, 5, 3487−3494. (22) Tarbuck, T. L.; Ota, S. T.; Richmond, G. L. J. Am. Chem. Soc. 2006, 128, 14519−14527. (23) Watry, M. R.; Richmond, G. L. J. Am. Chem. Soc. 2000, 122, 875−883. (24) Perry, A.; Neipert, C.; Space, B.; Moore, P. B. Chem. Rev. 2006, 106, 1234−1258. (25) Zhang, D.; Ward, R.; Shen, Y.; Somorjai, G. J. Phys. Chem. B 1997, 101, 9060−9064. (26) Chen, P.; Kung, K. Y.; Shen, Y. R.; Somorjai, G. A. Surf. Sci. 2001, 494, 289−297. (27) Holinga, G. J.; York, R. L.; Onorato, R. M.; Thompson, C. M.; Webb, N. E.; Yoon, A. P.; Somorjai, G. A. J. Am. Chem. Soc. 2011, 133, 6243−6253. (28) Kliewer, C. J.; Somorjai, G. A. Catal. Lett. 2010, 137, 118−122. (29) Aliaga, C.; Tsung, C. K.; Alayoglu, S.; Komvopoulos, K.; Yang, P. D.; Somorjai, G. A. J. Phys. Chem. C 2011, 115, 8104−8109. (30) Ye, H. K.; Gu, Z. Y.; Gracias, D. H. Langmuir 2006, 22, 1863− 1868.
ASSOCIATED CONTENT
S Supporting Information *
SFG spectral fitting parameters. This material is available free of charge via the Internet at http://pubs.acs.org. 6058
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
Langmuir
Article
(31) Ye, H. K.; Abu-Akeel, A.; Huang, J.; Katz, H. E.; Gracias, D. H. J. Am. Chem. Soc. 2006, 128, 6528−6529. (32) Kweskin, S. J.; Komvopoulos, K.; Somorjai, G. A. Langmuir 2005, 21, 3647−3652. (33) Lu, X. L.; Shephard, N.; Han, J. L.; Xue, G.; Chen, Z. Macromolecules 2008, 41, 8770−8777. (34) Lu, X. L.; Li, D. W.; Kristalyn, C. B.; Han, J. L.; Shephard, N.; Rhodes, S.; Xue, G.; Chen, Z. Macromolecules 2009, 42, 9052−9057. (35) Wilson, P. T.; Briggman, K. A.; Wallace, W. E.; Stephenson, J. C.; Richter, L. J. Appl. Phys. Lett. 2002, 80, 3084−3086. (36) Harp, G. P.; Rangwalla, H.; Yeganeh, M. S.; Dhinojwala, A. J. Am. Chem. Soc. 2003, 125, 11283−11290. (37) Li, Q.; Hua, R.; Chou, K. C. J. Phys. Chem. B 2008, 112, 2315− 2318. (38) Chen, Z. Poly. Inter. 2007, 56, 577−587. (39) Chen, Z. Prog. Polym. Sci. 2010, 35, 1376−1402. (40) Wang, J.; Chen, C.; Buck, S. M.; Chen, Z. J. Phys. Chem. B 2001, 105, 12118−12125. (41) Wang, J.; Paszti, Z.; Even, M. A.; Chen, Z. J. Am. Chem. Soc. 2002, 124, 7016−7023. (42) Chen, C. Y.; Even, M. A.; Chen, Z. Macromolecules 2003, 36, 4478−4484. (43) Chen, C.; Wang, J.; Chen, Z. Langmuir 2004, 20, 10186−10193. (44) Chen, C. Y.; Loch, C. L.; Wang, J.; Chen, Z. J. Phys. Chem. B 2003, 107, 10440−10445. (45) Loch, C. L.; Ahn, D.; Chen, C. Y.; Wang, J.; Chen, Z. Langmuir 2004, 20, 5467−5473. (46) Chen, C. Y.; Wang, J.; Loch, C. L.; Ahn, D.; Chen, Z. J. Am. Chem. Soc. 2004, 126, 1174−1179. (47) Loch, C. L.; Ahn, D.; Chen, Z. J. Phys. Chem. B 2006, 110, 914− 918. (48) Wang, J.; Clarke, M. L.; Chen, Z. Anal. Chem. 2004, 76, 2159− 2167. (49) Loch, C. L.; Ahn, D. C.; Vazquez, A. V.; Chen, Z. J. Colloid Interface Sci. 2007, 308, 170−175. (50) Vázquez, A. V.; Shephard, N. E.; Steinecker, C. L.; Ahn, D.; Spanninga, S.; Chen, Z. J. Colloid Interface Sci. 2009, 331, 408−416. (51) Löbau, J.; Wolfrum, K. J. Opt. Soc. Am. B 1997, 14, 2505−2512. (52) Fourkas, J. T.; Walker, R. A.; Can, S. Z.; Gershgoren, E. J. Phys. Chem. C 2007, 111, 8902−8915. (53) Moad, A. J.; Simpson, G. J. J. Phys. Chem. B 2004, 108, 3548− 3562. (54) Hirose, C.; Akamatsu, N.; Domen, K. J. Chem. Phys. 1992, 96, 997. (55) Hirose, C.; Akamatsu, N.; Domen, K. Appl. Spectrosc. 1992, 46, 1051−1072. (56) Rao, Y.; Comstock, M.; Eisenthal, K. B. J. Phys. Chem. B 2006, 110, 1727−1732. (57) Tyrode, E.; Johnson, C. M.; Baldelli, S.; Leygraf, C.; Rutland, M. W. J. Phys. Chem. B 2005, 109, 329−341. (58) de Beer, A. G. F.; Roke, S. Phys. Rev. B 2007, 75, 245438. (59) Xue, G.; Koenig, J. L.; Wheeler, D. D.; Ishida, H. J. Appl. Polym. Sci. 1983, 28, 2633−2646.
6059
dx.doi.org/10.1021/la300004x | Langmuir 2012, 28, 6052−6059
