pubs.acs.org/Langmuir © 2010 American Chemical Society
Direct Measurement of Acid-Base Interaction Energy at Solid Interfaces Anish Kurian,†,§ Shishir Prasad,‡,§ and Ali Dhinojwala*,† †
Department of Polymer Science, The University of Akron, Akron, Ohio 44325, United States, and ‡Universal Technology Corporation, Dayton, Ohio 45432, United States. § These authors contributed equally to this work Received September 7, 2010. Revised Manuscript Received October 20, 2010
We have studied acid-base interactions at solid-liquid and solid-solid interfaces using interface-sensitive sum frequency generation (SFG) spectroscopy. The shift of the sapphire hydroxyl peak in contact with several polar and nonpolar liquids and polymers was used to determine the interaction energy. The trend in the interaction energies cannot be explained by measuring only water contact angles. Molecular rearrangements at the sapphire interface, to maximize the interaction of the acid-base groups, play a dominant role, and these effects are not accounted for in the current theoretical models. These results provide important insights into understanding adhesion, friction, and wetting on solid interfaces.
Introduction Molecular interactions at interfaces govern the wetting, adhesion, friction, chemical reactions, and many other material and biological phenomena at interfaces.1,2 The progress in this field is hindered by the lack of direct experimental tools to measure molecular interactions at buried solid-solid interfaces. Currently, techniques such as the heat of mixing in dilute solutions are used to determine the strength of the donor-acceptor interactions,3 bulk infrared spectroscopic techniques that relate the shift of carbonyl or hydroxyl vibrational peaks (Δν) to the strength of the interactions,4-6 and vacuum-based techniques such as X-ray photoelectron spectroscopy (XPS)7 . Highresolution electron energy-loss spectroscopy (HREELS)8 have also been employed to study polar interactions. Although these techniques have provided valuable insight into polar (donoracceptor, hydrogen-bonding, and acid-base) interactions, they are not capable of capturing the complexity of the interactions at buried solid-solid contact interfaces. The roles of surface reconstruction, competition between multiple interactions, and steric constraints are not considered in macroscopic theories that are expected to play an important role in dictating the magnitude of the interaction strength. Here, we report the use of surface-sensitive sum frequency generation spectroscopy (SFG) to probe directly the shift in the vibrational peaks at buried interfaces. SFG is a second-order nonlinear optical effect and is forbidden in the centrosymmetric bulk and thus provides exclusive sensitivity to probing the molecular orientation and composition of the noncentrosymmetric interface.9,10 *To whom correspondence should be addressed. E-mail:
[email protected].
(1) Fowkes, F. M. J. Phys. Chem. 1953, 57, 98–103. (2) Chaudhury, M. K. Mater. Sci. Eng. R 1996, R16, 97–159. (3) Drago, R. S.; Wayland, B. B. J. Am. Chem. Soc. 1965, 87, 3571–3576. (4) Badger, R. M.; Bauer, S. H. J. Chem. Phys. 1937, 5, 839–851. (5) Drago, R. S.; O’Bryan, N.; Vogel, G. C. J. Am. Chem. Soc. 1970, 92, 3924– 3929. (6) Oldak, R. K.; Pearson, R. A. J. Adhes. Sci. Technol. 2007, 21, 775–793. (7) Watts, J. F.; Chehimi, M. M. Int. J. Adhes. Adhes. 1995, 15, 91–94. (8) Layman, K. A.; Ivey, M. M.; Hemminger, J. C. J. Phys. Chem. B 2003, 107, 8538–8546. (9) Shen, Y. R. The Principles of Nonlinear Optics; John Wiley: New York, 1984. (10) Du, Q.; Freysz, E.; Shen, Y. R. Phys. Rev. Lett. 1993, 70, 2313–2316. (11) Yeganeh, M. S.; Dougal, S. M.; Pink, H. S. Phys. Rev. Lett. 1999, 83, 1179– 1182.
17804 DOI: 10.1021/la103591f
SFG in total internal reflection geometry has been used to study the solid-liquid11 and solid-solid interfaces.12 By monitoring the shift in the vibrational peaks in conjunction with the formalism developed by Drago and Wayland3 and Badger and Bauer,4 we can determine the acid-base coefficients for the solid surface. The Drago-Wayland coefficients can then be used to determine the interactions of various liquids and polymers in contact with the solid substrate. In this letter, we have demonstrated the application of this spectroscopic technique to determine the interaction energies of various polar and nonpolar polymers in contact with the sapphire substrate. Because this technique does not require dilute solutions or the use of common solvents, it is widely applicable for determining interaction energies for many other solid-solid interfaces that are important in the areas of adhesion, tribology, and wetting.
Experimental Section High-purity organic liquids were either used as received (acetone, purchased from Prometheus Photonics, high-purity chromatographic grade; pyridine, from Sigma-Aldrich, anhydrous, 99.8% purity; pentadecane, from Fluka, g99.8% (GC) purity) or distilled in-house (triethylamine, from Sigma-Aldrich, g 99.5% purity). Poly(methyl methacrylate) (PMMA, from Scientific Polymer Products, Inc., molecular weight (MW) = 467 kg/mol, polydispersity index (PDI) = 1.06), polyisobutylene (PIB, from Scientific Polymer Products, MW=85 kg/mol), polystyrene (PS, from Polymer Source, Inc., MW = 118 kg/mol, PDI = 1.05), polyacrylonitrile (PAN, from Aldrich, MW=23 kg/mol, PDI=3.81), and poly(vinyl octadecylcarbamate) (PVNODC, a gift from 3M, MW = 70 kg/mol, PDI = 3.0) were obtained from commercial sources. The chemical structures of all materials (small molecules and polymers) are provided in the Supporting Information. Sapphire prisms, used as substrates, were sonicated twice in toluene, followed by air plasma cleaning for 5 min. Thin polymer films on the faces of sapphire prisms were prepared by spin coating a 2 wt % polymer solution in their respective good solvents (toluene for PMMA, PS, and PVNODC; dimethyl sulfoxide for PAN; and chloroform for PIB) at 2000 rpm for 1 min. The films were subsequently annealed overnight in a vacuum oven maintained at 29 in. of Hg. The melting/glass-transition (12) Gautam, K. S.; Schwab, A. D.; Dhinojwala, A.; Zhang, D.; Dougal, S. M.; Yeganeh, M. S. Phys. Rev. Lett. 2000, 85, 3854–3857.
Published on Web 11/08/2010
Langmuir 2010, 26(23), 17804–17807
Kurian et al.
Letter
Figure 1. (A) SFG interfacial spectra in the hydrocarbon region (A) and hydroxyl region (B) for sapphire in contact with air ()), pentadecane (3), acetone (O), pyridine (0), and triethylamine (Δ). All of the spectra were collected at room temperature using p-polarized input and SFG beams. The solid lines in A and B are the fits using the Lorentzian equation. temperature (the higher of the two temperatures for the polymers listed) and the annealing temperature of all polymers are provided in the Supporting Information. We have used the total internal reflection geometry12 to probe sapphire-air, -liquid, or -solid interfaces (diagram provided in the Supporting Information). Two input laser beams were incident on the prism face at 42° with respect to the surface normal to probe the sapphire-air interface. To probe the sapphireliquid or polymer interfaces, the incident angles depend on the refractive index of the material. (Values are provided in the Supporting Information.) The experiments were performed with incident visible beam of 798.5 nm wavelength and by scanning the incident IR wavenumber from 2750 to 3850 cm-1. The spectra were obtained in PPP (p-polarized SFG, p-polarized visible, and p-polarized IR) and SSP (s-polarized SFG, s-polarized visible, and p-polarized IR) polarization combinations at room temperature. The SFG spectra equation,12 P wereiφqfitted using a Lorentzian 2 I(SFG) µ |χeff,NRþ q(Aqe /(ωIR - ωq-iΓq))| , where Aq, Γq, ωq are the strength, damping constant, and angular frequency of a single resonance vibration, respectively, and φq is the phase.
Results and Discussion Drago and Wayland have proposed an empirical equation to determine the enthalpy of mixing in solutions (ΔH = EAEB þ CACB).3 EA and EB are the susceptibilities to undergoing acid and base electrostatic interactions, respectively. CA and CB represent their susceptibilities to forming covalent bonds. Calorimetry has been used to determine the enthalpy of mixing of various acidbase pairs, and this data has been used to determine the DragoWayland coefficients. The enthalpy of mixing can also be determined using infrared spectroscopy and an empirical equation proposed by Badger-Bauer4 (ΔH=mΔμ þ C). The values of two constants, m and C, are obtained by calibrating the peak shifts using standard liquids with the enthalpy of mixing determined using calorimetry. To determine the Drago-Wayland coefficients for the sapphire substrate, EA and CA, we have studied three liquids with known values of the Drago-Wayland coefficients, EB and CB. The SFG spectra in PPP polarization for these various liquids in contact with the sapphire substrate are shown in Figure 1. The experimental conditions of the SFG technique are provided as Supporting Information. The spectral region below 3100 cm-1 corresponds to aliphatic and aromatic C-H vibrations, and the peaks above 3100 cm-1 correspond to the O-H stretching vibrations. Here, we will discuss only the hydroxyl peak associated Langmuir 2010, 26(23), 17804–17807
Table 1. Peak Position (ωq) and Peak Width (Γq) of the Sapphire Surface -OH Peak in Contact with Air and Organic Liquidsa materials
ωq (cm-1)
Γq (cm-1)
ΔH (kcal/mol)
triethylamine 3565 102 1.72 ( 0.08 pyridine 3575 124 1.61 ( 0.09 acetone 3610 118 1.23 ( 0.07 pentadecane 3690 106 0.36 ( 0.05 air 3720 18 a Acetone: EB = 0.987, CB = 2.33. Pyridine: EB = 1.17, CB = 6.4. Triethylamine: EB = 0.991, CB = 11.09.15 EB and CB have units of (kcal/mol)0.5. The calculated enthalpy of interaction (ΔH) of liquids with the sapphire surface is also reported.
with the Al-OH molecules at the sapphire surface.13 In the sapphire-air spectrum, the surface hydroxyl peak is at 3720 cm-1 and this peak is red-shifted in contact with polar and nonpolar liquids. The peak position and the peak width (fwhm) of the surface hydroxyl peak are summarized in Table 1. Even for a nonpolar pentadecane we observe a red shift of 30 cm-1 that is due to dispersion interactions and has been reported previously.14 Combining the Drago-Wayland (ΔH = EAEB þ CACB) and Badger-Bauer (ΔH = mΔν þ C) equations leads to EAEB þ CACB=ΔH=mΔν þ C, where m and C are constants that depend on the chemical group. The values of m reported in the literature range from 0.0105 to 0.0115 kcal/mol cm and depend on the chemical group linked with the hydroxyl groups.5,16 We have used an average value of 0.0109 kcal/mol cm and Δν values from Table 1 to determine the Drago-Wayland coefficients (EA = 1.05 ( 0.08 (kcal/mol)0.5 and CA =0.06 ( 0.01 (kcal/mol)0.5) and C (= 0.03 ( 0.01 kcal/mol) for the sapphire substrate. We can also rearrange the Drago-Wayland and Badger-Bauer equations as follows: EB m Δν C CA ð1Þ ¼ þ E A CB CB E A E A CB Figure 2 shows a plot of EB/CB as a function of Δν/CB. The excellent linear fit indicates the validity of eq 1, proving that EA (13) Nanjundiah, K.; Hsu, P. Y.; Dhinojwala, A. J. Chem. Phys. 2009, 130, 024702/1–024702/7. (14) Shen, Y. R.; Ostroverkhov, V. Chem. Rev. 2006, 106, 1140–1154. (15) Arnett, E. M.; Mitchel, E. J.; Murty, T. S. S. R. J. Am. Chem. Soc. 1974, 96, 3875–3891. (16) Purcell, K. F.; Stikeleather, J. A.; Brunk, S. D. J. Am. Chem. Soc. 1969, 91, 4019–4027.
DOI: 10.1021/la103591f
17805
Letter
and CA are constants in eq 1, and also suggests that the magnitude of C is small. The low CA/EA ratio (0.06) indicates that the hydroxyl group on the sapphire substrate is a hard acid. We have also calculated the interaction energy between the sapphire substrate and these liquids (Table 1). On the basis of the knowledge of the Drago-Wayland coefficients for sapphire, we can now estimate the interaction energies with various molecules whose Drago-Wayland coefficients are tabulated in the literature.15 Next, we illustrate the application of this direct spectroscopic technique to determine the interaction energies of various polar and nonpolar polymers in contact with the sapphire substrate. The SFG spectra in PPP polarization for these polymers in contact with the sapphire substrate are shown in Figure 3. The positions of the hydroxyl peak are tabulated in Table 2. Figure 4 shows the SFG spectra for a clean sapphire surface along with selected polymers in contact with the sapphire substrate measured using SSP polarization. The positions of the hydroxyl peaks in SSP polarization (in Figure 4B) are similar to those obtained using PPP polarization (Table 2). Using the Badger-Bauer equation, we have also determined the interaction energies at
Kurian et al.
the sapphire interface (Table 2). Because we do not have EB and CB values for some of the polymers that we have used in this study, it is difficult to make a quantitative comparison with the interaction energies determined using the Drago-Wayland equation. Fowkes has estimated EB and CB for PMMA to be 1.1 and 2.7 (kcal/mol)0.5,17 respectively. Using these numbers, we obtain ΔH=1.32 ( 0.09 kcal/mol. This estimate compares very well with the value determined using SFG data and the Badger-Bauer equation (Table 2), particularly when considering the assumptions involved in the technique used by Fowkes to estimate the Drago-Wayland coefficients. We can also qualitatively compare the trends observed for other polymers on the basis of the Drago-Wayland coefficients reported for small molecular compounds.3 The interaction energy expected for acetonitrile (an appropriate model for PAN) based on the Drago-Wayland coefficients (CB =1.77(kcal/mol)0.5 and EB =0.533(kcal/mol)0.5) is lower than that for PMMA. In addition, the interaction energy for toluene (an appropriate model for PS) based on the DragoWayland coefficients3 (CB = 1.91(kcal/mol)0.5 and EB = 0.087(kcal/mol)0.5) is lower than that for acetonitrile. The methyl groups of PIB in contact with the sapphire substrate can interact only with the surface OH using dispersion forces, and they show the smallest shift of the surface OH peak. The interaction energies determined from the shift do not correlate with water-contact angles, which clearly demonstrates Table 2. Peak Position (ωq) and Peak Width (Γq) of the Sapphire Surface -OH Peak in Contact with Various Polymers and Aira materials
Figure 2. EB/CB as a function of Δν/CB. The solid line is a linear fit with a slope of 0.010 and an intercept of -0.048.
ωq (cm-1)
Γq (cm-1)
ΔH (kcal/mol)
PMMA 3580 96 1.56 ( 0.03 PVNODC 3590 106 1.45 ( 0.05 PAN 3605 123 1.28 ( 0.07 PS 3645 61 0.85 ( 0.04 PIB 3690 65 0.36 ( 0.03 air 3720 18 a We have also reported the calculated enthalpy of interaction (ΔH) for polymers with the sapphire surface.
Figure 3. (A) SFG spectra in the hydrocarbon region (A) and hydroxyl region (B) for sapphire in contact with air (side-facing 4), PIB (0), PS (O), PAN (Δ), PVNODC (3), and PMMA ()). All of the spectra were collected at room temperature using p-polarized input and SFG beams. The solid lines in A and B are the fits using the Lorentzian equation. 17806 DOI: 10.1021/la103591f
Langmuir 2010, 26(23), 17804–17807
Kurian et al.
Letter
Figure 4. (A) SFG interfacial spectra in SSP polarization in the hydrocarbon region (A) and hydroxyl region (B) for sapphire in contact with air (side-facing 4), PS (O), and PMMA ()). The solid lines in A and B are the fits using the Lorentzian equation. The obtained peak positions (peak widths) from fits for hydroxyl peak are 3720 cm-1 (17 cm-1), 3645 cm-1 (66 cm-1), and 3575 cm-1 (101 cm-1) for air, PS, and PMMA, respectively.
the importance of molecular rearrangement in contact with the sapphire substrate. For example, the octadecyl side-chain polymer (PVNODC) exhibits a very high water contact angle (110°, similar to that for PIB). However, the interaction energies of PVNODC with the sapphire substrate are closer to the ester interaction energy of poly(methyl methacrylate) (PMMA). These differences are due to the molecular rearrangement of the octadecyl side chains in contact with the sapphire substrate to increase the interactions of the polar groups (carbonyl or N-H) with the surface OH. This rearrangement is also observed in the differences in the orientation of the methyl and methylene groups.18 Another interesting example is that of polyacrylonitrile (PAN) in contact with sapphire. Even though PAN has the lowest water contact angle (60°) among the five polymers that we have studied, the interaction energies are not as high as for PMMA (water contact angle of 73°). This is due to the strong affinity between the CNCN groups (within the polymer) compared to the CN-surface OH interactions.19 We have also observed a shift in the CN peak due to the interactions of the CN group with the surface OH groups. The PIB is the most hydrophobic polymer that we have studied, and the small red shift of 30 cm-1 is similar to that of sapphire in contact with nonpolar pentadecane. Polystyrene (PS) is in between the more polar PMMA and PIB because of the interaction of the phenyl groups with the sapphire substrate.12 Finally, we reflect on the width of the hydroxyl peaks in contact with the sapphire substrate. In the case of acid-base interactions in solution, the peak shift is accompanied by peak broadening.5,6 We have also observed similar peak broadening with peak shifts (17) Fowkes, F. M. Tribol. Ser. 1982, 7, 119–137. (18) Gautam, K. S.; Dhinojwala, A. Phys. Rev. Lett. 2002, 88, 145501/1–145501/4. (19) Lachat, V.; Varshney, V.; Dhinojwala, A.; Yeganeh, M. S. Macromolecules 2009, 42, 7103–7107.
Langmuir 2010, 26(23), 17804–17807
at solid-solid interfaces. However, for the sapphire substrate, the broad peak could be due to a number of effects, including the charge of the Al group, surface heterogeneity, the distribution of bonded and nonbonded groups, and the orientation of the acidbase pairs. Molecular dynamics and density functional calculations are needed to elucidate the reasons for this broadening and its effect on the thermodynamics of these interactions at the buried interfaces, and these will be the focus of our future work. In summary, we have for the first time quantified the acid-base interactions at hidden solid-liquid and solid-solid interfaces using the interface sensitivity of SFG spectroscopy. We have determined the interaction energies for a series of polymers in contact with sapphire using solvent-free, nondestructive conditions. Molecular rearrangements played a dominant role in explaining the trend in the interaction energies. Direct measurements of the interaction energy will help in answering many outstanding questions in the areas of nucleation, the glass transition of nanometer thin films, corrosion, thermal transport, wetting, and adhesion hysteresis. Acknowledgment. We acknowledge financial support from the National Science Foundation (DMR-0512156 and DMR0526797), and S.P. acknowledges financial support from the Air Force Research Laboratory under the Collaborative Research and Development III Contract (FA 8650-07-D-5800, 0056). We also thank Manoj Chaudhury and Alan Gent for their helpful comments. Supporting Information Available: Molecular structure of materials, diagram of internal reflection geometry, and values of incident angles used in this study. This material is available free of charge via the Internet at http://pubs.acs.org.
DOI: 10.1021/la103591f
17807
