IR Spectroscopic Studies of Adsorption of Dithiol-Containing Ligands

Nov 17, 2007 - ... Biology, University of Otago, PO Box 56, Dunedin, New Zealand ... Citation data is made available by participants in Crossref's Cit...
0 downloads 0 Views 210KB Size
Download PDF
Langmuir 2007, 23, 12923-12931

12923

IR Spectroscopic Studies of Adsorption of Dithiol-Containing Ligands on CdS Nanocrystal Films in Aqueous Solutions Aidan G. Young,† David P. Green,‡ and A. James McQuillan*,† Departments of Chemistry and of Anatomy and Structural Biology, UniVersity of Otago, PO Box 56, Dunedin, New Zealand ReceiVed July 18, 2007. In Final Form: October 3, 2007 The adsorption of the ligands R-lipoic acid, dihydrolipoic acid, and dithiothreitol to films of deposited CdS nanoparticles was studied in situ by ATR-IR spectroscopy. For R-lipoic acid and dihydrolipoic acid, the spectra of the adsorbed species closely resemble those of the respective solution species. However, for dithiothreitol, the spectrum of the adsorbed species is significantly different from that of the solution species and is attributed to an interruption of intermolecular hydrogen bonding upon adsorption to the CdS. The S-H stretching absorption of the dihydrolipoic acid solution species at pH ) 8.6 is observed at 2542 cm-1. The corresponding absorptions for dithiothreitol occur at 2578 and 2528 cm-1 and are attributed to monomers and dimers. Adsorption of dihydrolipoic acid and dithiothreitol is found to occur via both thiol functional groups and an additional interaction between the carboxylate and the CdS surface. The adsorption of R-lipoic acid to CdS in the presence of light proceeds with photo-oxidation of the CdS surface and reductive cleavage of the disulfide bond of R-lipoic acid to produce some adsorbed dihydrolipoic acid and thiosulfate. The adsorption of R-lipoic acid to CdS in the absence of visible light shows no photo-oxidation and suggests that adsorption occurs via retention of the disulfide bond. The adsorption isotherm data for dihydrolipoic acid and dithiothreitol gave good fits to the Langmuir isotherm, with adsorption constants higher than those for monothiol-containing ligands on CdS. The Langmuir adsorption constant for n-octanoic acid on CdS indicates that the additional interaction between the carboxylate group of dihydrolipoic acid and the CdS is weak in comparison with the dithiol interaction with CdS.

Introduction Colloidal semiconductor nanocrystals that have size-dependent optical properties, often called quantum dots (QDs), have received much recent interest due to their potential in a variety of applications, including biological imaging,1-5 lasers,6,7 and lightemitting devices.8-11 Typical QD materials are semiconductors that readily oxidize in ambient environments. To prevent degradation and to solubilize the nanocrystal in different environments, QDs are often capped with bifunctional ligands, and the chemical nature of the QD surface exposed to solution can be altered by ligand exchange.12-17 However, QDs have a * Corresponding author. Fax: +64 3 479 7906. Tel: +64 3 479 7928. E-mail: [email protected]. † Department of Chemistry. ‡ Department of Anatomy and Structural Biology. (1) Lidke, D. S.; Arndt-Jovin, D. J. Physiology 2004, 19, 322-325. (2) Jaiswal, J. K.; Mattoussi, H.; Mauro, J. M.; Simon, S. M. Nat. Biotechnol. 2003, 21, 47-51. (3) Kim, S.; Lim, Y. T.; Soltesz, E. G.; De Grand, A. M.; Lee, J.; Nakayama, A.; Parker, J. A.; Mihaljevic, T.; Laurence, R. G.; Dor, D. M.; Cohn, L. H.; Bawendi, M. G.; Frangioni, J. V. Nat. Biotechnol. 2004, 22, 93-97. (4) Lidke, D. S.; Nagy, P.; Heintzmann, R.; Arndt-Jovin, D. J.; Post, J. N.; Grecco, H. E.; Jares-Erijman, E. A.; Jovin, T. M. Nat. Biotechnol. 2004, 22, 198-203. (5) Mattoussi, H.; Medintz, I. L.; Clapp, A. R.; Goldman, E. R.; Jaiswal, J. K.; Simon, S. M.; Mauro, J. M. Jala 2004, 9, 28-32. (6) Artemyev, M. V.; Woggon, U.; Wannemacher, R.; Jaschinski, H.; Langbein, W. Nano Lett. 2001, 1, 309-314. (7) Klimov, V. I.; Mikhailovsky, A. A.; Xu, S.; Malko, A.; Hollingsworth, J. A.; Leatherdale, C. A.; Eisler, H. J.; Bawendiz, M. G. Science 2000, 290, 314317. (8) Coe, S.; Woo, W.-K.; Bawendi, M.; Bulovic, V. Nature 2002, 420, 800803. (9) Tesster, N.; Medvedev, V.; Kazes, M.; Kan, S.; Banin, U. Science 2002, 295, 1506-1508. (10) Colvin, V. L.; Schlamp, M. C.; Allvisatos, A. P. Nature 1994, 370, 3547. (11) Zhu, L.; Zhu, M.-Q.; Hurst, J. K.; Li, A. D. Q. J. Am. Chem. Soc. 2005, 127, 8968-8970. (12) Querner, C.; Reiss, P.; Bleuse, J.; Pron, A. J. Am. Chem. Soc. 2004, 126, 11574-11582.

propensity to degrade, with the weakest point usually being the interaction between the capping ligand functional group and the inorganic core surface atoms.18,19 Potential applications like biological imaging, where fluorescent nanocrystals are tracked in vivo through aqueous environments, require high stability to avoid loss of the fluorophore. The surface chemistry of the nanocrystals has a crucial role in these applications. While common methods of surface analysis, such as contact angle and atomic force measurements, can provide inferences about the nature of surfaces, these methods generally need to be supplemented by spectroscopic methods that probe the molecular nature of interfacial structure and composition. However, addressing the molecular nature of chemical reactions at solid-solution interfaces remains a challenging area of research, as there are few applicable spectroscopic techniques. Infrared (IR) spectroscopy has long been applied to surface chemical reactions but has only recently been applied to particulate solid-solution interfaces.20-22 Use of deposited thin films of nanocrystals on internal reflection prism materials facilitates the collection of IR spectra of species adsorbed on the nanocrystals in an aqueous environment.23 The in situ IR spectra from such particle films reveal the molecular nature of surface chemical (13) Bullen, C.; Mulvaney, P. Langmuir 2006, 22, 3007-3013. (14) Shavel, A.; Gaponik, N.; Eychmueller, A. J. Phys. Chem. B 2004, 108, 5905-5908. (15) Kalyuzhny, G.; Murray, R. W. J. Phys. Chem. B 2005, 109, 7012-7021. (16) Bryant, G. W.; Jaskolski, W. J. Phys. Chem. B 2005, 109, 19650-19656. (17) Tsuruoka, T.; Akamatsu, K.; Nawafune, H. Langmuir 2004, 20, 1116911174. (18) Aldana, J.; Lavelle, N.; Wang, Y.; Peng, X. J. Am. Chem. Soc. 2005, 127, 2496-504. (19) Jeong, S.; Achermann, M.; Nanda, J.; Ivanov, S.; Klimov, V. I.; Hollingsworth, J. A. J. Am. Chem. Soc. 2005, 127, 10126-10127. (20) Lefevre, G. AdV. Colloid Interface Sci. 2004, 107, 109-123. (21) Gisler, A.; Burgi, T.; Baiker, A. J. Catal. 2004, 222, 461-469. (22) Chiem, L. T.; Huynh, L.; Ralston, J.; Beattie, D. A. J. Colloid Interface Sci. 2006, 297, 54-61. (23) McQuillan, A. J. AdV. Mater. 2001, 13, 1034-1038.

10.1021/la702165u CCC: $37.00 © 2007 American Chemical Society Published on Web 11/17/2007

12924 Langmuir, Vol. 23, No. 26, 2007

Young et al.

Figure 1. The dithiol-containing ligands (a) R-lipoic acid; (b) DHLA, with NMR numbering assignments; and (c) DTT.

interactions and reactions. While there has been much attention paid to the luminescence properties of nanocrystals and quantum dots in solution, there have been far fewer in situ IR spectroscopic studies addressing the surface chemistry of such particles. Awatani et al. used in situ IR spectroscopy to study the adsorption of lactic acid to CdS photocatalyst particles.24 Awatani and McQuillan25 identified thiosulfate adsorbed at the surface of CdS particles and characterized the thiosulfate adsorption behavior. More recently, Young et al.26 reported an IR study of the adsorption of some commonly used QD monothiol ligands, mercaptoacetic acid (MAA) and mercaptopropionic acid (MPA), on CdS particles. Mercaptoethanol (ME), which is chemically similar, although not often used for QD capping, was also included in the study. These three ligands, as expected, were found to adsorb to CdS via deprotonation of the thiol functional group. The above studies have led to a better understanding of factors influencing CdS capping reactions and adsorbed ligand stability. In the present work, we have used attenuated total reflection infrared (ATR-IR) spectroscopic techniques23 to probe the surface chemistry in aqueous solutions of CdS nanocrystal deposited films, focusing on the adsorption of dithiol-containing ligands. The capping of QDs with monothiol-containing ligands has been questioned due to decomposition of the capped particle in solution phase.18,27 This is generally considered to be due to cleavage of the capping ligand from the QD surface.18 Various research groups have employed dithiol-containing ligands such as dihydrolipoic acid (DHLA) in order to strengthen ligand-surface bonding and improve QD stability in solution.28 The ligands R-lipoic acid, DHLA, and DTT are shown in Figure 1. DHLA is typically prepared from the cyclic disulfide molecule R-lipoic acid, and the adsorption of R-lipoic acid to CdS is included in the current study to give a comparison with the adsorption of DHLA. Dithiothreitol (DTT) has also been suggested as a possible nanoparticle capping ligand.29 It is more commonly used as reducing agent but its adsorption to CdS provides a comparison with the adsorption of DHLA in this study. The similarity of the chelating/bridging thiol functional groups for DHLA and DTT suggests that the coordination ability of the two ligands to deformable coordination sphere metal ions may be similar. Materials and Methods Materials. Cadmium nitrate tetrahydrate (Scharlau, 99+%), hydrated sodium sulfide (Riedel-De Hae¨n, 60-62% Na2S), (()R-lipoic acid (Sigma-Aldrich, 99+ %), dithiothreitol (Acros, 99%), (24) Awatani, T.; Dobson, K. D.; McQuillan, A. J.; Ohtani, B.; Uosaki, K. Chem. Lett. 1998, 849-850. (25) Awatani, T.; McQuillan, A. J. J. Phys. Chem. B 1998, 102, 4110-4113. (26) Young, A. G.; Green, D. P.; McQuillan, A. J. Langmuir 2006, 22, 1110611112. (27) Schroedter, A.; Weller, H. Angew. Chem. Int. Ed. 2002, 41, 3218-3221. (28) Mattoussi, H.; Mauro, J. M.; Goldman, E. R.; Anderson, G. P.; Sundar, V. C.; Mikulec, F. V.; Bawendi, M. G. J. Am. Chem. Soc. 2000, 122, 1214212150. (29) Pathak, S.; Choi, S.-K.; Arnheim, N.; Thompson, M. E. J. Am. Chem. Soc. 2001, 123, 4103-4104.

Figure 2. 1H NMR spectrum of DHLA in D2O (containing 1 × 10-1 mol L-1 NaOH). and n-octanoic acid (Sigma-Aldrich, 99+ %) were all used as received. All water used in experiments was deionized (Millipore, Milli-Q RG, resistivity 18 MΩ cm). Preparation of DHLA. DHLA was prepared via a sodium borohydride reduction of R-lipoic acid as previously described30 with the extraction solvent changed to dichloromethane from the reported benzene. 1H NMR (300 MHz, CDCl3) δ (TMS): 2.92 (1H, m, H6), 2.69 (2H, m, H9), 2.37 (2H, t, H2), 1.89 (1H, m, 1/2H8), 1.80-1.40 (7H, m, H3, H4, H5, 1/2H8), 1.35 (1H, t, H10), 1.30 (1H, d, H7) ppm. 1H NMR (500 MHz, D2O, pD 13) δ (DSS) 3.65 (1H, m, H6), 3.17 (1H, m, 1/2H9), 3.13 (1H, m, 1/2H9), 2.43 (1H, m, 1/2H8), 2.11 (2H, t, H2), 1.93 (1H, m, 1/2H8), 1.69 (1H, m, 1/2H5), 1.58 (1H, m, 1/2H5), 1.51 (2H, m, H4), 1.35 (2H, m, H3) ppm. 13C NMR (500 MHz, D2O, pH 13) δ (DSS): 186.4, quaternary (C1), 59.2 (C6), 42.9 (C8), 40.7 (C9), 40.0 (C2), 36.5 (C5), 31.1 (C3), 28.2 (C4) ppm. The 1H NMR spectrum of DHLA in CDCl3 showed chemical shifts similar to those reported for similar molecules.31 DHLA was converted to the sodium salt to provide aqueous solution solubility. The NMR spectra of DHLA were recorded in D2O containing 1 × 10-1 mol L-1 NaOH, and the spectra show some large chemical shift changes compared with those in CDCl3. The chiral carbon at position six produced a complex 1H NMR spectrum (shown in Figure 2) with diastereomeric resonances observed for H5, H8, H9, and to a very small extent for H4. Preparation CdS Nanocrystals. Aqueous suspensions of uncapped CdS nanocrystals were prepared using a previously reported method.26,32,33 The UV-vis absorption spectrum of an aqueous nanocrystal suspension showed a CdS band edge at 478 nm (2.59 eV), indicating an average particle size of 4.5 nm.34 An X-ray diffraction pattern from the reported preparation indicated a cubic morphology.35 Electrophoretic mobility measurements of hexagonal phase CdS particles reported previously36,37 indicate that isoelectric point (iep) values occur in the range pH 1-2. Electrophoretic mobility measurements conducted on the CdS material employed in the current (30) Gunsalus, I. C.; Barton, L. S.; Gruber, W. J. Am. Chem. Soc. 1956, 78, 1763-6. (31) Uyeda, H. T.; Medintz, I. L.; Jaiswal, J. K.; Simon, S. M.; Mattoussi, H. J. Am. Chem. Soc. 2005, 127, 3870-3878. (32) Ramsden, J. J.; Graetzel, M. J. Chem. Soc. Faraday Trans. 1 1984, 80, 919-33. (33) Schindler, W.; Kisch, H. J. Photochem. Photobiol. A 1997, 103, 257264. (34) Brus, L. E. J. Chem. Phys. 1983, 79, 5566-71. (35) Olshavsky, M. A.; Allcock, H. R. Chem. Mater. 1997, 9, 1367-1376. (36) Nicolau, Y. F.; Menard, J. C. J. Colloid Interface Sci. 1992, 148, 551-70. (37) Guindo, M. C.; Zurita, L.; Duran, J. D. G.; Delgado, A. V. Mater. Chem. Phys. 1996, 44, 51-8.

IR of Adsorption of Dithiol-Containing Ligands

Langmuir, Vol. 23, No. 26, 2007 12925

work gave a comparable iep value. Above this pH the surface charge was found to be negative, and no obvious charge reversal points were apparent. Preparation of DHLA-Capped CdS Nanocrystals. DHLAcapped CdS nanocrystal suspensions were prepared by using a modification of a method reported for the preparation of MAAcapped CdS nanocrystals.38 A turbid blue solution containing 1 × 10-3 mol L-1 cadmium nitrate and 3.2 × 10-3 mol L-1 DHLA (50 mL) was prepared, with the pH adjusted to 8.5 with HCl. A solution of 1 × 10-3 mol L-1 sodium sulfide (50 mL) was added, and the suspension was stirred for 1 h and then stored in the absence of light. The UV-vis absorption spectrum of the suspension had a band edge at 390 nm (3.18 eV), indicating an average particle size of 3.5 nm.34 The emission spectrum (from excitation at 395 nm) consisted of a band centered at 590 nm and had a similar profile to emission spectra reported for MAA-capped CdS nanocrystals.38-41 An aliquot of the prepared suspension was adjusted to pH ) 2.0 and the nanocrystals removed using centrifugation. The nanocrystals were then resuspended in water. ATR-IR Method. A Harrick FastIR accessory containing a 45° single internal reflection ZnSe prism, a DuraSampleIR triplereflection 3-mm diameter diamond-faced ZnSe prism (ASI SensIR Technologies), and a Horizon accessory containing a 13-reflection 50 × 10 × 2 mm 45° ZnSe prism (Harrick Scientific) were used to collect spectral data. The observed absorbances are approximately proportional to the number of internal reflections within the accessory. Multiple reflection accessories were used when greater sensitivity was required. The diamond surface of the triple-reflection prism isolates the redox-active CdS and any potentially detrimental solution effects (e.g., low pH) from the ZnSe surface. The opacity of the diamond/ZnSe prism in the 2600-1900 cm-1 range makes it unsuitable to study S-H stretching absorptions. Prior to preparation of the CdS films for ATR-IR analysis, the prism surfaces were cleaned by polishing with 0.015 µm γ-alumina (BDH, polishing grade) on a wet polishing microcloth (Buehler) and then rinsed with water. For the single-reflection prism, the nanocrystal film was formed by depositing 300 µL of 0.5 × 10-3 mol L-1 CdS suspension onto the prism. The removal of most of the water using a water pump vacuum for ca. 30 min produced a ∼2 cm2 film. For the triplereflection prism, the film was formed in a similar manner by depositing 8 µL of 1 × 10-3 mol L-1 CdS suspension to produce a ∼0.25 cm2 film. All films were deposited on the prisms immediately following the CdS suspension preparation and were initially washed for at least 15 min with 1 × 10-2 mol L-1 NaOH to remove any surface S2O32- ligands.25 SEM images of deposited films prepared in this way have been reported previously and show these films to have a thickness of ∼500 nm.26 Each prism was interfaced via a rubber O-ring to a flow cell. The single- and triple-reflection prisms incorporated Teflon and glass flow cells, respectively, each of which was manufactured in this department.42 The solutions were delivered to the flow cells using a Masterflex peristaltic pump and Masterflex Tygon tubing at a constant flow rate of ca. 1 mL min-1. The IR analysis of the CdS films under solution flow was conducted in the absence of laboratory lighting. A Digilab FTS 4000 infrared spectrometer equipped with a KBr beamsplitter, Peltier cooled DTGS detector, and WinIR Pro version 3.4 software was used to collect and analyze spectra. The optical bench was purged with dried air. ATR-IR spectra were obtained from 10-64 scans at 4 cm-1 resolution (unless otherwise stated) and were not corrected for dependence of absorbance on frequency. Spectra are shown as recorded at room temperature and were not

further modified by baseline correction or subtraction. Reported spectra of adsorbed species were generally recorded after the system had reached equilibrium under constant solution flow conditions. Typically, for a ligand concentration of 1 × 10-4 mol L-1 this takes approximately 1 h. Adsorption isotherms were determined by flowing stepwiseincreasing concentrations of aqueous solution species over CdS films, at constant pH, with each solution given sufficient time to achieve adsorption equilibrium. Langmuir adsorption isotherm constants were obtained from nonlinear fitting of absorbance vs time data using Origin Pro version 7.5 software. Uncertainties in Langmuir adsorption constants were calculated using a 95% confidence interval based on this fit. General Methods. Solution pH measurements were made with a Mettler-Toledo MP220 meter to a precision of (0.1. UV-vis spectra of aqueous solutions were recorded on a Varian Cary 500 Scan UV-vis-NIR spectrophotometer. Emission spectra from aqueous solutions were recorded on optically dilute samples (absorbance < 0.05) using a Perkin-Elmer LS 50B spectrophotometer, with an excitation wavelength of 390 nm. Electrophoretic mobility measurements were conducted at a constant ionic strength of 1 × 10-2 mol L-1 NaNO3. CdS suspensions of different pH values were prepared in the range of pH ) 3-10, and the electrophoretic mobility was measured with a Malvern Zetasizer Nano ZS (Malvern Instruments, UK). Each sample was measured three times, and the average mobility was calculated.

(38) Chen, H. M.; Huang, X. F.; Xu, L.; Xu, J.; Chen, K. J.; Feng, D. Superlattices Microstruct. 2000, 27, 1-5. (39) Uchihara, T.; Fox, M. A. Inorg. Chim. Acta 1996, 242, 253-9. (40) Fu, Z.; Zhou, S.; Shi, J.; Zhang, S. Mater. Res. Bull. 2005, 40, 15911598. (41) Winter, J. O.; Gomez, N.; Gatzert, S.; Schmidt, C. E.; Korgel, B. A. Colloids Surf. A 2005, 254, 147-157. (42) Dickie, S. A.; McQuillan, A. J. Langmuir 2004, 20, 11630-11636.

(43) Cotton, F. A.; Wilkinson, G.; Bochmann, M.; Murillo, C. AdVanced Inorganic Chemistry, 6th ed.; Wiley-Interscience: New York, 1998; 1248 pp. (44) Porter, L. A., Jr.; Ji, D.; Westcott, S. L.; Graupe, M.; Czernuszewicz, R. S.; Halas, N. J.; Lee, T. R. Langmuir 1998, 14, 7378-7386. (45) ACD/Labs AdVanced Chemistry DeVelopment Software V8.14 for Solaris, 1994-2006. (46) Spildo, K.; Blokhus, A. M.; Andersson, A. J. Colloid Interface Sci. 2001, 243, 483-490.

Results and Discussion Adsorption of R-Lipoic Acid to CdS. Due to the length of the aliphatic hydrocarbon chain in R-lipoic acid and DHLA, there is the possibility that adsorption may occur via either the thiol groups, the carboxylic acid group, or both groups. It is expected that there would be a stronger interaction with CdS via the thiol groups rather than the carboxylic acid group, due to the deformable coordination sphere of Cd(II).43 In our previous study of the monothiol-containing carboxylic acid ligands, MAA and MPA, adsorption to CdS via both functional groups was not observed, probably due to the shortness of the aliphatic hydrocarbon chain. There have been numerous studies that employed DHLA adsorbed on cadmium chalcogenide nanoparticles; however, no studies of the adsorption of R-lipoic acid on these substrates have been reported. A study of the preparation of n-octadecyl disulfide-capped Au and Ag nanoparticles in the presence of light was reported by Porter et al.44 In their preparation of the capped nanoparticles, reducing conditions were used, and cleavage of the disulfide bond was expected. However, evidence from IR, Raman, and 1H NMR spectroscopy suggested that reduction of the disulfide to the thiol did not occur and that the actively adsorbing species contained a disulfide bond. Little further information has been reported on the specific nature of the R-lipoic acid surface interactions. Adsorption experiments with R-lipoic acid in aqueous media were able to be carried out only at sufficiently high pH, where the carboxylic acid functional group was deprotonated and solubility was increased. Only a calculated pKa(COOH) of 4.75 ( 0.10 for R-lipoic acid has been reported.45 The pKa for n-octanoic acid, which has a similar aliphatic chain length, has been reported as 4.8.46 Figure 3 shows the spectra of R-lipoic acid as a solid, in aqueous solution, and adsorbed on a CdS particle film. Spectrum a of Figure 3 shows the ATR-IR spectrum

12926 Langmuir, Vol. 23, No. 26, 2007

Young et al.

of solid R-lipoic acid. The spectrum shows the broad O-H stretch absorption of the carboxylic acid dimer across the 3300-2500 cm-1 range, the CdO stretch at 1687 cm-1, and the broad outof-plane O-H bend of the carboxylic acid dimer at 929 cm-1.47 The wavenumber of the carbonyl stretch is approximately 20 cm-1 less than that commonly observed for aliphatic carboxylic

acids, and this difference is attributed to intermolecular hydrogen bonding in the solid phase.47 Weaker C-H stretching bands are observed at 2927-2843 cm-1 superimposed on the broad O-H stretch. Other assigned peaks include the CH2 scissoring band at 1466 cm-1 and the CH2 rocking band at 733 cm-1 (not shown).47 The 1450-1200 cm-1 spectral region is complicated by solidstate interactions and contains bands due to C-O stretching and O-H bending combinations along with weaker CH2 twisting and wagging vibrations.47 The weak disulfide stretching S-S mode usually found in the 500-400 cm-1 region falls outside the spectral region that can be observed with a ZnSe prism. Spectrum b of Figure 3 is that of 1 × 10-1 mol L-1 R-lipoic acid in aqueous solution at pH ) 12. The strongest absorptions are the antisymmetric and symmetric carboxylate stretching modes at 1546 and 1408 cm-1, respectively, and the C-H stretching bands at 2937-2864 cm-1. A very weak shoulder on the symmetric carboxylate stretch peak is due to the CH2 scissoring mode. Spectrum c of Figure 3 shows the time evolution over 40 min of IR spectra during the adsorption of R-lipoic acid to a film of CdS nanocrystals from a 1 × 10-4 mol L-1 R-lipoic acid solution at pH ) 8.6, in the absence of light. The spectrum of the adsorbed species closely resembles that of the solution species at pH ) 12 shown in spectrum b. However, the frequency of the antisymmetric carboxylate stretch is lowered by approximately 11 cm-1 upon adsorption. This is likely to be due to the interaction of the terminal carboxylate group with the CdS surface. The observation of an absorption at 1533 cm-1 for adsorbed octanoate ions from flowing 1 × 10-4 mol L-1 solutions (pH ) 8.6) over a CdS particle film supports this conclusion. The adsorption via the carboxylate group is relatively weak, which is confirmed by adsorption isotherm measurements for n-octanoic acid adsorption on the CdS particle films and is discussed in the adsorption isotherm section later in the paper. A 1 × 10-1 mol L-1 solution spectrum of n-octanoic acid at the same pH shows the antisymmetric carboxylate stretch band at 1543 cm-1 (not shown). In spectrum c of Figure 3, there are also two broad absorption losses at 1462 and 1362 cm-1, which are due to displacement of adsorbed carbonate. This carbonate adsorption has arisen due to a trace of CO2 present in the water flow used to obtain the background spectrum. Spectrum e of Figure 3 shows the spectrum resulting from flowing a solution of carbonic acid (CO2-saturated water, pH ) 3.6) over a similarly prepared CdS film. Such spectra have been observed from several different substrates and have been attributed to adsorbed monodentate carbonate.20,48 The loss of monodentate carbonate distorts the spectrum of adsorbed R-lipoic acid, principally by reducing the observed intensity of the 1409 cm-1 peak relative to that at 1535 cm-1. Furthermore, it may also influence the observed wavenumber of the antisymmetric stretch peak at about 1535 cm-1. Figure 4 shows a plot of absorbance at 1534 cm-1 against the R-lipoic acid solution flow time. The rate of adsorption in the initial 40 min is relatively slow in the absence of light, which implies relatively weak adsorption. Adsorption saturation for 1 × 10-4 mol L-1 R-lipoic acid corresponds to absorbance of 0.005-0.006 at 1535 cm-1 with a triple-reflection accessory. After 40 min flow, the system was exposed to laboratory fluorescent lighting, which resulted in an abrupt increase in the rate of absorbance change at 1534 cm-1. Spectrum d of Figure 3 shows the time evolution of spectra following this exposure to fluorescent lighting over the subsequent 40 min. These spectra show a further increase in the carboxylate group absorptions

(47) Silverstein, R. M.; Webster, F. X. Spectrometric Identification of Organic Compounds, 6 ed.; John Wiley and Sons, Inc.: New York, 1998.

(48) Bargar, J. R.; Kubicki, J. D.; Reitmeyer, R.; Davis, J. A. Geochim. Cosmochim. Acta 2005, 69, 1527-1542.

Figure 3. ATR-IR spectra of (a) neat R-lipoic acid; (b) 1 × 10-1 mol L-1 solution of R-lipoic acid at pH ) 12; (c) time evolution adsorption (at 2.5, 23.5, and 40 min of solution flow time) of R-lipoic acid to a CdS film from a 1 × 10-4 mol L-1 R-lipoic acid solution at pH ) 8.6, without exposure to light; (d) time evolution adsorption (at 19, 71.5, and 103.5 min of solution flow time) of experiment d above that was exposed to light; and (e) adsorption of carbonate to a CdS film from a pH ) 3.6 carbonic acid solution. Final spectrum of time evolution spectra is in bold. The background spectra were from the bare ZnSe prism for spectrum a, water on ZnSe for spectrum b, water on CdS (pH ) 8.6) for spectra c and d, and water on CdS (pH ) 3.6) for spectrum e. All spectra were recorded using a triplereflection accessory.

IR of Adsorption of Dithiol-Containing Ligands

Langmuir, Vol. 23, No. 26, 2007 12927

Figure 4. The time dependence of absorbance at 1534 cm-1 of R-lipoic acid to a CdS film from a 1 × 10-4 mol L-1 solution at pH ) 8.6. The dashed line indicates the time at which exposure to laboratory fluorescent lighting began. Data were recorded at 1 min intervals using the triple-reflection accessory.

with no apparent change in peak wavenumber. This increase in absorbance upon exposure to light suggests the initiation of a photochemical or photocatalytic process giving an increased signal at the same wavenumber. During photocatalytic processes, an oxidation reaction is accompanied by a separate reduction reaction. CdS is an n-type semiconductor that under band gap excitation (λ