The Reaction between Tetrakis (diethylamino) tin and Indium Tin Oxide

Crissy Rhodes , Stefan Franzen , Jon-Paul Maria , Mark Losego , Donovan N. Leonard , Brian Laughlin , Gerd Duscher , Stephen Weibel. Journal of Applie...
0 downloads 0 Views 122KB Size
Download PDF
5696

Langmuir 2001, 17, 5696-5702

The Reaction between Tetrakis(diethylamino)tin and Indium Tin Oxide Eric L. Bruner, Amelia R. Span, Steven L. Bernasek,* and Jeffrey Schwartz* Department of Chemistry, Princeton University, Princeton, New Jersey 08544-1009 Received April 2, 2001. In Final Form: June 12, 2001

Indium tin oxide (ITO) is a common anodic material for modern optoelectronics,1-4 and it has been proposed5 that organizing a dipole at the ITO surface can change the work function of that material, directly affecting its hole injection propensity in a device. Surface adsorption of organics can create this surface dipole; systematic variation of these organics could provide a means to control device characteristics. For example, changes in the work function of gold6 or copper7 electrodes can be effected by self-assembly of variously substituted organic thiols,6 but thiols do not bind strongly to ITO.8 Our approach to surface modification of ITO with organics involves using “linker” complexes which react with surface hydroxyl groups of the mixed oxide 9,10 and which maintain substitutionally reactive functionality following surface chemisorption. Ligand exchange, or metathesis, by proton transfer in surface complexes proceeds rapidly and to completion, when the attacking reagent is of greater acidity than the “leaving group” and when kinetic barriers to proton transfer are low.11 Accordingly, simple Zr12 or Sn10 alkoxides are viable precursors of stable interfaces between ITO and carboxylic acids12 or phenols through sequences of surface deposition and ligand metathesis,10,13 carboxylic acids (pKa ≈ 5)14 or phenols (pKa ≈ 10)14 are more acidic than the tert-butyl alcohol “leaving group” (pKa ≈ 18),14 and kinetic barriers to proton transfer between the oxygens of reagent and ligand are low.11 By these precedents, metathesis should be possible with arylthiols (pKa ≈ 6),15 but surface modification with weakly acidic carbon acids may not succeed. For example, the * To whom correspondence should be addressed. (1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913. (2) Campbell, I. H.; Davids, P. S.; Smith, D. L.; Barashkov, N. N.; Ferraris, J. P. Appl. Phys. Lett. 1998, 72, 1863. (3) Antoniadis, H.; Miller, J. N.; Roitman, D. B.; Campbell, I. H. IEEE Trans. Electron Devices 1997, 44, 1289. (4) Burrows, P. E.; Bulovic´, V.; Shen, Z.; Forrest, S. R.; Thompson, M. E. IEEE Trans. Electron Devices 1998, 44, 1188. (5) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. Adv. Mater. (Weinheim, Ger.) 1999, 11, 605. (6) Zehner, R. W.; Parsons, B. F.; Hsung, R. P.; Sita, L. R. Langmuir 1999, 15, 1121. (7) Campbell, I. H.; Kress, J. D.; Martin, R. L.; Smith, D. L.; Barashkov, N. N.; Ferraris, J. P. Appl. Phys. Lett. 1997, 71, 3528. (8) Yan, C.; Zharnikov, M.; Go¨lzha¨user, A.; Grunze, M. Langmuir 2000, 16, 6208. (9) Purvis, K. L.; Lu, G.; Schwartz, J.; Bernasek, S. L. J. Am. Chem. Soc. 2000, 122, 1808. (10) Span, A. R.; Bruner, E. L.; Bernasek, S. L.; J., S. Langmuir 2001, 17, 948. (11) Miller, J. B.; Schwartz, J.; Bernasek, S. L. J. Am. Chem. Soc. 1993, 115, 8239. (12) VanderKam, S. K.; Gawalt, E. S.; Schwartz, J.; Bocarsly, A. B. Langmuir 1999, 15, 6598. (13) VanderKam, S. K.; Lu, G.; Bernasek, S. L.; Schwartz, J. J. Am. Chem. Soc. 1997, 119, 11639. (14) Carey, F. A. Organic Chemistry, 4th ed.; McGraw-Hill Higher Education: Boston, 2000. (15) Delby, K. N.; Jencks, W. P. J. Chem. Soc., Perkin Trans. 2 1997, 8, 1555.

preparation of a surface tin acetylide by metathesis between an acetylene (pKa ≈ 25)16 and a surface tin alkoxide would not be anticipated. However, the preparation of tin acetylides in solution from tin amides17-21 (for the amine “leaving group,” pKa ≈ 36),22 is well-known, and the reaction between tetrakis(diethylamino)tin (1) and silica, followed by treatment with an acetylene at elevated temperature, has been reported23,24 to yield surface bound tin monoacetylides. Unfortunately, although qualitative X-ray photoelectron spectroscopic (XPS) data were obtained, which do show the presence of C and Sn on the silica surface, no direct evidence was offered for the chemical reaction between 1 and silica, nor was the stoichiometry of any resulting surface bound materials reported. It was therefore important, as a starting point, to elucidate details of the chemistry of 1 with an hydroxylated surface, since metathesis would give surface bound products of comparable stoichiometry. We report herein details of the reaction between 1 and hydroxylated ITO, both under “normal” laboratory conditions and in ultrahigh vacuum (UHV). We find that surface bound tin amides are indeed formed and that the composition of these amides evolves with increasing temperature of the surface. We have also found that these surface tin amides can react with acetylenes to give the corresponding surface bound acetylide derivatives, but slowly. Experimental Section General. Our UHV chamber25 is equipped with a Mattson Research Series Fourier transform infrared spectrometer (FTIR), a quadrupole mass spectrometer (QMS), and an XPS spectrometer. ITO on glass (Colorado Concept Coatings, 15 Ω/0, 1500 Å) was cleaned by sonication in detergent and then by rinsing successively with water, hot trichloroethylene, acetone, and methanol. Slides thus treated were stored under vacuum and then bombarded with Ar+ ions after transfer into UHV to remove residual organics. XPS analysis of the cleaned ITO showed the Sn/In ratio to be ca. 6%; the In signal serves as an internal standard for determination of surface tin complex stoichiometries, as described previously.10 Deconvolution of XP spectra was done using “XPSPEAK” freeware, available at http://www.phy.cuhk.edu.hk/∼surface/. Hydroxylation of the ITO was accomplished in UHV through five cycles of water dose (180 K) desorption (290 K);9 XPS showed approximately 8% of the surface oxygen to be OH (O[1s] binding energy [BE] ) 533.1 eV). 1 was used as received (Aldrich) and was stored under nitrogen. Substituted phenylacetylenes (p-X-C6H4CtC-H; X ) H (Acros), F and OCH3 (Aldrich), and NO2 (Toronto Research)) were either used as received or distilled before use. Surface tin phenylacetylides were characterized by diffuse reflectance infrared spectroscopy (DRIFT) using a Nicolet 730 FTIR spectrometer equipped with a SpectraTech DRIFT accessory. The IR spectrum of 1 was acquired in (16) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic Chemistry, 3rd ed.; Harper & Row: New York, 1987. (17) Davies, A. G. Organotin Chemistry; VCH: New York, 1997; Chapter 9. (18) Jousseaume, B. J. Chem. Soc., Chem. Commun. 1984, 1452. (19) Williamson, B. L.; Stang, P. J. Synlett 1992, 199. (20) Jones, K.; Lappert, M. F. J. Organomet. Chem. 1965, 3, 296. (21) Cauletti, C.; Furlani, C. Gazz. Chim. Ital. 1988, 118, 1. (22) House, H. O. Modern Synthetic Reactions, 2nd ed.; W. A. Benjamin, Inc.: Menlo Park, CA, 1972. (23) Yam, C. M.; Kakkar, A. K. J. Chem. Soc., Chem. Commun. 1995, 8, 907. (24) Yam, C. M.; Dickie, A.; Malkhasian, A.; Kakkar, A. K.; Whitehead, M. A. Can. J. Chem. 1998, 76, 1766. (25) Miller, J. B.; Bernasek, S. L.; Schwartz, J. Langmuir 1994, 10, 2629.

10.1021/la010495l CCC: $20.00 © 2001 American Chemical Society Published on Web 08/04/2001

Notes

Langmuir, Vol. 17, No. 18, 2001 5697 Table 1. X-ray Photoelectron Spectral Analysis of the Reaction between ITO and 1 XPS

temp (K)

complex obsd.

185

multilayer of Sn(NEt2)4, 1

293

mixture of 2 and 3 (1:4) [ITO]-[O]1-Sn(NEt2)3, 2 [ITO]-[O]2-Sn(NEt2)2, 3

a

380

mixture of 3 and 4 (1:1.5) [ITO]-[O]3-Sn(NEt2)1, 4

450

mixture of 3 and 4 (1:9)

atomic species

binding energy (eV)a

Sn(3d5/2) C(1s) N(1s) Sn(3d5/2) C(1s) N(1s) C(1s) N(1s) Sn(3d5/2) C(1s) N(1s) Sn(3d5/2)

487.5 285.3, 286.0 399.1 487.5 286.7, 287.6 400.6 285.2, 286.3 399.3 486.9 284.6, 285.7 399.1 486.6

atomic ratiosb C/ Sn ) 16 C/ N ) 4 C/ Sn ) 9.3, C/ N ) 4.2, N/ Sn ) 2.1

C/ Sn ) 5.9, C/ N ) 4.0, N/ Sn ) 1.4 C/ Sn ) 4.6, C/ N ) 4.0, N/ Sn ) 1.1

b

Calibrated against In(3d5/2). Experimental sensitivity factors determined from atomic ratios in the defined species Sn(NEt2)4 (1).

a nitrogen-filled glovebox using a Midac Illuminator FTIR spectrometer equipped with a Surface Optics specular reflectance accessory. 1-Ethynyl-4-(trifluoromethyl)benzene.26 A solution of 4-iodo-(trifluoromethyl)benzene (3.0 mmol, 0.816 g), palladium dichloride (0.06 mmol, 0.0106 g), triphenylphosphine (0.42 mmol, 0.110 g), and copper(I) iodide (0.15 mmol, 0.029 g) in diisopropylamine (5 mL) was stirred under nitrogen at 0 °C. A second solution of trimethylsilylacetylene (3.6 mmol, 0.354 g) in diisopropylamine (2 mL) was added by syringe. The reaction was warmed to room temperature and stirred vigorously. Additional diisopropylamine (4 mL) was added after 1 h. Gas chromatography and mass spectrometry (GC-MS) analysis showed the coupling reaction to be complete after a total of 2.5 h. The reaction mixture was concentrated by rotary evaporation. The residue was suspended in hexanes (15 mL), filtered through Celite, and concentrated again by rotary evaporation. The resulting brown liquid was distilled to give 0.479 g (2 mmol, 66%) of 1-(trimethylsilyl)ethynyl-4-(trifluoromethyl)benzene as a colorless oil (1H NMR [300 MHz, CDCl3]: δ 7.50 [s, 4 H], 0.25 [s, 9 H]), which was then dissolved in methanol (2.5 mL) containing pulverized KOH (0.4 mmol, 0.026 g). GC-MS analysis indicated removal of the silyl protecting group within 30 min. The reaction mixture was diluted with water (4.8 mL), extracted with ether, and dried (MgSO4). The ether was removed by distillation under nitrogen to give 0.236 g of the product (1.4 mmol, 70%) as a clear liquid. 1H NMR (300 MHz, CDCl3): δ 7.58 (s, 4 H), 3.18 (s, 1 H). IR (neat): 3306, 1624, 1612, 1408, 1326, 1171, 1129, 1067, 1017, 843 cm-1. [ITO]-[O]1-Sn(NEt2)3/[ITO]-[O]2-Sn(NEt2)2 (2/3). A horizontal tube reaction chamber equipped with two stopcockregulated reservoirs (one for 1, and the other for the phenylacetylene) was fitted with samples of conventionally cleaned ITO/glass. The chamber was evacuated to 10-4 Torr at room temperature. The ITO/glass samples were exposed to vapor of 1 for three cycles each of 15 min with the chamber open to the vacuum, followed by 30 min with the chamber isolated from the vacuum source. The ITO substrates were cooled by external dry ice during the second and third cycles, and the reservoir of 1 was gently heated to 55 °C during the third cycle, after which time a film of 1 was visible on the glass slides. The reservoir of 1 was then closed, and excess 1 was removed from the ITO by evacuation at room temperature for at least 1 h, until no 1 was visible. [ITO]-[O]1-Sn(C≡C-C6H4)3/[ITO]-[O]2-Sn(C≡C-C6H4)2 (5a/6a). Phenylacetylene was vapor deposited onto 2/3 at reduced pressure. The reservoir containing phenylacetylene was opened to the reaction chamber, which was still under active vacuum following removal of excess 1. The stopcock between the reaction chamber and vacuum source was then immediately closed. Substrate slides were gently cooled by placing dry ice below the chamber until a film of phenylacetylene was visible on the slides. The dry ice was then removed, and the reservoir was isolated from the reaction chamber. Samples of 2/3 were exposed to phenylacetylene in this way for 4 h before excess phenylacetylene was removed by evacuation at room temperature. (26) Crisp, G. T.; Flynn, B. L. J. Org. Chem. 1993, 58, 6614.

Figure 1. Infrared spectra of 1 on ITO: (a) multilayer of 1 prepared in UHV; (b) film of 1 evaporated on ITO at -78 °C at 10-4 Torr. [ITO]-[O] 1 -Sn(C≡C-C 6 H 4 -F) 3 /[ITO]-[O] 2 -Sn(C≡CC6H4-F)2 (5b/6b). Samples of 2/3 were exposed to 1-ethynyl-4fluorobenzene as described for phenylacetylene, but the exposure time totaled only ca. 1 h. Excess 1-ethynyl-4-fluorobenzene was removed by evacuation at room temperature. [ITO]-[O]1-Sn(C≡C-C6H4-OCH3)3/[ITO]-[O]2-Sn(C≡CC6H4-OCH3)2 (5c/6c). Samples of 2/3 were exposed to 1-ethynyl4-methoxybenzene as described for phenylacetylene, but the exposure time was 3.5 h. Dry ice was placed beneath the slides only during the final hour of exposure. Excess 1-ethynyl-4methoxybenzene was removed by evacuation at room temperature.

5698

Langmuir, Vol. 17, No. 18, 2001

Notes

Scheme 1. Synthesis of ITO Surface Bound Tin Acetylides

Results and Discussion

tion at low temperature and to elucidate the thermal evolution9 of such surface tin amide species. Deposition of 1 was accomplished from the vapor phase onto the hydroxylated ITO surface at 185 K. XPS analysis showed complete attenuation of the In(3d5/2) and O(1s) core level signals, indicative of formation of a multilayer of 1 (see Table 1). Under these conditions, the Sn(3d5/2) signal measured is due entirely to 1, and C(1s)/Sn(3d5/2) and N(1s)/Sn(3d5/2) sensitivity factor ratios can be determined experimentally according to the stoichiometry of 1. The FT-RAIRS spectrum of multilayer 1 (recorded at 150 K) showed prominent peaks for νC-H at 2952 and 2927 cm-1 and two weaker, broader peaks centered at 2867 and 2819 cm-1 (Figure 1a); this pattern is similar to that recorded for a film of 1 evaporated onto conventionally prepared ITO at -78 °C and 10-4 Torr (2962, 2925, 2861, and 2831 [br] cm-1; Figure 1b). Thermal desorption of excess 1 was accomplished at 293 K over 15 min, and the composition of the chemisorbed tin amide was determined by XPS to be a mixture of tris-amide [ITO]-[O]1-Sn(NEt2)3, 2, and bis-amide [ITO]-[O]2-Sn(NEt2)2, 3 (C/Sn ) 9.3, N/Sn ) 2.1, 2:3 ) 1:4, Table 1). Whereas O(1s) signals for lattice oxygens of ITO were again apparent, the signal for surface OH was substantially lost, showing that this functionality was consumed on formation of 2/3. The FT-RAIRS spectrum of 2/3 (recorded at 150 K) showed a prominent peak for νC-H at 2970 cm-1 and two weaker, broader peaks centered at 2915 and 2876 cm-1 (Figure 2). Heating the substrate to 380 K for 10 min caused further surface OH group reaction, to yield a 1.5:1 mixture of 3 and monoamide species [ITO]-[O]3-Sn(NEt2)1, 4 (C/Sn ) 5.9, N/Sn ) 1.4, Table 1). Heating the initial mixture of 2 and 3 to 293 K for 10 h or to 450 K for 10 min gave 3 and 4 (C/Sn ) 4.6, N/Sn ) 1.1, 3:4 ) 1:9, Table 1). This stepwise reactivity of surface tin amides with OH groups on ITO (Scheme 1) closely parallels that of their surface Sn alkoxide analogues on this surface10 and on others of low hydroxyl group content.27 The C(1s) and N(1s) binding energies of component complexes were determined by deconvolution

The reaction between ITO and 1 was studied in UHV to determine the stoichiometry of surface complex forma-

(27) Lu, G.; Purvis, K.; Schwartz, J.; Bernasek, S. L. Langmuir 1997, 13, 5791.

Figure 2. Infrared spectra of 2/3 on ITO prepared on ITO in UHV. [ITO]-[O] 1-Sn(C≡C-C6H4-NO 2)3/[ITO]-[O] 2-Sn(C≡CC6H4-NO2)2 (5d/6d). Because of its relatively low volatility, treatment of 2/3 with 1-ethynyl-4-nitrobenzene was accomplished through four cycles of exposure as described for deposition of 1 onto ITO. The first cycle was performed at room temperature, and the second and subsequent cycles involved cooling the substrate with external dry ice. The 1-ethynyl-4-nitrobenzene reservoir was slowly warmed to 70 °C during the third and fourth cycles. Excess 1-ethynyl-4-nitrobenzene was removed by evacuation for 15 h at room temperature, followed by heating to 70 °C with continued evacuation. [ITO]-[O] 1 -Sn(C≡C-C 6 H 4 -CF 3 ) 3 /[ITO]-[O] 2 -Sn(C≡CC6H4-CF3)2 (5e/6e). Samples of 2/3 were exposed to 1-ethynyl4-(trifluoromethyl)benzene as described for phenylacetylene; exposure time was 4 h, but substrate cooling with dry ice was not required. Excess 1-ethynyl-4-(trifluoromethyl)benzene was removed by evacuation at room temperature.

Notes

Langmuir, Vol. 17, No. 18, 2001 5699

Figure 3. C(1s) XP spectra for surface tin amide species: (a) multilayer of 1; (b) substrate heated to 293 K, showing a mixture of 2 (higher binding energy peaks) and 3 (1:4); (c) substrate heated to 380 K, showing a mixture of 3 (higher binding energy peaks) and 4 (1:1.5); (d) substrate heated to 450 K, showing a mixture of 3 and 4 (1:9).

of the corresponding XP spectra of mixtures (Figures 3 and 4). Deconvolution of C(1s) binding energy data9,10 for 1 was done with the assumption that the intensities for the two types of carbon signals in the ethyl group, N-CH2CH3 and N-CH2CH3, were equal; binding energy differences obtained for these carbons were the same (0.7 eV) as those reported for ethylamine.28 Carbon(1s) binding energies for 3 and 4 were similarly obtained from deconvolution of the XP spectra measured for material heated to both 380 and 450 K, taking into account relative amounts of 3 and 4 present at these temperatures (Table 1) as determined from integration of total C(1s), N(1s), (28) Gelius, U.; Heden, P. F.; Hedman, J.; Lindberg, B. J.; Manne, R.; Nordberg, R.; Nordling, C.; Siegbahn, K. Phys. Scr. 1970, 2, 70.

and Sn(3d5/2) signals. Binding energies for 2 were obtained analogously from data for material warmed to 293 K. Nitrogen(1s) binding energies for individual amide complex components were also determined based on deconvolution of spectra weighted according to relative amounts of amide complexes present (Figure 4), though error in assignment for N(1s) binding energies is likely to be higher than that for C(1s) (for C[1s] error is (0.2 eV), given the 4-fold higher signal-to-noise ratios measured for the latter element. Fitting N(1s) peaks of 2 and 3 using binding energies of 1 and 4 served as an independent validation of peak position assignments made based on integration ratios; both deconvolution procedures gave similar peak position values, within experimental error.

5700

Langmuir, Vol. 17, No. 18, 2001

Notes

Figure 4. N(1s) XP spectra for surface tin amide species: (a) substrate heated to 293 K, showing a mixture of 2 (higher binding energy peaks) and 3 (1:4); (b) substrate heated to 380 K, showing a mixture of 3 (higher binding energy peaks) and 4 (1:1.5); (c) substrate heated to 450 K, showing a mixture of 3 and 4 (1:9).

In a parallel series of experiments, samples of ITO/ glass were cleaned and dried under conventional conditions and were placed in a simple reactor connected to reservoirs of 1 and a particular phenylacetylene.10 This tube could be evacuated and either heated with external heating tape or cooled by externally applied dry ice. A multilayer of 1 was deposited on the ITO substrate at -78° C by evaporation from a reservoir held at 55 °C at 10-4 Torr (IR: νC-H ) 2962, 2925, and 2831 cm-1; Figure 3a). Excess 1 was removed in vacuo at room temperature to give 2/3, by analogy with material produced by similar thermal processing in UHV. Acetylides are interesting targets for possible surface dipole introduction5 in the context of device behavior modification through work function change. Because the Sn-acetylide bond is linear,17,29 binding a substituted acetylide moiety to ITO might optimize the effective dipole

moment normal to the surface5,6,30,31 imparted by surface molecular coordination. (In contrast, because the SnS-C bond is bent,17,32,33 conformational variation in surface tin thiolate geometries might give rise to net surface dipole normal components that are far smaller than those expected based on the dipole moment of the free ligands.) Samples of 2/3 were exposed to vapor of each of a series (29) Khakin, L. S.; Grikina, O. E.; Sipachev, V. A.; Granovsky, A. A.; Nikitin, V. S. Russ. Chem. Bull. 2000, 49, 620. (30) Nu¨esch, F.; Rotzinger, F.; Si-Ahmed, L.; Zuppiroli, L. Chem. Phys. Lett. 1998, 288, 861. (31) Zuppiroli, L.; Si-Ahmed, L.; Kamaras, K.; Nu¨esch, F.; Bussac, M. N.; Ades, D.; Siove, A. Eur. Phys. J. B 1999, 11, 505. (32) Bravo, J.; Cordero, M. B.; Casas, J. S.; Sa´nchez, A.; Sordo, J.; Castellano, E. E.; Zukerman-Schpector, J. J. Organomet. Chem. 1994, 482, 147. (33) Labib, L.; Khalil, T. E.; Iskander, M. F. Polyhedron 1996, 15, 349.

Notes

Langmuir, Vol. 17, No. 18, 2001 5701 Scheme 2. Ligand Metathesis Illustrated for 3 via Putative Tin-Acetylene Complex 8

Table 2. X-ray Photoelectron Spectral Analysis of the Reaction between 3/4 and p-NO2(C6H4)CtCH XPS temp (K)

complex obsd.

185 multilayer p-NO2(C6H4)C≡CH 273 [ITO]-[O]2-Sn(C≡C[C6H4]NO2)2, [ITO]-[O]3-Sn(C≡C[C6H4]NO2)1 6d/7d

atomic species

binding energy (eV)a

C(1s) N(1s)c O(1s)c Sn(3d5/2) C(1s) N(1s)c O(1s)c

286.3, 285.3 407.9 535.3 X 285.2 407.3 534.1

atomic ratiosb O/N ) 2 Oc/Nc ) 2 Nc/Sn ) X Nd/Sn ) X Oc,e/Sn ) 1.1

a Calibrated against In(3d ). b Experimental sensitivity factors 5/2 determined from atomic ratios of defined species p-NO2(C6H4)CtCH. c N(1s) and O(1s) for the NO group. d For 3/4. e For complete 2 conversion of 3 and 4 (1:9) to 6d/7d (1:9), expected O/Sn ) 2.2:1.

Figure 5. DRIFT analysis of the reaction between phenylacetylene and 2/3 (lower trace) to give 5a/6a, and a control (upper trace).

of substituted phenylacetylenes (p-X-C6H4CtC-H; X ) H, F, CF3,26 OCH3, and NO2), with substrate cooling to form multilayers. The multilayers were then desorbed in vacuo, at room temperature, to yield the surface acetylides (for 1-ethynyl-4-nitrobenzene, the reservoir was warmed for initial deposition, as was the substrate for multilayer desorption). Conversion of 2/3 to the corresponding acetylides (5/6) was slow at 10-4 Torr, perhaps because of

Figure 6. XP spectra following partial conversion of 3/4 to 6d/7d by reaction with 1-ethynyl-4-nitrobenzene: (a) N(1s); (b) O(1s).

a relatively high kinetic barrier to proton transfer11 from the carbon acid to the amide ligand in a weakly bound tin-acetylene complex intermediate (8, Scheme 2);34,35 exchange was, however, complete at this reaction pressure. Product tin acetylides were identified by DRIFT,

5702

Langmuir, Vol. 17, No. 18, 2001

noting the introduction of peaks characteristic of the various acetylenes and the disappearance of νC-H signals both for 2/3 and for the terminal C-H bond of the free acetylenes (Figure 5). In control experiments, it was found that no surface acetylenic materials remained on the ITO surface in the absence of 2/3, following simple evacuation. Attempts were made to measure the stoichiometry of tin acetylide formation by metathesis in UHV, but they were hampered by slow ligand exchange rates, which are likely to be exacerbated by low residence times, expected for weak complexation of the acetylene to tin, and at operating background pressures e10-7 Torr. Nonetheless, partial metathesis could be observed for reasonable exposure times and pressures. For example, a sample of 3/4 (1:9) prepared from 1 by heating to 450 K was cooled to 190 K and exposed to 1-ethynyl-4-nitrobenzene for 20 min at 2 × 10-7 Torr, which gave a multilayer of the acetylene on the substrate (Table 2). Thermal desorption of excess acetylene was accomplished at 293 K over a period of 10 min. Following two further cycles of dose/desorption, XPS analysis (Figure 6) showed that metathesis to the corresponding surface tin acetylides, 6d/7d, was about one-third to one-half complete, based on comparative nitro group N(1s) and O(1s) data (Table 2); we have noted similar requirements for multiple dose/desorption cycles to ensure (34) For examples of isolable 5-coordinate tin complexes, see: Van Koten, G.; Noltes, J. G. In Adv. Chem. Ser. 1976, No. 157, 275. (35) For an example of an isolable 6-coordinate tin alkoxide complex, see: Reuter, H.; Kremser, M. Z. Anorg. Allg. Chem. 1991, 598/599, 259.

Notes

complete metathesis even using kinetically more reactive systems, such as carboxylic acids or phenols.13,36 Conclusions Synthesizing a surface tin amide complex on ITO enables strong bonding of substituted arylacetylenes to the ITO as surface tin acetylides. Since previous work has shown that acetylenic group substitution in a thiol can lead to large changes in the work function of gold,6 it is now important to determine how acetylide coordination can affect the work function of an ITO electrode to which it is indirectly bonded. Accordingly, ultraviolet photoelectron spectroscopic measurements are now in progress, and organic light-emitting devices are being constructed in UHV so that device-characteristic determinations can be made using these materials. Acknowledgment. The authors thank the National Science Foundation and the New Jersey Center for Optoelectronic Materials for their support of this research. Supporting Information Available: The N(1s) XPS spectrum of the multilayer of tetrakis(diethylamino)tin (1); the Sn(3d5/2) XPS spectra of [ITO]-[O]1-Sn(NEt2)3, 2, [ITO]-[O]2Sn(NEt2)2, 3, and [ITO]-[O]3-Sn(NEt2)1, 4; and the IR spectra (3800-800 cm-1) of multilayer 1, 5b/6b, 5c/6c, 5d/6d, and 5e/ 6e. These materials are available free of charge via the Internet at http://pubs.acs.org. LA010495L (36) Purvis, K. L.; Lu, G.; Schwartz, J.; Bernasek, S. L. Langmuir 1998, 14, 3720.