Background Art

At present, small organic molecules are finding increasing use in molecular electronics as active electronic components. Well ordered self-assembled monomolecular layers (SAMs) are a fundamental structure in the organization of such devices. They are formed by an ordered layer of amphiphilic molecules whose "head" terminus exhibits specific affinity for a substrate. The opposite terminus, usually a long aliphatic chain, usually carries a functional group, such as OH, NH ₂ , COOH. The most frequently used molecules are alkanethiols, whose alkyl chain is attached through the S-H head group to a noble metal substrate, for which a sulfur atom has high affinity. Very often the interaction is through a semicovalent bond between an atom of sulfur and an atom of gold (cf. Love et al, Chem. Rev. 2005, 105, 1 103-1 170). Thiol molecules adsorb on atoms of gold very easily from solutions (e.g., in ethanol), and the resulting densely packed monomolecular layers can posses a variety of chemical properties depending on the outward oriented chain end functional groups.

This type of attachment of an atom Z carrying a single organic group R, such as an alkyl, to the surface of a substrate yields surface structures of the type RZM _n , where M„ stands for one or more metal atoms. An example of atom Z are elements of column VIA of the periodic table of the elements, especially sulfur.

An alternative to the attachment of individual organic molecules to gold-coated substrates through a sulfur atom in thiocyanataes (Ciszek et al., J. Am. Chem. Soc. 2004, 126, 13172) is found in the attempts to form a SAM by adsorption from solution or solvent vapor by the intermediacy of an atom of selenium (Huang et al., Langmuir 1998, 14, 4802; Shaporenko et al., J Phys. Chem. 52005, 709, 3898; Monnell et al., J Phys. Chem. 52004, 108, 9834; Clark et al., Surface Science 2002, 498, 285; Protsailo et al., Langmuir 2002, 18, 9342; Han, S. W. a Kim, K., J Colloid Interface Sci. 2001, 240, 492), tellurium (Weidner et al., J Phys. Chem. C2007, 111, 1 1627), silicon (Owens et al., J Phys. Chem. B 2003, 107, 3177 and J ^m. Chem. Soc. 2002, 7 4, 6800; Katsonis et al., Chem. Eur. J. 2003, 9, 251 ), or a carbon atom in acetylenes (Zhang et al., J. Chem. Soc, 2007, 729, 4876), diazonium salts (Shewchuk a McDermott, Langmuir 2009, 25, 4556; Laforgue et al., Langmuir 2005, 27, 6855), and isocyanates (Stapleton et al., Langmuir 2005, 27, 1 1061).

The SAM-forming attachment of molecules to the surfaces of gold and other metals through an atom of divalent sulfur is facile and well documented (Love et al., Chem. Rev. 2005, 105, 1103; Ulman, Chem Rev. 1996, 96, 1533 and Ultrathin Organic Films, Academic Press: San Diego, 1991). This type of attachment offers numerous advantages, such as easy formation under ordinary laboratory conditions (ambient pressure and temperature), but also has some disadvantages. Examples are limited long-term resistance of thiolates to aerial oxidation (Joseph et al., Chem Mater. 2009, 27, 1670; Willey et al., Surface Science 2005, 576, 188; Huang et al, Langmuir 1998, 14, 4802) and somewhat marginal electron transport across the partially polar bond between the electronegative sulfur atom and a metal atom, which limits the use of these S AMs, for instance in molecular electronics, which is based on the attachment of individual molecules to metal surfaces (substrates).

The steric demands posed by simple alkyl chains attached to a substrate through sulfur or silicon atoms are conducive to the formation of essentially impenetrable wide domains composed of densely packed layers of long alkane chains, somewhat inclined away from the surface normal. The attachment of the organic group through a silicon atom (Owens et al., J. Phys. Chem. B 2003, 707, 3177 and J. Am. Chem. Soc. 2002, 724, 6800; Katsonis et al., Chetn. Eur. J. 2003, 9, 2574) utilizes monoalkylsilanes and the deposition takes place from the vapor phase.

Recently, also the adsorption of trifluoroacetates of certain organomercuric cations on gold, accompanied by a release of the trifluoroacetate anion, has been described (Zheng et al., J. Am. Chem. Soc. 2004, 126, 4540; Mulcahy et al., J. Phys. Chem. C 2009, 113, 20698; Mulcahy et al., J Phys. Chem. C 2010, 114, 14050). Infrared and X-ray photoelectron spectroscopy confirmed that this adsorption is mediated by the interaction of mercury atoms with the gold substrate (surface). Such adsorption has not been examined for other organometallic salts.

Disclosure of the Invention

The present invention provides a self-assembled monomolecular layer on a substrate covered with a metal selected from the group containing palladium, platinum, silver, and gold. The monomolecular layer consists of molecules of trialkyltin, wherein the tin atom is oriented towards the substrate and the alkyl chains are oriented away from the substrate.

In one aspect of the invention, the trialkyltin is of a general formula (C„H ₂ „+i) ₃ Sn, wherein n = 1 to 36. The three alkyl ponytails in one trialkyltin molecule may be the same or different.

In a preferred embodiment of the invention, the trialkyltin is of a general formula (CH ₃ ) Sn(C _! 8H3 ₇ ^, wherein x is in the range of from 0 to 2, y is in the range of from 1 to 3, and x + > = 3.

Preferably, the self-assembled monolayer of trialkyltin molecules has a thickness of 0.1 to 5 nm.

It is an advantage of the self-assembled monomolecular layer (monolayer) according to the present invention, relative to the self-assembled monolayers known from the state of the art, that it is stable to organic solvents such as rc-hexane or ethanol and resistant to reducing and oxidizing agents.

Another object of the present invention is the method of preparation of a self- assembled monomolecular layer on a substrate whose surface is covered with a metal selected from the group comprising palladium, platinum, silver, and gold, wherein said monomolecular layer is formed by molecules of trialkyltin, with the tin atom oriented towards the substrate and the alkyl chains oriented away from the substrate, and where the method of preparation consists of adsorbing a trialkylstannyl salt onto the metal-covered surface of the substrate.

In one aspect of the method of the present invention, the trialkylstannyl salt is of a general formula (C„H ₂ „ ₊₁ ) ₃ Sn-Y, wherein n is in the range of from 1 to 36 and Y is selected from the group comprising salts of strong acids, including triflate (OTf, salt of trifluoromethanesulfonic acid), trifluoroacetate (OCOCF ₃ , salt of trifluoroacetic acid), and tosylate (OTs, salt of 7-toluenesulfonic acid).

In a preferred embodiment of the method of the present invention, the trialkylstannyl salt is of a general formula (CH3) _¾ (Ci ₈ H ₃₇ )^Sn-Y, wherein x ranges from 0 to 2, y ranges from 1 to 3, where x +y = 3, and Y is selected from the group comprising triflate (OTf, salt of trifluoromethanesulfonic acid), trifluoroacetate (OCOCF ₃ , salt of trifluoroacetic acid), and tosylate (OTs, salt of j?-toluenesulfonic acid).

In another preferred embodiment the method of the preparation of the self- assembled monomolecular layer according to the present invention comprises the following steps:

(a) immersion of the substrate into a solution of the trialkylstannyl salt in an organic solvent,

(b) rinsing of the substrate with an organic solvent and drying of the substrate carrying an adsorbed layer of trialkyltin molecules.

Preferably, a solution of the trialkylstannyl salt of trifluoromethanesulfonic, trifluoroacetic, or /7-toluenesulfonic acid in a range of concentrations from 1 mol.l ^"1 to 10 ^"6 mol.l ^"1 is used in step (a). It is preferable to choose the organic solvent used in step (a) from the group comprising solvents based on hydrocarbons, ethers, halogenated alkanes, carbonitriles, or nitro compounds, and a solvent selected from the same group is then used in the rinsing step (b).

The present invention describes the attachment of alkyl residues to a substrate whose surface is covered with a metal, preferably gold, through the intermediacy of a tin atom. The electropositive nature of this ligand atom then represents an advantage for electron transfer and affects the properties of the SAM formed.

Synthesized were covalent trialkylstannyl salts of trifluoromethanesulfonic, trifluoroacetic, or /?-toluenesulfonic acids (triflates, trifluoroacetates, and tosylates, respectively), carrying three alkyl substituents in the molecule, which include one, two, or three C ₁₈ H ₃ chains and 0 to 2 methyl groups. These salts have demonstrated the ability to attach to gold from solution at ambient temperature and pressure to form a self-assembled monolayer on the gold-covered substrate. This ability was characterized through the use of ellipsometry, FTIR spectroscopy, contact angle measurements, and electron transfer blocking measurements.

All trialkylstannyl salts prepared split off the acid residue to form similar stable self- assembled monomolecular layers. The formation of multilayers was never observed. The monomolecular layers prepared in the manner described herein were distinct from the self- assembled layers formed from alkanethiols. The monolayers prepared according to the procedure described in the present invention were thinner, less well organized, less hydrophobic, and they blocked electron transfer only weakly. Their stability against certain solvents, bases, and acids was somewhat lower that that of monomolecular layers built from 1-octadecanethiol, but their resistance to elevated temperature and strong reduction and oxidation agents, including air, was higher.

Trialkylsilyl salts, prepared in a similar manner for comparison, did not exhibit any tendency to adsorb on gold under similar conditions.

There is no indication that the leaving group Y remains bound in the monomolecular layer, as is apparent from infrared spectra of the SAMs (see Example 8). Most likely, it leaves during the adsorption process in the form of an anion Y ^" and its nature does not affect the properties of the monomolecular layer formed, whose structure can have the form (CH ₃ ) _¾ (Ci8H ₃₇ )^Sn-Au or (CH^dsH^Sn-Au _o , where o is larger than 1. If the group Y departs in the form of an anion, it can be assumed that the electron needed for the generation of the anion originates in the metal, but this has not been explicitly proven.

The monomolecular self-assembled layers formed according to the present invention are much thinner than similar layers formed from 1 -octadecanethiol and are less organized, as was shown by measurements of contact angle with water (Example 7), by weaker electron transfer blocking (Example 9), by lower ability to withstand the removal of the monolayer using certain solvents (Example 4), by ellipsometric measurement of thickness (Example 6), and by measurement of infrared spectra (Example 8).

The permeability of the monomolecular layers prepared according to the present invention is higher than the permeability of a similar layer prepared from 1 -octadecanethiol, as is apparent particularly from Example 9, which shows the measurement of electron transfer blocking (Figure 1).

Monomolecular self-assembled layers formed according to the present invention are more stable to solvents such as rc-hexane and ethanol, but not to methylene chloride, water (Figure 2), or basic reagents. Surprisingly, they are not desorbed when exposed to 80 - 90% acids. Superior stability relative to layers prepared from 1 -octadecanethiol was demonstrated also for oxidants, especially hydrogen peroxide, but partly also for ambient air. Monomolecular layers formed according to the present invention resist strong reductants, such as sodium borohydride (Example 4).

The properties examined were almost identical for the self-assembled trialkylstannyl monolayers formed from molecules carrying one, two, or three long chains, and only electron transfer blocking displayed some differences (Example 9). The similarity extends to the ellipsometric layer thickness (Example 6). The results could be slightly misleading in that the index of refraction of the monolayers formed from 7, 6, and 5 (octadecyldimethylstannyl, dioctadecylmethylstannyl, trioctadecylstannyl) probably increases slightly in that order, but in general the observed thickness of 0.6 to 0.7 nm represents only about 1/3 of the thickness measured for a layer formed from 1- octadecanethiol, suggesting that the alkyl chains lie on the gold surface.

The substrates used were commercial glass plates coated with a 200 nm layer of gold purchased from Platypus Technologies, Inc., and similar results were obtained with other materials, such as gold on mica, platinum and silver on glass, and copper on glass.

In addition to using substrates coated with a thin layer of gold, substrates coated with thin layers of silver, palladium, and platinum can also be used.

Brief Description of the Figures

Fig. 1 : Cyclic voltammograms at 100 mV/s of Fe(CN) ₆ ^3"/4" on bare gold electrode (dotted line) and on an electrode with an adsorbed layer of 1 (black), 5c (gray), 6c (light gray), and 7c (dark gray) after 2 h immersion in a solution of the compound. The x axis shows the potential of the working electrode in mV relative to Ag/AgCl reference and they axis shows the current in μΑ.

Fig. 2: Stability of adsorbed monolayers of compounds 1 (black), 5 (gray), 6 (light gray), and 7 (dark gray) to overnight immersion in various reagents followed by rinsing and drying. The y axis displays R, the per cent monolayer remaining on the gold surface, calculated as the ratio of the final to the initial integrated intensity of the 2800-3000 cm ^"1 bands. The x axis contains the solutions used. A, dry CH ₂ C1 ₂ ; B, wet CH ₂ C1 ₂ ; C, rc-hexane; D, ethanol; E, water; F, 0.1 M H ₂ S0 ₄ ; G. 0.1 M NaOH; H, 1 mM KMn0 ₄ ; I, 30% H ₂ 0 ₂ ; J, 10 mM NaBH ₄ ; K, seven days in laboratory air.

Fig. 3 : Partial degradation of monolayer 7c upon exposure to reagents: CH ₂ C1 ₂ (A, ■), ethanol (D, V), water (Ε,·), 30% (v/v) H ₂ 0 ₂ (I,♦). The y axis shows R, the per cent monolayer remaining on the gold surface, calculated as the ratio of the final to the initial integrated intensity of the 2800-3000 cm ^" ' bands. The x axis shows time in h.

Fig. 4 A,B: Thermal desorption of monolayers of 1 (black), 5c (gray), 6c (light gray), and 7c (dark gray) on gold covered substrate after 1 h at 20, 80, 140 and 200 °C. A: IR spectra of adsorbed 1 (black) in the upper part and 7c (dark gray) in the lower part; B: per cent of adsorbed layer remaining on the gold surface, calculated as the ratio R of the final to the initial integrated intensity in the 2800-3000 cm ^"1 region (shown on the y axis) plotted against the temperature used (shown in °C on the x axis).

Fig. 5 A,B: Kinetics of monolayer formation at room temperature (20 °C) for compounds 5 - 7 compared with 1 -octadecanethiol (1, V). A: Effect of the leaving group: triflate 7a (_), trifluoroacetate 7b ( ), and tosylate 7c (■). B: Effect of the number of long chains on the Sn atom in the tosylates 5c (Δ), 6c (·), and 7c (■). The x axis shows time in min and the y axis shows the thickness of the monomolecular layer in nm.

Fig. 6 A,B: Ellipsometric thickness (A) and static contact angle of water (B) on adsorbed monolayers of 1 and 5 - 7 on gold covered substrate after 2 h immersion in solution of the compound.

Fig. 7: ATR-FTIR spectra of monolayers of the compounds 1 (black), 5 (gray), 6 (light gray), and 7 (dark gray) adsorbed on gold covered substrate. The x axis shows the wave number in crn 'and the y axis shows the absorbance. From top to bottom are shown the spectral curves of monolayers produced from 1, 7a, 7b, 7c, 6a, 6b, 6c, 5a, 5b, 5c, successively shifted by 0.01 absorbance units. Examples

Example 1

Synthesis of Trialkylstannyl Salts

Tetraalkyltin compounds 2 (Meals, J. Org. Chem. 1944, 9, 21 1), 3, and 4 were synthesized from commercial tin chlorides and the corresponding Grignard reagent.

2 x = 0, y = 4

(CH ₃ ) _x Sn(C-)eH3 ) _y 3 x = 2, y = 2

4 x = 1 , y = 3

1 = 1-octadecanethiol

Tin triflates (5a, 6a, and 7a), trifluoroacetates (5b, 6b and 7b), and tosylates (5c, 6c and 7c) were prepared by refluxing the tetraalkylstannanes 2 to 4 with the corresponding acid in dichloromethane solution (Sasin et al., J. Org. Chem. 1958, 23, 1366), cf. Scheme 1. A slight excess of trifluoroacetic acid was used to compensate for its volatility.

Scheme 1. Synthesis of trialkylstannyl salts

Y-H

(CH ₃ ) _x Sn(C (CH ₃ ) _x (C ₁₈ H ₃₇ ) _y Sn-Y

2 x = 0, y = 4 x = 0, y = 3 5a 5b 5c OTf

3 x = 2. y = 2 x = 1 , = 2 6a 6b 6c OCOCF ₃

4 x = 3, y = 1 x = 2, y = 1 7a 7b 7c OTs

Materials were purchased from Sigma-Aldrich or Alfa-Aesar and used as received unless otherwise specified. 1-Bromooctadecane was distilled over CaCl ₂ , tetrahydrofuran (THF), toluene and pentane were distilled from benzophenone ketyl, dichloromethane (DCM) and acetonitrile (ACN) were freshly distilled over CaH ₂ or P ₂ 0 ₅ , trifluoromethanesulfonic acid was distilled over a small amount of P ₂ 0 ₅ . The resulting salts were characterized as follows. NMR spectra were measured on a Bruker 400 MHz Ultrashield, a Bruker Avance II 500 MHz, and a Bruker Avance II 600 MHz and referenced to the solvent signal. FTIR spectra were recorded on a Bruker Equinox 55. Melting points were taken with Stuart SMP3 instrument or Kofler block and are uncorrected, mass spectrometry data were obtained with LTQ Orbitrap XL or LCQ Fleet (both Thermo Fisher Scientific, or Q-TOF micro (Waters), or Reflex IV (Bruker Daltonics) instruments. Elemental analyses were measured on Perkin Elmer 2400 II instrument.

Example 1A

Methyltrioctadecyltin (2), cf. Mulcahy et aL J Phys. Chem. 2010, 114, 14050. Mg turnings (0.96 g, 0.04 mol) and a catalytic amount of I ₂ were taken in a three neck round bottom flask fitted with a reflux condenser. The flask was heated with a heat gun until purple vapors of I ₂ appeared. THF (20 mL) was added, followed by 1-bromooctadecane (13.6 mL, 0.04 mol) and the mixture was heated to reflux under argon atmosphere for 5 h. After cooling to room temperature SnCl ₄ (0.7 mL, 0.006 mol) was added dropwise. The mixture was refluxed for 5 h and then stirred overnight at room temperature. Hexane (50 mL) was added and the reaction mixture was washed with a saturated NH ₄ C1 solution (15 mL), water (3 x 25 mL) and saturated NaCl solution (15 mL). The hexane layer was dried over anhydrous Na ₂ S0 ₄ , filtered and concentrated under reduced pressure to obtain a white solid which was crystallized from ethyl acetate (6.0 g, 90 % yield). Mp (melting point) 47 - 49 °C (lit. ²⁵ 47 °C). IR (KBr, cm ^-1 ): 2956, 2918, 2873, 2850, 1468, 1455, 1418, 1378, 1342, 721, 598, 513, 501. 1H NMR (499.8 MHz, CDC1 ₃ ): δ 0.72 (m, 8H), 0.81 (t, J= 7Hz, 12H), 1.05 - 1.35 (m, 120H), 1.41 (m, 8H). ¹³ C NMR (125.7 MHz, CDCl ₃ ): 6 9.17, 14.13, 22.72, 27.02, 29.32, 29.41, 29.71, 29.74, 29.77, 31.97, 34.49. ¹¹⁹ Sn NMR (186.4 MHz, CDC1 ₃ ): δ -13.45. Elemental analysis: calcd for C ₇₂ H ₁₄₈ Sn: C 76.35, H 13.17, found: C 77.03, H 13.45. Example IB

Dioctadecyldimethyltin (3). Mg turnings (0.76 g, 0.031 mol) and a catalytic amount of I ₂ were taken in a three neck round bottom flask fitted with a reflux condenser and heated with a heat gun until the purple color of I ₂ appeared. THF (20 mL) was added, followed by 1 -bromooctadecane (10.84 mL, 0.03 mol). The mixture was heated to reflux for about 5 h, and a solution of Me ₂ SnCl ₂ ( 2.5 g, 0.011 mol) in THF (5 mL) was added dropwise. The mixture was refluxed for 5 h and stirred overnight at room temperature. Hexane (50 mL) was added and the reaction mixture was washed with saturated NH ₄ C1 solution (15 mL), water (3 x 25 mL) and saturated NaCl solution (15 mL). The hexane layer was dried over anhydrous Na ₂ S0 ₄ , filtered and concentrated under reduced pressure to obtain a white solid, which was crystallized from ethyl acetate (6.92 g, 93 % yield). Mp 39

- 41 °C. IR (KBr, cm ^"1 ): 2955, 2917, 2871, 2849, 1472, 1463, 1452, 1440, 1421, 1413, 1377, 1193, 1 186, 761, 729, 720, 526, 518. Ή NMR (499.8 MHz, CDC1 ₃ ): δ 0.01 (s, 6H), 0.81 (m, 4H), 0.89 (t, J= 7 Hz, 6H), 1.21-1.35 (m, 60H), 1.49 (m, 4H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ): δ -11.14, 10.50, 14.13, 22.72, 26.80, 29.31, 29.40, 29.68, 29.70, 29.75, 31.96, 34.18. ¹¹⁹ Sn NMR (186.4 MHz, CDC1 ₃ ): δ -2.98. HRMS-EI calcd for C ₃₇ H ₇₇ Sn (M

- CH ₃ ): 641.5047, found: 641.5063. Example 1C

Octadecyltrimethyltin (4). Mg turnings (0.18 g, 7.53 x 10 ^"3 mol) and a catalytic amount of I ₂ were placed in a three neck round bottom flask fitted with a reflux condenser, which was then heated with a heat gun until the purple color of I ₂ appeared. THF (20 mL) was added, followed by 1 -bromooctadecane (2.57 mL, 7.53 x 10 ^"3 mol), and the mixture was heated to reflux for about 5 h. Then a solution of Me ₃ SnCl (1.0 g, 5.02 x 10 ^"3 mol) in THF (5 mL) was added dropwise. The mixture was refluxed for 5 h and stirred overnight at room temperature. Hexane (50 mL) was added, the reaction mixture was washed with a saturated NH ₄ C1 solution (15 mL), water (3 x 25 mL) and saturated NaCl solution (15 mL). The hexane layer was dried over anhydrous Na ₂ S0 ₄ , filtered and concentrated under reduced pressure to obtain a semisolid. Purification by column chromatography furnished the product (1.66 g, 79 % yield). IR (KBr, cm ^'1 ): 2956, 2923, 2852, 1466, 1455, 1444, 1433, 1417, 1378, 1200, 1 188, 764, 731, 720, 525, 510. ^! H NMR (499.8 MHz, CDC1 ₃ ): δ 0.06 (s, 9H), 0.85 (m, 2H), 0.90 (t, J= 7.1 Hz, 3H), 1.23 - 1.37 (m, 30H), 1.52 (m, 2H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ): δ -10.32, 11.15, 14.17, 22.78, 26.78, 29.38, 29.48, 29.74, 29.77, 29.80, 29.82, 32.04, 34.12. ¹¹⁹ SnNMR (186.4 MHz, CDC1 ₃ ): δ -1.21. HRMS-APCI calcd for C ₂₁ H ₄₆ SnNa (M + Na): 441.2525, found: 441.2647.

Example ID

Trioctadecyltin Triflate (5a). Tetraoctadecyltin (1 g, 8.85 x 10 ^"4 mol) was dissolved in warm dry DCM and solution of CF ₃ S0 ₃ H (0.025 g, 0.18 x 10 ^'3 mol) in dry DCM (1 mL) was added dropwise, followed by a 48 h reflux. DCM was evaporated on rotatory evaporator to obtain a white solid which was crystallized from DCM/ACN (0.82 g, 90 % yield). Mp 77 - 78 °C. IR (KBr, cm ^"1 ): 2956, 2918, 2872, 2850, 1468, 1378, 1321,1261, 1203, 1097, 1025, 804, 721, 656, 632, 582, 518. 1H NMR (499.8 MHz, THF- d _& ): δ 0.89 (t, J = 7 Hz, 9H), 1.29 (bs, 92H), ¹³ C NMR (125.7 MHz, THF-i/ ₈ ): δ -14.63, 20.78, 23.73, 26.56, 30.35,309.49, 30.79, 30.80, 30.83, 30.85, 30.86, 35.15, 121.07, (q, J _c- F=319.1). ¹⁹ F NMR (470.3 MHz, THF-<¾): δ -75.16. ¹¹⁹ SnNMR (186.4 MHz, CDC1 ₃ ): δ -

15.75. HRMS-ESI calcd for C ₅₄ Hi iiSn (M - OTf): 879.7702, obsd: 879.7713. Elemental analysis: calcd for C ₅₅ Hi _U F ₃ 0 ₃ SSn: C, 64.24; H, 10.88; F, 5.54; S, 3.12, found C 64.33; H

10.76, F 5.97, S 3.47.

Example IE

Trioctadecyltin Trifluoroacetate (5b). Tetraoctadecyltin (lg, 8.83 x 10 ^"4 mol) was dissolved in warm DCM (10 mL). A solution of trifluoroacetic acid (0.15g, 1.33 x 10 ^" mol) was added dropwise from a syringe. The mixture was refluxed for 48 h. An aliquot was removed and the product crystallized from DCM/ACN. ¹¹⁹ Sn NMR spectrum showed that some starting material was still present. The reaction was continued and again an aliquote was removed and the product crystallized from DCM/ACN. Again, ^{1 19} Sn NMR spectrum showed the presence of some starting material. Then, 0.5 equiv of trifluoroacetic acid was added and the reaction was continued for two more days. Again an aliquote was removed and the ¹¹⁹ Sn NMR spectrum of the product showed no indication of the presence of the starting material. The reaction mixture was concentrated under reduced pressure, the residue was dissolved in DCM, and ACN was added. A pale yellow solid was obtained (0.81 g, 92 % yield). Mp 70 - 72 °C. IR (KBr, cm ¹ ): 2956, 2919, 2850, 1670, 1650, 1468, 1379, 1213, 1191, 1159, 852, 840, 795, 728, 696, 675, 605, 523, 475. Ή NMR (499.8 MHz, CDCl ₃ ): 5 0.88 (t, J= 7.0 Hz, 9H), 1.20 - 1.36 (m, 90H), 1.40 (m, 6H), 1.67 (m, 6H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ): 6 14.11, 17.71, 22.70, 25.36, 29.14, 29.39, 29.52, 29.65, 29.68, 29.72, 29.73, 31.94, 33.99, 1 15.18, (q, J _c-F = 288.1), 161.33, (q, J _c-F = 39.9). ¹⁹ F NMR (470.3 MHz, CDC1 ₃ ): δ -71.02. ¹¹⁹ Sn NMR (186.4 MHz, CDC1 ₃ ): δ 172.31. HRMS- APCI calcd for C ₅₄ H ₇₇ Sn (M - CCOCF ₃ ): 879.7702, found: 879.7719. Elemental analysis: calcd for C ₅ 6H _{i n} F ₃ 0 ₂ Sn: C 67.79; H 1 1.28; F 5.74; found C 68.16; H 11.56; F 6.11.

Example IF

Trioctadecyltin Tosylate (5c). A mixture of tetraoctadecyltin (1 g, 0.88 x 10 ^"3 mol) and -toluenesulfonic acid (0.152 g, 0.88 x 10 ^"3 mol) was refluxed in DCM (10 mL) for 48 h. The reaction mixture was concentrated under reduced pressure to obtain a thick liquid, which was dissolved in hot ethyl acetate and left overnight. A white solid precipitated, was filtered off and dried overnight under reduced pressure at 60 °C. A white solid was obtained (0.90 g, 98 % yield) Mp 56 - 58 °C. IR (KBr, cm ^-1 ): 2955, 2919, 2850, 1601, 1497, 1468, 1378, 1 194, 1 131, 1044, 1015, 815, 721, 694, 569. 1H NMR (499.8 MHz, CDC1 ₃ ): δ 0.89 (t, J= 6.8 Hz, 9Hz), 1.10 - 1.35 (m, 96H), 1.49 (bm, 6H), 2.36 (s, 3H), 7.16 (m, 2H), 7.60 (m, 2H). ^I3 C NMR (125.7 MHz,CDCl ₃ ): δ 14.11, 20.55, 21.37, 22.70, 25.53, 29.41 , 29.70, 29.71, 29.78, 29.79, 29.82, 29.84, 29.86, 29.89, 31.95, 34.12, 126.06, 128.98, 139.46, 141.61. HRMS-APCI calcd for C ₅₄ HniSn (M - OTs): 879.7702, found: 879.7697. Elemental analysis: calcd for C ₆ iHii ₈ 0 ₃ SSn: C 69.75; H 1 1.32; S 3.05; obsd C 70.16; H 1 1.76; S 2.83.

Example 1G

Dioctadecylmethyltin Triflate (6a). The procedure described in Example 1 D gave a 73 % yield. Mp 56 - 58 °C. IR (KBr, cm ^-1 ): 2954, 2917, 72850, 1468, 1414, 1378, 1257, 1 179, 1034, 769, 721, 654, 644, 581, 523. 1H NMR (600.1 MHz, CDC1 ₃ ): δ 0.70 (s, 3H), 0.88 t, J= 7.0 Hz, 6H), 1.22 - 1.38 (m, 60H), 1.41 (m, 4H), 1.68 (m, 4H). ¹³ C NMR (150.9 MHz, CDC1 ₃ ): δ -0.20,14.12, 21.39, 22.70, 25.14, 29.22, 29.39, 29.65, 29.69, 29, 29.75, 29.76, 29.77, 29.78, 31.94, 33.72, 118.80, (q, J _c-F 317.8). ¹⁹ F NMR (376.5 MHz, CDC1 ₃ ): δ -78.25. ¹¹⁹ Sn NMR (186.4 MHz, CDC1 ₃ ): δ 160.74. HRMS-ESI calcd for C ₃₇ H ₇₇ Sn (M - OTf): 641.5042, found: 641.5049. Elemental analysis: calcd for C ₃₈ H ₇₇ F ₃ 0 ₃ SSn: C 57.79; H 9.83; F 7.22, S 4.06, found C 57.74; H 10.1 1 ; F 7.30; S 4.02. Example 1H

Dioctadecylmethyltin Trifluoroacetate (6b). The procedure described in Example IE gave a 82 % yield. Mp 60 - 62 °C. IR (KBr, cm ^"1 ): 2955, 2919, 2850, 1670, 1650, 1468, 1378, 1194, 1158, 854, 840, 796, 728, 672, 604, 536, 473. Ή NMR (499.8 MHz, CDC1 ₃ ): 8 0.62 (s, 3H), 0.88 t, J= 7.1 Hz, 6H), 1.21 - 1.35 (m, 60H), 1.42 (m, 4H), 1.67 (m, 4H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ): δ -3.88, 14.1 1, 18.17, 22.70, 25.23, 29.1 1, 29.37, 29.49, 29.63, 29.65, 29.67, 29.71, 31.93, 33.79, 115.12, 161.24. ⁱ⁹ F NMR (470.3 MHz, CDC1 ₃ ): δ -71.05. ¹¹⁹ Sn NMR (186.4 MHz, CDC1 ₃ ): δ 183.89. HRMS-EI calcd for C ₃₈ H ₇₄ F ₃ 0 ₂ Sn: 739.4663, found: 739.4639. Elemental analysis: calcd for C ₃₉ H ₇₇ F ₃ 0 ₂ Sn: C 62.15; H 10.30; F 7.56; found C 62.31 ; H 10.55; F 7.27.

Example II

Dioctadecylmethyltin Tosylate (6c). The procedure described in Example IF gave a 85 % yield. Mp 39 - 41 °C. IR (KBr, cm ^-1 ): 2954, 2917, 2850, 1601, 1468, 1378, 1 192, 1 132, 1045, 1015, 815, 720, 695, 607, 568, 525. Ή NMR (499.8 MHz, CDC1 ₃ ): δ 0.56 (s, 3H), 0.89 (t, J= 6.9 Hz, 6H), 1.10 - 1.35 (m, 64H), 1.48 (m, 4H), 2.37 (s, 3H), 7.17 (m, 2H), 7.56 (m, 2H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ) δ -0.32, 14.1 1, 21.10, 21.38, 22.70, 25.43, 29.39, 29.70, 29.76, 29.78, 29.80, 29.82, 29.84, 29.86, 31.94, 33.91 , 125.95, 129.05, 139.38, 141.77. HRMS-APCI calcd for C ₃₇ H ₇₇ Sn (M - OTs): 641.5042, found: 641.5042. Elemental analysis: calcd for C ₄₄ H ₈₄ 0 ₃ SSn: C 65.09; H 10.43; S 3.95; found C 65.27; H 10.68; S 3.65.

Example U

Octadecyldimethyltin Triflate (7a). The procedure described in Example 1 D gave a 75 % yield. Mp 49 - 51 °C. IR (KBr, cm ^"1 ): 2952, 2915, 2849, 1470, 1376, 1258, 1179, 1039, 1034, 840, 768, 718, 655, 644, 580, 544, 520. Ή NMR (499.8 MHz, CDC1 ₃ ): δ 0.75 (s, 6H), 0.88 (t, J= 7 Hz, 3H), 1.21 - 1.38 (m, 30H), 1.42 (m, 2H), 1.68 (m, 2H). ^I3 C NMR (125.7 MHz, CDC1 ₃ ): δ 0.82, 14.11, 21.56, 22.69, 24.99, 29.15, 29.37, 29.58, 29.67, 29.70, 29.72, 29.73, 31.93, 33.47, 118.67 (q, J _c-F = 317.5). ¹⁹ F NMR (470.3 MHz, CDC1 ₃ ): δ - 74.04. HRMS-ESI calcd for C ₂₀ H ₄₃ Sn (M - OTf): 403.2381, found: 403.2383. Elemental analysis: calcd for C ₂₁ H ₄₃ F ₃ 0 ₃ SSn: C 45.75; H 7.86; F 10.34; S 5.82; found C 45.60; H 7.99; F 10.32; S 5.69. Example IK

Octadecyldimethyltin Trifluoroacetate (7b). The procedure described in Example IE gave a 89 % yield. Mp 79 - 81 °C. IR (KBr, cm ^"1 ): 2953, 2921, 2851, 1678, 1652, 1467, 1453, 1379, 1215, 1201, 1 191 , 1154, 854, 839, 796, 770, 726, 667, 605, 555, 526, 476. 1H NMR (499.8 MHz, CDC1 ₃ ): δ 0.63 (s, 6H), 0.84 (t, J= 7.0 Hz, 3H), 1.20 -1.30 (m, 30H), 1.41 (m, 2H), 1.65 (m, 2H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ): δ -2.69, 14.12, 18.52, 22.69, 25.14, 29.08, 29.36, 29.46, 29.60, 29.63, 29.66, 29.69, 31.92, 33.63, 115.04, 161.18. ¹⁹ F NMR (470.3 MHz, CDC1 ₃ ): δ -71.09. ^{1 19} Sn NMR (186.4 MHz, CDC1 ₃ ): δ 191.57. HRMS- EI calcd for C ₂ ]H ₄₀ F ₃ O ₂ Sn: 501.2002, found: 501.2001. Elemental analysis: calcd for C ₂₂ H ₄₃ F ₃ 0 ₂ Sn: C 51.28; H 8.41; F 11.06, found C 51.29; H 8.53; F 10.89.

Example 1L

Octadecyldimethyltin Tosylate (7c). The procedure described in Example IF gave a 74 % yield. Mp 62 - 64 °C. IR (KBr, cm ^-1 ): 2953, 2919, 2850, 1600, 1497, 1470, 1397, 1378, 1191, 1131, 1044, 1014, 815, 781, 761 719, 695, 568, 542, 521, 496. 1H NMR (499.8 MHz, CDC1 ₃ ): δ 0.56 (s, 6H), 0.88 (t, J= 6.8 Hz, 3H), 1.10 - 1.33 (m, 32H), 1.46 (bm, 2H), 2.38 (s, 3H), 7.19 (m, 2H), 7.55 (m, 2H). ¹³ C NMR (125.7 MHz, CDC1 ₃ ) δ 0.77, 14.10, 21.24, 21.39, 22.68, 25.29, 29.31, 29.37, 29.64, 29.68, 29.73, 29.75, 29.76, 29.78, 31.92, 33.67, 125.30, 129.14, 139.09, 142.00. ^{1 19} Sn NMR (186.4 MHz, CDC1 ₃ ): δ 83.08. HRMS-EI calcd for C ₂₇ H ₅₀ O ₃ SSn:574.2503, found: 574.2493. Elemental analysis: calcd for C ₂₇ H ₅₀ O ₃ SSn: C 56.55; H 8.79; S 5.59; found C 56.26; H 8.74; S 5.20.

Example 2

Product Structure

One of the tetraalkylstananes was known (2, cf. Mulcahy et al., J. Phys. Chem. C 2010, 114, 14050) and the analytical data for the others were in agreement with a tetrahedral arrangement around the tin atom (3, 4).

The ¹¹⁹ Sn NMR chemical shifts for CDC1 ₃ solutions of such stannyl salts have values ranging from 75 to 165 ppm, and this is regarded as typical of a quasitetrahedral arrangement in trimethyltin(IV) compounds with a four-coordinate tin atom (Kapoor et al., Polyhedron 1995, 14, 489). All trifluoroacetates (5b, 6b, 7b), the triflate 6a, as well as the tosylate 7c belong to this category. The triflate 5a was measured in THF and the strong upfield shift (~ 200 ppm) is characteristic of the formation of a five-coordinate complex with solvent donor molecules (Chandrasekhar and Thirumoorthi, Eur. J. Inorg. Chem. 2008, 4578).

The spectral data of compounds 5 - 7 are summarized in Table 1. It was not possible to detect ¹¹⁹ Sn signals for the tosylates 5c and 6c and the triflate 7a. Table 1. NMR data for stannyl salts 5 - 7

a ⁾ In CDCI3 solvent in all cases except 5a, for which THF-d ₈ was used. ^b) NMR signal for ¹¹⁹ Sn not found

ppm = parts per million Example 3

Adsorbed Monolayer Formation

Glass substrates coated with 200 nm thick layer of gold were purchased from Platypus Technologies. Prior to use they were cleaned in a piranha solution (3 : 1 sulfuric acid:hydrogen peroxide) at 90 °C, rinsed with copious amounts of 18.2 ΜΩ H ₂ 0 and absolute ethanol and dried under a stream of nitrogen. Trialkylstannyl monolayers were formed by immersing the gold substrates in a 1 x 10 ^"5 M solution of a trialkylstannyl salt in dry CH ₂ C1 ₂ for ~2 h. After removal from solution, the gold substrates were rinsed thoroughly with either absolute ethanol or CH ₂ C1 ₂ and dried under a stream of nitrogen prior to analysis.

Monolayers of 1 -octadecanethiol (1) were formed on the gold substrates by immersing the substrate in a 1 x 10 ^"5 M solution in absolute ethanol for ~ 2 h. After removal from the solution, the gold substrates were rinsed thoroughly with either absolute ethanol or CH ₂ C1 ₂ and dried under a stream of nitrogen prior to analysis.

Example 4

Stability of Monolayers The self-assembled monolayers (SAMs) formed from the salts 5 to 7 were prepared as described in Example 3. For stability measurements the gold slide was immersed for 15 to 18 hours at room temperature in dry CH ₂ C1 ₂ , wet CH ₂ C1 ₂ , rc-hexane, ethanol, water, 0.1 M H ₂ S0 ₄ , 0.1 M NaOH, 1 mM KMn0 ₄ , 30% (v/v) H ₂ 0 ₂ , or 10 mM NaBH ₄ , removed from the solution, rinsed thoroughly with either appropriate solvent (CH ₂ C1 ₂ , hexane, ethanol) or with copious amounts of 18.2 ΜΩ H ₂ 0 and absolute ethanol in the case of water solutions, and dried under a stream of nitrogen, before an IR spectrum was recorded.

The samples were also exposed to the ambient laboratory atmosphere for 7 days, then rinsed with absolute ethanol and dried before IR spectrum was measured. The measurement involved monitoring the loss of their IR absorbance between 2800 and 3000 cm ^"1 as a function of time after exposure either to the laboratory atmosphere for a week or to various solvents for 15 to 18 h, thorough rinsing, and drying.

The results for the trialkylstannyl monolayers were compared with those obtained for a SAM built from 1 -octadecanethiol (1). Numerous initial tests with all three leaving groups showed that they all gave identical results, and those shown in Fig. 2 were obtained with tosylates as averages of at least three measurements.

Even in the cases in which the monolayers were partly or fully desorbed, we did not observe the appearance of any new absorption bands. In Fig. 3, we show the time dependence of the loss of IR intensity for the adsorbed layer of 7 in water, ethanol, methylene chloride and hydrogen peroxide.

Example 5

Thermal Desorption

The thermal desorption of the trialkylstannyl monolayers was followed by keeping the monolayer under nitrogen for 1 h each at 80, 140, and finally 200 °C. Before the measurement of an IR spectrum the gold covered substrates were rinsed with absolute ethanol and dried in a stream of nitrogen.

The observed decrease in overall IR signal intensity is shown in Fig. 4. No IR spectral shape changes resulted for the monolayers formed from 5 to 7, but the spectrum of the SAM of 1 changed after heating to 80 °C and became essentially identical with the spectra of 5 to 7. Example 6

Ellipsometric Determination of Layer Thickness

All measurements were made using a Variable Angle Stokes Ellipsometer (Gaertner Scientific) with a 633 nm HeNe laser with the incident angle adjusted to 70°. Optical constants of the gold substrates were taken for all freshly cleaned substrates. An index of refraction of 1.47 was assumed for the films. EUipsometry measurements were taken at a minimum of five different areas on each sample.

Figure 5 shows the gradual increase in the ellipsometric thickness of a surface layer when a gold substrate is immersed into a solution of trialkylstannyl salts 5 to 7. The adsorption was self-limiting and the growth of the adsorbed layer stops after about an hour under the conditions used. It appears that the tosylates form an adsorbed layer the fastest, but we have detected no significant difference in the ultimate thickness nor any other properties of the layers formed from the stannyl triflates, trifluoroacetates, and tosylates (Fig. 5A).

It seems that the compounds carrying only one long alkyl chain on the tin atom adsorb the fastest, but the number of long chains on the tin atom has little if any effect on the final observed ellipsometric thickness (Fig. 5B).

The limiting ellipsometric thickness of 0.6 to 0.7 nm for the adsorbed layers of the alkylstannyl derivatives 5 to 7 and the value of approximately 2.2 nm for 1 -octadecanethiol (1), reported in Figure 6A, have only relative significance since we do not know the optical constants of the trialkylstannyl structures. The values were determined with the refractive index assumed to be 1.47, a value typical of tetraalkylstannanes with relatively short alkyl chains (Mulcahy et al., J Phys. Chem. C 2010, 114, 14050).

Example 7

Contact Angle A static contact angle of H ₂ 0 (18.2 ΜΩ) was found with a CAM101 instrument

(KSV Instruments) using a 1 to 2 ih drop of water. Measurements were taken at a minimum of five different areas for each sample.

Contact angles of water on the trialkylstannyl monolayers 5 to 7 were 92° to 97°, distinctly lower than the 1 13° observed for a 1 -octadecanethiol monolayer (Figure 2B). There was again no significant dependence on the nature of the leaving group or the number of long chains on the tin atom. Example 8

Infrared Spectroscopy

FTIR-ATR spectra (1000 scans, 4 cm ^"1 resolution) were recorded using a Nicolet 6700 FT-IR spectrometer (Thermo Electron Corporation) with liquid N ₂ cooled MCT detector in the range 650 - 4 000 cm ^"1 . The data were collected using a Seagull variable- angle accessory (Harrick Scientific Inc.) and a Ge hemisphere (12.5 mm diameter). Prior to each measurement the Ge crystal was cleaned with ethanol and a reference spectrum of the crystal in contact with air was measured.

Single reflection attenuated total reflectance (ATR) Fourier transform infrared spectroscopy was used to confirm the presence of the monolayers on the gold surface. The recorded IR spectra of monolayers formed from the salts 5 to 7 on gold show only vibrations attributable to the CH ₃ and CH ₂ groups (stretching at 2855, 2926, 2961 cm ^"1 and bending at 1378, 1418 and 1468 cm ^"1 ) and no evidence for the typical peaks of the triflate, trifluoroacetate, or tosylate residues was detected.

The C-H stretching vibration region between 2800 and 3000 cm ^"1 is particularly informative, and Fig. 7 therefore shows the spectra of 5 to 7 on a gold covered substrate, along with that of a 1 -octadecanethiol (1) monolayer for comparison.

For 5 to 7, the v _as (CH ₂ ), v _s (CH ₂ ), v _as (CH ₃ ), and v _s (CH ₃ ) bands are observed at 2926, 2855, 2961, and 2871 cm ^"1 , respectively, regardless of the nature of the leaving group. The v _s (CH ₃ ) band is weak in the spectrum of 1 and present only as an indistinct shoulder in the spectra of 5 to 7. Example 9

Electrochemical Measurements

To provide information about permeability of the adsorbed layers, we measured their electrochemical blocking properties by cyclic voltammetry (CV) using the [Fe(CN) ] ^3" /[Fe(CN) ₆ ] ^4" system as a redox probe. Electrochemical measurements were performed using an AutoLab PGSTAT302N potentiostat (Metrohm Autolab). Electrochemistry was carried out in a conventional three-electrode glass cell at room temperature. A gold electrode, either bare or carrying an adsorbed layer, a platinum wire, and an Ag/AgCl (saturated KC1) were used as the working, auxiliary, and reference electrodes, respectively. The blocking experiments were performed with a 0.1 M KC1 solution containing 2 mM K ₃ [Fe(CN) ₆ ] purged with Ar before measurement. A scan rate 100 mV/s was used in all measurements.

Fig. 1 compares the responses to 2 mM [Fe(CN) ₆ ] ^" in 0.1 M KC1 observed on a bare gold electrode and on an electrode that had been immersed for 2 h in a solution of a trialkylstannyl tosylate. The adsorbed layer reduces the current response relative to the bare electrode, especially when the tin atom carries three long chains, but the blocking is not complete even after immersion overnight.

In contrast, a SAM of 1 suppresses the electrochemical response essentially completely and only the charging current remains.

Industrial Applicability

The self-assembled trialkyltin monolayers adsorbed on metallized substrate through a tin atom are useful for instance as active partners in electronic components.

Literature cited

1. Love, J. C; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103. 2. Ulman, A. Chem Rev. 1996, 96, 1533.

3. Ulman, A. Ultrathin Organic Films, Academic Press: San Diego, 1991.

4. Joseph, Y.; Guse, B.; Nelles, G. Chem Mater. 2009, 21, 1670.

5. Willey, T. M.; Vance, A. L.; van Buuren, T.; Bostedt, C; Terminello, L. J.; Fadley, C. S. Surface Science, 2005, 576, 188.

6. Huang, F. K.; Horton, R. C; Myles, Jr, D. C; Garrell, R. L. Langmuir 1998, 14, 4802.

7. Ciszek, J. W.; Stewart, M. P.; Tour, J. M. J. Am. Chem. Soc. 2004, 126, 13172.

8. Shaporenko, A.; Ulman, A.; Terfort, A.; Zharnikov, M. J. Phys. Chem. B 2005, 109, 3898.

9. Monnell, J. D.; Stapleton, J. J.; Jackiw, J. J.; Dunbar, T.; Reinerth, W. A.; Dirk, S. M; Tour, J. M.; Allara, D. L.; Weiss, P. S. J. Phys. Chem. B. 2004, 108, 9834.

10. Clark, B. K.; Standard, J. M.; Gregory, B. W.; Hall, A. D. Surface Science 2002, 498, 285.

11. Samant, M. G.; Brown C. A.; Gordon, J. G. II Langmuir 1992, 8, 1615.

12. Protsailo, L. V.; Fawcett, W. R.; Russell, D.; Meyer, R. L. Langmuir 2002, 18, 9342.

13. Han, S. W.; Kim, K. J. Colloid Interface Sci. 2001, 240, 492.

14. Weidner, T.; Shaporenko, A.; Mueller, J.; Hoeltig M.; Terfort, A.; Zharnikov, M. J. Phys. Chem. C 2007, 111, 1 1627.

15. Owens, T. M.; Suezer, S.; Banaszak Holl, M. M. J. Phys. Chem. B 2003, 107, 3177. 16. Owens, T. M.; Nicholson, K. T.; Banaszak Holl, M. M.; Suezer, S. J. Am. Chem. Soc.

2002, 124, 6800.

17. Katsonis, N.; Marchenko, A.; Taillemite, S.; Fichou, D.; Chauraqui, D.; Aubert, C; Malacria, M. Chem. Eur. J. 2003, 9, 251 A.

18. Zhang, S.; Chandra, K. L.; Gorman, C. B. J. Am. Chem. Soc. 2007, 129, 4876.

19. Shewchuk, D. M.; McDermott, M. T. Langmuir 2009, 25, 4556.

20. Laforgue, A.; Addou, T.; Belanger, D. Langmuir 2005, 21, 6855.

21. Stapleton, J. J.; Daniel, T. A.; Uppili, S.; Cabarcos, O. M.; Naciri, J.; Shashidhar, R.; Allara, D. L. Langmuir 2005, 21, 11061. Ί

24

22. Zheng, X.; Mulcahy, M. E.; Horinek, D.; Galeotti, F.; Magnera, T. F.; Michl, J. J. Am. Chem. Soc. 2004, 126, 4540.

23. Mulcahy, M. E.; Magnera, T. F.; Michl, J. J. Phys. Chem. C 2009, 113, 20698.

24. Mulcahy, M. E.; Bastl, Z.; Stensrud, K. F.; Magnera, T. F.; Michl, J., J. Phys. Chem. C 2010, 114, 14050

25. Meals, R. N. J. Org. Chem. 1944, 9, 21 1.

26. Sasin, G. S.; Borror, A. L.; Sasin, R. J. Org. Chem. 1958, 23, 1366.

27. Kapoor, R.; Sood, V.; Kapoor, P. Polyhedron 1995, 14, 489.

28. Chandrasekhar, V.; Thirumoorthi, R. Eur. J. Inorg. Chem. 2008, 4578.