Identification of (phosphine)gold(I) hydrates and their equilibria in wet solutions

Abstract

The elusive gold(I) hydrates [Ph₃PAu(OH₂)]⁺TfO⁻ and [((Ph₃PAu)₂OH)₂]²⁺(TfO⁻)₂ resulting from the most common gold(I) pre-catalyst Ph₃PAuOTf have been characterized, leading to disclosure of the equilibria between gold oxo species in wet solutions.

---

Gold(I) complexes have attracted tremendous attention in recent years, owing to their excellent catalytic activity in a broad range of chemical transformations¹ and their potential as therapeutic agents² and functional materials.³ A theoretical framework has been put forward to rationalize the unique properties of these complexes.^1c,4 As homogeneous catalysts, bicoordinate linear complexes LAuX (L is most commonly a tertiary phosphine R₃P and X represents a weakly coordinating counteranion such as ⁻OTf, ⁻NTf₂, BF₄⁻ or SbF₆⁻) are used predominantly as the pre-catalysts. In place of the counteranion X⁻, unsaturated C–C bonds could coordinate to [LAu]⁺ and are thus activated for nucleophilic addition.¹ A number of the resting intermediates containing LAu–C bonds en route to the final products have been identified recently, supporting the idea that [LAu]⁺ is the initial catalytic species.^1g Recent results indicate that the counteranion X⁻ or an insoluble silver salt (remaining from the preparation of the pre-catalyst) could exert a significant effect on certain transformations.⁵ However, the catalytic species in these reactions are yet to be investigated.

On the other hand, the gold(I) complexes and their catalyzed reactions are known to be moisture insensitive and are applied mostly without exclusion of moisture or even with water as a solvent. It is therefore important to know whether water reacts with the gold(I) complexes, thus affecting their catalytic activity and turnover rate. In fact, the reaction of [R₃PAu]⁺ with water was reported first in 1980 to afford [(Ph₃PAu)₃O]⁺ salt in alkaline or acid media.^6,7 Treatment of a relevant [(o-Tol)₃PAu)₃O]⁺BF₄⁻ (o-Tol = ortho-methylphenyl) with (o-Tol)₃PAu⁺BF₄⁻ led to [((o-Tol)₃PAu)₄O]²⁺(BF₄⁻)₂, which is isolobal to the elusive H₄O²⁺ species.⁸ Replacement of R₃P with more electron-donating N-heterocyclic carbene (NHC) as the ligand has enabled the preparation of IPrAuOH (IPr = N,N′-bis(2,6-diisopropylphenyl)-imidazol-2-ylidene).⁹ Subjection of IPrAuOH to aqueous HBF₄ solution resulted in [(IPrAu)₂OH]⁺BF₄.¹⁰ Based on these characterized gold(I) oxo species, a reaction pathway of [LAu]⁺ with water could be proposed, as shown in Fig. 1. In this pathway, the key intermediate [LAu(OH₂)]⁺ has yet to be identified.¹¹ In addition, with the most frequently used ligand R₃P, the corresponding LAuOH and (LAu)₂OH⁺ species have not been characterized.

Fig. 1 Proposed equilibria of the gold(I) oxo species in the presence of water. Note: [LAu(OH₂)]⁺ and [(LAu)₂OH]⁺ characterized in the present work represent the first examples of the isolobal series of [H₃O]⁺ with R₃P ligands; R₃PAuOH and (LAu)₂O are yet to be identified.

Preparation of Ph₃PAuOTf has become routine practice in our laboratory since the development of a powerful gold(I)-catalyzed glycosylation protocol with glycosyl ortho-alkynylbenzoates as donors.¹² Accidentally, crystalline needles (1a, Fig. S1, ESI†) were observed when a batch of newly prepared Ph₃PAuOTf in analytical grade CH₂Cl₂ (containing ∼100 ppm water) was concentrated slowly under reduced pressure. Compound 1a was suspected to be the elusive gold(I) hydrate [LAu(OH₂)]⁺OTf⁻ based on NMR analysis; however, numerous attempts to grow a crystal of 1a suitable for single-crystal X-ray crystallography failed. In fact, complex 1a is hygroscopic and decomposes readily at room temperature. Fortunately, when we switched Ph₃P with the bulkier (o-Tol)₃P as the ligand, the corresponding homolog 1b was to be found stable and could be obtained as single crystals (see ESI†). X-ray diffraction analysis confirmed it to be the gold(I) hydrate [(o-Tol)₃PAu(OH₂)]⁺OTf⁻ (Fig. 2). Thus, the coordination of a water molecule to the (phosphine)gold(I) atom is unambiguously confirmed. In crystal 1b, the Au–O bond length and the P–Au–O angle are 2.069(6) Å and 174.7(2)°, respectively.

Fig. 2 ORTEP diagram of complexes 1b, 2a, and 2b with 30% probability ellipsoids (hydrogen atoms and the TfO⁻ in 2a are omitted for clarity). Key bond lengths (Å) and angles (°) for 1b: Au–O 2.070(7), Au–P 2.208(2), P–Au–O 174.6(3); 2a: Au1–O1 2.041(4), Au2–O2 2.033(5), Au1–O1–Au2 126.3(4), Au3–O2–Au4 127.7(6); 2b: Au1–O1 2.047(3), Au2–O1 2.070(3), Au3–O2 2.070(3), Au4–O2 2.042(2), Au1–Au3 3.397, Au1–Au4 3.486, Au2–Au3 3.1035, Au2–Au4 3.2798(3), Au1–O1–Au3 129.09(16), Au2–O2–Au4 123.34(14).

An attempt at recrystallization of complex 1a in CH₂Cl₂/i-Pr₂O led to colorless tetragonal dipyramid shaped crystals (2a) and colorless rectangular shaped crystals (2b) (Fig. S7, ESI†) (Scheme 1). These crystals were separated manually and determined to be the stereoisomers of [((Ph₃PAu)₂OH)₂]²⁺(TfO⁻)₂ (Fig. 2). We later found that the formation of these products was time-dependent in the presence of 5 Å molecular sieves in the initial reaction system. Thus, stirring the mixture of Ph₃PAuCl and AgOTf (with 5 Å MS) for 1 h followed by crystallization led to 2a/2b in yields of up to 65%; while stirring for 6 h followed by addition of i-Pr₂O resulted in a white precipitate, which crystallized in various solvents to provide [(Ph₃PAu)₃O]⁺TfO⁻ (3) (Fig. S12 and S13, ESI†).¹³

Scheme 1 Preparation of the gold(I) oxo complexes 1–3.

Both complex 2a and 2b are tetranuclear dimers with a quasi-tetrahedral array of the four gold atoms in crystals. Complex 2a has a highly symmetrical structure (point group S₄). The Au–Au distance in each monomer unit is 3.65 Å, indicating the absence of any significant aurophilic interaction. The Au–Au bonds between the two monomers units are 3.1939(5) Å and 3.2191(6) Å, showing a typical aurophilic interaction. Complex 2b contains two dinuclear monomer units with different geometries. The Au–Au distances in and between the monomers are 3.71 Å/3.67 Å and 3.1035(3) Å/3.2798(3) Å, respectively, showing again that aurophilic interactions occur only between the two monomer units.

Previous structural studies of trigold oxonium cations [(R₃P)Au₃O]⁺ reveal a similar tetrahedral contact of the monomers with trimethylphosphine ligands (R = Me), whereas all other salts of this type (R = Ph etc.) form rectangular dimers, or are monomeric. Theoretical studies^13c show that for bulkier ligands such as triphenylphosphine (R = Ph), the rectangular structure is clearly favored, due to steric intermonomer repulsion. Our results here show that in the case of gold oxo complexes 2a and 2b, a “tetrahedral” dimer could be formed even with bulkier ligands such as triphenylphosphine. These structures provide new cases of isolobal species with an isostructural core.

It is noted that hydrogen bonds between the hydroxyl group and the CF₃SO₃⁻ anion were observed in all three complexes. This interaction may facilitate crystal growth. In fact, all attempts to obtain crystals of the (phosphine)gold(I) complexes with non-coordinating anions such as BF₄⁻ and SbF₆⁻ failed.¹⁴ In addition, the present dimeric (phosphine)gold(I) hydrates 2a/2b are different from the NHC based hydroxo-bridged dinuclear cation [((NHC)Au)₂OH]⁺, which exists as a monomer in crystals.^10b Another related example is the di[gold(I)]halonium salts X[Au(PR₃)]₂⁺ (X = Cl, Br, I). With large anion SbF₆⁻ they exist as tetranuclear dications, while with small anions such as BF₄⁻ and ClO₄⁻, they occur as a dinuclear monomer.¹⁵ These results indicate that the association between polyaurated onium salts depends on both ligand and anion, and hydrogen bonding may also play an important role.

The ³¹P and ¹H NMR spectra of gold(I) oxo complexes 1a, 2a/2b and 3 were measured (see ESI†). Interestingly, the ³¹P signals shift upfield gradually as the number of [Ph₃PAu]⁺ cations coordinated with oxygen increases (Fig. 3). The ¹H NMR signals of 1a, 2a/2b, and 3 in the aromatic region have different shapes and positions (Fig. S16, ESI†), despite the fact that they contain Ph₃P as the sole ligand. Both the ³¹P and ¹H NMR spectra of 2a and 2b are identical, implying that they adopt the same solution structure.

Fig. 3 ³¹P NMR chemical shifts of the gold oxo complexes (1a, 2a/2b, and 3) and Ph₃PAuOTf/Ph₃PAuNTf₂ in CD₂Cl₂ in the presence of varied amounts of water.

Based on the data of the authentic samples, the reaction of water with Ph₃PAuOTf and Ph₃PAuNTf₂,¹⁴ two widely used homogeneous gold(I) pre-catalysts,¹ was investigated by ³¹P and ¹H NMR spectrometry. Experimentally, water was added gradually to a freshly prepared CD₂Cl₂ solution of Ph₃PAuOTf (or Ph₃PAuNTf₂), and the ³¹P and ¹H NMR spectra were recorded after each addition (Fig. S18–21, ESI†).

Key results obtained from these experiments include: (a) in all cases, ³¹P NMR spectra showed only one single sharp signal, indicating the existence of a rapid equilibrium between interchanging species, as is also suggested by the crystallization experiments under various conditions (see ESI†). (b) For Ph₃PAuOTf bearing weakly coordinating anion TfO⁻, [Ph₃PAu(OH₂)]⁺TfO⁻ (1a) formed readily when small amounts of water were introduced into the solution. As the amount of water increased, [((Ph₃PAu)₂OH)₂]²⁺(TfO⁻)₂ (2) formed gradually, accompanied by the generation of H₃O⁺:

[Ph₃PAu]⁺ + H₂O → [Ph₃PAu(OH₂)]⁺

2[Ph₃PAu(OH₂)]⁺ → [(Ph₃PAu)₂OH]⁺ + H₃O⁺

When 100 μL water was added, both the ³¹P and ¹H NMR spectra indicated that the major species in the solution was 2. A titration experiment at this point showed the generation of 0.5 eq of H₃O⁺, in good agreement with the theoretical amount (Table S1, ESI†). (c) For Ph₃PAuNTf₂ bearing a relatively strong coordinating anion ⁻NTf₂,¹⁶ the coordination of nitrogen to the gold(I) atom greatly inhibited the hydration process, as indicated by the small change in the ³¹P NMR chemical shift against the continuous increase in water volume (Fig. 3).

The coordination of water to [Ph₃PAu]⁺ in solution was also observed by ¹H NMR measurement (see ESI†). For Ph₃PAuOTf in CD₂Cl₂ containing 1.0 eq water, the water signal shifted downfield from 1.52 (in pure CD₂Cl₂) to 4.56 ppm (Fig. S17, ESI†). In contrast, the water (1.0 eq) signal in AuPPh₃NTf₂ solution showed up only slightly downfield at 1.69 ppm. These results strongly suggest that the coordination of [Ph₃PAu]⁺ with NTf₂⁻ is much more favorable than with water in solution.

The existence of dynamic equilibria rather than a single species in solution was further supported by crystallization experiments under various conditions (Table S2, ESI†). Complex 1a was formed only in non-coordinating solvents, whereas in oxygen-containing solvents the major product was complex 2. Interestingly, recrystallization of complex 2 in wet non-coordinating solvents led to complex 3, with only a small amount of complex 2 remaining. These results show clearly that the reaction equilibrium is strongly affected by the nature of the solvent.

The role of molecular sieves (which is always used in the glycosylation reaction to remove moisture)¹² in the present reaction was briefly investigated. When 5 Å MS was added to a mixture of AuPPh₃Cl and AgOTf (1.0 eq) in CH₂Cl₂, the ³¹P signals shifted gradually upfield from 28.04 ppm to 25.95 ppm as the stirring time went on (Fig. S22, ESI†). Crystallization experiments confirmed that the major product changed gradually from 1a to 2a and then to 3 (Table S2, ESI†). Apparently, the molecular sieves serve as a mild base, which slowly neutralize the acid generated in situ, leading to a shift in equilibrium.

In conclusion, the formation of (phosphine)gold(I) hydrates could be a ready process in the presence of water. This finding provides a good explanation for the stability of the gold(I) complexes known to chemists, so that LAuNTf₂,¹⁴ [LAu-triazole]X¹⁷ and LAu(NCMe)X,^1e,18 which resist hydration, have been found stable, while the readily hydrolysable LAuX (X⁻ = ⁻OTf, BF₄⁻ and SbF₆⁻)¹⁴ are found unstable. More importantly, the present finding shall afford a new visual angle for looking into the mechanism of gold(I) catalysis in certain reactions. Thus, the dramatic difference between the catalytic behaviour of Ph₃PAuOTf and Ph₃PAuNTf₂ might correlate to their disparate reactivities toward moisture;^5a the previous proposal of new gold(I) species based on the ³¹P NMR signal might have overlooked this hydration process.^5a Additionally, awareness of the hydration process of gold(I) complexes will also be helpful in understanding their biological and material properties.

Acknowledgements

We thank the National Natural Science Foundation of China (20932009 and 20921091) and the Ministry of Science and Technology of China (2012ZX09502-002) for funding and X. B. Leng and J. Sun for X-ray crystallographic analysis.

References

1. For selected reviews, see: (a) A. S. K. Hashmi and G. J. Hutchings, Angew. Chem., Int. Ed., 2006, 45, 7896; (b) A. Fürstner and P. W. Davies, Angew. Chem., Int. Ed., 2007, 46, 3410; (c) D. J. Gorin and F. D. Toste, Nature, 2007, 446, 395; (d) D. J. Gorin, B. D. Sherry and F. D. Toste, Chem. Rev., 2008, 108, 3351; (e) E. Jiménez-Núñez and A. M. Echavarren, Chem. Rev., 2008, 108, 3326; (f) S. P. Nolan, Acc. Chem. Res., 2011, 44, 91; (g) L.-P. Liu and G. B. Hammond, Chem. Soc. Rev., 2012, 41, 3129.
2. I. Ott, Coord. Chem. Rev., 2009, 253, 1670.
3. J. C. Lima and L. Rodriguez, Chem. Soc. Rev., 2011, 40, 5442.
4. (a) P. Pyykkö, Angew. Chem., Int. Ed., 2004, 43, 4412; (b) H. Schmidbaur and A. Schier, Chem. Soc. Rev., 2012, 41, 370; (c) H. G. Raubenheimer and H. Schmidbaur, Organometallics, 2012, 31, 2507.
5. (a) D. Wang, R. Cai, S. Sharma, J. Jirak, S. K. Thummanapelli, N. G. Akhmedov, H. Zhang, X. Liu, J. L. Petersen and X. Shi, J. Am. Chem. Soc., 2012, 134, 9012; (b) H. Schmidbaur and A. Schier, Z. Naturforsch., B: J. Chem. Sci., 2011, 66b, 329.
6. A. N. Nesmeyanov, E. G. Perevalova, Yu. T. Struchkor, M. Yu. Antipin, K. I. Grandberg and V. P. Dyadchenko, J. Organomet. Chem., 1980, 201, 343.
7. This type of gold(I)-oxo complexes have been used in catalysis, see for example: (a) B. D. Sherry and F. D. Toste, J. Am. Chem. Soc., 2004, 126, 15978; (b) B. D. Sherry, L. Maus, B. N. Laforteza and F. D. Toste, J. Am. Chem. Soc., 2006, 128, 8132; (c) R. L. LaLonde, W. E. Brenzovich, D. Benitez, E. Tkatchouk, K. Kelley, W. A. Goddard III and F. D. Toste, Chem. Sci., 2010, 1, 226.
8. (a) H. Schmidbaur, S. Hofreiter and M. Paul, Nature, 1995, 377, 503; (b) K. Nomiya, T. Yoshida, Y. Sakai, A. Nanba and S. Tsuruta, Inorg. Chem., 2010, 49, 8247.
9. S. Gaillard, A. M. Z. Slawin and S. P. Nolan, Chem. Commun., 2010, 46, 2742.
10. (a) S. Gaillard, J. Bosson, R. S. Ramón, P. Nun, A. M. Z. Slawin and S. P. Nolan, Chem.–Eur. J., 2010, 16, 13729; (b) R. S. Ramón, S. Gaillard, A. Poater, L. Cavallo, A. M. Z. Slawin and S. P. Nolan, Chem.–Eur. J., 2011, 17, 1238.
11. G. N. Khairallah, R. A. J. O'Hair and M. I. Bruce, Dalton Trans., 2006, 3699.
12. (a) Y. Li, Y. Yang and B. Yu, Tetrahedron Lett., 2008, 49, 3604; (b) Y. Zhu and B. Yu, Angew. Chem., Int. Ed., 2011, 50, 8329; (c) J. Zhang, H. Shi, Y. Ma and B. Yu, Chem. Commun., 2012, 48, 8679; (d) B. Yu, J. Sun and X. Yang, Acc. Chem. Res., 2012, 45, 1227.
13. (a) Y. Yang, V. Ramamoorthy and P. R. Sharp, Inorg. Chem., 1993, 32, 1946; (b) K. Angermaier and H. Schmidbaur, Inorg. Chem., 1994, 33, 2069; (c) S.-C. Chung, S. Krüger, H. Schmidbaur and N. Rösch, Inorg. Chem., 1996, 35, 5387.
14. N. Mézailles, L. Ricard and F. Gagosz, Org. Lett., 2005, 7, 4133.
15. (a) R. Uson, A. Laguna and M. V. Castrillo, Synth. React. Inorg. Met.-Org. Chem., 1979, 9, 317; (b) A. Hamel, N. W. Mitzel and H. Schmidbaur, J. Am. Chem. Soc., 2001, 123, 5106; (c) H. Schmidbaur, A. Hamel, N. W. Mitzel, A. Schier and S. Nogai, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 4916.
16. S. Antoniotti, V. Dalla and E. Duñach, Angew. Chem., Int. Ed., 2010, 49, 7860.
17. H. Duan, S. Sengupta, J. L. Petersen, N. G. Akhmedov and X. Shi, J. Am. Chem. Soc., 2009, 131, 12100.
18. E. Herrero-Gómez, C. Nieto-Oberhuber, S. López, J. Benet-Buchholz and A. M. Echavarren, Angew. Chem., Int. Ed., 2006, 45, 5455.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, characterization data, and NMR spectra for new compounds. Crystallographic data for compounds 1b, 2a, 2b, 3 H₂O, and 3 1/2CH₂Cl₂. CCDC reference numbers 892405–892409. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c2ra22282e

---