Structural studies of gold(I, II, and III) compounds with pentafluorophenyl and tetrahydrothiophene ligands

Article abstract of DOI:10.1002/anie.200604592

(Figure Presented) A golden opportunity was seized to prepare an unbridged, dinucleur gold(II) compound, [Au₂(C₆F₅) ₄(tht)₂] (tht = tetrahydrothiophene), without any stabilizing chelating ligands (see picture; Au orange, S yellow, F green, C black) and to crystallographically characterize the gold(I) and gold(III) complexes participating in this conversion. A unique ligand scrambling of the latter gold(III) compound occurs in solution.

Full text of DOI:10.1002/anie.200604592

Angewandte
Chemie
DOI: 10.1002/anie.200604592
Gold Compounds
Structural Studies of Gold(I, II, and III) Compounds with
Pentafluorophenyl and Tetrahydrothiophene Ligands**
Jacorien Coetzee, William F. Gabrielli, Karolien Coetzee, Oliver Schuster, Stefan D. Nogai,
Stephanie Cronje, and Helgard G. Raubenheimer*
The first tetrahydrothiophene (tht) complex of gold was
reported by Cattalini et al. in 1968,[1] and Vaughan and
Sheppard were the first to obtain a gold complex that
contained the pentafluorophenyl (pfp) group as a ligand two
years later.[2] Since then and mainly up to the 1980s, Usón and
co-workers—with the Laguna brothers and Vicente prom-
inent—have developed the chemistry of such compounds. The
pfp group is now commonly used as a robust and generally
inert ligand (compared to other aryl groups), whereas tht, a
labile ligand towards gold, is commonly employed for
Scheme 1. Formation of crystalline products 2 and 3a from reaction
enabling substitution reactions. Several synthetic procedures
(a)(assuming 2 formed from 1 and 3)and 2 from reaction (b); in (c)
involving these ligands are described in two books edited by
3 rearranges to form 4 and also crystallizes as 3b.
King and Eisch.[3] Later applications are available in various
books and review articles.[4] It is surprising that no crystal
structure determinations of gold complexes that contain both
of these important ligands coordinated to a gold center have
been described. Herein, as a tribute to Usón and others, we
rectify that shortcoming by reporting for the first time the
crystal and molecular structures of both [Au(C6F5)(tht)]( 1)
and [Au(C6F5)3(tht)]( 3) as well as an unprecedented example
of an unbridged, unchelated dinuclear gold(II) compound,
[Au2(C6F5)4(tht)2] (2), that could be obtained in a rational
manner. Finally, we discuss a unique rearrangement of 3 in
solution that effects ionic compound formation between two
gold(III) complexes in [Au(C6F5)4][Au(C6F5)2(tht)2] (4). Such
mononuclear ligand scrambling, although fairly common for
gold(I) compounds, has never before been documented for
gold(III) complexes. It is noteworthy, however, that the
Fackler and Schmidbaur groups have independently de-
scribed the ligand dynamics (which involve isomerizations
and rearrangements) in various ylide-bridged gold(III) com-
plexes.[5]
oxidation states. From this mixture the gold(II) compound
[Au2(C6F5)4(tht)2] (2), representing the first example of an
unbridged complex with a formal [Au2]4+ core in which the
gold centers are not stabilized by any chelating ligands, as well
as the gold(III) complex [Au(C6F5)3(tht)]( 3a, one polymorph
of compound 3; see below) were crystallized. Upon standing
(ca. 24 h), gold metal was visible in the form of a gold mirror
on the walls of the Schlenk tube. In a similar reaction, 1 was
treated with 2-bromo-5-methylthiophene, which also resulted
in oxidation of the gold(I) precursor to eventually furnish
crystals of 3a. We subsequently prepared the gold(II)
compound 2 from an extremely slow redox reaction between
equimolar quantities of compounds 1 and 3. The formation of
compound 2 in this reaction could be confirmed by unit-cell
determinations with single-crystal X-ray diffraction of various
crystals isolated from the mixture. In addition, unreduced
[Au(C6F5)3(tht)]crystallized unaltered from the solution,
thereby effecting the structure 3b (a second polymorph of
compound 3; see below), while some of 3 underwent
spontaneous ligand rearrangement to afford ionic compound
4, which consists of two gold(III) complexes. Uncertainty as to
the existence of 2 in solution exists, since all attempts to
obtain NMR spectroscopic data were unsuccessful. Further-
more, FAB mass spectrometric data obtained for crystals of
compound 2 also failed to deliver any identifiable peaks.
In the most spectacular of the crystalline pfp–tht complex
As part of an ongoing investigation of preferential gold
coordination to various N-heterocycles, [Au(C6F5)(tht)]was
treated with the organic bromide 5-(5-bromothiophen-2-yl)-
2-phenethyl-2H-tetrazole in a 1:1 ratio (Scheme 1). This
reaction afforded a mixture that contained gold in various
[*] J. Coetzee, Dr. W. F. Gabrielli, K. Coetzee, Dr. O. Schuster,
Dr. S. D. Nogai, Dr. S. Cronje, Prof. H. G. Raubenheimer
Department of Chemistry and Polymer Science
University of Stellenbosch
À
products, [(C6F5)2(tht)Au Au(tht)(C6F5)2] (2), the gold atoms
occur in two distorted square-planar configurations, featuring
an essentially linear S-Au-Au-S axis and two orthogonally
oriented {C6F5-Au-C6F5} moieties with all aromatic rings in
parallel planes (Figure 1). The tht ligands are situated at a
torsion angle of 938 with respect to each other, on either side
Private Bag X1, Matieland, Stellenbosch 7602 (South Africa)
Fax: (+27)21-808-3850
E-mail: hgr@sun.ac.za
[**] The authors thank the National Research Foundation and the
Alexander von Humbolt Foundation for financial support.
II
II
À
of the S1-Au1-Au1’-S1’ backbone. The Au Au bond length
(2.5679(7) Š) is within the range of interactions displayed by
the only other reported examples of unbridged, but chelated,
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 2497 –2500
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2497

Communications
We have now established that ligand scrambling also takes
place for a mononuclear organogold(III) compound accord-
ing to Equation (2), and other examples may be discovered in
the future. The ionic compound [Au(C6F5)4][Au(C6F5)2(tht)2]
(4; Figure 2), the result of such a ligand scrambling, represents
the first gold(III) compound containing pfp and/or tht in both
of the composite ions. The coordination geometry of both
Figure 1. Molecular structure of 2; thermal ellipsoids are set at 25%
probability and hydrogen atoms are omitted for clarity. Selected bond
lengths [Š] and angles [8]: C111-Au1 2.110(2), C121-Au1 2.078(2),
S1-Au1 2.418(3), Au1-Au1’ 2.5679(7); S1-Au1-Au1’ 173.58 (7),
C111-Au1-S1 90.7(1), C121-Au1-S1 89.9(1), C111-Au1-Au1’ 91.53(6),
C121-Au1-Au1’ 88.14(7), C111-Au1-C121 177.6 (2).
II
II
À
Au Au compounds, namely, bis(quinolin-8-ylthio)digold
dichloride
(2.5355(5) Š),[6]
[Au2Cl2(dppn)2][PF6]2
Figure 2. ORTEP representation of 4. Thermal ellipsoids are set at
50% probability. Hydrogen atoms and pentane molecules are omitted
for clarity. Selected bond lengths [Š] and angles [8]: C111-Au1 2.066(6),
S1-Au1 2.335(2), C121-Au2 2.051(6), C131-Au2 2.050(6), C141-Au2
2.065(6), C151-Au2 2.055(7); C111-Au1-S1 91.1(2), C111-Au1-S1’
88.9(2), C111’-Au1-S1 88.9(2), C111’-Au1-S1’ 91.1(2), C111-Au1-C111’
180.0, S1-Au1-S1’ 180.0, C121-Au2-C131 91.2(3), C121-Au2-C151
87.8(3), C131-Au2-C141 92.3(2), C141-Au2-C151 88.8(2),
(2.6112(7) Š;
dppn = 1,8-bis(diphenylphosphino)naphtha-
lene),[7] [Au2Br2(dppn)2][PF6]2 (2.6035(8) Š), and [Au2I2-
(dppn)2][PF6]2 (2.6405(8) Š).[8] Furthermore, the Au C
À
bonds (2.110(2), 2.078(2) Š) are significantly longer than
those observed for the anionic gold(I) compound [Au(C6F5)2]-
[nBu4N](2.044(4) Š), [9] probably reflecting the higher s char-
acter of the overlapping orbitals in gold(I) compared to
gold(II).[10] The formation of the dinuclear compound 2 from
1 and 3 can formally be interpreted in terms of C6F5 radical
transfer from the gold(III) center to the gold(I) center and
C121-Au2-C141 176.5(3), C131-Au2-C151 176.4(3).
À
simultaneous Au Au bond formation. Few examples of C6F5
reactivity in gold complexes, at least in terms of ligand
gold atoms is square-planar. The homoleptic anion has the
same propeller-like appearance as reported for [Au(C6F5)4]
transfer, are known.[11]
with comparable Au C bond lengths,
[15]
= =
[Ph3P N PPh3]
À
Crystals of the gold(III) compound 3 were obtained in two
polymorphic forms with both types of crystals featuring
similar, equivalent molecules wherein the donor atoms of the
ligands occur in a square-planar arrangement about the
central gold atoms. Molecular-structure representations of 3a
and 3b with selected bond lengths and angles are available in
the Supporting Information. The only obvious difference
between the two polymorphs appears to be the orientation of
the tht ligand. Owing to the large mutual trans influences of
ranging from 2.050(6) to 2.065(6) Š. The cation, on the
other hand, contains a AuIII center of the type trans-[AuR2L2]
(L = neutral ligand, R = anionic organic ligand) comparable
to the complex [Au(C6F5)2(tht)2]OTf.[16] The coordination
arrangement in the formed cationic complex represents a rare
example of a symbiotic relationship between ligands in a
square-planar d8 complex. Furthermore, the very strong
preference for cis isomers in [AuR2X2]compounds (according
to the Tobias orbital rationalization)[17] is refuted.
À
the two opposing pfp ligands, their Au C bond lengths
To complete a series of gold(I, II, and III) compounds
[Aux(C6F5)y(tht)z], the structure of the important starting
compound 1[18] was determined. The asymmetric unit contains
two independent monomers with similar paddle-like geo-
metries. These monomers form an infinite chain along the
a axis of the crystal through alternating long (3.306 Š) and
two different short (3.191, 3.128 Š) aurophilic interactions
(Figure 3). All of these separations fall within the range of
typically observed separations for aurophilic interactions
between gold(I) centers.[16,19] Identical pairs of molecules
are alternatingly rotated by 1808. Wide angles (142.88, 139.28)
between symmetry-related molecules and small torsion
(ꢀ2.06 Š) are longer than when the pfp ligand is coordinated
trans to tht (2.035(5), 2.031(3) Š). Similar results have been
obtained for [Au(C6F5)3(S2CPEt3)][12] and [AuMe3PPh3].[13]
À
À
The Au C and Au S bond lengths in compound 3 are in
close agreement with those observed in [Au(C6F5)3(S2CPEt3)]
(2.067(4), 2.076(4), 2.037(3) Š, and 2.366(1) Š, respectively).
Well-documented homoleptic rearrangements of gold(I)
complexes occur according to Equation (1) (charges omitted)
where X and L correspond to combinations of soft ligands.[14]
2498 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 2497 –2500

Angewandte
Chemie
of a Patterson synthesis (3a, 3), which yielded the position of the
metal atoms, and conventional difference Fourier methods. All non-
hydrogen atoms were refined anisotropically by full-matrix least-
squares calculations on F2 using SHELXL-97[26] within the X-seed
environment.[27,28] The hydrogen atoms were fixed in calculated
positions. X-seed was used to generate the various figures of the
complexes.
CCDC-626605–626609 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.[29–31]
Received: November 10, 2006
Published online: February 28, 2007
Keywords: aurophilicity · fluorinated ligands · gold ·
.
ligand scrambling · radical transfer
Figure 3. The organization of 1 in an infinite chain along the a axis.
Thermal ellipsoids are set at 25% probability. Hydrogen atoms and
ether molecules are omitted for clarity. Selected bond lengths [Š] and
angles [8]: C111-Au1 2.014(9), C121-Au2 2.03(1), S1-Au1 2.317(3),
S2-Au2 2.320(3), Au1···Au2 3.306(1), Au2···Au2’ 3.191(2), Au1’···Au1’’
3.128(2); C111-Au1-S1 176.3(3), C121-Au2-S2 178.0(3).
[1]L. Cattalini, M. Martelli, G. Marangoni, Inorg. Chem. 1968, 7,
1492 – 1495.
[2]L. G. Vaughan, W. A. Sheppard, J. Organomet. Chem. 1970, 22,
739 – 742.
[3]a) R. Usón, A. Laguna in Organometallic Synthesis, Vol. 3 (Eds.:
R. B. King, J. J. Eisch), Elsevier, Amsterdam, 1986, pp. 322 – 342;
b) R. Usón, A. Laguna in Organometallic Synthesis, Vol. 4 (Eds.:
R. B. King, J. J. Eisch), Elsevier, Amsterdam, 1988, pp. 342 – 353.
[4]For example, see: a) M. C. Gimeno, A. Laguna in Comprehen-
sive Coordination Chemistry II, Vol. 6 (Eds.: J. A. McCleverty,
T. J. Meyer) 1st ed., Elsevier, Oxford, 2004, pp. 990 – 1145; b) A.
Grohmann, H. Schmidbaur in Comprehensive Organometallic
Chemistry II (Eds.: E. W. Abel, F. G. A. Stone, G. Wilkinson), 1st
ed., Elsevier Science, Oxford, 1995, pp. 1 – 56.
[5]a) S. Wang, J. P. Fackler, Jr. in The Chemistry of Organic
Derivatives of Gold and Silver, Vol. 6 (Eds.: S. Patai, Z.
Rappoport) 2nd ed., Wiley, Chichester, 1999, pp. 431 – 450;
b) H. Schmidbaur, A. Grohmann, M. E. Olmos, A. Schier in The
Chemistry of Organic Derivatives of Gold and Silver, Vol. 6
(Eds.: S. Patai, Z. Rappoport), 2nd ed., Wiley, Chichester, 1999,
pp. 271 – 272.
[6]S. A. Yurin, D. A. Lemenovskii, K. I. Grandberg, I. G. Llꢀina,
L. G. Kuzꢀmina, Russ. Chem. Bull. Int. Ed. 2003, 52, 2752 – 2753.
[7]V. W. W. Yam, S. W. K. Choi, K. K. Cheung, Chem. Commun.
1996, 1173 – 1174.
[8]V. W. W. Yam, C. K. Li, C. L. Chan, K. K. Cheung, Inorg. Chem.
2001, 40, 7054 – 7058.
[9]P. G. Jones, Z. Kristallogr. 1993, 208, 347 – 350.
[10]L. E. Orgel, J. Chem. Soc. 1958, 4186 – 4190.
[11]S. Cronje, H. G. Raubenheimer, H. S. C. Spies, C. Esterhuysen,
H. Schmidbaur, A. Schier, G. J. Kruger, Dalton Trans. 2003,
2859 – 2866.
angles (42.858) between the independent monomers are
observed.
The small torsion angles of the latter result in greater
steric hindrance and weaker Au···Au interactions, as is evident
À
from the observed bond lengths. The Au S bond lengths
present in the structure (2.317(3) and 2.320(3) Š) do not
deviate significantly from those observed for [Au(tht)2]
[C6H4NO4S2](2.292(2) Š) and [Au(tht) 2][AuI2](2.306(7),
2.335(6) Š).[20,21] The observed Au C bond lengths
À
(2.014(9), 2.03(1) Š) can be compared to those in [Au(C6F5)-
[11]
=
=
{S CK N(H)C(CH3) C(H)SL }](2.06(1) Š).
Herein we have reported the first aryl- and thiophene-
containing gold(II) complex, which could be formed by
radical transfer and gold–gold bond formation, as well as a
gold(I) and two gold(III) compounds that contain the same
ligands. One of the latter products resulted from a novel
ligand-scrambling (metathesis or disproportionation) reac-
tion. In all the complexes (independent of charge) that
À
contain a {C6F5-Au-C6F5} unit, the Au C bond lengths
decrease in the order AuII > AuIII > AuI (ca. 2.11, 2.06, and
À
2.04 Š). A similar variation is found for the Au S separation
in the {tht-AuIII-C6F5}, {tht-AuIII-tht}, {tht-AuII-AuII}, and {tht-
AuI-C6F5} fragments (ca. 2.36, 2.34, 2.42, and 2.32 Š). Theo-
retical studies are underway.
[12]R. Usón, A. Laguna, M. Laguna, M. Luz Castilla, P. J. Jones, C.
Fittschen, J. Chem. Soc. Dalton Trans. 1987, 3017 – 3022.
[13]J. Stein, J. P. Fackler, Jr., C. Paparizos, H. W. Chen, J. Am. Chem.
Soc. 1981, 103, 2192 – 2198.
Experimental Section
[14]For example, see: a) S. Ahmad, Coord. Chem. Rev. 2004, 248,
231 – 243; b) S. Onaka, Y. Katsukawa, M. Shiotsuka, O. Kane-
gawa, M. Yamashita, Inorg. Chim. Acta 2001, 312, 100 – 110;
c) A. L. Hormann-Arendt, C. F. Shaw, Inorg. Chem. 1990, 29,
4683 – 4687.
[15]a) P. G. Jones, E. Bembenek, Z. Kristallogr. 1994, 209, 690 – 692;
b) H. H. Murray, J. P. Fackler, Jr., L. C. Porter, D. A. Briggs,
M. A. Guerra, R. J. Lagow, Inorg. Chem. 1987, 26, 357 – 363.
[16]F. Mohr, E. Cerrada, M. Laguna, Organometallics 2006, 25, 644 –
648.
The preparation of compounds 1–4 is described in detail in the
Supporting Information.
X-ray structure determinations and data collection: Suitable
single crystals for X-ray structural analysis were obtained by
crystallization from concentrated diethyl ether solutions layered
with n-hexane or n-pentane. Data sets for 1, 2, 3, and 4 were collected
on a Bruker SMART Apex CCD difractometer[22] with graphite-
monochromated MoKa radiation (l = 0.71073 Š). Data reduction was
carried out with standard methods from the software package Bruker
SAINT.[23] SMART data were treated with SADABS.[24, 25] The
structures were solved by direct methods (1, 3b, 4) or interpretation
[17]a) R. J. Puddephatt in Comprehensive Organometallic Chemistry
I, Vol. 2 (Eds.: E. J. Abel, F. J. Stone, G. Wilkinson), 2nd ed.,
Angew. Chem. Int. Ed. 2007, 46, 2497 –2500
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2499

Communications
Pergamon, Oxford, 1982, p. 762, p. 789, p. 801; b) R. S. Tobias,
Inorg. Chem. 1970, 9, 1296 – 1298.
sured, 5819 unique (Rint = 0.1020), goodness of fit = 0.988, R1 =
0.0512, wR2 = 0.0863, R indices based on 3266 reflections with
I > 2s(I) (refinement on F2), 352 parameters, 0 restraints.
[18]R. Usón, A. Laguna, M. Laguna, Inorg. Synth. 1989, 26, 85 – 91.
[19]H. Schmidbaur, Chem. Soc. Rev. 1995, 24, 391 – 400.
[20]S. Friedrichs, P. G. Jones, Acta Crystallogr. Sect. C 2000, 56, 56 –
57.
[21]S. Ahrland, B. Nor›n, A. Oskarsson, Inorg. Chem. 1985, 24,
1330 – 1333.
[30]Crystal data for
2
at 100 K: C32H16Au2F20S2, Mr =
1238.50 gmolÀ1, yellow cube, tetragonal, space group P41212
(No. 92), a = 10.8156(8), c = 28.675(4) Š, V= 3354.3(6) Š3, Z =
4, 1calcd = 2.452 gcmÀ3
,
F(000) = 2312, m(MoKa) = 9.000 mmÀ1
,
2qmax = 51.48, 12595 reflections measured, 3102 unique (Rint
=
[22]SMART, Data collection software (version 5.629), Bruker AXS
Inc.(Madison), WI, 2003.
[23]SAINT, Data reduction software (version 6.45), Bruker AXS
Inc. (Madison), WI, 2002.
0.0624), goodness of fit = 1.127, R1 = 0.0458, wR2 = 0.0900, R
indices based on 2824 reflections with I > 2s(I) (refinement on
F2), 190 parameters, 24 restraints. Absolute structure parame-
ter= 0.06(2) (H. D. Flack, Acta Crystallogr. Sect. A 1983, 39,
876 – 881).
[24]R. H. Blessing, Acta Crystallogr. Sect. A 1995, 51, 33 – 38.
[25]SADABS (version 2.05), Bruker AXS Inc. (Madison), WI, 2002.
[26]G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Analysis, University of Göttingen (Germany), 1997.
[27]J. L. Atwood, L. J. Barbour, Cryst. Growth Des. 2003, 3, 3 – 8.
[28]L. J. Barbour, Supramol. Chem. 2003, 1, 189 – 191.
[29]Crystal data for 1 at 100 K: C10H8AuF5S1·0.25C4H10O, Mr =
470.72 gmolÀ1, colorless needle, orthorhombic, space group
Pbnb (No. 56), a = 11.851(5), b = 18.664(8), c = 22.576(9) Š,
V= 4994(4) Š3, Z = 16, 1calcd = 2.504 gcmÀ3, F(000) = 3496, m-
(MoKa) = 11.993 mmÀ1, 2qmax = 56.48, 29015 reflections mea-
[31]Crystal data for
4
at 100 K: C44H16Au2F30S2·C5H12, Mr =
1644.77 gmolÀ1
,
colorless needle, triclinic, space group P1
¯
(No. 2), a = 11.541(3), b = 13.675(3), c = 17.772(4) Š, a =
74.360(4), b = 79.626(4), g = 66.427(3)8, V= 2467.3(9) Š3, Z =
2, 1calcd = 2.214 gcmÀ3
,
F(000) = 1564, m(MoKa) = 6.176 mmÀ1
,
2qmax = 52.98, 26187 reflections measured, 10081 unique (Rint
=
0.0479), goodness of fit = 1.047, R1 = 0.0432, wR2 = 0.0946, R
indices based on 8117 reflections with I > 2s(I) (refinement on
F2), 747 parameters, 0 restraints.
2500 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 2497 –2500