Dinuclear and tetranuclear palladium(II) complexes of a thiolato-functionalized, benzannulated N-heterocyclic carbene ligand and their activities toward suzuki-miyaura coupling

Article abstract of DOI:10.1021/om1008023

The thiolato-bridged dimeric Pd(II) benzimidazolin-2-ylidene complex 1, with a [Pd₂S₂] core, was conveniently prepared by the reaction of the thiol-functionalized benzimidazolium salt C and Pd(OAc) ₂. More straightforwardly, 1 can also be synthesized by direct treatment of the thioester-functionalized benzimidazolium salt B with Pd(OAc)₂ in wet DMSO under in situ hydrolysis of the thioester function. A subsequent salt metathesis reaction of 1 with AgO ₂CCF₃ afforded the mixed dicarboxylato/NHC analogue 2 in quantitative yield, leaving the sulfur bridges of the [Pd₂S ₂] core unaffected despite the use of the soft Ag(I) ions. Treatment of 1 with Me₃OBF₄ resulted in an unexpected bromido abstraction of 1 leading to an unusual rearrangement/dimerization reaction to give the tetranuclear NHC complex 3, which features a [Pd₄S ₄] macrocylic square with sulfur corners. These reactions demonstrate the structural diversity of the thiolato-functionalized N-heterocyclic carbene complexes and may offer access to metallo-NHC-based supramolecular architectures. A comparative catalytic study revealed the superiority of NHC/thiolato complex 2 over complexes 1 and 3 in aqueous Suzuki-Miyaura couplings at very low catalyst loading.

Full text of DOI:10.1021/om1008023

6020 Organometallics 2010, 29, 6020–6027
DOI: 10.1021/om1008023
Dinuclear and Tetranuclear Palladium(II) Complexes of a
Thiolato-Functionalized, Benzannulated N-Heterocyclic Carbene Ligand
and Their Activities toward Suzuki-Miyaura Coupling
Dan Yuan and Han Vinh Huynh*
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543,
Republic of Singapore
Received August 18, 2010
The thiolato-bridged dimeric Pd(II) benzimidazolin-2-ylidene complex 1, with a [Pd2S2] core, was
conveniently prepared by the reaction of the thiol-functionalizedbenzimidazolium saltC and Pd(OAc)2.
More straightforwardly, 1 can also be synthesized by direct treatment of the thioester-functionalized
benzimidazolium salt B with Pd(OAc)2 in wet DMSO under in situ hydrolysis of the thioester function.
A subsequent salt metathesis reaction of 1 with AgO2CCF3 afforded the mixed dicarboxylato/NHC
analogue 2 in quantitative yield, leaving the sulfur bridges of the [Pd2S2] core unaffected despite the
use of the soft Ag(I) ions. Treatment of 1 with Me3OBF4 resulted in an unexpected bromido abstraction
of 1leadingtoanunusualrearrangement/dimerizationreactiontogivethetetranuclearNHC complex3,
which features a [Pd4S4] macrocylic square with sulfur corners. These reactions demonstrate the
structural diversity of the thiolato-functionalized N-heterocyclic carbene complexes and may offer
access to metallo-NHC-based supramolecular architectures. A comparative catalytic study revealed the
superiority of NHC/thiolato complex 2 over complexes 1 and 3 in aqueous Suzuki-Miyaura couplings
at very low catalyst loading.
Introduction
P-donor groups is relatively common, whereas NHCs bearing
S-donors are still rare. So far, only a few thiolato-NHCs,3
thioether-NHCs,4 and thiophene-NHCs5 have been reported.
Among these, we are particularly interested in the first type
due to the versatile and diverse coordination chemistry of the
soft and electron-rich thiolato ligands in general.6 The mono-
anionic thiolato function also forms stronger M-S bonds
compared to the neutral thioether or thiophene moieties.
Current methodologies employed to prepare thiolato-NHC
complexes include either (i) oxidative addition across a C-S
bond using low-valent metal presursors3b,c or (ii) the reaction
withfree thiols.3a Bothmethods suffer from the limitation that
air-sensitive precursors, such as metal(0) complexes or free
thiols, are required. Furthermore, method (i) is restricted to
ligand precursors with C2-S bonds. In addition, all reported
examples are based on a saturated five- or six-membered
N-heterocyclic ring system, while those based on benzimida-
zole remain to be explored. In relation to our research on
benzimidazole-derived NHCs and their functionalization, we
herein present the synthesis of the thiolato-bridged Pd(II)
benzimidazolin-2-ylidene complex 1 from a benzimidazolium
salt with a thiol function. More conveniently, 1 can also be
synthesized from an air-stable thioester-functionalized benz-
imidazolium salt via in situ hydrolysis under aerobic condi-
tions. Reactivity studies of this complex toward Ag salts and
electrophiles led to the isolation of the carboxylato derivative
2 and the tetranuclear [Pd4S4] molecular square 3. The iso-
lation of the latter demonstrates an unusual reactivity of
N-heterocyclic carbenes (NHCs) have become common
ligands in organometallic chemistry and catalysis due to their
unique properties.1 A particularly interesting feature is that
NHCs can be modified in many ways by introducing func-
tionalities at the nitrogen atoms of the N-heterocyclic ring.
Various complexes of donor-functionalized NHCs and their
use in catalysis have been explored and recently reviewed.2
Toward this point, N-functionalization with N-, O-, and
*To whom correspondence should be addressed. E-mail: chmhhv@
nus.edu.sg.
(1) For a recent review, see: Hahn, F. E.; Jahnke, M. C. Angew.
Chem., Int. Ed. 2008, 47, 3122.
(2) For a recent review, see: Normand, A. T.; Cavell, K. J. Eur. J.
Inorg. Chem. 2008, 2781.
(3) (a) Sellmann, D.; Geipel, F.; Heinemann, F. W. Z. Anorg. Allg.
Chem. 2001, 627, 1034. (b) Cabeza, J. A.; del Rio, I.; Sanchez-Vega, M. G.;
Suarez, M. Organometallics 2006, 25, 1831. (c) Iwasaki, F.; Yasui, M.;
Yoshida, S.; Nishiyama, H.; Shimamoto, S.; Matsumura, N. Bull. Chem. Soc.
Jpn. 1996, 69, 2759.
ꢀ
ꢀ
(4) (a) Seo, H.; Park, H.-J.; Kim, B. Y.; Lee, J. H.; Son, S. U.; Chung,
Y. K. Organometallics 2003, 22, 618. (b) Wolf, J.; Labande, A.; Daran,
J. -C.; Poli, R. Eur. J. Inorg. Chem. 2007, 5069. (c) Roseblade, S. J.; Ros, A.;
ꢀ
ꢀ
Monge, D.; Alcarazo, M.; Alvarez, E.; Lassaletta, J. M.; Fernandez, R.
Organometallics 2007, 26, 2570. (d) Huynh, H. V.; Yeo, C. H.; Chew, Y. X.
Organometallics 2010, 29, 1479. (e) Huynh, H. V.; Yuan, D.; Han, Y. Dalton
Trans. 2009, 35, 7262. (f) Huynh, H. V.; Yeo, C. H.; Tan, G. K. Chem.
ꢀ
Commun. 2006, 3833. (g) Ros, A.; Monge, D.; Alcarazo, M.; Alvarez, E.;
Lassaletta, J. M.; Fernandez, R. Organometallics 2006, 25, 6039. (h) Fliedel,
C.; Schnee, G.; Braunstein, P. Dalton Trans. 2009, 2474.
(5) (a) Nielsen, D. J.; Cavell, K. J.; Viciu, M. S.; Nolan, S. P.; Skelton,
B. W.; White, A. H. J. Organomet. Chem. 2005, 690, 6133. (b) McGuinness,
D. S.; Suttil, J. A.; Gardiner, M. G.; Davies, N. W. Organometallics 2008, 27,
4238. (c) Huynh, H. V.; Chew, Y. X. Inorg. Chim. Acta 2010, 363, 1979.
ꢀ
(6) Krebs, B.; Henkel, G. Angew. Chem., Int. Ed. Engl. 1991, 30, 769.
r
pubs.acs.org/Organometallics
Published on Web 10/13/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 22, 2010 6021
Scheme 1. Preparation of Sulfur-Functionalized Ligand Precursors B and C
Scheme 2. Two Synthetic Pathways to Dinuclear Complex 1
Me3OBF4 as a halide abstraction reagent, which represents a
very useful metal-free alternative to the commonly used Ag-
and toxic Tl-based reagents. The catalytic activities of all these
thiolato-functionalized NHC Pd(II) complexes in the Suzuki-
Miyaura coupling reaction have also been investigated and
compared.
a base peak in the ESI mass spectrum at m/z 269 for the
molecularcation[M - Br]þ alsosupports the formation of C.7
Synthesis of Pd(II) Complexes. The reaction of thiol-
functionalized salt C with 1 equiv of Pd(OAc)2 in degassed
CH3CN under reflux overnight yielded the thiolato-bridged
dimeric Pd(II) benzimidazolin-2-ylidene complex 1 in a good
yield of 76% (Scheme 2, method 1). In this case, the solvent
CH3CN was chosen over the commonly used DMSO to
avoid oxidation of the free thiol function to disulfide. Com-
plex 1 is insoluble in CH3CN and can be readily isolated by
filtration.
More straightforwardly, we found that the direct reaction
of thioester salt B with 1 equiv of Pd(OAc)2 in DMSO at
80 °C overnight also yielded complex 1 in a similar yield of
74% (Scheme 2, method 2). Isolation of 1 from this reac-
tion is also easy, involving only a simple filtration step due to its
low solubility in DMSO. Column chromatography of the
saturated filtrate afforded a second crop of the yellow product.
Apparently, the methyl thioester function undergoes in situ
hydrolysis, probably assisted by free HOAc liberated from the
deprotonation of B with Pd(OAc)2. The resulting thiolate
coordinates the Pd(II) center, and subsequent dimerization
affords the desired thiolato-bridged Pd(II) carbene complex 1.
In this reaction, the thioester function acts as a thiol-protecting
group, simplifying the synthesis of thiolato complexes to a
great extent. Method 2 is preferred over method 1, since it
shortens the reaction sequence and also eliminates the require-
ment of working under an inert atmosphere.
Results and Discussion
Synthesis of the Ligand Precursors. Recently, we have
reported the synthesis of the bromo-functionalized benzimid-
azolium salt A from N-alkylation of N-benzylbenzimidazole,
which offered access to CSC-pincer ligands.4e In a similar
way, A can undergo nucleophilic substitution with 1 equiv of
KSCOCH3 in CH3CN to afford the thioester-functionalized
benzimidazolium salt B in a yield of 75% (Scheme 1).
The formation of the ligand precursor B is supported by
a base peak in the ESI mass spectrum at m/z 311 for the molec-
ular cation [M - Br]þ. Moreover, the 1H NMR spectrum in
DMSO-d6 of B shows two triplets at 4.75 and 3.44 ppm
assignable to the two inequivalent CH2 groups in the ethylene
chain of the N-substituent and a singlet at 2.27 ppm corre-
sponding to the methyl thioester protons. In comparison to
precursor A, the signals for the ethylene protons shifted
upfield due to the replacement of the bromide with a less
electron withdrawing thioester group. The 13C NMR spec-
trum of B shows signals for the procarbene and carbonyl
carbon atoms at 142.7 and 194.7 ppm, respectively.
The thioester group of B was easily hydrolyzed by aqueous
HBr to form the thiol-functionalized salt C (Scheme 1). In the
1H NMR spectrum of salt C in CD3CN, the absence of the
singlet assignable to the methyl thioester group of B and the
appearance of a new signal at 2.72 ppm for the free thiol
function corroborate the successful hydrolysis. Furthermore,
Complex 1 is soluble in CH2Cl2, CHCl3, DMSO, and
DMF, but insoluble in less polar solvents such as hexane
and diethyl ether. Its formation is supported by ESI mass
spectrometry, which shows a base peak at m/z 829 for the
[M - Br]þ complex fragment. In the 1H NMR spectrum, the
benzylic protons become diastereotopic upon complexation
and resonate as two doublets centered at 6.63 and 5.58 ppm,
respectively, with a coupling constant of 2J(H-H) = 15.6 Hz.
All four methylene protons of each ethanethiolato bridge are
also diastereotopic but give rise to two pseudodoublets and
(7) In addition, a strong peak at m/z 617 was observed, which can be
assigned to the oxidized disulfide compound associated with one bro-
mide counteranion.

6022 Organometallics, Vol. 29, No. 22, 2010
Yuan and Huynh
Figure 1. Molecular structure of 1 showing 50% probability ellipsoids: (top) top view; (bottom) side view. Hydrogen atoms and one
part of the disordered phenyl ring are omitted for clarity. The phenyl rings of the benzyl substituents have been omitted in the side view
˚
for clarity as well. Selected bond lengths (A) and bond angles (deg): Pd1-C1 1.977(4), Pd1-S1 2.2952(11), Pd1-S2 2.3705(10),
Pd1-Br1 2.4401(5), Pd2-C17 1.997(4), Pd2-S2 2.2941(10), Pd2-S1 2.3721(10), Pd2-Br2 2.4652(5); C1-Pd1-S1 91.17(11),
S1-Pd1-S2 80.32(4), S2-Pd1-Br1 97.46(3), Br1-Pd1-C1 91.44(11), C1-Pd1-S2 170.93(11), S1-Pd1-Br1 170.26(3), C17-
Pd2-S2 90.75(12), S2-Pd2-S1 80.31(4), S1-Pd2-Br2 96.33(3), Br2-Pd2-C17 92.65(11), C17-Pd2-S1 171.01(12), S2-Pd2-Br2
171.21(3), Pd1-S1-Pd2 82.49(3), Pd1-S2-Pd2 82.55(3).
two pseudotriplets ranging from 4.88 to 2.10 ppm due to
insufficient resolution and accidental overlap. In the 13C
NMR spectrum, the two equivalent carbene carbon atoms
resonate at 175.4 ppm.
alkylate the thiolato bridges. The metathesis reaction of mixed
halo/NHC Pd(II) complexes with Ag carboxylates is an esta-
blished method to synthesize mixed carboxylato-carbene
complexes as potentially useful catalyst precursors.8 However,
this methodology has not been extended to sulfur-function-
alized NHCs, due to the potential interaction of the soft sulfur
donor with the soft Ag(I) ion, which would compete with the
desired halide abstraction. On the other hand, the bridging
mode of the thiolato donor reduces the number of lone pairs
available at sulfur, which may prevent or lower the interference
with silver. To test this hypothesis, complex 1 was reacted with
2 equiv of AgO2CCF3 in CH3CN, as depicted in Scheme 3.
Indeed, this reaction cleanly afforded the dinuclear mixed
carbene-carboxylato Pd(II) complex 2 in quantitative yield
without affecting the [Pd2S2] core. The reaction mixture,
initially a yellow suspension, turned greenish yellow with the
precipitation of AgBr after stirring overnight. Complex 2
shows an improved solubility as compared to its precursor
complex 1, which is a common phenomenon observed for
mixed carbene-carboxylato Pd(II) complexes.8 It is soluble in
most organic solvents, with the exception of hexane. In the 13C
NMR spectrum, the resonance for the Ccarbene in 2 amounts to
171.4 ppm, which is shifted upfield due to the stronger shield-
ing effect of the trifluoroacetato ligand compared to that of the
bromido ligand in the precursor 1. Furthermore, the identity of
2 was also supported by a base peak in the ESI mass spectrum
at m/z 861 for the monocation [M - O2CCF3]þ and a less
intense peak at m/z 373 for the dication [M - 2 O2CCF3]2þ due
to stepwise loss of trifluoroacetato ligands.
Single crystals of 1 were grown from CH2Cl2, and the
molecular structure determined by X-ray diffraction analysis
is shown in Figure 1. The two Pd(II) centers are coordinated and
bridged by two μ-thiolato-functionalized carbene ligands in a
chelating fashion. The square-planar coordination sphere at
each Pd(II) is completed by one terminal bromido ligand. The
dihedral angles between this [PdCS2Br] coordination plane and
the carbene ring planes amount to 55.07(9) and 54.21(9)°,
respectively, which deviate substantially from the favored per-
pendicular orientation (90°) due to the chelating binding mode
of the S-functionalized carbene ligand. There are two types of
Pd-S bonds in the complex, of which the Pd1-S2 bond
˚
(2.3705(10) A) trans to the carbene is significantly longer than
˚
the Pd1-S1 bond (2.2952(11) A) trans to the bromido ligand
due to the strong trans influence of the NHC. The [Pd2S2] core
of 1 is significantly bent with a hinge angle of ca. 119.26(3)°,
which is much smaller than the reported value of 144.7(2)° for a
similar but dicationic dimeric complex bearing S-functionalized
saturated NHCs and terminal phosphine ligands.3b
In order to investigate the robustness of the [Pd2S2] core,
attempts were made to abstract the bromido ligands or to
(8) (a) Huynh, H. V.; LeVan, D.; Hahn, F. E.; Hor, T. S. A.
J. Organomet. Chem. 2004, 689, 1766. (b) Huynh, H. V.; Neo, T. C.; Tan,
G. K. Organometallics 2006, 25, 1298. (c) Huynh, H. V.; Jothibasu, R.
Organometallics 2007, 26, 6852. (d) Han, Y.; Huynh, H. V.; Koh, L. L.
J. Organomet. Chem. 2007, 692, 3606. (e) Huynh, H. V.; Seow, H. X. Aust. J.
Chem. 2009, 62, 983. (f) Han, Y.; Huynh, H. V. Chem. Commun. 2007, 1089.
Single crystals of 2 suitable for X-ray diffraction analysis
were grown from a concentrated CH2Cl2 solution. There are

Article
Organometallics, Vol. 29, No. 22, 2010 6023
Scheme 3. Formation of Complexes 2 and 3
Figure 2. Molecular structure of one independent molecule
of 2 showing 50% probability ellipsoids; hydrogen atoms and
the CH2Cl2 molecules are omitted for clarity. Selected bond
˚
lengths (A) and bond angles (deg): Pd1-C1 1.988(4), Pd1-
S1 2.2754(11), Pd1-S2 2.3611(11), Pd1-O1 2.080(3), Pd2-
C19 1.980(4), Pd2-S2 2.2689(11), Pd2-S1 2.3709(11), Pd2-
O3 2.080(3); C1-Pd1-S1 91.52(12), S1-Pd1-S2 81.23(4),
S2-Pd1-O1 94.42(9), O1-Pd1-C1 92.67(14), C1-Pd1-S2
172.39(12), S1-Pd1-O1 174.44(9), C19-Pd2-S2 90.06(13),
S2-Pd2-S1 81.15(4), S1-Pd2-O3 94.96(9), O3-Pd2-C19
93.72(15), C19-Pd2-S1 170.95(13), S2-Pd2-O3 175.26(9), Pd1-
S1-Pd2 81.00(3), Pd1-S2-Pd2 81.34(4).
two independent molecules of complex 2 in the unit cell. Due
to the similarities, the parameters of only one molecule of 2
are discussed. Its molecular structure depicted in Figure 2
shows the successful replacement of the bromido ligands in 1
by monodentate trifluoroacetato ligands. The dihedral an-
gles between the carbene ring planes and the [PdCS2O] co-
ordination planes have further decreased to 45.1(1) and
49.7(1)° compared to those in the precursor 1. The Pd-O
the formation of 3. Its1H NMR spectrumshowsa complicated
pattern consisting of 11 signals for all methylene groups,
indicating the presence of inequivalent chelating thiolato-
NHC ligands in the complex. In line with this proposal, two
carbene peaks appear in the 13C NMR spectrum at 168.1 and
166.0 ppm, respectively.
˚
bond distances of 2.080(3) A are quite comparable with
reported values for other mixed Pd-carboxylato/NHC com-
plexes.8a-e More important, the halo-carboxylato exchange
leads to a notable shortening of the trans-disposed Pd-S
˚
bonds to 2.2754(11) and 2.2689(11) A, respectively. Presum-
ably, this ligand substitution leads to more Lewis acidic
metal centers, which in turn results in a higher electron
donation of the thiolato functions and consequently
strengthening of the Pd-S bonds. The stronger binding of
the thiolato donors may also explain the aforementioned
decrease in dihedral angles. Other parameters, including the
hinge angle (117.89(4)°), remain largely unaffected.
The successful S-alkylation of thiolato-bridged complexes
of type 1 would allow for a convenient template-directed
synthesis of thioether-functionalized NHC complexes. Thus,
a series of electrophiles, including iodomethane and dimethyl
sulfate, has been used in an attempt to alkylate the sulfur
atoms of dimer 1; however, this was to no avail, and only
starting materials were reisolated. Again, the lower electron
density at the sulfur atoms as a result of the bridging mode
may be the reason they cannot be alkylated. Unexpectedly,
the reaction of 1 with an excess of the strong alkylating agent
Me3OBF4 led to the formation of the novel tetranuclear
complex 3 (Scheme 3). Apparently, electrophilic attack oc-
curredpreferably atthe bromido ligand, supposedly liberating
CH3Br and resulting in a rearrangement of the unsaturated
complex fragments to tetranuclear 3.
The solid-state molecular structure of 3 was finally con-
firmed by X-ray diffraction analysis on single crystals ob-
tained by diffusing ether into a concentrated DMF solution
(Figure 3). Each of the four Pd(II) centers is coordinated by
one carbene, one μ-bromido, and two μ-thiolato ligands in a
square-planar fashion. As found in solution, there are two
types of thiolato-NHC ligands in the molecule. The removal
of one bromido ligand in 1 leads to the cleavage of one Pd-S
bond and subsequent rearrangement/aggregation to form
tetrameric 3. An interesting feature of complex 3 is an eight-
membered square consisting of alternating four palladium and
˚
˚
four sulfur atoms with dimensions of 4.653(5) A ꢀ 4.662(5) A.
The sulfur atoms form the corners of this [Pd4S4] macrocycle,
which resembles a Pd4 system bearing remote dicarbene
ligands reported recently.9 Two carbene moieties are each
found above and below the plane spanned by this square. As
a consequence, the two bridging bromido ligands trans to these
carbenes are found on opposite sides. The presence of two
pairs of inequivalent carbenes is reflected in the significantly
different orientation of their ring planes with respect to the
[PdCS2Br] coordination (56.6(4) and 51.0(3)°) and the [Pd4S4]
macrocyclic planes (59.5(3) and 55.1(4)°). The Pd-Br bonds
To verify the unusual role of Me3OBF4 as a halide-
abstracting agent, 1 was treated with 1 equiv of AgBF4.
Indeed, this reaction furnished complex 3 as well. Conse-
quently, Me3OBF4 can be regarded as a very useful metal-
free alternative to the commonly used Ag-based and toxic
Tl-based reagents. A dicationic base peak at m/z827 correspond-
ing to the [M - 2BF4]2þ fragment in the ESI spectrum confirms
˚
of 2.5382(15) and 2.5268(16) A, respectively, are as expected
˚
elongated compared to those in 1 (2.4401(5) and 2.4652(5) A),
since the μ-bromido ligands experience the trans influence of
two carbenes.
(9) Han, Y.; Lee, L. J.; Huynh, H. V. Chem. Eur. J. 2010, 16, 771.

6024 Organometallics, Vol. 29, No. 22, 2010
Yuan and Huynh
Table 1. Suzuki-Miyaura Coupling Reactionsa Catalyzed
by Complexes 1-3
temp
(°C)
t
(h)
yield
(%)b
TOF
(h-1
)
entry catalyst
solvent
TON
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
1
1
2
3c
140
120
100
100
80
80
60
60
40
dioxane
dioxane
6
6
>99
>99
>99
>99
48
>99
0
84
4
100 000 16 667
100 000 16 667
dioxane 22
H2O 22
dioxane 24
H2O 24
dioxane 24
H2O
H2O
H2O
H2O
100 000
100 000
48 000
100 000
0
4 545
4 545
2 000
4 167
0
24
48
24
24
84 000
4 000
3 500
83
9
10
11
60
60
>99
48
100 000
96 000
4 167
4 000
a Reaction conditions: 1 mmol of 4-bromoacetophenone; 1.4 mmol of
phenylboronic acid; 1.4 mmol of K2CO3; 1 mL of solvent; 0.001 mol %
of precatalyst. b Yields were determined by 1H NMR spectroscopy for an
average of two runs. c 0.0005 mol % of 3.
Figure 3. Molecular structure of 3 showing 50% probability
ellipsoids; hydrogen atoms, BF4- counteranions, DMF mole-
cules, H2O molecules, and phenyl rings are omitted for
˚
clarity. Selected bond lengths (A) and bond angles (deg):
Pd1-C1 1.956(12), Pd1-S2A 2.334(4), Pd1-Br3 2.5382(15),
Pd1-S1 2.321(4), Pd2-S1 2.318(3), Pd2-Br3 2.5268(16),
Pd2-S2 2.350(4), Pd2-C11 1.986(13); C1-Pd1-S1 90.5(4),
C1-Pd1-S2A 91.6(4), S2A-Pd1-Br3 94.60(9), Br3-Pd1-
S1 83.26(9), Br3-Pd2-S1 83.58(9), Br3-Pd2-S2 94.79(10), S2-
Pd2-C11 88.5(4), C11-Pd2-S1 93.9(4), Pd1-Br3-Pd2 76.48(4),
Pd1-S1-Pd2 85.02(12), Pd1-S2A-Pd2A 103.31(13).
Catalysis. Despite often being regarded as catalyst poisons,10
sulfur-containing compounds have become versatile donors
in transition-metal chemistry. Nevertheless, some complexes
of thioether-functionalized phosphite, phosphine, and NHC
ligands have shown catalytic activities in reactions, such as
allylic alkylations,11 hydroformylations,12 hydrogenations,13
and Suzuki-Miyaura4d and Mizoroki-Heck reactions.4e
More recently, dimeric phosphino-thiolato Pd(II) com-
plexes were also reported to be highly active in the catalytic
Suzuki-Miyaura coupling reaction at low catalyst load-
ing.14 For example, using 0.001 mol % catalyst, 91% yield
was achieved with 4-bromoacetophenone and phenylboro-
nic acid in dioxane at 140 °C in the presence of K2CO3 after
6 h. Encouraged by this finding, dimeric 1 was tested for
its catalytic activity under the same reaction conditions,
affording a quantitative yield (Table 1, entry 1) and show-
ing the superiority of NHC complex 1 over its phosphine
analogue. Further investigation revealed that the tempera-
ture can be lowered to 120 °C (entry 2) and even to 100 °C
without compromising the yield when the reaction time was
prolonged to 22 h (entry 3).
Figure 4. Concentration/time diagram (yield (%), time (h)) for
the Suzuki-Miyaura coupling reaction of 4-bromoacetophe-
none and phenylboronic acid catalyzed by 2.
Since water has been reported to be beneficial for the
Suzuki-Miyaura coupling reaction,15 we also tested its
suitability as an environmentally benign solvent. Indeed, the
coupling at 100 °C gave quantitative yield (entry 4). At a lower
temperature of 80 °C, H2O proves to be a superior solvent, as
the yield remained quantitative (entry 6), while a drop to 48%
was observed in dioxane (entry 5). When the temperature
was further decreased to 60 °C, no coupling reaction was
detected in dioxane (entry 7), while a good yield of 84% was
still obtained in H2O (entry 8). At temperatures below 60 °C,
the yield dropped dramatically in H2O as well (entry 9).
A comparative study on the activities of all three thiolato-
NHC complexes 1-3 (entries 8, 10, and 11) under these mild
reaction conditions revealed that 2 is the best precatalyst.
Apparently, the replacement of the bromido ligands in 1
with more labile trifluoroacetato ligands results in enhanced
(10) Dunleavy, J. K. Platinum Met. Rev. 2006, 50, 110.
(11) (a) Evans, D. A.; Campos, K. R.; Tedrow, J. S.; Michael, F. E.;
ꢀ
Gagne, M. R. J. Am. Chem. Soc. 2000, 122, 7905. (b) Barbaro, P.; Currao,
A.; Herrmann, J.; Nesper, R.; Pregosin, P. S.; Salzmann, R. Organometallics
1996, 15, 1879. (c) Herrmann, J.; Pregosin, P. S.; Salzmann, R. Organome-
tallics 1995, 14, 3311.
ꢁ
ꢀ
(12) Pamies, O.; Dieguez, M.; Net, G.; Ruiz, A.; Claver, C. Organo-
metallics 2000, 19, 1488.
(13) (a) Guimet, E.; Dieguez, M.; Ruiz, A.; Claver, C. Dalton Trans.
ꢀ
2005, 2557. (b) Evans, D. A.; Michael, F. E.; Tedrow, J. S.; Campos, K. R.
J. Am. Chem. Soc. 2003, 125, 3534. (c) Hauptman, E.; Fagan, P. J.; Marshall,
W. Organometallics 1999, 18, 2061.
(14) Dervisi, A.; Koursarou, D.; Ooi, L.; Horton, P. N.; Hursthouse,
M. B. Dalton Trans. 2006, 5717.
(15) (a) Huynh, H. V.; Yuan, H.; Ho, J. H. H.; Tan, G. K. Organo-
metallics 2006, 25, 3267. (b) Gu, S.; Xu, H.; Zhang, N.; Chen, W. Chem.
Asian J. 2010, 5, 1677. (c) Bedford, R. B.; Blake, M. E.; Butts, C. P.; Holder,
D. Chem. Commun. 2003, 466. (d) Tu€rkmen, H.; Can, R.; C-etinkaya, B.
Dalton Trans. 2009, 7039.

Article
Organometallics, Vol. 29, No. 22, 2010 6025
Table 2. Suzuki-Miyaura Coupling Reactionsa Catalyzed by Complex 2
entry
aryl halide
catalyst load (mol %)
temp (°C)
t (h)
yield (%)b
TON
TOF (h-1
)
1
2
3
4
5
6
7
8
4-bromobenzonitrile
1-bromo-4-nitrobenzene
4-bromobenzaldehyde
2,6-dibromopyridinec
4-bromotoluene
4-bromoanisole
1-bromo-4-chlorobenzene
4-chlorobenzaldehyde
0.001
0.001
0.001
0.0025
0.0025
0.0025
0.0025
0.1
60
60
80
100
100
100
100
100
24
42
23
48
20
24
24
72
>99
90
>99
>99
97
100 000
90 000
100 000
20 000
38 800
34 400
28 800
410
4 167
2 143
4 348
417
1 940
1 433
1 200
5.7
86
72d
41e
a Reaction conditions: 1 mmol of aryl halide; 1.4 mmol of phenylboronic acid; 1.4 mmol of K2CO3; 1 mL of H2O; the desired amount of precatalyst.
b Yields were determined by 1H NMR spectroscopy for an average of two runs. c 0.5 mmol of 2,6-dibromopyridine. d Isolated yield. e With addition of
1.0 equiv of [N(n-C4H9)4]Br.
activities of 2.8 The lability of such ligands allows an easier
ligand displacement, which in turn facilitates the formation
of catalytically active species and, thus, promotes catalyst
initiation. The poor performance of the higher aggregate 3,
on the other hand, is probably due to the strongly bridging
bromido and thiolato ligands hampering the formation of
active Pd(0) species.
of 41% was obtained when 0.1 mol % of 2 was employed in
the presence of 1 equiv of TBAB as a promoter (entry 8).
Conclusion
We have reported two methods (1 and 2) for synthesis of
the new thiolato-bridged dimeric Pd(II) benzimidazolin-2-
ylidene complex 1, of which the pathway through in situ
hydrolysis of a thioester (method 2) is more desirable due
to its ease and convenience. Subsequent reaction of 1 with
AgO2CCF3 gave rise to the mixed carboxylato-thiolato car-
bene complex 2. The treatment of 1 with excess Me3OBF4 or 1
equiv of AgBF4 led to the abstraction of one bromido ligand
and subsequent rearrangement/dimerization to afford the tet-
ranuclear species 3 featuring a macrocyclic [Pd4S4] square. The
unusual reactivity of Me3OBF4 as a halide-abstracting agent
is remarkable and is believed to be of great synthetic use in cases
where soft Agþ and toxic Tlþ ions are interfering and are unde-
sirable. All complexes have been fully characterized and their
molecular structures determined by X-ray diffraction analyses.
Catalytic studies reveal that dimeric complexes 1 and 2 show
good activities at low loadings toward aqueous Suzuki-
Miyaura couplings despite carrying thiolato functions,
which are usually regarded as catalyst poisons. Research in
our laboratory is ongoing to extend the straightforward and
versatile coordination chemistry of thiolato-functionalized
NHCs to other transition metals and also to aim at bigger
thiolato-bridged aggregates as an entry to metallo-NHC-
based supramolecular chemistry. We are also further explor-
ing the feasibility of Meerwein’s salts as alternative halide-
abstracting agents.
Having identified the best precatalyst, 2, a more detailed
kinetic study was carried out on the coupling of 4-bromo-
acetophenone and phenylboronic acid at 0.001 mol % catalyst
loading (Figure 4) to gain more insights into its catalytic
behavior. The concentration/time profile shows that the
reaction rate is rather low during a long period of about
22 h, after which the reaction exponentially completes within
2 h. Although the exact reason for the sluggish reaction at the
beginning is not known yet, it may be due to a complicated
dimer-monomer equilibrium14,16 prior to catalyst initia-
tion. Furthermore, the shape of the reaction profile suggests
the involvement of various catalytically active species prior
to the formation of the most active catalyst. It is commonly
agreed that catalytically active species are generally mono-
meric,16 which would also indirectly explain the poor per-
formance observed for the tetranuclear complex 3. The latter
can be regarded as a tetramer, in which a tetramer-monomer
dissociation is greatly hampered due to multiple strong brid-
ging ligands.
To further test its efficiency, precatalyst 2 was subjected
to the coupling of other aryl halides in pure water, and the
results are summarized in Table 2. Generally, complex 2
showed good activities with activated bromo substrates
(entries 1-3) at low loading (0.001 mol %) and under mild
reaction conditions (60 or 80 °C). The coupling of 2 equiv of
phenylboronic acid with 2,6-dibromopyridine was equally
successful, giving quantitative yield (entry 4). For deacti-
vated aryl bromides, the catalyst load and temperature were
raised to 0.0025 mol % and 100 °C, respectively, to obtain
good yields (entries 5-7). The catalyst, however, shows its
limitation in the coupling with aryl chlorides. The best yield
Experimental Section
General Considerations. Unless otherwise noted, all opera-
tions were performed without taking precautions to exclude air
and moisture, and all solvents and chemicals were used as
received. Salt A was prepared according to a previously reported
method.4e 1H, 13C, and 19F NMR spectra were recorded on a
Bruker ACF 300 spectrometer or AMX 500 spectrophotometer,
and the chemical shifts (δ) were internally referenced to the
residual solvent signals relative to tetramethylsilane (1H, 13C) or
(16) Rosner, T.; Bars, J. L.; Pfaltz, A.; Blackmond, D. G. J. Am.
Chem. Soc. 2001, 123, 1848.

6026 Organometallics, Vol. 29, No. 22, 2010
Yuan and Huynh
Table 3. Selected X-ray Crystallographic Data for Complexes 1, 2 CH2Cl2 and 3 2DMF Oa
3
3
3
1
2 CH2Cl2
3
3 2DMF Oa
3
3
formula
fw
C32H30Br2N4Pd2S2
907.34
C36H30F6N4O4Pd2S2 CH2Cl2
3
C64H60B2Br2F8N8Pd4S4 C6H14N2O3
3
1058.49
yellow, block
1990.67
yellow, needle
color, habit
cryst size (mm)
temp (K)
cryst syst
space group
yellow, block
0.40 ꢀ 0.12 ꢀ 0.10
0.36 ꢀ 0.36 ꢀ 0.06
0.60 ꢀ 0.07 ꢀ 0.06
223(2)
orthorhombic
100(2)
orthorhombic
100(2)
monoclinic
P212121
9.6224(3)
15.5833(5)
21.3048(7)
Pbca
20.9496(7)
21.0979(7)
36.6305(13)
P21/n
12.922(3)
13.527(3)
23.326(5)
˚
a (A)
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
90
90
90
90
90
91.314(5)
90
4076.4(14)
2
1.622
Mo KR
2.018
90
3194.63(18)
4
1.887
Mo KR
3.787
90
16 190.4(10)
16
1.737
Mo KR
1.196
3
˚
V (A )
Z
Dc (g cm-3
radiation used
)
μ (mm-1
)
θ range (deg)
1.62-27.50
7216
0.7032, 0.3126
R1 = 0.0322, wR2 = 0.0621
R1 = 0.0376, wR2 = 0.0636
1.000
1.11-27.50
18 607
0.9317, 0.6727
R1 = 0.0490, wR2 = 0.1094
R1 = 0.0667, wR2 = 0.1163
1.083
1.74-25.00
7177
0.8885, 0.3772
R1 = 0.0869, wR2 = 0.2116
R1 = 0.1525, wR2 = 0.2448
1.027
no. of unique data
max, min transmissn
final R indices (I > 2σ(I))
R indices (all data)
goodness of fit on F2
-3
˚
peak/hole (e A
)
1.060/-0.389
1.478/-0.827
1.089/-0.711
a No hydrogen atoms were added to this water molecule.
externally to CF3CO2H (19F NMR). ESI mass spectra were
measured using a Finnigan MAT LCQ spectrometer. Elemental
analyses were performed on a Perkin-Elmer PE 2400 elemental
analyzer at the Department of Chemistry, National University
of Singapore.
vacuum to give the product as a yellow solid (66 mg, 0.073 mmol,
76%).
Method 2. A mixture of salt B (117 mg, 0.3 mmol) and
Pd(OAc)2 (67 mg, 0.3 mmol) in DMSO (5 mL) was stirred at
80 °C overnight. The resulting yellow solid product was filtered
and washed with H2O (3 ꢀ 20 mL). A second crop of the product
can be isolated from the filtrate after vacuum distillation,
washing with H2O (3 ꢀ 20 mL), and subsequent purification
by chromatography on a short plug of silica using ethyl acetate
(101 mg, 0.111 mmol, 74%). 1H NMR (300 MHz, CDCl3):
δ 7.48-7.40 (m, 6 H, Ar H), 7.36-7.19 (m, 12 H, Ar H), 6.63
(d, 2J(H,H) = 15.6 Hz, 2 H, NCHHPh), 5.58 (d, 2J(H,H) = 15.6
Hz, 2 H, NCHHPh), 4.88 (ps-d, 2 H, NCHH), 4.63 (ps-t, 2 H,
NCHH), 3.90 (ps-d, 2 H, CHHS), 2.10 (ps-t, 2 H, CHHS).
13C{1H} NMR (75.47 MHz, CDCl3): 175.4 (s, NCN), 136.4,
134.9, 134.3, 129.5, 128.7, 128.3, 124.44, 124.40, 112.7, 111.2 (s,
Ar H), 53.9 (s, NCH2Ph), 52.3 (s, NCH2), 26.8 (s, CH2S). Anal.
Calcd for C32H30Br2N4Pd2S2: C, 42.36; H, 3.33; N, 6.17. Found:
C, 42.50; H, 3.50; N, 6.06. MS (ESI): m/z 829 [M - Br]þ.
Dinuclear Complex 2. A mixture of 1 (87 mg, 0.096 mmol)
and AgO2CCF3 (42 mg, 0.192 mmol) was suspended in CH3CN
(10 mL) and heated at 70 °C overnight shielded from light. The
reaction mixture was filtered over Celite, and the solvent of the
filtrate was evaporated off to afford the product quantitatively
Thioester-Functionalized Benzimidazolium Bromide B. A mix-
ture of salt A (792 mg, 2 mmol) and KSCOCH3 (274 mg,
2.4 mmol) in CH3CN (20 mL) was stirred at ambient tempera-
ture overnight. The resulting suspension was filtered, and the
solvent of the filtrate was removed in vacuo. The resulting solid
was washed with THF (3 ꢀ 20 mL) to give the product as an off-
1
white solid (584 mg, 1.49 mmol, 75%). H NMR (300 MHz,
DMSO-d6): δ 10.11 (s, 1 H, NCHN), 8.17 (d, 3J(H,H) = 8.1 Hz,
1 H, Ar H), 7.97 (d, 3J(H,H) = 8.1 Hz, 1 H, Ar H), 7.68 (m, 2 H,
Ar H), 7.52 (d, 2 H, Ar H), 7.40 (m, 3 H, Ar H), 5.84 (s, 2 H,
NCH2Ph), 4.75 (t, 3J(H,H) = 6.3 Hz, 2 H, NCH2), 3.44 (t, 3J(H,
H) = 6.3 Hz, 2 H, CH2S), 2.27 (s, 3 H, COCH3). 13C{1H} NMR
(75.47 MHz, DMSO-d6): 194.7 (s, CO), 142.7 (s, NCHN), 134.0,
131.2, 130.7, 128.9, 128.6, 128.1, 126.7, 113.9, 113.8 (s, Ar C),
49.8 (s, NCH2Ph), 46.2 (s, NCH2), 30.4 (s, CH3), 27.8 (s, CH2S).
MS (ESI): m/z 311 [M - Br]þ.
Thiol-Functionalized Benzimidazolium Bromide C. To a
MeOH (5 mL) solution of salt B (82 mg, 0.21 mmol) was added
aqueous HBr solution (2 mL, 2 M). The mixture was degassed
and heated to reflux under nitrogen overnight. The mixture
was cooled to ambient temperature, and all the volatiles were
removed in vacuo to give the product as a white solid (67 mg,
0.19 mmol, 92%). 1H NMR (500 MHz, CD3CN): δ 9.69 (s, 1 H,
NCHN), 7.97 (d, 3J(H,H) =8.2Hz, 1H,ArH),7.81(d, 3J(H,H) =
8.2 Hz, 1 H, Ar H), 7.68 (m, 2 H, Ar H), 7.51 (m, 2 H, Ar H), 7.46
(m, 3 H, Ar H), 5.71 (s, 2 H, NCH2Ph), 4.67 (t, 3J(H,H) = 6.3
Hz, 2 H, NCH2), 3.07 (t, 3J(H,H) = 6.3 Hz, 2 H, CH2S), 2.72 (br
s, 1 H, SH). 13C{1H} NMR (125.77 MHz, CD3CN): 142.7
(s, NCHN), 134.0, 132.4, 132.2, 130.0, 129.9, 129.3, 127.96,
127.95, 114.59, 114.57 (s, Ar C), 51.6 (s, NCH2Ph), 50.4 (s,
NCH2), 24.4 (s, CH2S). MS (ESI): m/z 269 [M - Br]þ.
Dinuclear Complex 1. Method 1. A mixture of salt C (67 mg,
0.19 mmol) and Pd(OAc)2 (43 mg, 0.19 mmol) was heated under
reflux in degassed CH3CN under nitrogen overnight. The result-
ing brown solid was filtered and dissolved in CH2Cl2 (10 mL).
The mixture was filtered, the residue was washed with CH2Cl2
(2 ꢀ 10 mL), and the filtrate was collected and dried under
1
as a yellow solid (94 mg, 0.096 mmol). H NMR (500 MHz,
CDCl3): δ 7.43-7.30 (m, 14 H, Ar H), 7.23 (m, 4 H, Ar H), 5.95
(d, 2J(H,H) = 15.8 Hz, 2 H, NCHHPh), 5.56 (d, 2J(H,H) = 15.8
Hz, 2 H, NCHHPh), 4.87 (ps-t, 2 H, NCHH), 4.77 (ps-d, 2 H,
NCHH), 2.69 (ps-d, 2 H, CHHS), 2.01 (ps-t, 2 H, CHHS).
13C{1H} NMR (125.76 MHz, CDCl3): 171.4 (s, NCN), 161.5
(q, 2J(C,F) = 35.7 Hz, COO), 135.1, 133.6, 133.2, 128.9, 128.2,
127.2, 124.1, 124.0 (s, Ar H), 115.7 (q, 1J(C,F) = 291.4 Hz, CF3),
112.1, 110.7 (s, Ar H), 51.4 (s, NCH2Ph), 50.6 (s, NCH2), 24.9 (s,
CH2S). 19F NMR (282.37 MHz, CDCl3): 2.02 (s, CF3). Anal.
Calcd for C36H30F6N4O4Pd2S2: C, 44.41; H, 3.11; N, 5.75.
Found: C, 44.35; H, 3.05; N, 5.59. MS (ESI): m/z 861 [M -
O2CCF3]þ, 373 [M - 2O2CCF3]2þ
.
Tetranuclear Complex 3. Method 1. A mixture of 1 (91 mg,
0.1 mmol) and Me3OBF4 (44 mg, 0.3 mmol) in dry CH2Cl2
(10 mL) was heated under reflux overnight. The resulting mix-
ture was filtered, and the product was obtained as a yellow
powder (52 mg, 0.028 mmol, 57%).

Article
Method 2. A mixture of 1 (91 mg, 0.1 mmol) and AgBF4
Organometallics, Vol. 29, No. 22, 2010 6027
The reaction mixture was stirred at elevated temperature. After
the desired reaction time, the mixture was cooled to ambient
temperature, and dichloromethane (10 mL) was added. The
organic layer was then washed with water (6 ꢀ 8 mL) and dried
over Na2SO4. The solvent was allowed to evaporate, and the
residue was analyzed by 1H NMR spectroscopy.
(19.5 mg, 0.1 mmol) in dry CH2Cl2 (10 mL) was heated under
reflux overnight shielded from light. The precipitate was filtered
and washed with CH2Cl2 (2 ꢀ 10 mL). The residue was triturated
with DMF (3 ꢀ 10 mL) and the DMF solutions filtered through
Celite. The solvent of the filtrate was removed under vacuum to
afford the product as a yellow solid (62 mg, 0.034 mmol, 68%).
1H NMR (500 MHz, DMSO-d6): δ 8.06 (d, 2 H, Ar H),
7.80-7.74 (m, 4 H, Ar H), 7.55-7.26 (m, 30 H, Ar H), 6.17 (d,
2J(H,H) = 15.8 Hz, 2 H, CHHPh), 5.96 (d, 2J(H,H) = 15.8 Hz,
2 H, CHHPh), 5.85 (br s, 4 H, CH2Ph), 5.61 (ps-d, 2 H, NCHH),
5.32 (ps-d, 2 H, NCHH), 5.24 (ps-t, 2 H, NCHH), 4.92 (ps-t, 2 H,
NCHH), 3.96 (ps-d, 2 H, CHHS), 2.64 (ps-t, 2 H, CHHS), 2.21
(ps-t, 2 H, CHHS), 1.91 (ps-d, 2 H, CHHS). 19F{1H} NMR
(282.37 MHz, DMSO-d6): -72.28 (s, 10BF4), -72.33 (s, 11BF4).
13C{1H} NMR (125.76 MHz, DMSO-d6): 168.1 (s, NCN), 166.0
(s, NCN), 135.7, 135.6, 134.3, 133.8, 133.4, 129.4, 129.2, 128.8,
128.5, 128.0, 127.8, 127.3, 125.3, 125.2, 125.0, 112.9, 112.7 (s, Ar
H), 52.4 (s, CH2Ph), 51.7 (s, CH2Ph), 51.6 (s, NCH2), 51.3 (s,
NCH2), 33.3 (s, CH2S), 33.1 (s, CH2S). Anal. Calcd for
C64H60B2Br2F8N8Pd4S4: C, 42.04; H, 3.31; N, 6.13. Found: C,
41.92; H, 3.30; N, 6.09. MS (ESI): m/z 827 [M - 2BF4]2þ, 1136
[M - BF4]þ.
X-ray Diffraction Studies. X-ray data for 1, 2 CH2Cl2, and
3 2DMF O were collected with a Bruker AXS SMART APEX
3
3
3
diffractometer, using Mo KR radiation at 223(2) K (for 1) or at
100(2) K (for 2 CH2Cl2 and 3 2DMF O) with the SMART
3
3
3
suite of programs.17 Data were processed and corrected for
Lorentz and polarization effects with SAINT18 and for absorp-
tion effects with SADABS.19 Structural solution and refinement
were carried out with the SHELXTL suite of programs.20 The
structure was solved by direct methods to locate the heavy
atoms, followed by difference maps for the light, non-hydrogen
atoms. All non-hydrogen atoms were generally given aniso-
tropic displacement parameters in the final model. All H atoms
were put at calculated positions with exception of the water
molecule in 3 2DMF O. A summary of the most important
3
3
crystallographic data is given in Table 3.
Acknowledgment. We thank the National University
of Singapore for financial support (No. WBS R-143-000-
327-133). Technical support from staff at the CMMAC of
our department is appreciated. In particular, we thank
Ms. Geok Kheng Tan and Prof. Lip Lin Koh for deter-
mining the X-ray molecular structures.
Suzuki-Miyaura Catalysis. In a typical run, a reaction tube
was charged with a mixture of aryl halide (1.0 mmol for mono-
halides, 0.5 mmol for dihalides), K2CO3 (1.4 mmol), phenyl-
boronic acid (1.4 mmol), precatalyst, and solvent (1 mL).
(17) SMART version 5.628; Bruker AXS Inc., Madison, WI, 2001.
(18) SAINTþ version 6.22a; Bruker AXS Inc., Madison, WI, 2001.
Supporting Information Available: CIF files giving crystal-
lographic data for 1, 2 CH2Cl2 and 3 2DMF O. This material
is available free of charge via the Internet at http://pubs.acs.org.
€
(19) Sheldrick, G. W. SADABS version 2.10; University of Gottingen,
€
Gottingen, Germany, 2001.
3
3
3
(20) SHELXTL version 6.14; Bruker AXS Inc., Madison, WI, 2000.