Efficient cycloisomerization of propargyl amides by electrophilic gold(I) complexes of KITPHOS monophosphines: A comparative study

Article abstract of DOI:10.1021/om1006769

Electrophilic gold(I) complexes of diphenyl- and dicyclohexylphosphino- based KITPHOS monophosphines catalyze the 5-exo-dig cycloisomerization of a range of propargyl amides to afford the corresponding alkylidene oxazolines; in all cases catalysts formed from diphenylphosphino-substituted KITPHOS monophosphines outperformed their dicyclohexylphosphino counterparts as well as that based on triphenylphosphine, an indication that the biaryl/biaryl-like framework may be responsible for imparting high catalyst efficiency.

Full text of DOI:10.1021/om1006769

Organometallics 2010, 29, 4139–4147 4139
DOI: 10.1021/om1006769
Efficient Cycloisomerization of Propargyl Amides
by Electrophilic Gold(I) Complexes of KITPHOS Monophosphines:
A Comparative Study
Simon Doherty,*,† Julian G. Knight,*,† A. Stephen K. Hashmi,‡ Catherine H. Smyth,†
Nicholas A. B. Ward,† Katharine J. Robson,† Sophie Tweedley,† Ross W. Harrington,† and
William Clegg†
†School of Chemistry, Bedson Building, Newcastle University, Newcastle upon Tyne, NE1 7RU, U.K., and
‡
€
Organisch-Chemisches Institut, Universitat Heidelberg, Im Neuenheimer Feld 270,
69120 Heidelberg, Germany
Received July 12, 2010
Electrophilic gold(I) complexes of diphenyl- and dicyclohexylphosphino-based KITPHOS mono-
phosphines catalyze the 5-exo-dig cycloisomerization of a range of propargyl amides to afford the
corresponding alkylidene oxazolines; in all cases catalysts formed from diphenylphosphino-sub-
stituted KITPHOS monophosphines outperformed their dicyclohexylphosphino counterparts as
well as that based on triphenylphosphine, an indication that the biaryl/biaryl-like framework may be
responsible for imparting high catalyst efficiency.
Introduction
phosphines and N-heterocyclic carbenes are proving to be
the ligands of choice for gold(I)-catalyzed transformations,
and there is increasing evidence that both reactivity and
selectivity depend on the steric and/or electronic properties
of the supporting ligand.3 For example, Asensio has reported
that the gold(I)-catalyzed cyclization of 3-substituted 1-(o-
ethynylaryl)ureas occurs via either a 6-exo-dig or 5-endo-dig
pathway depending on the choice of ligand, with bulky
N-heterocyclic carbenes giving high 6-exo-dig selectivity,4
and Toste has demonstrated that the stability of gold(I) car-
bene intermediates can be modulated by adjusting the elec-
tronic properties of the supporting ligand; in this case an
electron-rich biaryl monophosphine provided the optimum
combination of electronic and steric properties.5 In this regard,
although electron-rich biaryl monophosphines have had their
greatest impact on palladium-catalyzed cross-coupling,6 they
are rapidly emerging as highly successful/efficient ligands for a
host of gold(I)- and gold(III)-catalyzed transformations.7
We have recently developed an architecturally related
class of electron-rich biaryl-like monophosphines, KITPHOS
The catalytic activation of carbon-carbon π-bonds by
carbophilic Lewis acid complexes of gold(I) and gold(III) has
evolved into an extremely powerful tool for the construction
of diverse and complex mono- and polycylic architectures.1
In the vast majority of these transformations an alkyne is
activated toward nucleophilic addition by coordination to an
electrophilic gold complex; the fate of such complexes in-
cludes intramolecular carbocyclizations, skeletal rearrange-
ments, inter- and intramolecular addition of X-H bonds
(X = O, N, S), and cyclizations initiated by addition of weak
nucleophiles such as a carbonyl oxygen atom.2 In many cases,
*To whom correspondence should be addressed. E-mail: simon.doherty@
newcastle.ac.uk; j.g.knight@newcastle.ac.uk.
(1) (a) Hashmi, A. S. K. Chem. Soc. Rev. 2008, 37, 1766. (b) Widenhoefer,
M. Chem.;Eur. J. 2008, 14, 5382. (c) Hashmi, A. S. K. Chem. Rev. 2007,
107, 3180. (d) F€urstner, A.; Davies, P. W. Angew. Chem., Int. Ed. 2007, 46,
ꢀ
3410. (e) Michelet, V.; Toullec, P. Y.; Genet, J.-P. Angew. Chem., Int. Ed.
2008, 47, 4268. (f) F€urstner, A. Chem. Soc. Rev. 2009, 38, 3208.
(g) Soriano, E.; Mar-Contelles, J. Acc. Chem. Res. 2009, 42, 1026.
(h) Muzart, J. Tetrahedron 2008, 64, 5815. (i) Hashmi, A. S. K. Angew.
Chem., Int. Ed. 2008, 47, 6754. (j) Crone, B.; Kirsh, S. F. Chem.;Eur. J.
2008, 14, 3514. (k) Das, A.; Sohel, S. M. A.; Liu, R.-S. Org. Biomol. Chem.
2010, 8, 960.
(3) Gorin, D. J.; Sherry, B. D.; Toste, D. F. Chem. Rev. 2008, 108,
3351.
ꢀ
(4) Gimeno, A.; Medio-Simon, M.; de Arellano, C. R.; Asensio, G.;
(2) (a) Patil, N. T.; Yamamoto, Y. Chem. Rev. 2008, 108, 3395.
Cuenca, A. B. Org. Lett. 2010, 12, 1900.
ꢀ
~
ꢀ
ꢀ
(b) Arcadi, A. Chem. Rev. 2008, 108, 3266. (c) Jimenez-Nꢀunez, E.;
Echavarren, A. Chem. Rev. 2008, 108, 3326. (d) Li, Z.; Brouwer, C.; He,
C. Chem. Rev. 2008, 108, 3239. For a selection of relevant transformations
initiated by nucleophilic addition of the oxygen atom of a carbonyl group see:
(e) Kusama, H.; Ishida, K.; Funami, H.; Iwasawa, N. Angew. Chem., Int. Ed.
2008, 47, 4903. (f) Ishida, K.; Kusama, H.; Iwasawa, N. J. Am. Chem. Soc.
2010, 132, 8842. (g) Oh, C. H.; Lee, S. M.; Hong, C. S. Org. Lett. 2010, 12,
1308. (h) Teng, T.-M.; Liu, R.-S. J. Am. Chem. Soc. 2010, 132, 9298.
(i) Bian, M.; Yao, W.; Ding, H.; Ma, C. J. Org. Chem. 2010, 75, 269.
(j) Meng, J.; Zhao, Y.-L.; Ren, C.-Q.; Li, Y.; Li, Z.; Liu, Q. Chem.;Eur. J.
2009, 15, 1830. (k) Robles-Machín, R.; Adrio, J.; Carretero, J. J. Org. Chem.
2006, 71, 5023. Belting, V.; Krause, N. Org. Biomol. Chem. 2009, 7, 1221.
(l) Buzas, A.; Gagosz, F. Synlett 2006, 2727. (m) Hashmi, A. S. K.; Salathe,
R.; Frey, W. Synlett 2007, 1763.
(5) (a) Mauleon, P.; Zeldin, R. M.; Gonzalez; Toste, D. F. J. Am.
Chem. Soc. 2009, 131, 6348. (b) Benitez, D.; Tkatchouk, E.; Gonzalez, A. Z.;
Goddard, W. A., III; Toste, D. F. Org. Lett. 2009, 11, 4798.
(6) For highly informative reviews see: (a) Martin, R.; Buchwald,
S. L. Acc. Chem. Res. 2008, 41, 1461. (b) Surry, D. S.; Buchwald, S. L.
Angew. Chem., Int. Ed. 2008, 47, 6338. (c) Zapf, A; Beller, M. Chem.
Commun. 2005, 431.
ꢀ
ꢀ
(7) For a selection of recent examples see: (a) Barabe, F.; Betournay,
G.; Bellavance, G.; Barriault, L. Org. Lett. 2009, 11, 4236. (b) Nieto-
ꢀ
Oberhuber, C.; Lopez, S.; Echavarren, A. M. J. Am. Chem. Soc. 2005, 127,
6178. (c) Kuram, M. R.; Bhanuchandra, M.; Sahoo, A. K. J. Org. Chem.
2010, 75, 2247. (d) Mukherjee, P.; Widenhoefer, R. A. Org. Lett. 2010, 12,
1184. (e) Zhang, Z.; Lee, S. D.; Widenhoefer, R. A. J. Am. Chem. Soc. 2009,
131, 5372. (f) Leyva, A.; Corma, A. J. Org. Chem. 2009, 74, 2067.
r
2010 American Chemical Society
Published on Web 08/17/2010
pubs.acs.org/Organometallics

4140 Organometallics, Vol. 29, No. 18, 2010
Doherty et al.
Scheme 1
Scheme 2
(11-dicyclohexylphosphino-12-phenyl-9,1-ethenoanthracenes),
which also form highly efficient catalysts for C-C and C-N
cross-coupling of a range of aryl chlorides.8 The similarity
between the basic skeletal framework of KITPHOS mono-
phosphines and Buchwald’s electron-rich biaryl monophos-
phines coupled with their competitive performance in palla-
dium-catalyzed cross-coupling prompted us to explore their
efficiency in gold-catalyzed cyclizations.9 Our preliminary
studies were encouraging as electrophilic Lewis acid gold(I)
complexes of the corresponding electron-rich KITPHOS
monophosphines efficiently catalyzed a range of intramole-
cular cyclizations to afford phenols, tetrahydropyrans, lac-
tones, and alkylidene oxazolines; in the majority of cases
these catalysts were superior to those previously reported.
Having restricted our initial study to electron-rich KIT-
PHOS monophosphines, we subsequently initiated a sys-
tematic comparison between their PPh2- and PCy2-based
counterparts with the aim of beginning to establish what
factors control catalyst performance. In this regard, PPh2-
based biaryl monophosphines have received much less atten-
tion than their PCy2 and PtBu2 analogues on the basis that
they are less electron rich and lack steric bulk, requirements
commonly accepted to be necessary for efficient platinum
group metal catalyzed C-C and C-N bond formation.10
The cycloisomerization of propargyl amides was selected as
the benchmark reaction for this study since these substrates
are easily accessible in a single step11 and the resulting alkyli-
dene oxazoline is a potentially useful intermediate, while its
isomeric oxazole counterpart is a substructure of a number
of naturally occurring products and bioactive compounds;12
both products can be selectively prepared using gold(I)- and
gold(III)-based catalysts, respectively.13 Herein we disclose
the results of this study, which has conclusively shown that
catalysts based on diphenylphosphino-containing KIT-
PHOS monophosphines outperform their dicyclohexylpho-
sphino counterparts and that the former are markedly more
efficient than [Au(PPh3)]þ,13b with the corollary that the
biaryl fragment appears be necessary for efficient catalysis;
cationic gold(I) complexes containing a KITPHOS mono-
phosphine and a thioether have also been shown to be
practical and highly efficient precatalysts.
Results and Discussion
Ligand Synthesis and Coordination Chemistry. KITPHOS
monophosphines 3a,b required for this study were prepared
in good yield following the procedure previously reported for
their dicyclohexylphosphino counterparts 3c,d,8 which in-
volved the Diels-Alder cycloaddition between 1-alkynyl
phosphine oxides 1a,b and anthracene to construct the
bicyclic framework followed by reduction of the resulting
KITPHOS monoxide 2a,b with chlorosilane/triethylamine
in toluene (Scheme 1). The identity and purity of 3a,b have
been established using conventional spectroscopic and ana-
lytical techniques.
With the intention of undertaking comparative testing of
KITPHOS monophosphines 3a-d in the gold(I)-catalyzed
cycloisomerization of a range of propargyl amides, precur-
sors 4a-d were prepared by reaction of [(tht)AuCl]14 with
the corresponding KITPHOS monophosphine in dichloro-
methane (Scheme 2). As relatively few gold(I) complexes of
KITPHOS monophosphines have been crystallographically
characterized,9 single-crystal X-ray analyses of 4b and 4d
were undertaken to compare their key structural features;
perspective views of the molecular structures are shown in
Figures 1 and 2, respectively.
(8) (a) Doherty, S.; Knight, J. G.; Smyth, C. H.; Jorgensen, G. A. Adv.
Synth. Catal. 2008, 350, 1801. (b) Doherty, S.; Knight, J. G.; McGrady, J. P.;
Ferguson, A. M.; Ward, N. A. B.; Harrington, R. W.; Clegg, W. Adv. Synth.
Catal. 2010, 352, 201.
(9) Hashmi, A. S. K.; Loos, A.; Littmann, A.; Braun, I.; Knight, J. G.;
Doherty, S.; Rominger, F. Adv. Synth. Catal. 2009, 351, 576.
(10) So, C. M.; Chow, W. K.; Choy, P. Y.; Lau, C. P.; Kwong, F. Y.
Chem.;Eur. J. 2010, 16, 7996.
(11) (a) Wipf, P.; Aoyama, Y.; Benedum, T. E. Org. Lett. 2004, 6,
3593. (b) Wipf, P.; Hopkins, C. R. J. Org. Chem. 1999, 64, 6881. (c) Merkul,
E.; Mueller, T. J. J. Chem. Commun 2006, 4817.
(12) (a) Pattenden, G. J. Heterocycl. Chem. 1992, 29, 607. (b) Palmer,
D. C.; Venkatraman, S. In The Chemistry of Heterocyclic Compounds, A
Series of Monographs, Oxazoles: Synthesis, reactions and Spectroscopy,
Part A; Palmer, D. C., Ed.; John Wiley: Hoboken, 2003.
(13) (a) Hashmi, A. S. K.; Rudolph, M.; Schymura, S.; Visus, J.; Frey,
W. Eur. J. Org. Chem. 2006, 4905. (b) Weyrauch, J. P.; Hashmi, A. S. K.;
Schuster, A.; Hengst, T.; Schetter, S.; Littmann, A.; Rudolph, M.; Hamzic,
M.; Visus, J.; Rominger, F.; Frey, W.; Bats, J. W. Chem.;Eur. J. 2010, 16,
956. (c) Hashmi, A. S. K. Pure Appl. Chem. 2010, 82, 657. (d) Hashmi,
A. S. K.; Weyrauch, J. P.; Frey, W.; Bats, J. W. Org. Lett. 2004, 6, 4391.
(e) Hashmi, A. S. K.; Schwarz, L.; Choi, L.-H.; Frost, T. M. Angew. Chem.,
Int. Ed. 2000, 39, 2285. (f) Hashmi, A. S. K.; Schuster, A. M.; Rominger
Angew. Chem., Int. Ed. 2009, 48, 8247. (g) Aguilar, D.; Contel, M.; Navarro,
R.; Soler, T.; Urriolabeitia, E. P. J. Organomet. Chem. 2009, 694, 486.
Figures 1 and 2 show that the gold atoms in 4b and 4d
adopt a near-linear geometry with P(1)-Au(1)-Cl(1) bond
angles of 176.31(4)° and 178.09(3)°, respectively, well within
(14) (a) Whittle, I. R.; Humphrey, M. G.; Samoc, M.; Luther-Davies,
B.; Hockless, D. C. R. J. Organomet. Chem. 1997, 544, 189. (b) Hashmi,
A. S. K.; Hengst, T.; Lothsch€ultz; Rominger, F. Adv. Synth. Catal. 2010,
352, 1315.

Article
Organometallics, Vol. 29, No. 18, 2010 4141
6000 Au-P bonds recorded in the Cambridge Structural
Database.17 There is a weak η1-interaction between the gold
atom and the ipso carbon atom of the anisyl ring in 4d; the
Au(1) C(17) distance of 3.150 A is less than the sum of the
3 3 3
van der Waals radii of gold and carbon and similar to the
Au Cipso interactions reported for gold(I) chloride com-
3 3 3
plexes of PCy2(o-biphenyl) and PBut2(o-biphenyl).15,16,18 In
stark contrast, even though the anisyl ring in diphenylpho-
sphino-based 4b adopts a similar orientation with respect to
the Au-P bond, the corresponding Au Cipso distance of
3 3 3
3.367 A is significantly longer and well outside the range
defined by the sum of the van der Waals radii. The Au(1)-
O(1) distances of 4.411 and 4.066 A in 4b and 4d, respectively,
are markedly longer than that of 3.27 A in [{PCy2(20,60-
dimethoxybiphenyl)}AuCl]. In this regard, a weak interac-
tion between palladium and the oxygen atom of a methoxy
group on the non-phosphine-containing aryl ring of
electron-rich biaryl monophosphines has been suggested
to contribute to the stability and hence efficiency of Pd/
SPHOS-based catalysts, and as such, the structures of a
number of Pd(0) and Pd(II) complexes of this ligand have
been determined; Pd-O interactions as short as 2.973 A have
been identified.19
Figure 1. Molecular structure of [{11-(diphenylphosphino)-
12-(2-methoxyphenyl)-9,10-dihydro-9,10-ethenoanthracene}-
AuCl] (4b). Hydrogen atoms and the dichloromethane molecule
of crystallization have been omitted for clarity. Ellipsoids are at
the 40% probability level.
Although it is operationally straightforward to generate
electrophilic gold(I) catalysts by abstraction of chloride with
a silver salt of a noncoordinating anion immediately prior to
the addition of substrate, it has recently become common
practice to isolate cationic complexes, as they are generally
robust and air-stable, which means they can be prepared on a
large scale, stored, and used directly, thus avoiding the use of
cocatalysts that are hygroscopic and difficult to weigh
accurately on a small scale.20 With the intention of compar-
ing the efficiency of catalyst generated in situ with the
corresponding isolated complex, cation 5 was generated by
reaction of 4d with AgOTf in dichloromethane, isolated, and
characterized by a single-crystal X-ray study; the molecular
structure is shown in Figure 3. The geometry about the gold
is close to linear, with a P(1)-Au(1)-O(2) bond angle of
175.52(10)°. The Au(1)-P(1) bond length of 2.2112(11) A is
shorter than those in related cationic complexes such as
[{PCy2(o-biphenyl)}Au(MeCN)][SbF6] (2.2466(3) A)15 and
[{PBut2(o-biphenyl)}Au(MeCN)][SbF6] (2.2539(7) A)15 as
well as that of 2.2244(10) A in the bis(trifluoromethane)-
sulfonimide complex of the same KITPHOS monophosp-
hine.9 The anisyl ring in 5 also adopts an orientation similar
to that in 4d such that it is in close proximity to the gold atom;
Figure 2. Molecular structure of [{11-(dicyclohexylphosphino)-
12-(2-methoxyphenyl)-9,10-dihydro-9,10-ethenoanthracene}AuCl]
(4d). Hydrogen atoms and the dichloromethane molecule of
crystallization have been omitted for clarity. Ellipsoids are at the
40% probability level.
the range reported for related complexes.15 The Au(1)-Cl(1)
bond length of 2.2794(11) A in 4b is slightly shorter than that
of 2.2935(7) A in 4d, which most likely reflects the different
trans influence of the dicyclohexylphosphino group com-
pared with its diphenylphosphino counterpart. The Au(1)-
P(1) bond length of 2.2348(7) A in 4d is comparable to those
reported for gold(I) complexes of electron-rich biaryl mono-
phosphines such as [{PCy2(o-biphenyl)}AuCl] (2.2364(6) A),15
[{PCy2(20,60-dimethoxybiphenyl)}AuCl] (2.2378(5) A),15 and
[{PCy2(20,40,60-triisopropylbiphenyl)}AuCl] (2.2349(11) A),16
whereas the corresponding bond length of 2.2188(9) A in 4b
is markedly shorter and is among the shortest 5% of over
the Au Cipso distance of 3.147 A is essentially the same as
3 3 3
the corresponding distance in 4d.
Complexes of the type [(R3P)AuL]X, which contain a
neutral weakly coordinating ligand L such as a nitrile, have
also been used as precursors for a host of gold(I)-catalyzed
skeletal rearrangements and transformations.21 In contrast,
(18) For a recent relevant review that discusses weak gold-arene
interactions see: Schmidbaur, H.; Schier, A. Organometallics 2010, 29, 2.
(19) (a) Biscoe, M. R.; Barder, T. E.; Buchwald, S. L. Angew. Chem.,
Int. Ed. 2007, 36, 7232. (b) Walker; Barder, T. E; Martinelli, J. R.; Buchwald,
S. L. Angew. Chem., Int. Ed. 2004, 43, 1871. (c) Barder, T. E.; Walker, S. D.;
Martinelli, J. R.; Buchwald, S. L. J. Am. Chem. Soc. 2005, 127, 5685.
ꢀ
ꢀ
(15) (a) Herrero-Gomez, E.; Nieto-Oberhuber, C.; Lopez, S.; Benet-
Buchholz, J.; Echavarren, A. M. Angew. Chem., Int. Ed. 2006, 45, 5455.
(b) Perez-Galan, P.; Delpont, N.; Herrero-Gomez, E.; Maseras, F.; Echavar-
ren, A. M. Chem.;Eur. J. 2010, 16, 5324.
ꢀ
(20) (a) Mezailles, N.; Ricard, L.; Gagosz, F. Org. Lett. 2005, 7, 4133.
For recent examples of a new class of thermally stable cationic gold(I)
precatalysts see: (b) Duan, H.; Sengupta, S.; Petersen, J. L.; Akhmedov,
N. G.; Shi, X. J. Am. Chem. Soc. 2009, 131, 12100. (c) Wang, D.; Ye, X.;
Shi, X. Org. Lett. 2010, 12, 2088. (d) Chen, Y.; Yam, W.; Akhmedov, N. G.;
Shi, X. Org. Lett. 2010, 12, 344. See also comment in: (e) Hashmi, A. S. K.;
Lothsch€utz, C. ChemCatChem. 2010, 2, 133.
(16) Partyka, D. V.; Robilotto, T. J.; Zeller, M.; Hunter, A. D.; Gray,
T. G. Organometallics 2008, 27, 28.
(17) Allen, F. H. Acta Crystallogr. Sect. B 2002, 58, 380. Cambridge
Structural Database, version 5.31 of 2009 plus 3 updates.

4142 Organometallics, Vol. 29, No. 18, 2010
Doherty et al.
Figure 4. Molecular structure of [{2-(dicyclohexylphosphino)-
12-phenyl-9,10-dihydro-9,10-ethenoanthracene}Au(tht)][SO3CF3]
(6). Hydrogen atoms, the chloroform molecule of crystalliza-
tion, and the trifluoromethanesulfonate anion have been omit-
ted for clarity. Ellipsoids are at the 40% probability level.
Figure 3. Molecular structure of one of the two independent
molecules of [{11-(dicyclohexylphosphino)-12-(2-methoxyphenyl)-
9,10-dihydro-9,10-ethenoanthracene}Au(SO3CF3)] (5). Hydro-
gen atoms have been omitted for clarity. Ellipsoids are at the
40% probability level.
molecular cyclizations, we chose the cycloisomerization of
propargyl amides13 from our initial study to systematically
compare the performance of diphenylphosphine-based KIT-
PHOS monophosphines 3a,b against their dicyclohexylpho-
sphine-based counterparts 3c,d, full details of which are
listed in Table 1. Our preliminary comparison focused on
cycloisomerization of the tert-butyl-based propargyl amide
7a in dichloromethane using 2 mol % catalyst generated
from 4a-d and silver triflate. Under these conditions each of
the catalysts gave excellent conversion to the corresponding
methylene oxazoline 8a with no evidence for the formation of
its oxazole counterpart; this is entirely consistent with recent
studies that have firmly established that Au(I)-based cata-
lysts are selective for the formation of the methyleneoxazo-
line, while Au(III) systems afford the corresponding oxa-
zole.13b,24 As near-quantitative conversions were obtained
for each catalyst, the comparison was extended to include a
range of alkyl-, aryl-, and heteroraryl-substituted substrates,
with the aim of revealing any differences in performance.
While each catalyst gave good to excellent conversions for all
substrates tested, those based on 3a,b consistently outper-
formed their dicyclohexylphosphino counterparts. The dif-
ference in performance between systems based on 3a,b and
3c,d manifests itself most evidently in the cycloisomerization
of the cyclohexyl- and phenyl-containing propargyl amides,
7b and 7c, respectively. For example, 2 mol % 4b/AgOTf
catalyzed cycloisomerization of the phenyl substrate to give
85% conversion after 1.5 h compared with only 51% conver-
sion for 4d/AgOTf in the same time. This trend in catalyst
efficiency also extended to the heteroaromatic propargyl
amides, although longer reaction times were required for each
substrate to reach acceptable conversions. For comparison,
the 4b/AgOTf-catalyzed cycloisomerization of thienyl substrate
there are very few reports of gold(I) complexes that contain
both a phosphine and a thioether and none that describe
their use in catalysis.22 Since we were interested in exploring
whether gold(I) complexes of tetrahydrothiophene (tht), or
indeed other thioethers could be an alternative class of
catalyst precursor, a dichloromethane solution of 4c, gener-
ated in situ by reaction of 3c with Au(tht)Cl in the presence
of a slight excess of tht, was treated with one equivalent of
silver triflate to give 6 in near-quantitative yield (eq 1). We
note here that our initial rationale for preparing 6 was the
low solubility of 4c in all common chlorinated and non-
chlorinated solvents; we reasoned that, as a cation, 6 would
be more soluble, easier to purify, and therefore a more
practical precatalyst, provided that the tetrahydrothiophene
is weakly coordinated and does not affect catalyst efficiency
(vide infra). As 6 is the first example of a gold(I) complex
supported by an electron-rich biaryl monophosphine and a
thioether,23 a single-crystal X-ray study was undertaken to
compare the key geometrical features with those of 4d and 5.
Figure 4 clearly shows that the phenyl ring is orientated
parallel to the Au-P bond; however, the Au(1) C(17) dis-
3 3 3
tance of 3.283 A is markedly longer than that of 3.150 and
3.147 A in 4d and 5, respectively. The gold atom in 6 has a
near-linear geometry, with a P(1)-Au(1)-S(1) bond angle of
175.60(7)°; for comparison the corresponding bond angles in
[(PPh3)Au(SMe2)][OTf]22b and [{PPh2(o-C6H4NH2)}Au(tht)]-
[ClO4]22c are 172.98° and 179.20°, respectively.
~
(22) (a) Bonide, E. V.; Grealis, J. P.; Muller-Bunz, H.; Ortin, Y.;
Casey, M.; Mendicute-Fierro, C.; Lagunas, M. C.; McGlinchey, M. J.
J. Organomet. Chem. 2008, 693, 1759. (b) Sladek, A.; Schmidbaur, H. Z.
Naturforsch. B. Chem. Sci. 1996, 51, 1207. (c) Lopez de Luzuriaga, J. M.;
Schier, A.; Schmidbaur, H. Chem. Ber. 1997, 130, 647.
Having already demonstrated that KITPHOS monopho-
sphines 4c,d are effective ligands for gold-catalyzed intra-
(23) For an example of a gold(I) complex of a triethylphosphine and
tetrahydrothiophene see: El-Etri, M. M.; Scovell, W. M. Inorg. Chem.
1990, 29, 480.
(24) The gold(I)-catalyzed intramolecular hydroamination of propar-
gyl alcohol-derived trichloroacetimidates selectively affords the methy-
lene dihydrooxazole with no evidence for the corresponding oxazole.
Kang, J.-E.; Kim, H.-B.; Lee, J.-W.; Shin, S. Org. Lett. 2006, 8, 3537.
(21) (a) Bolte, B.; Odabachian; Gagosz, F. J. Am. Chem. Soc. 2010,
ꢀ
ꢀ
132, 7294. (b) Amijs, C. H. M.; Lopez-Carrillo, V.; Raducan, M.; Perez-
ꢀ
Galan, P.; Ferrer, C.; Echavarren, A. M. J. Org. Chem. 2008, 73, 7721.
~
ꢀ
ꢀ
~
(c) Bieto-Oberhuber, C.; Munoz, M. P.; Lopez, S.; Jimenez-Nꢀuner, E.;
ꢀ
Nevado, C.; Herrero-Gomez, E.; Raducan, M.; Echavarren, A. M.
Chem.;Eur. J. 2006, 12, 1677.

Article
Organometallics, Vol. 29, No. 18, 2010 4143
Table 1. Gold(I)-Catalyzed Cycloisomerization of Propargyl Amides 7a-f
a Reaction conditions: (i) 2 mol % 4a-d, 2 mol % AgOTf, CH2Cl2, stir 30 min, (ii) 0.5 mmol propargyl amide, room temperature. b Reaction
conditions: (i) 2 mol % isolated 5, 0.5 mmol propargyl amide, CH2Cl2, room temperature. c Reaction conditions: (i) 2 mol % isolated 6, 0.5 mmol
propargyl amide, CH2Cl2, room temperature. d Reaction conducted in 1,2-dichloroethane at 50 °C. e See ref 13d for full experimental details.
f Conversions determined by 1H NMR analysis of the reaction mixture. Average of two runs.
7d reached 84% conversion after 4 h, whereas its furyl counter-
part 7e required heating at 50 °C for 5 h to achieve a similar
conversion; the corresponding dicyclohexylphosphino-based
catalysts gave 59% and 61% conversion, respectively. Finally,
while 4a-b/AgOTf and 4c-d/AgOTf both gave good conversions
of cinnamyl-based substrate 7f to methylene oxazoline 8f, the
former pair was consistently more efficient, albeit only margin-
ally. Reassuringly, there is an extremely close correlation bet-
ween the performance of catalyst generated by activation of 4d
with silver triflate and its isolated cation 5. As a slight excess of
cocatalyst is typically used to ensure quantitative activation of
the precatalyst and Lewis acid silver salts have been shown to
activate alkynes toward cyclization,25 parallel control experi-
ments were conducted for each substrate tested using 2 mol %
AgOTf in the absence of 4a-d. Although these experiments
conclusively showed that AgOTf is not an active cycloisome-
rization catalyst for propargyl amides,26 they do not exclude a
(26) Silver salts such as AgNO3 and AgSbF6 have been reported to
catalyze the cycloisomerization of N-propargylamides; however, sub-
strates lacking gem-dialkyl substitution showed no reaction in dichloro-
methane. Harmata, M.; Huang, C. Synlett 2008, 1399.
(25) (a) Oh, C. H.; Karmaker, S.; Park, H.; Ahn, Y.; Kim, J. W.
J. Am. Chem. Soc. 2010, 132, 1792. (b) Su, S.; Porco, J. A., Jr. J. Am. Chem.
Soc. 2007, 129, 7744. (c) Verniest, G.; Padwa, A. Org. Lett. 2008, 10, 4379.

4144 Organometallics, Vol. 29, No. 18, 2010
Doherty et al.
more subtle role for adventitious Agþ such as interception
elaborate the products of cycloisomerization by incorpora-
ting them into novel tandem/sequential reaction sequences.
Moreover, on the basis that new bulky triaryl-type mono-
phosphines have recently been reported to outperform their
triphenylphosphine counterparts in the hydrosilylation of
bulky ketones28 as well as the selective telomerization of
1,3-butadiene,29 we anticipate that triaryl-like monopho-
sphines based on the KITPHOS architecture may well find
use in a wide range of platinum group metal-catalyzed
transformations.
ꢀ
of Au(I) intermediates, as recently described by Gagne
and co-workers.27 Gratifyingly, the mixed phosphine-
tetrahydrothiophene cation 6 gave conversions that matched
those obtained with the corresponding system generated
from 4c. Thus, even though 6 was initially prepared to
overcome the low solubility of precatalyst 4c, it appears that
cationic gold(I)-phosphine complexes of tetrahydrothio-
phene may prove to be an alternative, practical, and highly
efficient class of precatalyst that are air-stable as well as easy
to prepare and handle.
Experimental Section
In an extension to our description of 3c,d as electron-rich
biaryl-like monophosphines, considering 3a,b as architec-
tural analogues of the corresponding triarylphosphines
prompted us to compare the efficiency of Au(I)/3a against
[Au(PPh3)]X (X = OTf, NTf2).13b In all cases 4a/AgOTf is
markedly more efficient than its triphenylphosphine coun-
terpart, which suggests that this analogy may have limited
validity. The disparate performance of these system is
clearly evident for the phenyl-substituted propargyl amide
7c, which reached 74% conversion after only 30 min with 2
mol % 4a/AgOTf, whereas [Au(PPh3)]NTf2 required 36 h
to reach a comparable conversion, albeit with a slightly
lower catalyst loading (1 mol %). Similarly, the cinnamyl
substrate 7f reached good conversion after 1.5 h using
catalyst based on 3a, whereas [Au(PPh3)]NTf2 required 36
h. At this stage we can only speculate about the origin of the
disparate performance of these two systems but note that
the non-phosphine-containing aryl ring in 3a forms a weak
interaction with gold, which is not possible for triphenyl-
phosphine. The importance of the biaryl unit in facilitating
palladium-catalyzed cross-coupling is well documented and
has been shown to be associated at least in part with the
formation of a weak stabilizing interaction between palla-
dium and the non-phosphine-containing aryl ring. A corol-
lary of this might be that a biaryl/biaryl-like unit is necessary
for efficient catalysis regardless of the basicity of the phos-
phine, which may have only a minor influence on perfor-
mance; in this regard a systematic comparison between 3a,
triphenylphosphine, PCy2(o-biphenyl), PCy3, and Ph2P(o-
biphenyl) might prove insightful.
General Comments. All manipulations involving air-sensitive
materials were carried out using standard Schlenk line techni-
ques under an atmosphere of nitrogen or argon in oven-dried
glassware. Dichloromethane and chloroform were distilled from
calcium hydride, THF was distilled from sodium, and hexane
and diethyl ether were distilled from Na/K alloy, under an atmo-
sphere of nitrogen. All amines were purchased from commercial
suppliers and purified by passing through a short column of
alumina immediately prior to use. Phenylacetylene, 2-ethynyl-
anisole, chlorodiphenylphosphine, and anthracene were purchased
from commercial suppliers and used without further purifica-
tion. (Diphenylphosphinoylethynyl)benzene (1a),30 11-(dicyclohe-
xylphosphino)-12-phenyl-9,10-dihydro-9,10-ethenoanthracene
(2c),8 11-(dicyclohexylphosphino)-12-(2-methoxyphenyl)-9,10-
dihydro-9,10-ethenoanthracene (2d),8 AuCl(tht),14 and the pro-
pargyl amides11 were prepared as previously described. 1H and
13C{1H} NMR spectra were recorded on JEOL LAMBDA-500
or ECS-400 instruments. Thin-layer chromatography was car-
ried out on aluminum sheets precoated with silica gel 60F 254,
and column chromatography was performed using Merck Kie-
selgel 60. Gas chromatography-mass spectrometry was per-
formed on a Saturn 2220 GC-MS system using a factorFour VF-
5 ms capillary column, 30 m, 0.25 mm, and 0.25 μm.
Synthesis of 2-(Diphenylphosphinoylethynyl)anisole (1b). To a
solution of 2-ethynylanisole (2.50 g, 18.9 mmol) in THF (65 mL)
cooled to -78 °C was added BuLi (7.56 mL, 2.5 M, 18.9 mmol).
The reaction was allowed to warm to 0 °C, stirred for 30 min,
and then cooled to -78 °C. Chlorodiphenylphosphine (3.40 mL,
18.4 mmol) was added dropwise, and the solution was allowed
to warm to room temperature and stirred for a further 2.5 h. The
reaction was then cooled to 0 °C and hydrogen peroxide (35%
aqueous solution, 8.32 mL, 24.6 mmol) added. The solution was
allowed to warm to room temperature and stirred for 30 min.
Water (90 mL) was added and the product extracted with diethyl
ether (3 ꢀ 40 mL). The organic fractions were combined and
dried over MgSO4, and the solvent was removed in vacuo to
leave a yellow solid. Purification by column chromatography
eluting with CH2Cl2/ethylacetate (2:3) gave the desired product
as a white crystalline solid in 84% yield (5.2 g). 31P{1H} NMR
(202.5 MHz, CDCl3, δ): 9.0. 1H NMR (399.78 MHz, CDCl3, δ):
7.33 (ddd, J = 14.2, 8.7, 1.8 Hz, 4H, Ar-H), 7.44-7.39 (m, 7H,
Ar-H), 7.32 (td, J = 8.7, 1.9 Hz, 1H, Ar-H), 6.85 (t, J = 7.7 Hz,
1H, Ar-H), 6.83 (d, J = 6.4 Hz, 1H, Ar-H), 3.82 (s, 3H, OCH3).
13C{1H} NMR (125.76 MHz, CDCl3, δ): 161.7 (C6H4), 134.1
(C6H5), 133.5 (d, J = 121 Hz, C6H4), 132.3 (C6H4), 132.0 (d, J =
2.9 Hz, C6H5), 131.0 (d, J = 11.5 Hz, C6H5), 128.7 (d, J = 13.4
Hz, C6H5), 120.6 (C6H4), 110.7 (C6H4), 109.2 (d, J = 3.1 Hz,
C6H4), 102.8 (d, J = 30.5 Hz, CtCP), 85.6 (d, J = 172.6 Hz,
Conclusions
In conclusion, electrophilic gold(I) complexes of KIT-
PHOS monophosphines catalyze the 5-exo-dig cycloiso-
merization of a range of propargyl amides. Comparative
catalyst testing has revealed that systems based on diphenyl-
phosphino-substituted KITPHOS monophosphines out-
perform their electron-rich dicyclohexylphosphino-based
counterparts and that the former are markedly more effi-
cient than [Au(PPh3)]þ, the benchmark complex previously
used to catalyze this reaction. The first air-stable cationic
gold(I) complex coordinated by an electron-rich monopho-
sphine and tetrahydrothiophene to be isolated is a highly
efficient cycloisomerization catalyst, giving conversions
that match those obtained with the corresponding catalyst
generated by chloride abstraction. Further studies are cur-
rently underway to (i) extend these cyclizations to the syn-
thesis of a broader range of heterocycles, (ii) develop gold-
(I)-catalyzed cascade carbocyclizations for the construction
of complex heteropolycyclic architectures, and (iii) further
(28) (a) Fujihara, T.; Samba, K.; Terao, J.; Tsuji, Y. Angew. Chem.,
Int. Ed. 2010, 49, 1472. (b) Ohta, H.; Tokunaga, M.; Obora, Y.; Iwai, T.;
Iwasawa, T.; Fujihara, T.; Tsuji, Y. Org. Lett. 2007, 9, 89.
´ ´
(29) Tschan, M. J.-L.; Garcıa-Suırez, E. J.; Zoraida, F.; Launay, H.;
Hagen, H.; Benet-Buchholz, J.; van Leeuwen, P. W. N. M. J. Am. Chem.
Soc. 2010, 132, 6463.
(30) Liu, B.; Wang, K. K.; Petersen, J. L. J. Org. Chem. 1996, 61,
8503.
ꢀ
(27) Weber, D.; Gagne, M. R. Org. Lett. 2009, 11, 4962.

Article
Organometallics, Vol. 29, No. 18, 2010 4145
CtCP), 55.7 (OCH3). LRMS (EIþ): m/z 333 [M þ H]þ. HRMS
(ESIþ): exact mass calcd for C21H18O2P [M þ H]þ requires m/z
333.1044, found m/z 333.1043. Anal. Calcd for C21H17O2P: C,
75.90; H, 5.16. Found: C, 76.32; H, 5.44.
was charged with 2a (4.0 g, 8.33 mmol), toluene (70 mL), and
triethylamine (40 mL, 287 mmol). Trichlorosilane (8.35 mL,
82.6 mmol) was added slowly and the mixture heated at 110 °C
for 2 days. The reaction mixture was diluted with diethyl ether
(100 mL) and added slowly to a mixture of ice (60 g) and 20%
aqueous NaOH (110 mL). After stirring vigorously at room
temperature for 30 min, the organic layer was removed and the
aqueous phase extracted with diethyl ether (3 ꢀ 100 mL). The
organic fractions were combined, washed with saturated NaH-
CO3 (2 ꢀ 100 mL), water (2 ꢀ 100 mL), and brine (2 ꢀ 100 mL),
dried over MgSO4, and filtered, and the solvent was removed in
vacuo. The product was purified by column chromatography
eluting with hexane/ethyl acetate (9:1) to afford 3a as a spectro-
scopically pure white solid in 67% yield (2.59 g). An analytically
pure sample was obtained by slow diffusion of a chloroform
solution layered with hexane at room temperature. 31P{1H}
NMR (202.5 MHz, CDCl3, δ): -12.8. 1H NMR (500.16 MHz,
CDCl3, δ): 7.38-7.31 (m, 13H, Ar-H), 7.22 (br t, J = 6.9 Hz,
4H, Ar-H), 6.98 (t, J = 6.9 Hz, 2H, Ar-H), 6.91 (t, J = 6.9 Hz,
2H, Ar-H), 6.81 (d, J = 7.3 Hz, 2H, Ar-H), 5.54 (d, J = 2.6 Hz,
1H, bridgehead CH), 5.06 (d, J = 1.5 Hz, 1H, bridgehead CH).
13C{1H} NMR (125.76 MHz, CDCl3, δ): 163.6 (d, J = 22.9 Hz,
C=CP), 145.8 (C6H4Q), 145.2 (C6H4Q), 139.3 (d, J = 22.9 Hz,
CdCP), 138.7 (d, J = 5.7 Hz, C6H5Q), 137.0 (d, J = 11.4 Hz,
C6H5Q), 133.4 (C6H5), 133.2 (C6H5), 128.5 (d, J = 5.7 Hz,
C6H5), 128.4 (C6H5), 128.2 (C6H5), 128.1 (C6H5), 124.7 (C6H4),
124.6 (C6H4), 123.3 (C6H4), 122.8 (C6H4), 59.4 (d, J = 6.7 Hz,
bridgehead CH), 55.2 (d, J = 5.7 Hz, bridgehead CH). LRMS
(EIþ) m/z 465 [M þ H]þ. HRMS (ESIþ): exact mass calcd for
C34H26P [M þ H]þ requires m/z 465.1722, found m/z 465.1772.
Anal. Calcd for C34H25P: C, 87.91; H, 5.42. Found: C, 88.24; H,
5.55.
Synthesis of 2-(Diphenylphosphinoyl)-12-phenyl-9,10-dihydro-
9,10-ethenoanthracene (2a). (Diphenylphosphinoylethynyl)ben-
zene (2.0 g, 6.58 mmol) and anthracene (4.72 g, 26.5 mmol) were
mixed in a flask, which was gradually heated to 220 °C using a
DrySyn multiposition heating block. The temperature was then
lowered to 200 °C and the mixture heated for a further 10 h. The
resulting dark solid residue was purified by column chromato-
graphy eluting with CH2Cl2/ethyl acetate (2:3) to afford 2a as an
off-white solid in 81% yield (2.55 g). 31P{1H} NMR (202.5
MHz, CDCl3, δ): 25.5. 1H NMR (500.16 MHz, CDCl3, δ): 7.43
(d, J = 7.8 Hz, 2H, Ar-H), 7.41 (d, J = 7.8 Hz, 2H, Ar-H), 7.36
(m, 4H, Ar-H), 7.25 (t br, J = 9.5 Hz, 4H, Ar-H), 7.11 (t br, J =
6.3 Hz, 4H, Ar-H), 7.06-6.98 (m, 7H, Ar-H), 5.42 (d, J = 2.8
Hz, 1H, bridgehead CH), 5.26 (d, J = 8.7 Hz, 1H, bridgehead
CH). 13C{1H} NMR (125.76 MHz, CDCl3, δ): 166.2 (d, J = 7.6
Hz, CdCP), 144.5 (C6H4Q), 144.3 (C6H4Q), 137.6 (d, J = 4.8
Hz, C6H5Q), 136.5 (d, J = 101 Hz, C=CP), 132.9 (d, J = 105
Hz, C6H5Q), 131.7 (C6H5), 131.6 (C6H5), 131.4 (C6H5), 128.3 (d,
J = 12.4 Hz, C6H5), 128.1 (C6H5), 127.7 (C6H5), 125.3 (C6H4),
125.1 (C6H4), 123.7 (C6H4), 123.4 (C6H4), 61.0 (d, J = 9.5 Hz,
bridgehead CH), 53.7 (d, J = 10.4 Hz, bridgehead CH). LRMS
(EIþ): m/z 481 [M þ H]þ. HRMS (ESIþ): exact mass calcd for
C34H26OP [M þ H]þ requires m/z 481.1721, found m/z
481.1724. Anal. Calcd for C34H25OP: C, 84.98; H, 5.24. Found:
C, 85.31; H, 5.62.
Reduction of 2b to 11-(Diphenylphosphino)-12-(2-methoxyphe-
nyl)-9,10-dihydro-9,10-ethenoanthracene (3b). Compound 2b
(3.5 g, 6.86 mmol) was reduced according to the procedure
described above for 2a to afford 3b as an off-white solid in 66%
yield (2.24 g) after purification by column chromatography,
eluting with hexane/ethyl acetate (85:15). An analytically and
spectroscopically pure sample was obtained by slow diffusion of
a chloroform solution layered with hexane at room temperature.
31P{1H} NMR (202.5 MHz, CDCl3, δ): -13.9. 1H NMR (500.16
MHz, CDCl3, δ): 7.35-7.19 (br m, 12H, Ar-H), 6.97 (t, J = 7.3
Hz, 2H, Ar-H), 6.92 (d, J = 8.3 Hz, 1H, Ar-H), 6.90 (br t, J =
7.3 Hz, 2H, Ar-H), 6.83 (t, J = 7.4 Hz, 1H, Ar-H), 6.80 (br, 4H,
Ar-H), 5.46 (d, J = 2.8 Hz, 1H, bridgehead CH), 5.02 (d, J = 1.8
Hz, 1H, bridgehead CH), 3.72 (s, 3H, OCH3). 13C{1H} NMR
(125.76 MHz, CDCl3, δ): 163.2 (d, J = 30.5 Hz, CdCP), 157.4
(C6H4OMe), 146.1 (br, C6H5), 145.7 (2 ꢀ C6H4), 139.9 (d, J =
21.1 Hz, C6H4OMe), 137.3 (d, J = 11.4 Hz, C6H4OMe), 133.4
(C6H5), 133.2 (C6H5), 130.8 (d, J = 6.7 Hz, C6H5), 129.7 (C6H5),
128.4 (d, J = 5.8 Hz, C6H5), 128.2 (C6H5), 128.0 (d, J = 6.7 Hz,
C6H4OMe), 124.4 (C6H4), 124.3 (C6H4), 123.5 (C6H4), 123.1
(C6H4), 120.0 (C6H4OMe), 111.0 (C6H4OMe), 58.8 (d, J = 6.7
Hz, bridgehead CH), 55.3 (OCH3), 54.9 (d, J = 5.7 Hz, bridge-
head CH). LRMS (EIþ): m/z 495 [M þ H]þ. HRMS (ESIþ)
exact mass calcd for C35H28OP [M þ H]þ requires m/z 495.1878,
found m/z 495.1858. Anal. Calcd for C35H27OP: C, 85.00; H,
5.50. Found: C, 85.35; H, 5.73.
Synthesis of 11-(Diphenylphosphinoyl)-12-(2-methoxyphenyl)-
9,10-dihydro-9,10-ethenoanthracene (2b). Compound 2b was
prepared from 2-(diphenylphosphinoylethynyl)anisole (2.0 g,
6.01 mmol) according to the procedure described above for 2a
and isolated as an analytically pure off-white solid in 77% yield
(2.36 g) after purification by column chromatography eluting
with CH2Cl2/ethyl acetate (1:1). 31P{1H} NMR (202.5 MHz,
CDCl3, δ): 25.8. 1H NMR (500.16 MHz, CDCl3, δ): 7.5-6.87
(br m, 19H, Ar-H), 6.69 (dd, J = 7.4, 1.9 Hz, 1H, Ar-H), 6.43 (d,
J = 8.2 Hz, 1H, Ar-H), 6.39 (t, J = 7.6 Hz, Ar-H), 5.20 (d, J =
4.1 Hz, 1H, bridgehead CH), 5.18 (d, J = 9.7 Hz, 1H, bridge-
head CH), 3.57 (s, 3H, OCH3). 13C{1H} NMR (125.76 MHz,
CDCl3, δ): 165.6 (d, J = 11.4 Hz, CdCP), 156.2, 145.3 (br),
144.4 (br), 137.0 (d, J = 103 Hz), 133.2 (br), 132.2 (br), 131.5
(br), 130.7, 129.7, 128.0 (br), 126.6, 124.9 (br), 124.8, 123.6,
119.9, 109.7, 60.1 (d, J = 7.6 Hz, bridgehead CH), 54.7 (OCH3),
53.6 (d, J = 7.4 Hz, bridgehead CH). LRMS (EIþ) m/z 511 [M þ
H]þ. HRMS (ESIþ): exact mass calcd for C35H28O2P [M þ H]þ
requires m/z 511.1827, found m/z 511.1816. Anal. Calcd for
C35H27O2P: C, 82.34; H, 5.33. Found: C, 82.66; H, 5.41.
Synthesis of [{2-(Diphenylphosphino)-12-phenyl-9,10-dihydro-
9,10-ethenoanthracene}AuCl] (4a). To a solution of Au(tht)Cl
(0.25 g, 0.78 mmol) in dichloromethane (8-10 mL) was added a
solution of 3a (0.37 g, 0.80 mmol) in dichloromethane (2-3 mL).
After stirring for 1 h the solvent was removed under reduced
pressure and the resulting white solid triturated with diethyl
Reduction of 2a to 11-(Diphenylphosphino)-12-phenyl-9,10-
dihydro-9,10-ethenoanthracene (3a). A flame-dried Schlenk flask

4146 Organometallics, Vol. 29, No. 18, 2010
Doherty et al.
ether (2 ꢀ 5 mL). Crystallization by slow diffusion of a chloro-
form solution layered with hexane at room temperature gave 4a
in 79% yield (0.43 g). 31P{1H} NMR (202.5 MHz, CDCl3, δ):
23.2. 1H NMR (500.16 MHz, CDCl3, δ): 7.54 (t, J = 7.4 Hz, 2H
Ar-H), 7.46-7.30 (m, 13H, Ar-H), 7.07 (m, 4H, Ar-H), 6.99 (t,
J = 7.4 Hz, 2H, Ar-H), 6.87 (t, J = 6.9 Hz, 2H, Ar-H), 5.50 (d,
J = 4.3 Hz, 1H, bridgehead CH), 4.98 (d, J = 7.8 Hz, 1H,
bridgehead CH). 13C{1H} NMR (125.76 MHz, CDCl3, δ): 169.1
(d, J = 30.5 Hz, CdCP), 144.3 (C6H4Q), 144.2 (C6H4Q), 137.3
(d, J = 6.7 Hz, C6H5Q), 133.9 (d, J = 14.3 Hz, C6H5), 132.1 (d,
J = 57.2 Hz, CdCP), 131.8 (C6H5), 129.7 (C6H5), 129.4 (C6H5),
129.2 (C6H5), 129.1 (C6H5), 128.8 (C6H5), 125.6 (C6H4), 125.3
(C6H4), 123.6 (C6H4), 123.3 (C6H4), 60.8 (d, J = 8.6 Hz,
bridgehead CH), 54.8 (d, J = 4.7 Hz, bridgehead CH). Anal.
Calcd for C34H25AuClP: C, 58.59; H, 3.62. Found: C, 58.81; H,
3.97.
Synthesis of [{11-(Diphenylphosphino)-12-(2-methoxyphenyl)-
9,10-dihydro-9,10-ethenoanthracene}AuCl] (4b). Compound 4b
was prepared according to the procedure described above for 4a
on the same scale and isolated as a white solid in 63% yield (0.36
g) by slow diffusion of hexane into a chloroform solution at
room temperature. 31P{1H} NMR (202.5 MHz, CDCl3, δ): 23.7.
1H NMR (500.16 MHz, CDCl3, δ): 7.51 (m br, 2H Ar-H),
7.45-7.34 (m, 8H, Ar-H), 7.35 (d br, J = 11.7 Hz, 1H, Ar-H),
7.28 (d br, J = 7.3 Hz, 2H, Ar-H), 7.06-6.99 (m br, 4H, Ar-H),
6.95 (br d, J = 7.3 Hz, 1H, Ar-H), 6.91 (d, J = 7.8 Hz, 1H, Ar-
H), 6.79 (td, J = 7.3, 0.9 Hz, 1H, Ar-H), 6.70 (br d, J = 11.7 Hz,
1H, Ar-H), 6.66 (dd, J = 7.4, 1.4 Hz, 1H, Ar-H), 5.44 (d, J = 4.1
Hz, 1H, bridgehead CH), 4.95 (d, J = 7.8 Hz, 1H, bridgehead
CH), 3.63 (s, 3H, OCH3). 13C{1H} NMR (125.76 MHz, CDCl3,
δ): 167.8 (d, J = 13.4 Hz, CdCP), 156.8 (C6H4OMe), 144.9
(C6H5Q), 144.7 (C6H4Q), 143.9 (C6H4Q), 133.9 (C6H5), 133.8
(C6H5), 132.5 (d, J = 60.1 Hz, CdCP), 131.5 (br d, J = 37.2 Hz,
C6H5), 131.1 (C6H4OMe), 129.9 (C6H4OMe), 129.2 (br d, J =
12.4 Hz, C6H5), 128.9 (br d, J = 11.4 Hz, C6H5), 126.3 (d, J =
7.6 Hz, C6H4OMe), 125.2 (br d, J = 28.6 Hz, C6H4), 124.9 (br d,
J = 33.3 Hz, C6H4), 123.7 (C6H4), 123.0 (br d, J = 2.4 Hz,
C6H4), 120.5 (C6H4OMe), 111.5 (C6H4OMe), 60.1 (d, J = 8.5
Hz, bridgehead CH), 54.0 (OCH3), 54.6 (d, J = 5.7 Hz, bridge-
head CH). Anal. Calcd for C35H27AuClOP: C, 57.82; H, 3.74.
Found: C, 58.12; H, 4.03.
7.22 (dd, J = 9.6, 1.9 Hz, 1H, Ar-H), 7.02-6.99 (m, 4H, Ar-H),
6.94 (d, J = 8.7 Hz, 2H, Ar-H), 6.67 (dd, J=7.3, 1.4Hz, 1H, Ar-H),
5.34 (d, J = 5.5 Hz, 1H, bridgehead CH), 5.16 (d, J = 3.2 Hz,
1H, bridgehead CH), 2.16 (br m, 2H, Cy-H), 1.93-1.50 (m, 8H,
Cy-H), 1.34-1.05 (m, 12H, Cy-H). 13C{1H} NMR (125.76 MHz,
CDCl3, δ): 168.3 (d, J = 11.4 Hz, CdCP), 156.7 (C6H4OMe),
144.9 (C6H4Q), 144.9 (C6H4Q), 144.1 (C6H4Q), 143.5 (C6H4Q),
132.6 (d, J = 49.6 Hz, CdCP), 130.4 (C6H4OMe), 129.3
(C6H4OMe), 127.7 (d, J = 6.7 Hz, C6H4OMe), 125.4 (C6H4),
125.2 (C6H4), 125.1 (C6H4), 124.7 (C6H4), 123.8 (C6H4), 123.7
(C6H4), 122.8 (C6H4), 122.6 (C6H4), 121.0 (C6H4OMe), 111.2
(C6H4OMe), 60.3 (d, J = 7.6 Hz, bridgehead CH), 54.9 (OCH3),
54.6 (d, J = 4.2 Hz, bridgehead CH), 35.5 (d, J = 35.1 Hz, Cy),
35.2 (d, J = 35.3 Hz, Cy), 31.1 (d, J = 3.8 Hz, Cy), 29.9 (Cy), 29.7
(d, J = 2.9Hz, Cy), 29.7(Cy), 26.9 (d, J =12.4Hz, Cy), 26.7 (Cy),
26.6 (Cy), 26.4 (d, J = 13.3 Hz, Cy), 25.8 (Cy), 25.6 (Cy). Anal.
Calcd for C35H39AuClOP: C, 56.88; H, 5.32. Found: C, 57.09;
H, 5.44.
Synthesis of [{11-(Dicyclohexylphosphino)-12-(2-methoxyph-
enyl)-9,10-dihydro-9,10-ethenoanthracene}Au(SO3CF3)] (5). A
flame-dried Schlenk flask charged with 4d (0.30 g, 4.06 mmol),
AgOTf (0.104 g, 4.06 mmol), and CH2Cl2 (10 mL) was stirred at
room temperature for 60 min, after which time the cloudy solu-
tion was filtered, the solvent removed in vacuo, and the residue
crystallized by slow diffusion of hexane into a dichloromethane
solution to give 5 as colorless crystals in 64% yield (0.22 g).
31P{1H} NMR (161.8 MHz, CDCl3, δ): 34.3. 1H NMR (399.78
MHz, CDCl3, δ): 7.45 (td, J = 8.8, 1.9 Hz, 1H, Ar-H), 7.34 (td,
J = 8.3, 1.4 Hz, 2H, Ar-H), 7.31 (d, J = 7.8 Hz, 1H, Ar-H), 7.25
(d, J = 7.8 Hz, 1H), 7.08-7.03 (m, 4H, Ar-H), 6.97 (t, J = 8.2
Hz, 2H, Ar-H), 6.67 (dd, J = 7.4, 1.9 Hz, 1H, Ar-H), 5.35 (d,
J = 5.5 Hz, 1H, bridgehead CH), 5.20 (d, J = 3.7 Hz, 1H,
bridgehead CH), 3.61 (s, 3H, OCH3), 2.18-1.53 (m, 12H, Cy-
H), 1.36-1.04 (m, 10H, Cy-H). 13C{1H} NMR (125.76 MHz,
CDCl3, δ): 169.8 (d, J = 10.5 Hz, CdCP), 156.6 (C6H4OMe),
144.6 (C6H4Q), 144.5 (C6H4Q), 143.8 (C6H4Q), 143.2 (C6H4Q),
130.9 (C6H4OMe), 132.6 (d, J = 56.2 Hz, CdCP), 128.9
(C6H4OMe), 127.2 (d, J = 7.6 Hz, C6H4OMe), 125.6 (C6H4),
125.4 (C6H4), 125.3 (C6H4), 124.9 (C6H4), 123.9 (C6H4), 123.9
(C6H4), 122.7 (C6H4), 122.5 (C6H4), 120.7 (C6H4OMe), 120.1 (q,
J = 316 Hz, CF3SO3), 111.5 (C6H4OMe), 60.2 (d, J = 7.6 Hz,
bridgehead CH), 54.9 (OCH3), 54.1 (d, J = 3.82 Hz, bridgehead
CH), 35.1 (d, J = 37.1 Hz, Cy), 34.8 (d, J = 36.2 Hz, Cy), 31.2
(d, J = 3.9 Hz, Cy), 29.9 (2 ꢀ Cy), 29.6 (Cy), 26.9 (d, J = 12.3 Hz,
Cy), 26.5(d, J =3.8Hz, Cy), 26.4(d, J = 2.8 Hz, Cy), 26.3 (d, J =
13.3 Hz, Cy), 25.7 (Cy), 25.5 (Cy). Anal. Calcd for C36H39Au-
F3O4PS: C, 50.71; H, 4.61. Found: C, 51.03; H, 4.78.
Synthesis of [{2-(Dicyclohexylphosphino)-12-phenyl-9,10-di-
hydro-9,10-ethenoanthracene}AuCl] (4c). Compound 4c was
prepared according to the procedure described above for 4a
on the same scale and isolated as a white solid in 92% yield
(0.507 g) after trituration with diethyl ether (2 ꢀ 5 mL). 31P{1H}
1
NMR (202.5 MHz, CDCl3, δ): 37.1. H NMR (500.16 MHz,
CDCl3, δ): 7.40 (t, J = 6.4 Hz, 1H, Ar-H), 7.33 (td, J = 7.8, 1.4
Hz, 2H, Ar-H), 7.30-7.24 (m, 4H, Ar-H), 7.01-6.96 (m, 4H,
Ar-H), 6.84 (dd, J = 6.9, 1.4 Hz, 2H, Ar-H), 5.32 (d, J = 5.5 Hz,
1H, bridgehead CH), 5.17 (d, J = 3.6 Hz, 1H, bridgehead CH),
2.12 (br m, 2H, Cy-H), 1.86 (br, 2H, Cy-H), 1.73 (br, 2H, Cy-H),
1.56 (br m, 6H, Cy-H), 1.30-1.02 (m, 10H, Cy-H); 13C{1H}
NMR (125.76 MHz, CDCl3, δ): 167.8 (d, J = 14.3 Hz, C=CP),
144.2 (C6H4Q), 143.8 (C6H4Q), 139.0 (d, J = 6.7 Hz, C6H5Q),
132.5 (d, J = 49.6 Hz, CdCP), 129.3 (C6H5), 128.8 (C6H5),
126.9 (C6H5), 125.6 (C6H4), 125.3 (C6H4), 123.6 (C6H4), 123.0
(C6H4), 61.2 (d, J = 7.6 Hz, bridgehead CH), 54.6 (d, J = 4.6
Hz, bridgehead CH); 35.2 (d, J = 35.3 Hz, Cy), 30.4 (d, J = 4.7
Hz, Cy), 29.8 (Cy), 26.6 (d, J = 7.6 Hz, Cy), 26.4 (d, J = 5.7 Hz,
Cy), 25.6 (Cy). Anal. Calcd for C34H37AuClP: C, 57.59; H, 5.26.
Found: C, 57.71; H, 5.32.
Synthesis of [{2-(Dicyclohexylphosphino)-12-phenyl-9,10-di-
hydro-9,10-ethenoanthracene}Au(tht)][SO3CF3] (6). To a solu-
tion of Au(tht)Cl (0.200 g, 0.63 mmol) and tetrahydrothiophene
(0.11 mL, 1.26 mmol) in dichloromethane (10 mL) was added a
solution of 3c (0.302 g, 0.65 mmol) in dichloromethane (2-3
mL), and the resulting mixture was stirred for 1 h, after which
time silver trifluoromethanesulfonate (0.16 g, 0.63 mmol) was
added. The reaction mixture was stirred for an additional 30 min
and filtered, and the solvent was removed under reduced
pressure to afford 6 as a white powder. Crystallization by slow
diffusion of a chloroform solution layered with hexane at room
temperature gave 6 in 71% yield (0.406 g). 31P{1H} NMR (202.5
MHz, CDCl3, δ): 40.6. 1H NMR (500.16 MHz, CDCl3, δ): 7.45
(m, 5H, Ar-H), 7.30 (dd, J = 5.9, 1.9 Hz, 2H, Ar-H), 7.04 (m,
4H, Ar-H), 6.96 (dd, J = 5.9, 1.9 Hz, 2H, Ar-H), 5.58 (d, J = 5.1
Hz, 1H, bridgehead CH), 5.22 (d, J = 7.1 Hz, 1H, bridgehead
CH), 3.0 (br t, J = 6.4 Hz, 4H, SCH2), 2.55 (br q, J = 9.4 Hz,
2H, Cy-H), 1.99 (m, 6H, SCH2CH2 þ Cy-H), 1.82 (br d, J =
11.4 Hz, 2H, Cy-H), 1.69 (br t, J = 11.4 Hz, 4H, Cy-H), 1.47 (br
q, J = 12.8 Hz, 2H, Cy-H), 1.34 (br, 2H, Cy-H), 1.25 (br q, J =
12.8 Hz, 2H, Cy-H), 1.09 (m, 6H, Cy-H). 13C{1H} NMR (125.76
MHz, CDCl3, δ): 170.4 (d, J = 11.4 Hz, CdCP), 143.9 (C6H4Q),
143.1 (C6H4Q), 139.8 (d, J = 6.7 Hz, C6H5Q), 133.2 (d, J =
Synthesis of [{11-(Dicyclohexylphosphino)-12-(2-methoxyph-
enyl)-9,10-dihydro-9,10-ethenoanthracene}AuCl] (4d). Compound
4d was prepared according to the procedure described above for
4a onthesame scale andisolatedasawhitesolidin84%yield(0.48
g) by slow diffusion of hexane into a chloroform solution at room
1
temperature. 31P{1H} NMR (161.8 MHz, CDCl3, δ): 37.5. H
NMR (399.78 MHz, CDCl3, δ): 7.44 (td, J = 7.8, 1.8 Hz, 1H, Ar-
H), 7.32 (d, J = 8.2Hz, 1H, Ar-H), 7.30 (t, J = 7.8Hz, 2H, Ar-H),

Article
Organometallics, Vol. 29, No. 18, 2010 4147
Table 2. Crystallographic Data
4d
4b
5
6
chem form
form wt
cryst syst
space group
a (A)
C
811.9
monoclinic
P21/c
13.5161(3)
17.2666(4)
15.3084(4)
112.921(3)
3290.54(14)
4
35H27AuClOP CH2Cl2
3
C35H39AuClOP 0.5CH2Cl2
3
C36H39AuF3O4PS
852.7
monoclinic
P21/c
20.9477(4)
19.3370(3)
16.9133(2)
96.282(1)
6809.86(19)
8
C38H45AuPSþ CF3O3S- CHCl3
1030.2
monoclinic
P21/n
9.0277(6)
19.2721(11)
24.1482(14)
98.851(6)
4151.3(4)
4
3
3
781.5
monoclinic
P21/n
11.7404(3)
20.3209(4)
13.8375(3)
103.601(2)
3208.71(12)
4
4.828
19 318
6721
0.0331
b (A)
c (A)
β (deg)
V (A3)
Z
μ (mm-1
)
4.791
22 102
7151
0.0347
4.483
51 986
11 965
0.0525
3.926
34 976
7295
0.0735
reflns measd
unique reflns
Rint
refined params
R (F, F2 > 2σ)
Rw (F2, all data)
GoF (F2)
353
0.0304
0.0690
0.920
353
0.0226
0.0451
0.929
831
0.0295
0.0557
0.850
478
0.0505
0.1311
1.010
diff map extremes (e A-3
)
2.82, -1.36
0.87, -0.91
1.94, -2.07
1.71, -2.47
47.6 Hz, CdCP), 129.6 (C6H5), 128.8 (C6H5), 127.3 (C6H5),
125.8 (C6H4), 125.6 (C6H4), 123.6 (C6H4), 123.5 (C6H4), 121.2
(q, J = 318.6 Hz, CF3SO3), 61.0 (d, J = 7.6 Hz, bridgehead
CH), 54.1 (bridgehead CH), 39.0 (SCH2), 34.4 (d, J = 34.3 Hz,
Cy), 30.9 (SCH2CH2 þ Cy), 30.5 (Cy), 26.4 (Cy), 26.2 (Cy), 25.3
(Cy). Anal. Calcd for C39H45AuF3O3PS2: C, 51.43; H, 4.98.
Found: C, 51.67; H, 5.05.
X-ray Crystallography. All data were measured on an Oxford
Diffraction Gemini A Ultra diffractometer at 150 K, with
Mo KR radiation (λ = 0.71073 A). Semiempirical absorption
corrections were applied, based on symmetry-equivalent and
repeated reflections. Structures were solved by direct methods
and refined on all unique F2 values, with anisotropic non-H
atoms and constrained riding isotropic H atoms. Disordered
dichloromethane solvent molecules could not be modeled as
discrete atoms for 4b and 4d and were treated by the SQUEEZE
procedure of PLATON.31 High anisotropy of some atoms
of 6 indicated possible disorder, but this could not be resolved
and has been ignored. Programs were CrysAlisPro for data
collection, integration, and absorption corrections32 and
SHELXTL for structure solution, refinement, and graphics.33
A summary of key crystallographic experimental information is
provided in Table 2, and full details are in the Supporting
Information.
General Procedure for the Gold-Catalyzed Cycloisomeriza-
tions Using Precursors 4a-d. A flame-dried Schlenk flask charged
with 4a-d (0.01 mmol), AgOTf (0.0026 g, 0.01 mmol), and
dichloromethane (1.0 mL) was stirred for 30 min, after which
propargyl amide (0.5 mmol) was added and the resulting mixture
stirred for the allocated time. The reaction mixture was diluted with
diethyl ether, 1,3-dinitrobenzene was added (0.084 g, 0.5 mmol),
and the resulting mixture was passed through a short silica plug.
1
The solvent was removed and the residue analyzed by H NMR
spectroscopy to determine conversions before being purified by
column chromatography, eluting with hexane/ethyl acetate.
Known products were characterized by NMR spectroscopy and
mass spectrometry and unknown products by NMR spectroscopy,
mass spectrometry, and high-resolution mass spectrometry.
General Procedure for the Gold-Catalyzed Cycloisomeriza-
tions Using 5 and 6. A flame-dried Schlenk flask was charged
with 5 or 6 (0.01 mmol), propargyl amide (0.5 mmol), and dich-
loromethane (1.0 mL), and the resulting mixture was stirred for the
allocated time. The reaction mixture was diluted with diethyl ether,
1,3-dinitrobenzene was added (0.084 g, 0.5 mmol), and the result-
ing mixture was passed through a short silica plug. The solvent was
Acknowledgment. We gratefully acknowledge the
EPSRC for a studentship (C.H.S.) and an equipment
grant (W.C.).
Note Added after ASAP Publication: This paper was pub-
lished on the Web on Aug 17, 2010, with errors in the fifth
paragraph of the Results and Discussion section and in Equa-
tion 1. The corrected version was reposted on Aug 23, 2010.
1
Supporting Information Available: Full details of experimen-
tal procedures, characterization data for all new compounds,
details of catalyst testing, and, for compounds 4b, 4d, 5, and 6,
full details of crystal data, structure solution and refinement,
atomic coordinates, bond distances, bond angles, and anisotro-
pic displacement parameters in CIF format. This material is
available free of charge via the Internet at http://pubs.acs.org.
removed and the residue analyzed by H NMR spectroscopy to
determine conversions before being purified by column chroma-
tography, eluting with hexane/ethyl acetate. Known products were
characterized by NMR spectroscopy and mass spectrometry and
unknown products by NMR spectroscopy, mass spectrometry,
and high-resolution mass spectrometry (HRMS).
(31) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(32) CrysAlisPro; Oxford Diffraction Ltd.: Oxford, UK, 2008.
(33) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.