Cationic gold(I) complexes: Highly efficient catalysts for the addition of alcohols to alkynes

Full text of DOI:10.1002/(SICI)1521-3773(19980605)37:10<1415::AID-ANIE1415>3.0.CO;2-N

COMMUNICATIONS
acetylene[2a] as well as to mono-[2b] and disubstituted acetyle-
nes.[2c] However, it has the serious drawback that under the
reaction conditions the mercury(ii) is quickly reduced to
metallic mercury, which is catalytically inactive, so 100 moles
of product at most can be produced per mole of mercury(ii)
salt introduced.
An alternative mercury-free catalyst for this reaction has
long been sought, but with little success.[3] Two recent papers
describe the use of gold(iii) catalysts (NaAuCl4) for the
addition of water and methanol to nonactivated alkynes,[4] but
this catalyst shows the same disadvantage: It is quickly
reduced to inactive metallic gold, so no more than 50 moles of
product can be synthesized per mole of gold. The fact that the
literature on homogeneous catalysis with gold complexes is
extremely sparse,[4, 5] and that until recently gold was thought
to be ªcatalytically dead,º[6] made the quest for an efficient
gold(i) catalyst all the more interesting.
[9] T. M. Klapötke, I. C. Tornieporth-Oetting, Nichtmetallchemie, VCH,
Weinheim, 1994, pp 254 ± 255.
[10] C. Boisson, J. C. Berthet, M. Ephritikhine, M. Lance, M. Nierlich, J.
Organomet. Chem. 1997, 531, 115 ± 119.
[11] See also: E. D. Jemmis, B. Kiran, J. Am .Chem. Soc. 1997, 119, 4076 ±
4077 (B3LYP/6-31G(d) geometry of Si2B10H12).
[12] For application to borane chemistry see: X. Y. Yang, H. Jiao, P. von R.
Schleyer, Inorg. Chem. 1997, 36, 4897 ± 4899, and references therein.
[13] a) A. D. Becke, Phys. Rev. 1988, A38, 3098 ± 3100; b) A. D. Becke, J.
Chem. Phys. 1993, 98, 5648 ± 5652; c) C. Lee, W. Yong, R. G. Parr,
Phys. Rev. 1988, B37, 785 ± 789.
[14] All calculations were performed with Gaussian 94; M. J. Frisch, G. W.
Trucks, H. B. Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J. R.
Cheeseman, T. Keith, G. A. Petersson, J. A. Montgomery, K. Ragha-
vachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Fores-
man, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challacombe,
C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S.
Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J.
Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, J. A.
Pople, Gaussian, Inc., Pittsburgh, PA, 1995.
[15] K. Wade, Adv. Inorg. Chem. Radiochem. 1976, 18, 1 ± 66; R. E.
Williams, Advances in Boron Chemistry (Ed.: W. Siebert), The Royal
Society of Chemistry, Cambridge, 1997.
[16] F. Meyer, J. Müller, P. Paetzold, R. Boese, Angew. Chem. 1992, 104,
1221 ± 1222; Angew. Chem. Int. Ed. Engl. 1992, 31, 1227 ± 1228.
[17] M. E. Jung, H. Xia, Tetrahedron Lett. 1988, 297 ± 300; K. Tamao, T.
Hayashi, Y. Ito, J. Am. Chem. Soc. 1990, 112, 2422 ± 2423.
Here we report a new, very efficient class of gold(i) catalysts
for the addition of alcohols to alkynes under mild conditions
‡
(Tˆ 20 ± 508C, H as cocatalyst).[7] In numerous studies
carried out in our laboratory we have found that cationic
‡
gold(i) complexes of the general type [L ± Au ] (where L is a
[18] The 11B and 29Si NMR shifts are reported relative to BF3 ´ OEt2 and
TMS, respectively. It is inconvenient to compute the magnetic
shielding of BF3 ´ OEt2, and so B2H6 was calculated instead and set
to 16.6 ppm (T. P. Onak, H. L. Landesman, R. E. Williams, I. Shapiro,
J. Phys. Chem. 1959, 63, 1533 ± 1535). The absolute chemical shieldings
at the B3LYP-GIAO/6-311 ‡ G(d, p)//B3LYP/6-31G(d) level are:
84.11 ppm (11B of B2H6) and 338.80 ppm (29Si of TMS).
phosphane, a phosphite, or an arsine)[8] are excellent catalysts
for the addition of alcohols to alkynes. These catalysts achieve
total turnover numbers of up to 105 moles of product per mole
1
of catalyst, with turnover frequencies of up to 5400 h . They
are neither water nor air sensitive, and the reaction can
usually be conducted without a solvent. Scheme 1 shows
examples of the addition of alcohols to unsubstituted alkynes
with methyl(triphenylphosphane)gold(i) and methanesulfonic
acid as precursors for in situ generation of the catalyst.
[19] Correct C,H analyses for 5 and 6 were obtained.
Cationic Gold(i) Complexes:
Highly Efficient Catalysts for the Addition
of Alcohols to Alkynes**
J. Henrique Teles,* Stefan Brode, and
Mathieu Chabanas
In memory of Dr. Marco Häser
The mercury(ii)-catalyzed addition of alcohols to alkynes
has been known for almost three-quarters of a century.[1] The
original catalyst has been further developed,[2] but its prep-
aration by the heating of red mercury(ii) oxide with boron
trifluoride etherate and trichloroacetic acid in methanol still
retains a touch of alchemy, and very little is known about the
reaction mechanism.[2g] This material catalyzes the addition of
primary, secondary, and even some tertiary alcohols to
[*] Dr. J. H. Teles, Dr. S. Brode, M. Chabanas
BASF AG, Ammoniaklaboratorium
D-67056 Ludwigshafen (Germany)
Fax: (‡49)621-60-56116
E-Mail: joaquim-henrique.teles@zag.x400.basf-ag.de
[**] We are obliged to Dr. M. Schulz and Dr. E. Zeller for helpful
discussions and to Dr. N. Walker for the X-ray structure determi-
nation of 12.
Scheme 1. Addition of alcohols to unsubstituted alkynes in the presence
of the catalyst prepared from methyl(triphenylphosphane)gold(i) and
methanesulfonic acid.
Angew. Chem. Int. Ed. 1998, 37, No. 10
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3710-1415 $ 17.50+.50/0
1415

COMMUNICATIONS
In the case of internal, symmetrical alkynes such as 3-
hexyne (1), the only product formed in the presence of excess
alcohol is 3,3-dimethoxyhexane (2).[9] With unsymmetrical
alkynes such as methylisopropylacetylene (3) only acetal 4 is
formed as a result of addition to the less sterically hindered
position, together with a small amount of enol ether 5. The
regioselectivity of the gold catalysts is thus higher than that of
mercury catalysts.[10] With diphenylacetylene (6) the main
product is enol ether 7, even in the presence of a large excess
of methanol. At low conversions the Z isomer predominates
(Z:E ꢀ 8:1), but in the course of the reaction it partially
isomerizes (probably through the corresponding acetal), and
at the end of the reaction the Z:E ratio is about 2:1, which is
close to the equilibrium ratio.[11]
Terminal alkynes such as 8a ± e are also suitable substrates.
In this case addition occurs almost exclusively at the more
highly substituted carbon atom to produce the expected
acetals 9a ± e. Phenylacetylene yields a mixture of the enol
ether 10 f and the acetal 9 f in a ratio of 2:1.
Scheme 2. Gold(i)-catalyzed addition of methanol to propargyl alcohols.
Propargyl alcohols also react smoothly. 2-Propynol (11), 1-
butyne-3-ol (13), and 2-butyne-1,4-diol (15) react with excess
methanol to give the expected products 12,[12] 14 (as a mixture
of isomers), and 16.[2f, 13] However, 1,4-dichloro-2-butyne and
propynoic acid methyl ester do not react at all (Scheme 2).
The reactivity of the alkynes increases with increasing
electron density of the triple bond and decreases with
increasing steric hindrance. The reactivity of the alcohols
decreases by a factor of about ten when going form primary to
secondary alcohols. Tertiary alcohols and phenols are un-
reactive.[14]
progresses from soft to hard anions (initial turnover frequen-
1
cies [h ] in parentheses): I (2)< Cl (7)< NO3 ꢀ CF3COO ꢀ
CH3SO3 (700).[18] For very hard anions the turnover frequen-
cy levels off, probably because of the presence of excess
fluoride from the hydrolysis of BF3. This interpretation is
supported by the fact that the catalyst generated from
1
[Ph3PAu(CH3SO3)] and BF3 ´ OEt2 (TOF ˆ 700 h ) is less
active than that generated under halide-free conditions from
1
[Ph3PAuCH3] and CH3SO3H (TOF ˆ 1500 h ).
The influence of catalyst structure on activity was also
studied, with the addition of methanol to propyne as a model
reaction (8a ‡ 2MeOH !9a). The nature of the ligand L in
the cationic gold complex has a considerable influence upon
catalytic activity. Initial turnover frequencies for the addition
of methanol to propyne were measured with different [L ±
Based on these experimental results and ab initio calcu-
lations[19] we propose the following mechanism for catalytic
addition of alcohols to alkynes (illustrated in Scheme 3 for the
addition of methanol to propyne). The cationic gold(i)
complex 17, generated for example by protonolysis of a
methylgold complex, will not exist in free form but will
instead coordinate with whatever donors are present in
solution, as shown by 31P NMR spectroscopy. If [Ph3PAuCH3]
is treated with one equivalent of CH3SO3H in [D2]dichloro-
‡
Au ] catalysts.[15] The activity order for the ligands tested was
1
as follows (initial turnover frequencies [h ] in parentheses):
Ph3As (430) < Et3P (550) < Ph3P (610) < (4-F-C6H4)3P (640) <
(MeO)3P (1200) < (PhO)3P (1500).[16] As ex-
pected, electron-poor ligands lead to an in-
crease in activity, but the stability of the
complexes decreases. The catalyst with L ˆ
(PhO)3P is more than twice as active as that
with L ˆ Ph3P; however, the former is com-
pletely deactivated after only 2500 turnovers,
whereas the latter is still active after 5000
turnovers.
‡
As mentioned above, H is necessary as a
cocatalyst,[2h] but a Lewis acid such as boron
trifluoride can also be utilized because it is
rapidly hydrolyzed to trimethyl borate and HF
under the reaction conditions.[17] The turnover
frequency increases in a roughly linear fashion
‡
with the concentration of H .
Anions present in the solution are also a
very important factor with respect to catalytic
activity. With the combination [Ph3PAuX] and
BF3 ´ OEt2 the reactivity increases as one
Scheme 3. Proposed mechanism for the addition of methanol to propyne catalyzed by the
trimethylphosphanegold(i) cation.
1416
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3710-1416 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1998, 37, No. 10

COMMUNICATIONS
methane at 408C and then cooled to 808C, one observes
in the 31P NMR spectrum two broadened peaks at d ˆ 36 and
28, which we attribute to dichloromethane and methanesul-
fonate complexes of the triphenylphosphanegold(i) cation.[20]
Addition of five equivalents of methanol at 808C gives rise
to a single broad peak at d ˆ 38, tentatively assigned to the
methanol complex. If five equivalents of 3-hexyne are instead
added at 808C a single broadened peak is observed at d ˆ 28
(tentatively assigned to the alkyne complex), which remains
almost unchanged up to 08C. Ab initio calculations provide
the following relative stabilities for complexes between the
trimethylphosphanegold(i) cation (17) and several neutral
Experimental Section
Typical procedure for the addition of methanol to propyne: A mixture of
methanol (2.5 mol) and a Lewis or Brùnsted acid (1.25 mmol) was heated
to 408C and saturated with propyne before the gold catalyst (0.125 mmol
Au) was added.[8] During the reaction a continuous stream of propyne was
introduced at a constant rate. Aliquots were taken at regular intervals, and
the reaction was quenched by addition of an excess of solid sodium
methanolate. The concentrations of methanol, 2,2-dimethoxypropane, and
2-methoxypropene were determined by GC and used to calculate turnover
numbers and frequencies.
trans-2,5-Dimethyl-2,5-dimethoxy-1,4-dioxane (12): Propargyl alcohol
(15.1 mol, 848 g, freshly distilled), methanol (59.6 mol, 1908 g), and
concentrated sulfuric acid (15 mmol, 1.47 g) were mixed and heated to
558C.
A
solution of methyl(triphenylphosphane)gold(i)[22] (148 mmol,
1
ligands (energies [kJmol ] relative to the 2-butyne complex
70.4 mg) in 125 mL of dioxane was then added within 10 h. The mixture
was stirred for an additional 10 h, and then most of the remaining methanol
was removed by distillation at reduced pressure. The residual solution was
neutralized with 30% sodium methanolate in methanol and cooled in an
ice bath. Precipitated product was collected by filtration and dried. A
second batch of product can be obtained by further concentration of the
mother liquor. Total yield of isolated product: 1238 g (93%). Colorless
crystals, m.p. 127 ± 1298C (literature value: 126 ± 1288C).[12, 23]
are given in parentheses): dichloromethane (‡63) < water
(‡44) < acetylene (‡38) < methanol ꢀ 1,4-dioxane (‡24) <
propyne (‡18) < tetrahydrofuran (‡2) < 2-butyne (0) < di-
methylsulfide ( 18) < triphenylphosphane ( 114). As ex-
pected, the soft bases triphenylphosphane and dimethyl-
sulfide give the most stable complexes. Somewhat
surprisingly, mono- and disubstituted alkynes are better
ligands than methanol or dioxane, which favor the formation
of 18.
Received: December 23, 1997 [Z11290IE]
German version: Angew. Chem. 1998, 110, 1475 ± 1478
The gold(i) propyne complex (18) is then attacked by a
molecule of methanol. Ab initio calculations predict this
attack to occur by an associative mechanism that involves
coordination of methanol to gold to give the intermediate
complex 19.[21] This precoordination is computed to
Keywords: additions
catalysis
´ alkynes ´ gold ´ homogeneous
[1] J. S. Reichert, J. H. Bailey, J. A. Niewland, J. Am. Chem. Soc. 1923, 45,
1553.
1
be exothermic by 24 kJmol . The calculated activation
1
[2] a) H. D. Hinton, J. A. Niewland, J. Am. Chem. Soc. 1930, 52, 2892;
b) D. B. Killian, G. F. Hennion, J. A. Niewland, ibid. 1934, 56, 1384;
c) G. F. Hennion, J. A. Niewland, ibid. 1935, 57, 2006; d) D. B. Killian,
G. F. Hennion, J. A. Niewland, ibid. 1936, 58, 1658; e) ibid. 1936, 58,
80; f) G. F. Hennion, W. S. Murray, ibid. 1942, 64, 1220; g) M. Bassetti,
B. Floris, J. Chem. Soc. Perkin Trans. 2 1988, 227; h) under basic
conditions the reaction becomes stoichiometric in mercury, and enol
ethers are produced instead of acetals: J. Barluenga, F. Aznar, M.
Bayod, Synthesis 1988, 144.
[3] a) I. K. Meier, J. A. Marsella, J. Mol. Catal. 1993, 78, 31; b) P. W.
Jennings, J. W. Hartman, W. C. Hiscox, Inorg. Chim. Acta 1994, 222,
317; c) Y. Kataoka, O. Matsumoto, K. Tani, Organometallics 1996, 15,
5246, and references therein.
energy for rearrangement of 19 to 20 is only 43 kJmol , and
1
the overall reaction is exothermic by
37 kJmol .
Both 19 and the transition state between 19 and 20 are
sterically quite crowded, and this may be the reason why
secondary alcohols react almost ten times more slowly than
primary alcohols. Interestingly, the isomer of 20 shown (the Z
isomer) is stabilized by almost 10 kJmol 1 with respect to the
E isomer as a result of a close contact between gold and
hydrogen (2.26 Š) within a planar five-membered chelating
ring.
The rearrangement of 20 to 21 could in principle proceed
along either of two pathways: deprotonation at oxygen with
reprotonation at carbon, or a 1,3-hydrogen migration. The
relative importance of these two pathways cannot easily be
predicted with ab initio calculations. Nevertheless, we were
able to locate the transition state for 1,3-hydrogen migration.
This pathway, in which gold functions as a proton shuttle, has a
[4] a) Y. Fukuda, K. Utimoto, J. Org. Chem. 1991, 56, 3729; b) Y. Fukuda,
K. Utimoto, Bull. Chem. Soc. Jpn. 1991, 64, 2013.
[5] a) Y. Ito, M. Sawamura, T. Hayashi, J. Am. Chem. Soc. 1986, 108, 6405;
b) Y. Fukuda, K. Utimoto, H. Nozaki, Heterocycles 1987, 25, 297; c) Y.
Fukuda, K. Utimoto, Synthesis 1991, 975; d) S. Komiya, T. Sone, Y.
Usui, M. Hirano, A. Fukuoka, Gold Bull. (London) 1996, 29, 131; e) Q.
Xu, Y. Imamura, M. Fujiwara, Y. Souma, J. Org. Chem. 1997, 62, 1594.
[6] H. Schmidbaur, Naturwiss. Rundsch. 1995, 48, 443.
1
[7] J. H. Teles, M. Schulz (BASF AG), WO-A1 9721648, 1997 [Chem.
Abstr. 1997, 127, 121499].
computed activation energy of only 67 kJmol , and the
migrating hydrogen atom is transferred to the position cis to
the methoxyl group. However, rotation about the C ± C
double bond in 21 has a low activation energy (less than
[8] These cationic gold(i) complexes can be generated in situ by several
methods: a) protonation of [LAuCH3] with a strong acid whose anion
does not coordinate strongly to gold (e.g., CH3SO3H, H2SO4, HBF4).
This is the preferred method, which was used in all cases unless
otherwise noted; b) [LAuX] ‡ BF3 ´ OEt2, where X is a hard anion
such as NO3 , CF3COO , CH3SO3 , Cl ; c) [LAuHal] ‡ AgX, where
1
70 kJmol ). The catalytic cycle is then closed by a ligand
exchange that again produces 18 (this ligand exchange is
1
calculated to be endothermic by 41 kJmol ). Another
Hal is Cl, Br, or I, and X is a noncoordinating anion such as BF4
;
possibility is that intermediate 21, whose structure resembles
that of protonated 2-methoxypropene, adds a second mole of
methanol and then undergoes protodeauration to give 2,2-
dimethoxypropane directly. These pathways are indistinguish-
able, because under the reaction conditions there is very rapid
equilibrium between methanol, 2-methoxypropene, and 2,2-
dimethoxypropane.
‡
‡
d) [(LAu)3O ] or [(LAu)2Cl ] ‡ BF3 ´ OEt2.
[9] Under the reaction conditions the acetal is in rapid equilibrium with
the corresponding enol ethers (four isomers). Enol ethers can become
the major products in the presence of excess alkyne.
[10] Mercury-catalyzed addition of water to 3 leads to a 2:1 mixture of the
two isomeric ketones 4-methyl-2-pentanone and 2-methyl-3-pentan-
one; J. W. Hartman, W. C. Hiscox, P. W. Jennings, J. Org. Chem. 1993,
58, 7613.
Angew. Chem. Int. Ed. 1998, 37, No. 10
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3710-1417 $ 17.50+.50/0
1417

COMMUNICATIONS
[11] E. Taskinen, M. Anttila, Tetrahedron 1977, 33, 2423.
[12] Compound 12 is formed as a mixture of two isomers that rapidly
equilibrate under the reaction conditions. The trans isomer with
methoxy groups in axial positions crystallizes preferentially from the
reaction mixture and can be obtained pure in greater than 90% yield
with only 10 4 mol% catalyst.
Synthesis and Biological Activity of
Sarcodictyins**
K. C. Nicolaou,* Sanghee Kim, Jeffrey Pfefferkorn,
Jinyou Xu, Takashi Ohshima, Seijiro Hosokawa,
Dionisios Vourloumis, and Tianhu Li
[13] W. Reppe, Justus Liebigs Ann. Chem. 1955, 596, 1 (see pages 62
and 64).
[14] These catalysts are also able to catalyze the addition of water and
carboxylic acids to alkynes. A preliminary report of this work appears
in reference [7].
Isolated from certain species of soft corals, the sarcodictyins
(1, 2),[1, 2] eleutherobin (3),[3, 4] and the eleuthosides (4, 5)[5]
have become important synthetic targets because of their
novel molecular architectures, biological activities, and me-
dicinal potential (Figure 1). Of special interest is their taxol-
like mechanism[6] of action, which involves tubulin polymeri-
[15] Generated from [LAu(NO3)] and ten equivalents of BF3 ´ OEt2. The
nitrates were synthesized from the known chlorides through meta-
thesis with AgNO3. For a typical procedure, see L. Malatesta, L.
Naldini, G. Simonetta, F. Cariati, Coord. Chem. Rev. 1966, 1, 255.
[16] Nucleophilic carbenes such as 1,3,4-triphenyl-4,5-dihydro-1,2,4-tria-
zol-(1H)-ylidene are also suitable ligands. We prepared the corre-
sponding gold(i) chloride complex by treating the isolated carbene
with dimethylsulfidegold(i) chloride. This carbenegold(i) complex
proved to be 3.5 times more active than triphenylphosphanegold(i)
chloride. Unfortunately we did not succeed in the preparation of
either the carbenegold nitrate or the carbenegold methyl complexes.
[17] Amazingly, even HBF4 is quickly hydrolyzed to trimethylborate under
the reaction conditions.
O
O
N
O
N
O
H
H
N
H
8
N
6
5
O
10
O
1
OR1
[18] The same turnover rate (700 h 1) is observed when the catalyst is
OH
CO2R
3
‡
H
generated from the oxonium salt [(Ph3PAu)3O BF4 ] and BF3 ´ OEt2.
O
OAc
[19] All ab initio calculations were performed with the TURBOMOLE
program package. For an overview of TURBOMOLE, see R.
Ahlrichs, M. von Arnim in Methods and Techniques in Computational
Chemistry: METECC-95 (Eds.: E. Clementi, G. Corongiu) STEF,
Cagliari, 1995, P. 509 ± 554. Geometries were calculated at the RI-DFT
level (K. Eichkorn, O. Treutler, H. Ihm, M. Häser, R. Ahlrichs; Chem.
Phys. Lett. 1995, 242, 652) with the Becke ± Perdew functional and the
TURBOMOLE SV(P) basis sets (K. Eichkorn, F. Weigend, O.
Treutler, R. Ahlrichs, Theor. Chem. Acc. 1997, 97, 119). For gold, the
60-mwb relativistic pseudopotential was utilized (D. Andrae, U.
Haeussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 1990,
77, 123). Energies were calculated at the MP2 level with TZVP basis
sets through the recently developed RI-MP2 implementation (F.
Weigend, M. Häser, Theor. Chem. Acc. 1997, 97, 331).
15
OR2
O
OR3
3: R1 = Me, R2 = H,
1:R = Me
2:R = Et
R3 = H
R3 = H
R3 =Ac
1
2
4:
R = H, R = Ac,
5: R1 = H, R2 = H,
Figure 1. Structures of sarcodictyins A (1) and B (2), eleutherobin (3), and
eleuthosides A (4) and B (5).
zation and microtubule stabilization and results in tumor-cell
death. The combination of the scarcity and the appealing
biological activity of these materials prompted us to initiate a
program directed at their chemical synthesis. We recently
disclosed the first total syntheses of sarcodictyin A (1)[7] and
eleutherobin (3).[8] Here we report the first synthesis of
sarcodictyin B (2), the construction of a sarcodictyin library,
and the tubulin-polymerization and cytotoxic properties of
members of that library, including their action against a
number of taxol-resistant tumor-cell lines.
[20] The peaks coalesce at 408C to a single broad peak at d ˆ 32. Above
208C this peak becomes narrower, and at 08C only the sharp signal
of [Ph3PAuCl] remains at d ˆ 31. The independently synthesized
triphenylphosphanegold(i) methanesulfonate (from triphenylphos-
phanegold(i) iodide and silver methanesulfonate) shows a broadened
peak at d ˆ 27 (31P NMR in CD2Cl2 at 408C).
[21] Coordination of methanol with 18 to form the T-shaped complex 19
disturbs the structure only slightly. The angle between phosphorus,
gold, and the center of the triple bond changes from 1798 in 18 to 1758
in 19. The Au ± O bond in 19 is very long (3.7 Š compared to 2.1 Š for
a Au ± O single bond), and the P-Au-O angle is 838.
To conveniently access a sarcodictyin library, an improved
method for their construction was devised (Scheme 1), which
[22] A. Tamaki, J. K. Kochi, J. Organomet. Chem. 1973, 61, 441.
[23] The stereochemistry was determined by X-ray crystallography.
Crystallographic data (excluding structure factors) for the structure
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication no.
CCDC-100957. Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax:
(‡44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
[*] Prof. Dr. K. C. Nicolaou, Dr. S. Kim, J. Pfefferkorn, Dr. J. Xu,
Dr. T. Ohshima, Dr. S. Hosokawa, Dr. D. Vourloumis, Dr. T. Li
Department of Chemistry and
The Skaggs Institute for Chemical Biology
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (‡1)619-784-2469
E-mail: kcn@scripps.edu
and
Department of Chemistry and Biochemistry
University of California, San Diego
9500 Gilman Drive, La Jolla, CA 92093 (USA)
[**] This work was supported by The Skaggs Institute for Chemical
Biology, the U.S. National Institutes of Health, Novartis, and the CaP
CURE Foundation, and by fellowships from Novartis (D.V.), the
Japanese Society for the Promotion of Science for Young Scientists
(S.H.), and the U.S. Department of Defense (J.P.).
1418
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3710-1418 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1998, 37, No. 10