The condensation of carbonyl compounds with electron-rich arenes: Mercury, thallium, gold or a proton?

Article abstract of DOI:10.1002/adsc.200505464

Gold(III) chloride, mercury(II) perchlorate, thallium(III) perchlorate and p-toluenesulfonic acid were found to efficiently catalyze the condensation of two furans with aldehydes or acetone. The olefinic unit of α,β- unsaturated carbonyl compounds reacts faster than the carbonyl group, other olefins do not react selectively. The first (addition) step is rate-limiting, the second (substitution) step is much faster.

Full text of DOI:10.1002/adsc.200505464

COMMUNICATIONS
DOI: 10.1002/adsc.200505464
The Condensation of Carbonyl Compounds with Electron-Rich
Arenes: Mercury, Thallium, Gold or a Proton?
A. Stephen K. Hashmi,* Lothar Schwarz, Philipp Rubenbauer, M. Carmen Blanco
Institut für Organische Chemie, Universität Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
Fax: (þ49)-711-685-4321, e-mail: hashmi@hashmi.de
Received: November 30, 2005; Accepted: February 2, 2006
Abstract: Gold(III) chloride, mercury(II) perchlo-
rate, thallium(III) perchlorate and p-toluenesulfonic
acid were found to efficiently catalyze the condensa-
tion oftwo ufrans with aldehydes or acetone. The
olefinic unit of a,b-unsaturated carbonyl compounds
reacts faster than the carbonyl group, other olefins
do not react selectively. The first (addition) step is
rate-limiting, the second (substitution) step is much
faster.
Starting point was the gold-catalyzed hydroarylation
of a,b-unsaturated ketones.[7–11] When the reaction of
2-methylfuran 5 with methyl vinyl ketone was conduct-
ed not in acetonitrile but in acetone as solvent, a product
ofthe type 4 was obtained as a significant side-product.
Monitoring the reaction of 5 without a,b-unsaturated
ketone in hexadeuteroacetone 6 with 0.7 mol % AuCl3
as the catalyst in a NMR tube by 1H NMR revealed three
facts (Figure 1): a) The reaction was very selective, no
side-products were observed; from the starting materi-
als (furyl signals at 6.3 and 7.3 ppm) only 7 is formed (sig-
nals at 2.2, 6.0 and 6.2 ppm). b) No relevant concentra-
tion ofa species oftype 3 was built up during the reaction
(a transient peak was not observed), which clearly shows
that in this case the initial step leading to intermediate 3
was rate-limiting and the subsequent conversion of 3 to 4
was a much easier step (especially ifone takes into ac-
Keywords: arylation; furans; gold; mercury; thallium;
ytterbium
The reaction ofelectron-rich aromatic compounds
1
with aldehydes or ketones 2 is catalyzed by acid under count the high concentration ofketone 6, which as the
relatively harsh conditions, prominent examples are solvent should significantly accelerate the first step). c)
[1]
the synthesis ofBakelite or furanophanes.[2] Overall, The water set free as a by-product (broad peak at
a two-fold reaction is observed – in these condensation 3.2 ppm) quickly exchanged H versus D with the sol-
reactions one equivalent water is produced as a by-prod- vent, thus the peak disappearedagain and apeak for par-
uct. Thus any alternative and potentially milder method tially deuterated acetone simultaneously appeared at
not dependent on Brønstedt acids must use water-toler- 2.1 ppm. The time/concentration profile of these species
ant Lewis acids. Since in the first step benzyl alcohols 3 was obtained by integration ofthe spectra and is shown
are produced, the second step in which the latter reacts in Figure 2.
with a second molecule ofthe electron-rich arene is
closely related to an iron-catalyzed reaction reported
by Beller et al.[3] this year. Other activated alcohols re-
cently shown to undergo similar reactions are propargyl-
ic alcohols, here Uemura et al.[4] investigated ruthenium
catalysts and Toste et al.[5] rhenium catalysts.[6]
Here we report our results on the comparison ofmer-
cury, thallium and gold compounds with p-toluenesul-
fonic acid.
Scheme 2.
Another Lewis acid, known to be water-tolerant, is
Yb(OTf)3. With the same amount ofthis catalyst after
2 days only 5% conversion to 7 were observed. Since
we had good experience with Hg(ClO4)2 as a cata-
lyst,[12,13] we tested the reaction of 5 and 6 with
Scheme 1.
1 mol % ofthis catalyst, too. Again 7 was formed selec-
Adv. Synth. Catal. 2006, 348, 705 – 708
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
705

A. Stephen K. Hashmi et al.
COMMUNICATIONS
Marshall et al.[14,15] have reported the cycloisomeriza-
tion ofsuch allenyl ketones with silver(I) catalysts in
acetone, the related cycloisomerization/dimerization is
usually conducted in acetonitrile with palladium[13,16–19]
or gold[7] catalysts. Different from the silver-catalyzed
reactions in acetone, with the AuCl3 catalyst in acetone
even after a short reaction time of only 2 h at room tem-
perature a 23% yield of 10 was observed (after this short
time the cycloisomerization is finished, 9 is still the ma-
jor product). After two days all 9 was consumed and in a
clean reaction converted to 10.
Next we tested the more reactive aldehyde group. In
the reaction with two equivalents of 5 the aldehyde 11
with aslittleas 0.3 mol % ofAuCl 3 in acetonitrileinstead
ofacetone at room temperature a quantitative yield of
the known compound 12 was obtained.
Figure 1. 1H NMR spectra taken during the gold-catalyzed
conversion of 5 and 6 to 7.
tively, after 14 hours at 208C the reaction was complete.
A similar behaviour was observed with 1 mol %
Tl(ClO4)3 as the catalyst, after only 5 hours at room tem-
perature the conversion was complete. Finally, InCl3 was
tested, only with 33 mol % ofthis catalyst and at 70 8C
did a slow and less selective reaction take place.
As Brønsted acids, 11 mol % ofHClO 4 and 1 mol %
ofTsOH were tested, both gave a clean reaction to 7,
too.
A similar behaviour was observed when the cycloiso-
merization ofallenyl ketone 8 was conducted in acetone. Scheme 3.
Figure 2. Time-concentration profile of the gold-catalyzed conversion of 5 and 6 to 7.
706
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 705 – 708

Condensation ofCarbonyl Compounds with Electron-Rich Arenes
COMMUNICATIONS
Scheme 4.
Scheme 6.
Experimental Section
Synthesis of 1,1,1,3,3,3-Hexadeutero-2,2-bis-
(5-methylfuran-2-yl)propane (7)
2-Methylfuran (183 mg, 2.23 mmol) was dissolved in acetone-
d6 (2.00 g), then AuCl3 (2.4 mg, 0.7 mol %) in acetone-d6
(250 mg) was added, after one day the starting material was
consumed and no further conversion was observed. The sol-
vent was removed and the crude product was purified by col-
umn chromatography (hexanes/ethyl acetate, 20:1) to afford
7; yield: 213 mg (91%). Rf (hexanes/ethyl acetate, 20:1)¼
0.50; 1H NMR (250 MHz, CDCl3): d¼2.26 (d, J¼0.9 Hz,
6H), 5.84–5.89 (m, 4H); 1H NMR (250 MHz, acetone-d6):
d¼2.18 (d, J¼0.9 Hz, 6H), 5.86–5.90 (m, 4H); 13C NMR
(62.9 MHz, CDCl3): d¼13.42 (q, 2C), 25.65 (m, CD3, 2C),
37.21 (m, C), 104.31 (d, 2C), 105.53 (d, 2C), 150.34 (s, 2C),
158.45 (s, 2C); IR (NaCl, film): n¼3106, 2924, 2231, 1612,
1559, 1448, 1220, 1045, 1023, 952, 781 cmꢀ1; anal. calcd. for
C13H10D6O2 (210.30): H 7.90, C 74.25; found: H 7.76, C 74.00.
Scheme 5.
With the more sensitive stilbene aldehyde 13 and two
equivalents of 5 under the similar conditions only a 35%
yield of 14 was achieved. Presumably, some unselective
side-reactions at the olefinic, styryl-like unit take place.
Finally, we investigated the a,b-unsaturated aldehyde
acrolein 15. With 1 mol % ofAuCl and three equiva-
3
lents of 5 a highly selective reaction to 17 was observed,
the 1H NMR spectra taken during the reaction indicated
that first in a very fast step the hydroarylation product 16
is formed, which subsequently reacts with two more
molecules of 5 as indicated for the other aldehydes 11
and 13. The absence of 18 and the survival ofthe double
bond in 14 are in accord with the observation that unac-
tivated alkenes in gold-catalyzed reactions react at high-
er temperatures in non-coordinating solvents only;[20] if
5 would react with the aldehyde group of 15 first, it
should be possible to isolate or at least observe 18.
In analogy to our recent report,[8] the conversions
leading to 12, 14 and 16 can also be conducted with
Brønsted acids with non-nucleophilic counterions.
While mercury and thallium salts due to their toxicity
are not an alternative to Brønsted acids with non-nucle-
ophilic counterions, and ytterbium salts suffer from ac-
tivity problems, the mild conditions ofthe gold catalysis
are an interesting synthetic alternative. This investiga-
tion furthermore showed that in a,b-unsaturated car-
bonyl compounds the olefin reacts faster than the alde-
hyde group, while other alkenes remain intact. The first
reaction ofthe carbonyl group with the electron-rich
arene is the rate-limiting-step, the reaction ofthe ben-
zylic alcohol is much faster.
2,2-Bis-[5-(4-methoxybenzyl)furan-2-yl]propane (10)
Compound 8 (298 mg, 1.60 mmol) was dissolved in acetone
(4.00 g), then AuCl3 (1.2 mg, 0.5 mol %) in acetone (125 mg)
was added at room temperature. After 2 days the reaction
was complete, the acetone was removed under reduced pres-
sure and the residue was purified by column chromatography
(hexanes:ethyl acetate, 10:1) to furnish 20.9 mg (7%) of 9[17]
and 263 mg (79%) of 10.
Product 10: Rf (hexanes/ethyl acetate, 5:1)¼0.30; mp 608C;
1H NMR (250 MHz, CDCl3): d¼1.60 (s, 6H), 3.79 (s, 6H), 3.87
(s, 4H), 5.79–5.81 (m, 2H), 5.86 (d, J¼3.1 Hz, 2H), 6.81–6.86
(m, 4H), 7.11–7.17 (m, 4H); 13C NMR (62.9 MHz, CDCl3):
d¼26.30 (q, 2C), 33.54 (t, 2C), 37.25 (s), 55.09 (q, 2C), 104.43
(d, 2C), 106.17 (d, 2C), 113.68 (d, 4C), 129.54 (d, 4C), 130.38
(s, 2C), 153.40 (s, 2C), 158.03 (s, 2C), 158.88 (s, 2C); IR (film):
n¼3104, 2975, 2935, 2907, 2835, 1611, 1554, 1512, 1462, 1299,
1247, 1176, 1103, 1034, 968, 788 cmꢀ1; MS (EI, 70 eV): m/z
(%)¼416 (34) [Mþ], 401 (100), 121 (37); anal. calcd. for
C27H28O4 (416.52): H 6.78, C 77.86; found: H 6.90, C 77.67.
Adv. Synth. Catal. 2006, 348, 705 – 708
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
707

A. Stephen K. Hashmi et al.
COMMUNICATIONS
Tris(5-methylfuran-2-yl)methane (12)
Acknowledgements
2-Methylfuran (800 mg, 9.74 mmol) and 5-methylfurfural
(390 mg, 4.87 mmol) were dissolved in acetonitrile (4 mL)
and 4.4 mg (0.3 mol %) AuCl3 were added. After two days
the conversion was quantitative, the usual work-up afforded
We thank the Fonds der Chemischen Industrie and the Euro-
pean Union for financial support (AURICAT EU-RTN,
HPRN-CT-2002-00174).
1
the known 12;[21] yield: 98%. H NMR (300 MHz, CD3CN):
d¼2.24–2.27 (m, 9H), 5.43 (s, 1H), 5.99–6.01 (m, 6H).
References and Notes
[1] L. H. Bakeland, Fr. Patent 386,627, 1908; Chem. Abstr.
1909, 3: 12896; for a review, see: P. J. Lemstra, Chemie-
AnlagenþVerfahren 1997, 30, 62–66.
Synthesis of 2-Methyl-5-[(5-methylfuran-2-yl)(4-
styrylphenyl)methyl]furan (14)
[2] M. de Sousa Healy, A. J. Rest, J. Chem. Soc. Perkin
Trans. 1 1985, 973–982. S. Tanaka, H. Tomokuni, J. Het-
erocycl. Chem. 1991, 28, 991–994.
To trans-4-stilbenecarboxaldehyde (73.0 mg, 350 mmol) and 5
(78.0 mg, 708 mmol) in acetonitrile (0.5 mL) AuCl3, (2.1 mg,
7.0 mmol, 2 mol %) was added. After 24 h the solvent was re-
moved under vacuum and the crude product was purified by
column chromatography on silica gel (petrol ether/ethyl ace-
tate, 20:1) to afford 14; yield: 44.0 mg (35%). Rf (petrol
ether/ethyl acetate, 20:1)¼0.30; mp 1088C; IR (film): n¼
[3] I. Iovel, K. Mertins, J. Kischel, A. Zapf, M. Beller, An-
gew. Chem. 2005, 117, 3981–3985; Angew. Chem. Int.
Ed. 2005, 44, 3913–3917.
[4] Y. Nishibayashi, Y. Inada, M. Yoshikawa, M. Hidai, S.
Uemura, Angew. Chem. 2003, 115, 1533–1536; Angew.
Chem. Int. Ed. 2003, 42, 1495–1498.
;
3020, 1214, 1021, 962, 748, 668, 629 cmꢀ1 1H NMR
(500 MHz, CDCl3): d¼2.16 (s, 6H), 5.25 (s, 1H), 5.80 (s, 4H),
7.00 (d, J¼1.1 Hz, 2H), 7.14–7.17 (m, 3H), 7.24–7.27 (m,
2H), 7.36–7.38 (m, 2H), 7.40–7.42 (m, 2H); 13C NMR
(500 MHz, CDCl3): d¼13.7 (q, 2C), 45.0 (d), 106.2 (d, 2C),
108.3 (d, 2C), 126.6 (d, 2C), 126.7 (d, 2C), 127.7 (d), 128.5 (d),
128.6 (d), 128.8 (d, 2C), 128.9 (d, 2C), 136.2 (s), 137.5 (s),
139.6 (s), 151.6 (s, 2C), 152.8 (s, 2C); MS (70 eV): m/z (%)¼
355 (47) [MþHþ], 354 (100) [Mþ], 312 (84), 311 (100), 272
(32), 178 (45), 175 (80); HRMS (EI 70 eV): m/z¼354.1620;
calcd. for C25H22O2: 354.1620;.
[5] J. J. Kennedy-Smith, L. A. Young, F. D. Toste, Org. Lett.
2004, 6, 1325–1327.
[6] For other related reactions, see: M. T. Reetz, K. Sommer,
in: Handbook of C-H Transformations, (Ed.: G. Dyker),
Vol. 1, Wiley-VCH, Weinheim, 2005, pp 157–166.
[7] A. S. K. Hashmi, L. Schwarz, J.-H. Choi, T. M. Frost, An-
gew. Chem. 2000, 112, 2382–2385; Angew. Chem. Int. Ed.
2000, 39, 2285–2288.
[8] G. Dyker, E. Muth, A. S. K. Hashmi, L. Ding, Adv.
Synth. Catal. 2003, 345, 1247–1252.
[9] A. S. K. Hashmi, L. Grundl, Tetrahedron 2005, 61, 6231–
6236.
Synthesis of 1,1,3-Tris(5-methylfuran-2-yl)propane
(17)
´
[10] A. S. K. Hashmi, R. Salathe, T. M. Frost, L. Schwarz, J.-
To acrolein (110 mg, 2.00 mmol) and 5 (492 mg, 6.00 mmol) in
1.00 g MeCN, AuCl3 (1.8 mg, 0.3 mol%) in 49 mg MeCN was
added. Purification of the crude product by column chroma-
tography (hexanes/ethyl acetate, 10:1), afforded 17; yield:
557 mg (98%).
H. Choi, Appl. Catal. A 2005, 291, 238–246.
[11] A. S. K. Hashmi, Angew. Chem. 2005, 117, 7150–7154;
Angew. Chem. Int. Ed. 2005, 44, 6990–6993.
[12] A. S. K. Hashmi, L. Schwarz, M. Bolte, Tetrahedron Lett.
1998, 39, 8969–8972.
In addition traces of 16 were obtained (less than 1 mg), with
an excess ofacrolein, 16 was prepared independently.
Product: 16: Rf (hexanes/ethyl acetate, 5:1)¼0.20; 1H NMR
(250 MHz, CDCl3): d¼2.24 (s, 3H), 2.73–2.79 (m, 2H), 2.90–
2.96 (m, 2H), 5.83–5.85 (m, 1H), 5.88–5.89 (m, 1H), 9.82 (t,
J¼1.4 Hz, 1 H); 13C NMR (62.9 MHz, CDCl3): d¼13.24 (q),
20.66 (t), 41.82 (t), 105.80 (d), 105.94 (d), 150.65 (s), 151.79
(s), 201.05 (d); IR (NaCl, film): n¼3105, 2949, 2923, 2828,
[13] A. S. K. Hashmi, L. Schwarz, J. W. Bats, J. Prakt. Chem.
2000, 342, 40–51.
[14] J. A. Marshall, E. D. Robinson, J. Org. Chem. 1990, 55,
3450–3451.
[15] J. A. Marshall, G. S. Bartley, J. Org. Chem. 1994, 59,
7160–7171.
[16] A. S. K. Hashmi, J.-H. Choi, J. W. Bats, J. Prakt. Chem.
1999, 341, 342–357.
[17] A. S. K. Hashmi, T. L. Ruppert, T. Knçfel, J. W. Bats, J.
Org. Chem. 1997, 62, 7295–7304.
[18] A. S. K. Hashmi, L. Schwarz, Chem. Ber./Recueil 1997,
130, 1449–1456.
[19] A. S. K. Hashmi, Angew. Chem. 1995, 107, 1749–1751;
Angew. Chem. Int. Ed. Engl. 1995, 34, 1581–1583.
[20] C.-G. Yang, C. He, J. Am. Chem. Soc. 2005, 127, 6966–
6967.
[21] A. Riad, Z. Mouloungui, M. Delmas, A. Gaset, Synth.
Commun. 1989, 19, 3169–3173.
¼
2726, 1726 (C O), 1616, 1570, 1436, 1387, 1218, 1162, 1021,
941, 784 cmꢀ1; MS (80 eV): m/z (%)¼138 (30) [Mþ], 95
(100), 82 (58); anal. calcd. for C8H10O2 (136.16): H 7.30, C
69.55; found H: 7.27, C 75.74.
Product 17: Rf (hexanes/ethyl acetate, 5:1)¼0.56; 1H NMR
(250 MHz, CDCl3): d¼2.22–2.31 (m, 2H), 2.25 (s, 9H), 2.55–
2.61 (m, 2H), 3.99 (t, J¼7.6 Hz, 1H), 5.83–5.88 (m, 4H), 5.96
(d, J¼3.0 Hz, 2 H); 13C NMR (62.9 MHz, CDCl3): d¼13.32
(q), 13.40 (q, 2C), 25.73 (t), 31.06 (t), 38.12 (d), 105.48 (d),
105.60 (d), 105.75 (d, 2C), 106.26 (d, 2C), 150.13 (s), 150.63 (s,
2C), 153.28 (s, 2C), 153.49 (s); IR (NaCl, film): n¼3104, 2947,
2922, 2884, 1616, 1568, 1452, 1357, 1219, 1131, 1020, 958,
779 cmꢀ1; anal. calcd. for C18H20O3 (284.35): H 7.09, C 76.03;
found: H 7.26, C 69.77.
708
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 705 – 708