Gold catalysis: Deuterated substrates as the key for an experimental insight into the mechanism and selectivity of the phenol synthesis

Article abstract of DOI:10.1002/chem.200701795

The second phase of the gold-catalyzed phenol synthesis, the ring opening of the intermediate arene oxide, follows general acid catalysis. The product selectivity is determined by the substrate only and can be explained by the stability of the intermediat

Full text of DOI:10.1002/chem.200701795

FULL PAPER
DOI: 10.1002/chem.200701795
Gold Catalysis: Deuterated Substrates as the Key for an Experimental Insight
into the Mechanism and Selectivity ofthe Phenol Synthesis
A. Stephen K. Hashmi,*[a] Matthias Rudolph,[b] Hans-Ullrich Siehl,*[c] Masato Tanaka,[c]
Jan W. Bats,[d] and Wolfgang Frey[b]
Abstract: The second phase of the
gold-catalyzed phenol synthesis, the
ring opening of the intermediate arene
oxide, follows general acid catalysis.
The product selectivity is determined
by the substrate only and can be ex-
plained by the stability of the inter-
mediate arenium ions. Thus, even
remote substitutents can be used to
control the chemoselectivity of the
overall reaction by electronic influen-
ces and their influence is stronger than
the steric influence of neighboring sub-
stituents. This is supported by quantum
chemical calculations of the intermedi-
ates. The lack of exchange of deuteri-
um labels excludes even equilibria with
acetylide or vinylidene intermediates
and the observed deuterium distribu-
tion in the final products is in accord
with the NIH-shift reaction. In addi-
tion, these findings now explain previ-
ously obtained results.
Keywords: gold
catalysis · quantum chemical calcu-
lations reaction mechanisms
selectivity
·
homogeneous
·
·
Introduction
The gold-catalyzed synthesis of highly substituted arenes 2
from w-alkynyl furans 1 is a very efficient tool for organic
synthesis (Scheme 1).[1–4] Early 18O labeling experiments
proofed an intramolecular migration of the furan oxygen
Scheme 1. Gold-catalyzed phenol synthesis.
[a]Prof. Dr. A. S. K. Hashmi
Organisch-Chemisches Institut
Ruprecht-Karls-Universität Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
Fax (+49)6221-544-205
atom (which finally becomes the phenolic oxygen atom),
which is evidence for a 1,2-migration via an arene oxide in-
termediate.[1a]
E-mail: hashmi@hashmi.de
In attempts to establish catalysts that are more stable and
reactive than simple gold halides,[2] we found a catalyst
which allowed the detection of the intermediate arene
oxides G or oxepines H (Scheme 2) as a transient species by
in situ NMR spectroscopy.[3] Furthermore, it was possible to
trap the intermediate G as the product of a [4+2]cycloaddi-
tion. While the AuIII-catalyzed reactions were highly selec-
tive, with the less-selective PtII catalysts Echavarren ob-
served side products which could be formed by hydrolysis of
an intermediate of type F (pathway II).[4] Accompanying
theoretical work favored pathway II, passing through a car-
+
[b]Dipl.-Chem. M. Rudolph, Dr. W. Frey
Institut für Organische Chemie
Universität Stuttgart, Pfaffenwaldring 55
70569 Stuttgart (Germany)
[c]Prof. Dr. H.-U. Siehl, Dr. M. Tanaka
Institut für Organische Chemie I
Universitaet Ulm, Albert Einstein Allee 11
89069 Ulm (Germany)
Fax (+49)731-502-2787
E-mail: Ullrich.Siehl@uni-ulm.de
+
[d]Dr. J. W. Bats
Institut für Organische Chemie und Chemische Biologie
Johann Wolfgang Goethe-Universität Frankfurt
Marie-Curie-Strasse 11
bene complex intermediate, over pathway I, proceeding by
[1b]
an initial [4+2]reaction.
Only recently, similar side prod-
60439 Frankfurt (Germany)
[+]Authors responsible for the crystallographic investigation.
ucts have been reported for special substrates in gold-cata-
lyzed reactions.[4c] Other conceivable pathways include vinyl-
idene complexes M or alkynyl complexes P (pathways IV
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2008, 14, 3703 – 3708
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3703

Scheme 2. Possible pathways of the gold-catalyzed phenol synthesis.
and V), or, if the catalyst system would be reduced in situ,
P from pathway V (Scheme 2) would eliminate the deuteri-
um. For a vinylidene species M the same problem would
occur, even if M would be formed by a 1,2-D-shift and thus
keep the deuterium label, it would be finally eliminated
during formation of the sp2-carbon atom of the oxepine
system (for example, from N to O).
2
À
an insertion into the C
by a second insertion of the alkyne moiety, to deliver J.
A
Results and Discussion
As the deuterium was still present in H, we could now
monitor its way right up to the product 5a. As expected for
a 5-substituted furan, the deuterium was finally eliminated
in the aromatization step, thus we decided to use unsubsti-
tuted furans, which were known to give a mixture of two re-
gioisomeric phenols, at the 5-position.[1a,b] The deuterated
alkyne 6 in the reaction with AuCl3 led to the two phenols
7a and 7b (Figure 2). Analysis
With the ability to investigate the intermediate arene
oxide[3a] or oxepine,[3b] we first used the deuterated alkyne 3
(Figure 1) and catalyst 4. 1H NMR spectroscopy clearly
showed that even in water containing solvent almost all the
deuterium was still present at the corresponding position in
the oxepine intermediate H (Figure 1). The alkynyl complex
of the proton NMR spectrum
shows that in product 7a, in
which the oxygen atom did not
migrate to another carbon
atom, the deuterium also main-
tained its position. More than
90% of the deuterium was re-
tained. Product 7b, in which
the oxygen atom migrated to
the carbon atom bearing the
deuterium, still possessed 45%
of the deuterium, but it had
moved to the neighboring
carbon atom by a 1,2-shift. So,
in the pathway to 7b, in addi-
tion to the simple loss of D+, a
NIH-shift[5] was observed. En-
couraged by these results, we
decided to deuterate the 5-posi-
tion of the furan by deprotona-
Figure 1. Deuterium at the corresponding oxepine position in 5a.
tion of diethyl acetal-protected
3704
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3703 – 3708

Phenol Synthesis
FULL PAPER
observed for the deuterium
labels (pathway Ia), also trans-
fer the R group by means of
Wagner–Meerwein shifts (espe-
cially for the easily migrating
aryl substituent in 3, but only
phenol 2 (R=4-bromophenyl)
was observed), which would
lead to products M and N in
about a 2:1 ratio. As this was
not observed at all, the only ex-
planation was that the inter-
mediate Wheland-type arenium
ion B is not passed before a re-
gioselective formation of the
arene oxide occurs, which was
only in accord with pathway II.
The selectivity of the ring open-
ing of the arene oxide seemed
to be determined by the intrin-
sic stability of the pentadienyl
cation formed.
Figure 2. Deuterium distribution in the products obtained from 6.
To support this assumption,
we investigated the AuCl3-cata-
lyzed reaction of substrate 10a
(Scheme 3), the latter bearing a
methyl group at the 3-position
but as in 6, possessing no sub-
stituent at the 5-position of the
Figure 3. Deuterium distribution in the products obtained from 8.
furfural with nBuLi and quenching with D2O.[6] Deprotec-
tion of the acetal, imine formation with propargylamine, re-
duction, and tosylation delivered 8. After gold catalysis,
product 9b (Figure 3) still held 80% of the deuterium,
whereas two isotopomers of 9a with deuteration levels of 40
and 20% were obtained.
These experiments showed that in the second phase of the
reaction, the transformation of the arene oxide/oxepine to
the phenol, a NIH-shift reaction was involved. As in all sub-
strates, two different NIH-shifts were possible and a direct
loss of a proton in general is a third possibility, it was diffi-
cult to interpret the product distribution itself. But if one
compared the H/D ratios of products 7b and 9b, the lower
H/D ratios for 9b should result from an acid-catalyzed iso-
merization rather than a spontaneous isomerization.[7] It was
obvious that the isomerization of the arene oxide to the
phenols follows general acid catalysis (the metal might only
play a role in Lewis catalysis, but the selectivity is deter-
mined by the substrate itself). Indeed, the addition of
5 mol% of pTsOH (pTs: p-toluenesulfonyl) to the enriched
arene oxide G (R=methyl) led immediately to the corre-
sponding phenol. Concerning the mechanistic aspects, it is
very interesting that product 9a showed two deuterated po-
sitions and, furthermore, the 5-position was strongly fa-
vored. If pathway I (Scheme 2) would be passed, intermedi-
ate B should consequently, in analogy to the NIH-shifts as
Scheme 3. Competing pathways for 10a/b.
furan ring. After column chromatography, 90% of 11b (X-
ray structure:[8] Figure 4) and only 4% of the other regioiso-
mer 11a [conversion of 10b (X-ray structure:[8] Figure 4)
yielded 88% of 11c as the only product]could be isolated!
This inversion of the chemoselectivity by introduction of
a methyl group far away from the reaction center must be
an electronic and not a steric effect, the stabilizing effect of
Chem. Eur. J. 2008, 14, 3703 – 3708
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3705

A. S. K. Hashmi et al.
Table 1. Comparison of energy differences[a] of the intermediates (DEint
and the products (DEpr) between structures of type I and II in Figure 5.
)
Y=H Y=CH3
DEpr DEpr
gas phase À2.03 (À1.89) À0.82 (À0.69) À5.13 (À4.68) À0.07 (0.05)
PCM[c]
[b]
[b]
[b]
[b]
DEint
DEint
À0.59 (À0.41) À0.42 (À0.36) À4.17 (À3.97) 0.82 (0.93)
[a]Energy differences are shown in kcalmol À1. Relative energies with
ZPE corrected are shown in parentheses. [b] DE=E(structure of type
II)ÀE(structure of type I). [c]PCM model calculation with e=78.6
(water).
CH3). The intermediate structures of type II are more stable
than the intermediate structures of type I. The larger energy
difference (DEint) for the methyl-substituted compounds
(Y=CH3) as compared to those with Y=H is a result of the
charge-stabilizing substituent effect of the CH3 group in the
intermediate structure of type II. The calculated energy dif-
ference of the intermediates determines the regioselectivity
of the reaction and leads to kinetically controlled product
formation. Thus, our calculations support the experimentally
observed product ratio, the excess yield of 11b (Y=CH3,
structure of type II) over 11a.
As one could argue that a methyl oxirane ring might un-
dergo an oxygen migration rather than a NIH shift of the
methyl group, we finally used substrate 12 that contains
methyl groups at the 4- and 5-positions of the furan ring
(Scheme 4).
Figure 4. X-ray structures of A) 10b and B) 11b.[8]
the methyl group being only effective in the pentadienyl
cation 10d and not in 10c.
In addition to these experi-
mental results, the energy dif-
ferences between the two possi-
ble intermediates and products
(Figure 5, further computation-
al details can be found in the
Scheme 4. Product distribution in the conversion of 12.
Supporting Information) were
quantified by using quantum
chemical calculations (HF/6-
Here a ring opening of the arene oxide in both directions
was observed and for the first opening mode the two possi-
ble NIH-shifts of the methyl group led to ketone 13a (10%)
and phenol 13b (10%) besides the other expected phenol
13c (51%), which resulted from a ring opening at the
higher substituted carbon–oxygen bond (structure determi-
nation of 13b/13c is based on HSQC-, HMQC- and
ROESY-NMR spectra). In addition, the crystal structures of
the penta-substituted phenol 13c and the ketone 13a (Fig-
ure 6A and B, respectively)[8] could be determined.
31G*).[9] Table 1 shows the energy differences between iso-
meric structures of type I and II for the intermediates and
products (Figure 5). The energy differences of the products
(DEpr) are less than 1 kcalmolÀ1, while the energy differen-
ces of the intermediates (DEint) are relatively large, up to
5 kcalmolÀ1 for the methyl-substituted intermediates (Y=
Having learned all this, in a retrospective view even prod-
uct 15b as a side product in the Jungianol synthesis[1d] could
be explained by this theory—there the electron-withdrawing
substituent in the tether destabilized the usually formed
pentadienyl cation and induced a less-selective ring opening
of the arene oxide in both directions and a subsequent NIH-
shift of the methyl group in the reaction leading to 15b
(Scheme 5).
Finally, our interest focused on the influence of a sterical-
ly demanding substituent at the propargylic position of the
starting material. Conversion of substrate 16 (Scheme 6)
Figure 5. Molecular structures of the intermediates and products (Y=H,
CH3).
3706
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3703 – 3708

Phenol Synthesis
FULL PAPER
All these results are in accordance with the mechanism
depicted in Scheme 7.
Conclusion
Overall, we could provide a direct experimental proof that
the reaction does not proceed via an alkynyl or a vinylidene
complex. Furthermore, the observed distribution of deuteri-
um labels or aryl substituents is in accord with the penta-
dienyl cation formed only after the arene oxide/oxepine in-
termediate and not before (see pathway I; Scheme 2); the
selectivity is intrinsic to the substrate and not to the Lewis/
Brøndsted acid catalyst, the ring opening leading to the
more stable pentadienyl cation being preferred.
Acknowledgement
Figure 6. X-ray structures of A) 13c and B) 13a.[8]
We thank the BMBF for their support of this work (01 RI 051177). Gold
salts were generously donated by Umicore AG & Co. KG.
shifted the product distribution from the 1:1.6 ratio observed
with the substrate lacking the diethyl substitution to a 1:4
ratio of 17a/17b, controlled by ring opening to the penta-
dienyl-cation with decreased steric interaction.
[1]a) A. S. K. Hashmi, T. M. Frost, J. W. Bats, J. Am. Chem. Soc. 2000,
122, 11553–11554; b) A. S. K. Hashmi, T. M. Frost, J. W. Bats, Org.
Lett. 2001, 3, 3769–3771; c) A. S. K. Hashmi, T. M. Frost, J. W. Bats,
Catal. Today 2002, 72, 19–27;
d) A. S. K. Hashmi, L. Ding, J. W.
Bats, P. Fischer, W. Frey, Chem.
Eur. J. 2003, 9, 4339–4345;
e) A. S. K. Hashmi, L. Grundl, Tet-
rahedron 2005, 61, 6231–6236;
f) A. S. K. Hashmi, M. C. Blanco,
E. Kurpejovic, W. Frey, J. W. Bats,
Adv. Synth. Catal. 2006, 348, 709–
713; g) A. S. K. Hashmi, P. Haufe,
Scheme 5. Product distribution in the key step of the Jungianol synthesis.
C. Schmid, A. Rivas Nass, W. Frey,
Chem. Eur. J. 2006, 12, 5376–5382;
h) A. S. K. Hashmi, J. P. Weyrauch,
E. Kurpejovic, T. M. Frost, B. Miehlich, W. Frey, J. W. Bats, Chem.
Eur. J. 2006, 12, 5806–5814; i) S. Carrettin, M. C. Blanco, A. Corma,
A. S. K. Hashmi, Adv. Synth. Catal. 2006, 348, 1283–1288; j) A. S. K.
Hashmi, R. SalathØ, W. Frey, Chem. Eur. J. 2006, 12, 6991–6996;
k) A. S. K. Hashmi, E. Kurpejovic, W. Frey, J. W. Bats, Tetrahedron
2007, 63, 5879–5885; l) A. S. K. Hashmi, M. Wçlfle, J. H. Teles, W.
Frey, Synlett 2007, 1747–1752; m) A. S. K. Hashmi, F. Ata, E. Kurpe-
jovic, J. Huck, M. Rudolph, Top. Catal. 2007, 44, 245–251.
[2]A. S. K. Hashmi, J. P. Weyrauch,
Scheme 6. Influencing the product distribution by steric interactions.
M. Rudolph, E. Kurpejovic,
Angew. Chem. 2004, 116, 6707–
6709.
[3]a) A. S. K. Hashmi, M. Rudolph,
J. P. Weyrauch, M. Wçlfle, W. Frey,
J. W. Bats, Angew. Chem. 2005,
117, 2858–2861; Angew. Chem.
Int. Ed. 2005, 44, 2798–2801;
b) A. S. K. Hashmi, E. Kurpejovic,
M. Wçlfle, W. Frey, J. W. Bats,
Adv. Synth. Catal. 2007, 349, 1743–
1750.
[4]a) B.
Martín-Matute,
A. M.
C. D. J.
Echavarren,
Cµrdenas,
Scheme 7. Pathway supported by the experimental findings (A=Lewis or Brøndsted acid). The intrinsic stabil-
ity of the cation determines the regioselectivity of the epoxide ring opening and, therefore, the product selec-
tivity.
Angew. Chem. 2001, 113, 4890–
4893; Angew. Chem. Int. Ed. 2001,
40, 4754–4757; b) B. Martín-
Chem. Eur. J. 2008, 14, 3703 – 3708
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3707

A. S. K. Hashmi et al.
Matute, C. Nevado, D. J. Cµrdenas, A. M. Echavarren, J. Am. Chem.
Soc. 2003, 125, 5757–5766; c) A. S. K. Hashmi, M. Wçlfle, F. Ata, M.
Hamzic, R. SalathØ, W. Frey, Adv. Synth. Catal. 2006, 348, 2501–
2508.
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G.
Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V.
Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C.
Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford CT, 2004.
[5]G. Guroff, J. W. Daly, D. M. Jerina, J. Renson, B. Witkop, S. Uden-
friend, Science 1967, 158, 1524–1527.
[6]D. J. Chadwick, J. Chambers, P. K. G. Hodgson, G. D. Meakins, R. L.
Snowden, J. Chem. Soc. Perkin Trans. 1 1974, 1141–1145.
[7]D. R. Boyd, J. W. Daly, D. M. Jerina, Biochemistry 1972, 11, 1961–
1966.
[8]CCDC 606485 ( 10b), 666815 (11b), 666816 (13a), and 666817 (13c)
contain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif
[9]Gaussian 03 (Revision D.02), M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery,
Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
Received: November 14, 2007
Published online: March 3, 2008
3708
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3703 – 3708