Generation and reactivity of the dehydrotropylium-Co2(CO) 6 ion

Article abstract of DOI:10.1039/b812174e

Dehydrotropylium-Co₂(CO)₆ ion (2) has been generated by the Lewis acid mediated ionization of alcohol (3); it is attacked by relatively strong nucleophiles (N > 1), but undergoes a radical homocoupling in the presence of weak nucleophiles (N < 1). The Royal Society of Chemistry.

Full text of DOI:10.1039/b812174e

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Generation and reactivity of the dehydrotropylium-Co2(CO)6 ionw
Sheida Amiralaei and James R. Green*
Received (in College Park, MD, USA) 16th July 2008, Accepted 7th August 2008
First published as an Advance Article on the web 17th September 2008
DOI: 10.1039/b812174e
Dehydrotropylium-Co2(CO)6 ion (2) has been generated by the
Lewis acid mediated ionization of alcohol (3); it is attacked
by relatively strong nucleophiles (N 4 1), but undergoes a
radical homocoupling in the presence of weak nucleophiles
(N o 1).
The chemistry of propargyliumdicobalt cations, more com-
monly known as the Nicholas reaction, has seen widespread
use in organic synthesis due a to balance of high stability and
sufficient reactivity of the cations, the predictable site selectiv-
ity and stereochemical features of the reaction, and the
straightforward handling or decomplexation of the precursor
and product alkyne complexes.1 The stability of these cations
has normally been interpreted in terms of a fluxional structural
represented by 1, which has found support in spectroscopic
and crystallographic studies.2 Nevertheless, there are impor-
tant features of the reactivity of these cations that are poorly
understood; among these, the most notable is the effect of
substituents on cation stability and reactivity.3 Substitution at
the propargylic site has been shown to have a minimal effect
on the aforementioned properties. This has significant advan-
tages, in that generation of cations substituted by electron
withdrawing groups4 and in antiaromatic systems5 is often
possible. This feature, however, complicates an in depth
understanding of these cations, as by some measures classical
standard conjugative and hyperconjugative stabilizing
groups actually result in a very slight de-stabilization of the
cation.3b,6 In our own group, solvolysis and allylation reac-
tions have often shown inverted trends in selectivity towards
substitution.
Access to ion 2 comes from precursor alcohol 3, whose
synthesis originated from propargyl diacetate complex 4, avail-
able by ring closing metathesis chemistry.7 Subjecting 4 to
H2SO4 treatment in the presence of acetic acid gave rearranged
diacetate 5 in excellent yield (96%). Removal of the acetate
functions was best accomplished reductively (DIBAL-H), and
the resulting diol 6 was oxidized selectively by MnO2 to give the
b-hydroxy ketone 7 (62% yield, 2 steps). Acid induced elimina-
tion of the alcohol function afforded the dienone complex 8 in
fair yield (50%). Subsequent reduction of 8 gave the dienol 3
(82% yield, 86% based on recovered starting material), with a
small amount of rearranged dienynol complex 9 (5% yield) and
a small amount of recovered 8 (5% recovery) (Scheme 1).
Our group has been engaged in the development of
a number of methods for the preparation of cycloheptyne-
dicobalt systems, many of which are capable of giving
precursors to substituted propargyliumdicobalt cations.7–9 In
particular, we believed the study of the nominally aromatic,
‘dehydrotropylium’ ion 2 would give information regarding
strongly stabilizing influences on such cations, and
therefore be important to their understanding. This
communication reports the preparation and reactivity studies
of this cation.
Scheme 1 Synthesis of cycloheptadienynol complex 3.
Subjecting alcohol 3 to HBF4 in Et2O at ꢀ75 1C, or in
CH2Cl2 followed by precipitation by Et2O, afforded a solid
1
whose H NMR spectrum (CD2Cl2) possessed resonances at
d 8.47 (d, J = 9.5 Hz, 2H), 8.36 (t, J = 9.4 Hz, 1H), and
8.19 ppm (apparent t, J = 9.5 Hz, 2H), but which gradually
degraded even at ꢀ20 1C. With the limited stability of 2, we
decided to investigate the Nicholas reactions of 3 with nucleo-
philes of varying strength, particularly as reflected by Mayr’s
N values,10 in order to obtain a measure of the electrophilicity
Department of Chemistry and Chemistry, University of Windsor,
Windsor, ON, Canada N9B 3P4. E-mail: jgreen@uwindsor.ca;
Fax: +1-519-973-7098; Tel: +1-519-253-3000 Ext. 3545
w Electronic supplementary information (ESI) available: Experimental
conditions for the preparation of 3, 10a–e, and 11; the spectral data
and copies of the 13C NMR spectra of 3, 10a–e, 11. See DOI: 10.1039/
b812174e
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4971–4973 | 4971

View Article Online
Table 1 Condensation reactions of 3 with N 4 1 nucleophiles
3-methylanisole (N = 0.13), these dimers could be isolated
in 26% yield (11aa : 11ag = 1 : 2.0). For thiophene (N = ꢀ1.01)
the dimers were formed in 34% yield (11aa : 11ag = 1 : 2.0),
and co-eluted with a trace (4%) of the condensation product
10fa. Most successfully, the use of mesitylenez allowed for-
mation of the dimers in 50% yield (11aa : 11ag = 1 : 2.0)
(Table 2).
Table 2 Reactions in the presence of N o 1 nucleophile
Entry
Nu (Na)
Yield (%)b
11aa : 11ag
1
2
3
4
a
3-Methylanisole (0.13)
Thiophene (ꢀ1.01)
Mesitylene (o ꢀ2.5)
None
26
34c
50
9
33 : 67
33 : 67
33 : 67
33 : 67
Yield
Compound (%)
Entry Nu (Na)
a- : g-
1
2
3
1,3,5-Trimethoxybenzene (3.40) 10a
1,3,5-Trimethoxybenzene (3.40) 10ab
2-Methallyltrimethylsilane
(4.41)
88
54
73
100 : 0
100 : 0
67 : 33
b
Nucleophilicity value. Yield based in all cases on 1 mmol 3 giving a
10b
c
maximum of 0.5 mmol 11. Isolated with 4% of condensation product
10fa.
4
5
6
a
Allyltrimethylsilane (1.79)
Methylenecyclohexane (1.66)
2-Methylthiophene (1.26)
b
10c
10d
10e
70
50
83
83 : 17
91 : 9
100 : 0
Nucleophilicity value. Starting from 9.
As propargyldicobalt radicals and their dimerization pro-
ducts are known results of the single electron reduction of
propargyliumdicobalt cations,12,13 the isolation of these dimers
is attributed to the intermediacy of a cycloheptadienynyl-
dicobalt radical; this process is also reminiscent of nucleophile
induced electron transfer of pentadienylirons.14 Nevertheless,
the samples generated for spectroscopic studies of the cation 2,
in the absence of these arenes but in the presence of Et2O, also
contained significant amounts of the dimer 11 (11aa : 11ag =
1 : 2.0). Furthermore, Lewis acid mediated ionization of alcohol
3 with no added nucleophile present gave some dimer, albeit in
lower yield (9%, 11aa : 11ag = 1 : 2.0). As a result, it appears
that the electron transfer source may not only be the added
arene, but also a second alkynedicobalt unit or the ethereal
solvent; both of these types of propargyldicobalt cation to
radical mediators have been observed by Melikyan.12,13 The
trace amount of condensation product observed in the thio-
phene case may be accounted for by a free radical substitution
process on thiophene.15 Dimer 11aa is isolated as a single
diastereomer, which we are assigning as the syn diastereomer,
based on extensive precedent for its predominance in
propargyldicobalt radical dimerization reactions.12,13,16
and reactivity pattern of 2 (Table 1). In the presence of
BF3–OEt2 (3 equiv., 0 1C, CH2Cl2), 1,3,5-trimethoxybenzene
(N = 3.40) reacted rapidly with 3 to give the 7-arylated
(a-arylated) dienyne complex 10aa exclusively in excellent
yield; under analogous conditions, rearranged dienynol com-
plex 9 also afforded 10aa (54% yield). Methallyltrimethyl-
silane (N = 4.41) reacted with 3 similarly to afford mixtures of
7-(a-) and 5-(g-) methallylated condensation products 10ba
and 10bg (73% yield, 2.0 : 1 ratio), while the less nucleophilic
allyltrimethylsilane (N = 1.79) gave condensation products
with an increased proportion of a- attack (70% yield,
10ca : 10cg = 4.9 : 1). Methylenecyclohexane (N = 1.66)
reacted to give a condensation product mixture containing
only a small amount of the g-product (50% yield, 10da : 10dg
= 10 : 1), while 2-methylthiophenene (N = 1.26) gave good
condensation yields but no observable g-product whatsoever
(10ea, 83% yield). The increasing amounts of conjugated
1,3-dien-5-yne products with decreasing nucleophile N value
is reminiscent of the previously reported increased amount of
conjugated enyne products with decreasing nucleophilicity in
Nicholas reactions of acetoxycycloheptenyne complexes.11
The apparently inconsistent results with 1,3,5-trimethoxy-
benzene may be attributed to its demonstrated ability to react
reversibly with propargyldicobalt cations.11
The distinction between the behaviour of the dehydro-
tropylium-Co2(CO)6 ion (2) and normal hexacarbonyl-
propargyldicobalt cations is quite striking. With electrophilicity
strengths of E = ꢀ1.2–ꢀ2.2,6,17 propargyl cation-Co2(CO)6
complexes in general may enter into electrophile–nucleophile
reactions with nucleophiles as weak as N = ꢀ4; conversely, 2
undergoes analogous reactions only with relatively strong (N 4 1)
nucleophiles, and switches to a radical dimerization process in the
presence of weaker nucleophiles (N o 1). In terms of judging the
stability of 2, however, these results are complex. While the relative
With nucleophiles less reactive than 2-methylthiophene, the
reaction product profiles underwent a distinct change. Con-
densation products were absent or at most present in
trace amounts, and the major isolated products were those
of dimerization of cycloheptadienyne unit, as mixtures of
a,a- and a,g-isomers (11aa and 11ag). In the case of
ꢁc
This journal is The Royal Society of Chemistry 2008
4972 | Chem. Commun., 2008, 4971–4973

View Article Online
slowing of the electrophile–nucleophile combination reactions is
consistent with a greater stability, and perhaps aromatic stabiliza-
tion, of the cation, the ready conversion of such a cation to a
radical has been cited as evidence of lowered stability of the cation
relative to the radical.12 Consequently, no definitive statement on
the aromatic stabilization of 2 can be made as of yet.
8 (a) J. R. Green, Chem. Commun., 1998, 1751; (b) M. M. Patel and
J. R. Green, Chem. Commun., 1999, 509; (c) Y. Lu and J. R. Green,
Synlett, 2001, 243; (d) Y. Ding and J. R. Green, Synlett, 2005, 271;
(e) S. Djurdjevic and J. R. Green, Org. Lett., 2007, 9, 5505.
9 For lead references to the contributions by other groups’ work on
cycloheptynedicobalt complexes, see ref. 8e and: (a) S. L.
Schreiber, T. Sammakia and W. E. Crowe, J. Am. Chem. Soc.,
1986, 108, 3128; (b) T. Nakamura, T. Matsui, K. Tanino and I.
Kuwajima, J. Org. Chem., 1997, 62, 3032; (c) N. Iwasawa and H.
Satoh, J. Am. Chem. Soc., 1999, 121, 7951; (d) K. Tanino, T.
Shimizu, M. Miyama and I. Kuwajima, J. Am. Chem. Soc., 2000,
122, 6116; (e) V. B. Golovko, L. J. Hope-Weeks, M. J. Mays, M.
McPartlin, A. M. Sloan and A. D. Woods, New J. Chem., 2004, 28,
527.
10 (a) H. Mayr, B. Kempf and A. R. Ofial, Acc. Chem. Res., 2003, 36,
66; (b) H. Mayr, Angew. Chem., Int. Ed. Engl., 1994, 33, 938.
11 (a) J. DiMartino and J. R. Green, Tetrahedron, 2006, 62, 1402;
(b) R. Gibe and J. R. Green, Chem. Commun., 2002, 1550.
12 G. G. Melikyan, F. Villena, S. Sepanian, M. Pulido, H. Sarkissian
and A. Florut, Org. Lett., 2003, 5, 3395.
In summary, we have been able to make the precursor
alcohol 3 to the dehydrotropylium-Co2(CO)6 cation (2), and
to generate the cation itself and observe both electrophilic and
electron transfer–radical dimerization reactions. Further stu-
dies on comparative reactions of 3 with acyclic analogues, and
calculational studies on 2 would shed light on the question of
stability/reactivity of this nominally aromatic cation, and will
be reported in due course.
13 A number of ethers, thioethers, acetals, dithianes, and ortho
estershave been reported to be radical mediators with propargyl-
dicobalt cations, but diethyl ether is not included in this group: (a)
G. G. Melikyan and A. Deravalian, J. Organomet. Chem., 1997,
544, 143; (b) G. G. Melikyan, A. Deravakian, S. Myer, S. Yadegar,
K. I. Hardcastle, J. Ciurash and P. Toure, J. Organomet. Chem.,
1999, 578, 68; (c) G. G. Melikyan, F. Villena, A. Florut, S.
Sepanian, H. Sarkissian, A. Rowe, P. Toure, D. Mehta, N.
Christina, S. Myer, D. Miller, S. Scanlon, M. Porazik and M.
Gruselle, Organometallics, 2006, 25, 4680; (d) G. G. Melikyan, A.
Floruti, L. Devletyan, P. Toure, N. Dean and L. Carlson, Orga-
nometallics, 2007, 26, 3173, and references therein.
14 A. J. Birch, P. E. Cross, D. A. Lewis, D. A. White and S. B. Wild,
J. Chem. Soc. A, 1968, 332.
15 (a) L. H. Klemm and J. Dorsey, J. Heterocycl. Chem., 1991, 28,
1153; (b) O. S. Herrera, J. D. Nieto, S. I. Lane and E. V. Oexler,
Can. J. Chem., 2003, 81, 1477.
Notes and references
z The N value is unknown, but based on the values for toluene and m-
xylene, its estimated maximum value is ꢀ2.5.
1 (a) D. D. Diaz, J. M. Betancort and V. S. Martin, Synlett, 2007,
343; (b) B. J. Teobald, Tetrahedron, 2002, 58, 4133; (c) J. R. Green,
Curr. Org. Chem., 2001, 5, 809.
2 (a) G. G. Melikyan, S. Bright, T. Monroe, K. I. Hardcastle and J.
Ciurash, Angew. Chem., Int. Ed., 1998, 37, 161; (b) S. L. Schreiber,
M. T. Klimas and T. Sammakia, J. Am. Chem. Soc., 1987, 109, 5749.
3 (a) M. Gruselle, C. Cordier, M. Salmain, H. El Amouri, C. Guerin, J.
Vaissermann and G. Jaouen, Organometallics, 1990, 9, 2993; (b) R. E.
Connor and K. M. Nicholas, J. Organomet. Chem., 1977, 125, C45.
4 H. Amouri, J.-P. Be
´
Gruselle and B. Male
´
gue, A. Chennoufi, D. Bonnet-Delpon, M.
zieux, Org. Lett., 2000, 2, 807.
´
16 C. S. Vizniowski, J. R. Green, T. L. Breen and A. V. Dalacu,
J. Org. Chem., 1995, 60, 7496.
17 For an extensive list of electrophilicity E values, see ref. 10b and:
H. Mayr and O. R. Arfin, in Carbocation Chemistry, ed. G. A.
Olah and G. K. S. Prakash, John Wiley and Sons, 2004,
pp. 331–358.
5 J. A. Dunn, W. J. Hunks, R. Ruffolo, S. S. Rigby, M. A. Brook
and M. J. McGlinchey, Organometallics, 1999, 18, 3372.
6 O. Kuhn, D. Rau and H. Mayr, J. Am. Chem. Soc., 1998, 120, 900.
7 (a) J. R. Green, Synlett, 2001, 353; (b) see also: D. G. J. Young, J.
A. Burlison and U. Peters, J. Org. Chem., 2003, 68, 3494.
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4971–4973 | 4973