Borrowing hydrogen: Indirect "Wittig" olefination for the formation of C-C bonds from alcohols

Article abstract of DOI:10.1002/ejoc.200600070

The successful development of an indirect three-step domino sequence for the formation of C-C bonds from alcohol substrates is described. An iridium-catalysed dehydrogenation of alcohol 1 affords the intermediate aldehyde 2. The desired C-C bond can then be formed by a facile Wittig olefination, yielding the intermediate alkene 3. In the final step the alkene is hydrogenated to afford the indirect Wittig product, the alkane 4. The key to this process is the concept of borrowing hydrogen; hydrogen removed in the initial dehydrogenation step is simply borrowed by the iridium catalyst. Functioning as a hydrogen reservoir, the catalyst facilitates C-C bond formation before subsequently returning the borrowed hydrogen in the final step. Herein we present full details of our examination into both the substrate and reaction scope and the limitations of the catalytic cycle. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejoc.200600070

FULL PAPER
DOI: 10.1002/ejoc.200600070
Borrowing Hydrogen: Indirect “Wittig” Olefination for the Formation of C–C
Bonds from Alcohols
Phillip J. Black,[a] Michael G. Edwards,[a] and Jonathan M. J. Williams*[a]
Keywords: Domino reactions / Dehydrogenation / Iridium / Transfer hydrogenation / Wittig reactions
The successful development of an indirect three-step domino
sequence for the formation of C–C bonds from alcohol sub-
strates is described. An iridium-catalysed dehydrogenation
of alcohol 1 affords the intermediate aldehyde 2. The desired
C–C bond can then be formed by a facile Wittig olefination,
yielding the intermediate alkene 3. In the final step the al-
kene is hydrogenated to afford the indirect Wittig product,
the alkane 4. The key to this process is the concept of bor-
rowing hydrogen; hydrogen removed in the initial dehydro-
genation step is simply borrowed by the iridium catalyst.
Functioning as a hydrogen reservoir, the catalyst facilitates
C–C bond formation before subsequently returning the bor-
rowed hydrogen in the final step. Herein we present full de-
tails of our examination into both the substrate and reaction
scope and the limitations of the catalytic cycle.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
Introduction
Recent research in these laboratories has explored the use
of domino reaction sequences[1] for the development of new
reactions. Inspired by the powerful possibilities of domino
chemistry, we envisioned that a sequence of domino trans-
formations could accomplish an otherwise “impossible” re-
action (Scheme 1). We reasoned that if an unreactive sub-
strate A could be temporarily activated towards reaction
(into electronically activated A*) then the desired bond for-
mation could proceed indirectly upon intermediate A* to
afford the activated intermediate A*–B. The reaction cycle
would then be completed by return of A*–B to the initial
oxidation level, yielding the desired product A–B. We
termed this concept Catalytic Electronic Activation.[2]
We were ultimately successful in employing this concept
for the 1,4-addition of carbon nucleophiles to allylic
alcohols[2] and the β-functionalisation of alcohols.[3,4] The
key to these systems was an aluminium-catalysed (Meer-
wein–Ponndorf–Verley/Oppenauer[5]) oxidation–reduction
couple to shuttle the alcohol–ketone equilibrium necessary
for the indirect reaction (upon the ketone). In order to de-
velop the scope of this concept, it was necessary to develop
a more general oxidation/reduction couple. We believed that
a transition-metal catalyst could function as a “hydrogen
reservoir”, borrowing hydrogen from the substrate to in-
duce the reaction sequence and then subsequently return it
Scheme 1.
in the final step (Scheme 2). In addition to the opportunity
for reduction of a wider range of functional groups this
would also eliminate the requirement for external hydride
donors and/or acceptors.
We tested our “borrowing hydrogen” hypothesis by
studying C–C bond formation upon alcohol substrates
(Scheme 2). An alcohol 1 would undergo metal-catalysed
dehydrogenation to afford an activated intermediate car-
bonyl compound 2. The formation of a C=C bond should
then proceed readily to provide the alkene intermediate 3.
The final hydrogenation step then returns the borrowed hy-
drogen to the intermediate alkene 3, which generates the
alkane product 4. In contrast to most transfer hydrogena-
tion reactions, this process requires the use of only
one equivalent of the alcohol.[6] In addition, because the hy-
drogen is only borrowed temporarily, no overall net-oxi-
dation transformation occurs.
[a] Department of Chemistry, University of Bath,
Claverton Down
Bath BA2 7AY, UK
Fax: +44-01225-826231
E-mail: j.m.j.williams@bath.ac.uk
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2006, 4367–4378
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4367

P. J. Black, M. G. Edwards, J. M. J. Williams
FULL PAPER
[Ir(COD)Cl]2 was used in all the reactions described herein,
we have insufficient evidence to propose the structure of the
species that is catalytically active under the reaction condi-
tions.
Scheme 2.
We communicated our preliminary findings[7] on this
novel three-step domino sequence[8] in 2002 and sub-
sequently reported an improved ruthenium-catalysed sys-
tem that proceeds under milder conditions.[9] A related pro-
cess for the α-alkylation of ketones or nitriles with alcohol
substrates (albeit with an excess of alcohol) has been re-
ported by various researchers.[10] In these reports the C–C
bond is formed by a base-promoted aldol addition to the
intermediate aldeyhde. Furthermore, the synthesis of
amines from alcohols can also be achieved by an oxidation/
imination/reduction sequence.[11] Subsequently we also ap-
plied the concept of borrowing hydrogen to the formation
of C–N bonds with aza-Wittig and imine chemistry.[12]
Herein we wish to report our full findings into the scope
and pathway of iridium-catalysed C–C bond formation
upon alcohols.
Scheme 3.
The results of these two initial experiments were gratify-
ing; 74% of benzyl dihydrocinnamate 8b and 70% of methyl
dihydrocinnamate (8a) were obtained from the ylides 9 and
10, respectively. Most importantly, these reactions gener-
ated considerably more product than our previous attempts
with phosphonates. Furthermore, although some transes-
terification was still observed with the methyl ester ylide 10
(7% of benzyl ester 8b was also obtained), this is minimal
in comparison with the almost complete transesterification
obtained with phosphonate nucleophiles.[15]
We undertook a brief optimisation of the reaction condi-
tions and found that the conditions used previously pro-
Results and Discussion
In our initial efforts, we used Wadsworth–Emmons ole- vided the best yields of product. Of note was the observa-
fination for the formation of the desired C–C bond.[7,13] Al- tion that the use of excess phosphorane ylide inhibited the
though we were able to obtain proof of principle from this reaction, presumably by complexation to iridium; ylide
chemistry, the low yields and complicated workup required complexes of metals do have literature precedent.[16] Re-
made it less than ideal. The incompatibility of phosphonate cently, microwave irradiation of reactions has been shown
nucleophiles made it apparent that an alternative ole- to be beneficial to the promotion of many organic reac-
fination method was required. In theory, any C–C bond- tions,[17] the Wittig reaction included.[17,18] Although we
forming method could be used. However, this method must were able to shorten the reaction times considerably under
be tolerant of the reaction conditions and functional groups microwave irradiation (typically to less than one hour), con-
involved. After careful consideration, we turned our atten- siderable amounts of benzaldehyde and benzyl cinnamate
tion towards Wittig olefination for formation of the crucial were obtained. This was ultimately symptomatic of a prob-
C–C bond. The added benefit of using stabilised phos- lem we were to discover later in our synthetic runs; namely
phorane ylide nucleophiles is that no additional base is re- loss of hydrogen from the system by undetectable pathways
quired. Thus, the domino sequence presented in Scheme 2 (vide infra); thus we reverted to conventional heating.
was examined.
We used the optimal reaction conditions to prepare a
In our previous studies,[7,13] we identified that modifica- series of dihydrocinnamate derivatives (Scheme 4, Table 1).
tion of the iridium system reported by Ishii and co- We were pleased to observe that the domino indirect Wittig
workers[14] was optimal for the crossover transfer hydrogen- process successfully afforded dihydrocinnamate derivatives
ation reaction. The starting point was to subject benzyl 8a–8e in 47–71% yield following chromatography and oxi-
alcohol to the optimised conditions (5 mol-% Ir dimer, dative workup (to remove the otherwise inseparable alkene).
150 °C, 72 hours, 0.333  concentration) in the presence of The overall yields achieved were reasonable given that a
the ester ylides 9 and 10 (Scheme 3). Although 5 mol-% of three-step 51% yield equates to 80% yield for each discrete
4368
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4367–4378

Borrowing Hydrogen: Indirect “Wittig” Olefination for C–C Bonds from Alcohols
FULL PAPER
step. In all cases however, small amounts of both the alde- cinic anhydride (not shown). When heated in the presence
hyde and alkene remained. The standard reaction with ben- of alcohols, acrylate esters were obtained as the products
zyl (triphenylphosphoranylidene)acetate (9) (Entry 2) together with triphenylphosphane. This observation sub-
proved to be best affording benzyl dihydrocinnamate (8b) sequently led us to develop a phosphane-catalysed synthesis
in 71% isolated yield. When alternative ylides were used, a of acrylates from maleic anhydride.[22]
small amount of transesterification (4–13%) occurred un-
Before we could solve the problem of transesterification
der the reaction conditions, resulting in the formation of it was necessary to ascertain the cause. Thus an equimolar
both (E)-benzyl cinnamate (7b) and benzyl dihydrocinna- mixture of methyl dihydrocinnamate (8a) and benzyl
mate (8b). The steric bulk of the ester group was found to alcohol (5) was subjected to three experiments in which the
be unimportant in this process, even with the tert-butyl (En- influence of the reaction components upon this process
try 3) and neopentyl esters (Entry 4) transesterification re- were examined (Scheme 5, Table 2). When the reagents were
sulted. More surprisingly, even amide groups were affected heated in the absence of additives (Entry 1), no transesteri-
(Entries 5–6). In view of the harsh reaction conditions and fication occurred, even under the harsh reaction conditions.
transesterification observed, the preparation of the syntheti- However, the addition of 5 mol-% of caesium carbonate
cally useful Weinreb amide derivative 8f in 47% yield (Entry caused considerable transesterification to occur (Entry 2).
6) was notable.
When all the reaction components were added (Entry 3)
it was notable that a lower level of transesterification was
observed. This is slightly misleading however, as some of
the benzyl alcohol (5) underwent oxidation to benzaldehyde
(6). Although, it is difficult to isolate the effect of the base
from that of the catalyst, it is likely that both are involved in
this process. It is also possible that transesterification occurs
upon the phosphorane ylide. This is however impossible to
verify since when the ylide and alcohol are heated together
decomposition occurs.[23] Furthermore, a literature search
revealed no reports of ylide esters undergoing transesterifi-
cation. If transesterification does occur on the ylide, it is
probably a relatively minor process. By way of comparison,
the ester group of phosphonates which are known to un-
dergo this process readily[15] exhibited almost complete
transesterification.[7,13]
Scheme 4.
Aside from the observed transesterification these reac-
tions proved to be remarkably clean; this is indicative of the
chemoselectivity of the reduction reaction. Only the alkene
intermediate undergoes hydrogenation; benzyl esters, cyano
groups and all the carboxylic acid derivatives proved to be
inert to hydrogenation under these conditions.[20] The reac-
tion with the tert-butyl ester ylide (Entry 3) proved to be
an exception. Unsurprisingly, under the reaction conditions
the tert-butyl ester underwent some thermal deprotection,
resulting in (E)-cinnamic acid and dihydrocinnamic acid
formation.[13,21] During these studies we also identified a
novel fragmentation of (triphenylphosphoranylidene)suc- Scheme 5.
Table 1. Synthesis of dihydrocinnamate derivatives.[a]
Entry
Ylide
R
Product
Conv.[b]
[%]
6
[%]
7
[%]
7b[c]
[%]
8b[c]
[%]
8
[%]
Yield
[%][d]
1
2
3
4
5
6
OMe
OBn
OtBu
8a
8b
8c
8d
8e
8f
100
100
100[e]
96
100
99
0
2
0
4
0
21
18
15
20
22
11
3
–
3
2
4
2
7
–
10
4
6
2
68
80
63
71
67
75
51
71
45
OCH2tBu
NMe2
NMe(OMe)
54
32[f]
47
11
[a] Reactions were carried out on 2.00-mmol scale in toluene (6 mL). [b] Total conversion of benzyl alcohol (5) into compounds 6, 7
and 8 as determined by 1H NMR spectroscopy. [c] Conversion into transesterification products (E)-benzyl cinnamate (7b) and benzyl
dihydrocinnamate (8b). [d] Yield of isolated product after flash column chromatography and von Rudloff oxidative workup.[7,13,19] [e] The
tert-butyl deprotection products (E)-cinnamic acid (1%) and dihydrocinnamic acid (8%) were also obtained. [f] Compound 8e decomposed
on silica; the product was therefore isolated by bulb-to-bulb distillation.
Eur. J. Org. Chem. 2006, 4367–4378
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4369

P. J. Black, M. G. Edwards, J. M. J. Williams
FULL PAPER
Table 2. Results of transesterification experiments.[a]
cantly, purification was facile; in all cases the product could
be obtained by simple column chromatography without the
need for oxidative workup. The yields obtained from these
reactions were in the range 31–66%, similar to those ob-
tained with the dihydrocinnamate series. The reaction of
simple aryl alcohols (Entries 1–4) proceeded to afford the
dihydrocinnamonitrile analogues 13a–d in 46–66% yield.
Most notably, pyren-1-ylmethanol afforded 3-(pyren-1-yl)-
propionitrile (13c) in 52% yield (Entry 3). The reaction
could also be extended to heteroaryl alcohols (Entries 5 and
6). Indol-3-ylmethanol proved to be a more suitable sub-
strate than furfuryl alcohol. Presumably, in the case of fur-
furyl alcohol the reaction is retarded by chelation of the
substrate to the metal centre, a structural impossiblity in
the case of indol-3-ylmethanol. Attempts to use ferrocenyl-
methanol (Entry 7) proved entirely unsuccessful. This reac-
tion converted all of the alcohol substrate into the ether
product bis(ferrocen-1-ylmethyl) ether 17; an 89% isolated
yield of the ether being obtained.
Entry Cs2CO3 [Ir(COD)Cl]2
dppp
Transesterification ratio[b]
[mol-%]
[mol-%]
[mol-%]
[8a/8b]
1
2
3
–
5
5
–
–
5
–
–
5
100:0
17:83
59:41
[a] Reactions were carried out on 1.00-mmol scale in toluene
(3 mL). [b] Ratio of methyl dihydrocinnamate (8a) to benzyl dihy-
1
drocinnamate (8b) as determined by H NMR analysis.
In view of the successful application of this domino pro-
cess towards the synthesis of dihydrocinnamate derivatives,
we wished to expand the methodology to a range of
alcohols. We reasoned that this was unlikely with phos-
phorane ester ylides for two reasons: (i) transesterification
issues and (ii) the tedious workup procedures required. In
order to circumvent these issues we elected to remove the
possibility of transesterification with the alternative stabi-
lised ylide (triphenylphosphoranylidene)acetonitrile (11) as
the olefinating reagent. We hoped the product(s) from these
reactions would be amenable to purification by simple col-
umn chromatography; this we supposed would also improve
the product yields.
The initial reaction with the cyano ylide 11 (Scheme 6,
Ar = Ph) proved entirely successful, 80% conversion into
the product dihydrocinnamonitrile (13a) being observed.
The identification of 8% of dibenzyl ether was surprising,
but can be attributed to the alcohol-promoted degradation
of the ylide (vide infra). Most importantly, under the reac-
tion conditions there was no evidence for any competing
Pinner[24] or Ritter[25] reactions with the cyano function-
ality.
In all of the domino experiments it is evident that there
is generally a 10–15% gap between the conversion into the
product and the isolated yield obtained. This phenomenon
was puzzling until work upon alternative alcohols with the
cyano ylide 11 established that two distinct paths for hydro-
gen/substrate loss existed. From the reaction with naphtha-
lene-2-ylmethanol, an inseparable 1.33:1 mixture of 2-meth-
ylnaphthalene (18) (7%) and naphthalene (19) (5%) was
isolated. These components were identified upon the basis
1
of H NMR, 13C NMR and GC-MS data. The isolation of
these two components indicated that material can be lost
by either a C–OH hydrogenolysis pathway or a decarbon-
ylation pathway (Scheme 7). Hydrogenolysis of the C–OH
bond removes hydrogen from the system, generating 2-
methylnaphthalene (18) and water; this process is well
known for heterogeneous systems.[26] The alternative path-
way, decarbonylation removes alcohol from the system. The
2-naphthaldehyde intermediate formed by dehydrogenation
is decarbonylated by the metal catalyst to yield naphthalene
(19) and a carbonyl iridium complex.[27] Decarbonylation
is well known for systems with ruthenium and particularly
rhodium,[28] but it is less common for iridium. Hydrogen/
substrate loss products were also isolated for additional ex-
amples (pyren-1-ylmethanol, indol-3-ylmethanol and un-
decan-1-ol all afforded such products – see supporting in-
formation; for supp. inf. see also the footnote on the first
page of this article) thus it appears likely that this is a gene-
ral cause of the loss of hydrogen and substrate.
The presence of ether in all of these reactions results
from a process which has been known since the 1960s. In
1960 Grayson and Keough[23] reported that the alcohols 20
reacted with the phosphorane ylides 21 to afford the sym-
Scheme 6.
The application of cyano ylide 11 to a range of activated metric ether 22, alkane 23 and phosphane oxide 24 as prod-
alcohols was then examined. Scheme 6 and Table 3 illus- ucts (Scheme 8). Somewhat surprisingly ether formation is
trate the substituted propionitriles obtained under these observed only in the reactions with the cyano ylide 11; no
conditions.
ether product or acetate ester 23 was detected in any of the
These reactions afforded in all cases (in addition to the reactions with phosphorane ester ylides. During experi-
desired product) mixtures of the aldehyde, alkene, and ether ments to probe the extent of ylide degradation, we discov-
derived from two molecules of alcohol substrate. Signifi- ered an alternate un-catalysed pathway for formation of the
4370
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4367–4378

Borrowing Hydrogen: Indirect “Wittig” Olefination for C–C Bonds from Alcohols
FULL PAPER
Table 3. Synthesis of substituted propionitriles.[a]
[a] Reactions were carried out on 2.00-mmol scale in toluene (6 mL). [b] Total conversion of alcohol 12 into compounds 13, 14, 15 and
1
16 as determined by H NMR analysis. [c] Yield of isolated product after chromatographic purification. [d] Reaction was performed on
a 1.5-mmol scale. [e] An 89% yield of bis(ferrocen-1-ylmethyl) ether (17) was obtained following chromatography.
Scheme 7.
C–C bond resulting from direct reaction of the alcohol with a separate report in parallel with our development of an
the ylide. This pathway proceeds only to a small extent with uncatalysed Wittig alkylation.
the cyano ylide; for the ester ylides only trace amounts of
At the outset of this project, we envisaged that a success-
product were observed. Subsequent work established that ful domino process would enable the use of a wide range of
this was a process which proceeded only at high tempera- alcohols and allow broad access to alkane products.
ture and in the absence of possible alternate reactions; at Towards this goal, we decided to examine the full scope of
110 °C, and at lower temperatures, no reaction was ob- this reaction in regard to the identity of the alcohol sub-
served. It is likely that this pathway contributes only mini- strate and the phosphorane ylide. To glean the maximum
mally towards the product obtained from the domino reac- possible information from these experiments we elected to
tions. We intend to publish our findings in this respect in use the cyano ylide 11 for the alternate alcohol runs and the
Eur. J. Org. Chem. 2006, 4367–4378
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4371

P. J. Black, M. G. Edwards, J. M. J. Williams
FULL PAPER
Scheme 8.
benzyl alcohol (5) for the alternate ylide runs. The results
obtained under these conditions are illustrated in Scheme 9
and Table 4.
Unsurprisingly, aliphatic alcohol undecan-1-ol (Entry 1)
proved to be somewhat unreactive, and significant amounts
of unreacted alcohol remained; however tridecanenitrile
Scheme 9.
(26a) was isolated in 32% yield. This is consistent with the discriminatory reduction suggests that allylic alcohols will
lower oxidation potential of aliphatic alcohols.[29] Allylic probably not have any future synthetic application in this
alcohols proved to be poor substrates, the reactions generat- chemistry.
ing extremely complex mixtures. Presumably this is ex-
We examined two complementary approaches for the
plained by the facile isomerisation of allylic alcohols;[30] it construction of tri-substituted alkanes. Firstly, the use of a
is also possible that π-allyliridium complexes are culpable. secondary alcohol substrate, ( )-phenethyl alcohol (Entry
Careful purification of the reaction mixture from cinnamyl 3), resulted only in a modest, albeit unsurprising, 14% yield
alcohol[31] (Entry 2) afforded 15% of the expected alkane of 3-phenylbutyronitrile (26c). The sluggish performance of
product (E)-5-phenyl-4-pentenenitrile (26b), and 26% of the ketones in Wittig olefination is well known, and many dif-
alternative alkene reduction product 5-phenyl-2-penteneni- ferent approaches have been examined to overcome this
trile [as a mixture of (E)/(Z) isomers].[32] The observed non- problem.[33] In contrast, when the α-substituted ylide ethyl
Table 4. Examination of substrate scope.[a]
[a] Reactions were carried out on 2.00-mmol scale in toluene (6 mL). [b] Total conversion of alcohol 25 into compounds 26, 27, and 28
1
as determined by H NMR spectroscopy. [c] Conversion of alcohol 25 into components 27 and 28 respectively. [d] Conversion of alcohol
25 into product 26 and yield of isolated product after chromatography in parenthesis. [e] Conversions could not be determined due to
the complex nature of the crude reaction mixture. [f] 26% Of a 1.15:1 (E)/(Z) mixture of 5-phenyl-2-pentenenitrile was also obtained.
[g] 3% Of (E)-benzyl α-methylcinnamate and 4% of benzyl α-methylhydrocinnamate was also present. [h] No identifiable components
were obtained from the reaction. [i] The phosphorane ylide was pre-formed from 1 equiv. of phosphonium halide and 1 equiv. of n-
butyllithium (PhMe, 25 °C, 0.5 hour).
4372
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4367–4378

Borrowing Hydrogen: Indirect “Wittig” Olefination for C–C Bonds from Alcohols
FULL PAPER
Table 5. Examination of alternative catalysts.
Entry[a]
Catalyst
Conv.[b]
[%]
14
[%]
15
[%]
13
[%]
Yield[c]
[%]
1
2[d,e]
3[e]
4
[Ru(η6-p-cymene)(S,S-TsDPEN)]
[Ru(PCy3)2Cl2(=CHPh)]
[Ir(COD)(Py)(PCy3)]PF6
RhCl(PPh3)3
100
100
90
12
13
0
4
82
73
76
54
48
43
31
0
14
13
46
100
0
[a] Reactions were carried out on 2.00-mmol scale in toluene (6 mL) with 5 mol-% of the indicated catalyst. [b] Total conversion of benzyl
1
alcohol (5) into compounds 13, 14 and 15 as determined by H NMR spectroscopy. [c] Yield of isolated product after chromatographic
purification. [d] Reaction was performed on a 0.376-mmol scale. [e] 5 mol-% of Cs2CO3 was also added.
2-(triphenylphosphoranylidene)propionate was used (Entry the case of Wilkinson’s catalyst the intermediate aldehyde
4) ethyl α-methylhydrocinnamate (26d) was isolated in 46% is presumably decarbonylated before any olefination can oc-
yield; small amounts of the transesterification products cur; the efficacy of rhodium catalysts for this process is well
were also obtained. Thus, this is a much more feasible ap- known.[28] These results were ultimately to lead to the devel-
proach to tri-substituted alkanes. We expected that the use opment of a highly active ruthenium N-heterocyclic car-
of a ketone ylide would lead to chemoselectivity problems bene catalyst for the domino indirect Wittig process; the
due to competitive ketone reduction[14] (Entry 5). However, results of which we have recently disclosed.[9]
we were pleased to observe a clean reaction with no evi-
dence for any reduction of the ketone. Despite the high con-
version observed, only 16% of the product 26e could be
obtained after purification; this presumably indicates that
Conclusions
alternative hydrogen loss pathways are operative here. In
view of this result, we attempted the even more challenging
reaction with a pyruvate-derived ylide (Entry 6); unfortu-
nately, only decomposition resulted.
In summary, we have described the successful develop-
ment of a first-generation system for the indirect formation
of C–C bonds with alcohols. Through the application of
our borrowing hydrogen concept, we were able to couple
(in a domino sequence) a Wittig olefination reaction with a
crossover transfer hydrogenation. The concept of borrowing
hydrogen, which we have further developed in the interim
period, is under further investigation within these laborato-
ries.
The final aspect of reactivity to be explored was the use
of unstabilised and semi-stabilised phosphorane ylides. The
use of these air- and moisture-sensitive ylides required in
situ ylide formation. This was accomplished by pre-forma-
tion of the required ylide in toluene solution from phospho-
nium halide and n-butyllithium; the addition of the remain-
der of the reagents then enabled the domino reaction to
proceed. Unstabilised ylides are reported to undergo facile
decomposition in the presence of alcohols.[23,34] We were
therefore delighted to observe that the corresponding dom-
Experimental Section
All reactions were performed under dry argon in oven-dried
ino product could be obtained from such reactions. Al- (150 °C) ACE pressure tubes. Toluene was distilled from sodium
wire, rigorously freeze/thaw degassed and stored in sealed Young’s
ampoules. All phosphorane ylides were freshly prepared prior to
use and used immediately. Although several of these are commer-
cially available, superior results were obtained with freshly prepared
ylides. Flash chromatography was carried out with Davisil LC 60A
silica gel (35–70 micron) purchased from Flurochem; thin-layer
chromatographic detection (abbreviated det.) by the methods given.
NMR spectra were performed in CDCl3 with either a Bruker Av-
ance 300 (300 MHz) or Varian WH-400 (400 MHz) instrument and
recorded at the following frequencies: proton (1H – 300/400 MHz)
though the yield obtained is modest, despite the unop-
timised nature of these experiments, the transformation of
benzyl alcohol (5) into pentylbenzene (26g) in 14% yield[35]
(Entry 8) is remarkable. The preparation of 1,2-diphenyl-
ethane (26f) from the semi-stabilised benzyl ylide was also
achieved in 20% yield (Entry 8). When we attempted in situ
ylide generation with KOtBu, inferior results were obtained.
From the results presented above it is evident that a more
active catalyst for the crossover hydrogenation was re-
quired.[36] The high temperatures and extended reaction and carbon (13C – 75.4/100.5 MHz). IR spectra were recorded with
a Perkin–Elmer 1600 series FT-IR spectrophotometer (with in-
ternal background scan) in the range 600–4000 cm–1. Mass spectra,
including high-resolution spectra, were recorded with a Micromass
Autospec spectrometer or a Finnigan MAT 8340 spectrometer
using electron impact (EI+) ionisation, chemical ionisation (CI+
using isobutane or ammonia) and/or Fast Atom Bombardment
(FAB+) ionisation. Melting points were measured with a Büchi 535
series instrument and are uncorrected. GC-MS was performed with
a Fisons series 8000GC/Finnigan MAT 8340 spectrometer setup
using an achiral 5% PH-ME siloxane, 20 m×0.2 mm, column. n-
times led also to the dual problems of transesterification
and hydrogen loss. We were convinced that if a more active
catalyst could be identified, then these problems could be
eliminated and higher yields achieved. A preliminary step
towards this goal was the examination of a number of alter-
native transfer-hydrogenation catalysts in the domino reac-
tion between benzyl alcohol (5) and the cyano ylide 11
(Scheme 6, Table 5, Ar = Ph). Unfortunately, none of these
catalysts proved to be superior to the original iridium sys-
tem, although some product was obtained (Entries 1–3). In Butyllithium was purchased from Acros organics, stored in a
Eur. J. Org. Chem. 2006, 4367–4378
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4373

P. J. Black, M. G. Edwards, J. M. J. Williams
FULL PAPER
Young’s ampoule and titrated according to the procedure of Suf-
CH2CO2Me). 13C NMR (100.5 MHz, CDCl3, 25 °C, ppm): δ =
173.5 (C=O), 140.7, 128.7, 128.4, 126.5, 52.0 (OCH3), 36.1
fert[37] before use.
(CH CO Me), 31.4 (PhCH ). IR (liquid film): ν = 3086, 3025,
˜
2
2
2
General Procedure 1: Preparation of Hydrocinnamate Derivatives:
To an argon-purged ACE pressure tube containing [Ir(COD)-
Cl]2 (67 mg, 0.1 mmol, 0.05 equiv.), dppp (41 mg, 0.1 mmol,
0.05 equiv.), caesium carbonate (33 mg, 0.1 mmol, 0.05 equiv.) and
the phosphorane ylide (2.2 mmol, 1.1 equiv.) was added benzyl
alcohol (5) (216 mg, 2.0 mmol, 197 µL, 1.0 equiv.) followed by de-
gassed anhydrous toluene (6.0 mL). The tube was sealed, stirred
vigorously and heated at 150 °C for 72 hours. The cooled reaction
mixture was quenched by the addition of wet diethyl ether (50 mL)
and concentrated in vacuo to afford the crude product. Conversion
was determined by analysis of the 1H NMR spectrum. Pre-adsorp-
tion of the crude mixture on silica and purification by flash column
chromatography [SiO2, petroleum ether (b.p. 40–60 °C)/diethyl
ether eluent] afforded an inseparable mixture of the dihydrocinna-
mate product and cinnamate intermediate. This mixture was sus-
pended in 50 mL of a 3:2 water/tert-butyl alcohol solution, and
treated with potassium permanganate (20 mg), sodium metaperiod-
ate (1.625 g) and potassium carbonate (125 mg) in a single portion.
The resulting suspension was stirred for 2 hours at room tempera-
ture and then poured into water (50 mL). This mixture was ex-
tracted with diethyl ether (3×20 mL), the combined organic ex-
tracts washed with 1  sodium hydroxide (2×25 mL), saturated
brine (50 mL), dried (MgSO4), filtered and concentrated in vacuo
to yield the desired alkane product.
2951, 1734 (C=O), 1603, 1496, 1452, 1161, 1078 cm–1. MS (EI+ =
70 eV): m/z (%) = 164 (39) [M·+], 133 (12), 105 (36), 104 (100), 91
(62), 77 (13). HRMS (EI+ = 70 eV): C10H12O2 requires 164.0837,
found 164.0838.
Benzyl Dihydrocinnamate (8b): According to general procedure 2
using benzyl (triphenylphosphoranylidene)acetate (9) (903 mg,
2.2 mmol, 1.1 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl
ether (40:1) as the eluent afforded the title compound 8b as a
colourless oil (341 mg, 80% conversion, 71% yield). Rf = 0.73 (pe-
troleum ether (b.p. 40–60 °C)/diethyl ether, 7:3, det. with p-anisal-
dehyde [red]). 1H NMR (400 MHz, CDCl3, 25 °C, ppm): δ = 7.24–
7.37 (m, 7 H, ArH), 7.16–7.21 (m, 3 H, ArH), 5.10 (s, 2 H,
OCH2Ph), 2.96 (t, J = 7.4 Hz, 2 H, CH2CO2Bn), 2.67 (t, J =
7.4 Hz, 2 H, PhCH2). 13C NMR (100 MHz, CDCl3, 25 °C, ppm):
δ = 172.8 (C=O), 140.6, 136.1, 128.8, 128.7, 128.5, 128.5, 126.5,
66.7 (OCH2Ph), 36.3 (CH2CO2Bn), 31.4 (PhCH2). IR (liquid film):
ν = 3061, 3026, 2948, 1735 (C=O), 1603, 1495, 1453, 1159,
˜
1077 cm–1. MS (FAB+): m/z = 240 [M·+]. HRMS (FAB+):
C16H16O2 requires 240.1150, found 240.1139.
tert-Butyl Dihydrocinnamate (8c): According to general procedure
2 using tert-butyl (triphenylphosphoranylidene)acetate (828 mg,
2.2 mmol, 1.1 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl
ether (40:1) as the eluent afforded the title compound 8c as a
colourless oil (200 mg, 63% conversion, 49% yield). Rf = 0.54 (7:3
petroleum ether (b.p. 40–60 °C)/diethyl ether, det. with p-anisal-
dehyde [black]); 1H NMR (300 MHz, CDCl3, 25 °C, ppm): δ =
7.19–7.23 (m, 2 H, ArH), 7.08–7.12 (m, 3 H, ArH), 2.83 (t, J =
7.1 Hz, 2 H, PhCH2), 2.46 (t, J = 7.1 Hz, 2 H, CH2CO2tBu), 1.34
[s, 9 H, OC(CH3)3]. 13C NMR (75.4 MHz, CDCl3, 25 °C, ppm): δ
= 171.2 (C=O), 139.8, 127.4, 127.3, 125.1, 79.3 [OC(CH3)3], 36.1
General Procedure 2: Preparation of Substituted Propionitriles: To
an argon-purged ACE pressure tube containing [Ir(COD)Cl]2
(67 mg, 0.1 mmol, 0.05 equiv.), dppp (41 mg, 0.1 mmol,
0.05 equiv.), caesium carbonate (33 mg, 0.1 mmol, 0.05 equiv.) and
(triphenylphosphoranylidene)acetonitrile (11) (663 mg, 2.2 mmol,
1.1 equiv.) was added the required alcohol (2.0 mmol, 1.0 equiv.)
followed by degassed anhydrous toluene (6.0 mL). The tube was
sealed, stirred vigorously and heated at 150 °C for 72 hours. The
cooled reaction mixture was quenched by the addition of wet di-
ethyl ether (50 mL) and concentrated in vacuo to afford the crude
(CH CO tBu), 30.1 (PhCH , 27.0 [OC(CH ) ]. IR (liquid film): ν
˜
2
2
2)
3 3
= 3086, 3063, 2976, 2930, 1729, 1604, 1496, 1453, 1146, 1078, 846,
743, 698 cm–1. MS (EI+ = 70 eV): m/z (%) = 206 (1.5) [M·+], 150
(82), 133 (37), 105 (49), 104 (52), 91 (60), 57 (100). HRMS (EI+ =
70 eV): C13H18O2 requires 206.1305, found 206.1306.
1
product. Conversion was determined by analysis of the H NMR
spectrum. Pre-adsorption of the crude mixture on silica and purifi-
cation by flash column chromatography [SiO2, petroleum ether
(b.p. 40–60 °C)/diethyl ether eluent] afforded the substituted pro-
pionitrile product.
Neopentyl Dihydrocinnamate (8d): According to general procedure
2 using neopentyl (triphenylphosphoranylidene)acetate (859 mg,
2.2 mmol, 1.1 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl
ether (100:1) as the eluent afforded the title compound 8d as a
colourless oil (237 mg, 71% conversion, 54% yield). Rf = 0.76 (7:3
petroleum ether (b.p. 40–60 °C)/diethyl ether, det. with p-anisal-
dehyde [black]). 1H NMR (300 MHz, CDCl3, 25 °C, ppm): δ =
7.11–7.20 (m, 5 H, ArH). 3.69 (s, 2 H, OCH2tBu), 2.89 (t, J =
7.5 Hz, 2 H, PhCH2), 2.58 (t, J = 7.5 Hz, 2 H, CH2CO2CH2tBu),
0.83 [s, 9 H, C(CH3)3]; 13C NMR (75.4 MHz, CDCl3, 25 °C, ppm):
δ = 173.2 (C=O), 140.7, 128.6, 128.4, 126.4, 73.9 (OCH2tBu), 36.0
(CH2CO2CH2tBu), 31.4 [CH2C(CH3)3], 31.2 (PhCH2), 26.5
General Procedure 3: Domino Indirect Wittig Reactions with Unsta-
bilised Ylides: To an argon-purged ACE pressure tube containing
a solution of the required phosphonium halide salt (2.20 mmol,
1.1 equiv.) in anhydrous toluene (6 mL) was added dropwise a solu-
tion of n-butyllithium [2.27  in hexanes] (141 mg, 2.20 mmol,
1.1 equiv., 969 µL), and the mixture was stirred for 30 minutes at
room temperature to form the phosphorane ylide.[38] [Ir(COD)Cl]2
(67 mg, 0.1 mmol, 0.05 equiv.), dppp (41 mg, 0.1 mmol, 0.05 equiv.)
and caesium carbonate (33 mg, 0.1 mmol, 0.05 equiv.) were added
in a single portion followed by benzyl alcohol (5) (216 mg,
2.0 mmol, 197 µL, 1.0 equiv.) The reaction was then conducted in
agreement with general procedure 1.
[C(CH ) ]. IR (liquid film): ν = 3063, 3028, 2959, 2869, 1732, 1604,
˜
3 3
1497, 1478, 1456, 1377, 1366, 1291, 1244, 1160, 1078, 1014, 749,
694 cm–1. MS (EI+ = 70 eV): m/z (%) = 220 (57) [M·+], 205 (7), 150
(61), 133 (47), 105 (77), 104 (69), 91 (84), 71 (100), 57 (25), 43
(47). HRMS (EI+ = 70 eV): C14H20O2 requires 220.1463, found
220.1464.
Methyl Dihydrocinnamate (8a): According to general procedure 2
using methyl (triphenylphosphoranylidene)acetate (10) (735 mg,
2.2 mmol, 1.1 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl
ether (40:1) as the eluent afforded the title compound 8a as a
colourless oil (168 mg, 68% conversion, 51% yield). Rf = 0.67 (pe-
troleum ether (b.p. 40–60 °C)/diethyl ether, 7:3, det. with p-anisal-
dehyde [orange]). 1H NMR (400 MHz, CDCl3, 25 °C, ppm): δ =
7.26–7.30 (m, 2 H, ArH), 7.18–7.22 (m, 3 H, ArH), 3.66 (s, 3 H,
OCH3), 2.95 (t, J = 8.2 Hz, 2 H, PhCH2), 2.63 (t, J = 8.2 Hz, 2 H,
N,N-Dimethyl-3-phenylpropionamide (8e): The reaction was per-
formed according to general procedure 2 using N,N-dimethyl-
(triphenylphosphoranylidene)acetamide
(764 mg,
2.2 mmol,
1.1 equiv.). Purification was achieved by bulb-to-bulb distillation
(85 °C, 1 Torr) to yield a mixture of the desired product and trans-
esterification product benzyl dihydrocinnamate. The product
4374
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4367–4378

Borrowing Hydrogen: Indirect “Wittig” Olefination for C–C Bonds from Alcohols
FULL PAPER
proved to be unstable to silica, undergoing decomposition. Thus
the mixture was dissolved in diethyl ether (15 mL), palladium on
carbon [10 w/w%] (50 mg, 0.047 mmol) added and the mixture
placed under 1 atm of hydrogen. The reaction mixture was stirred
for 12 hours, diluted with diethyl ether (35 mL) and then filtered
through a plug of celite. The filtrate was washed with 1  sodium
hydroxide (50 mL), saturated brine (50 mL), dried (MgSO4), fil-
tered and concentrated in vacuo to yield the title compound, 8e, a
colourless liquid (106 mg, 67% conversion, 32% yield). B.p. 85 °C/
1 Torr (ABT). 1H NMR (300 MHz, CDCl3, 25 °C, ppm): δ = 7.19–
7.24 (m, 2 H, ArH), 7.11–7.16 (m, 3 H, ArH), 2.87 [s, 3 H,
N(CH3)2], 2.85 [s, 3 H, N(CH3)2], 2.85 (partially obscured t, J =
25 °C, ppm): δ = 135.6, 133.6, 132.6, 127.8, 127.8, 126.9, 126.5,
126.4, 126.0, 119.3 (CϵN), 31.8 (ArCH2), 19.4 (CH2CN). IR
(KBr): ν = 3054, 2954, 2926, 2885, 2243 (CϵN), 1597, 1507, 1448,
˜
1422, 1368, 1273, 1202, 1126, 965, 951, 918, 900, 863, 828, 810,
749, 653 cm–1. MS (EI+ = 70 eV): m/z (%) = 181 (33) [M·+], 141
(100), 115 (15). HRMS (EI+ = 70 eV): C13H11N requires 181.0891,
found 181.0892.
3-(Pyrene-1-yl)propionitrile (13c): According to General procedure
3 using pyren-1-ylmethanol (345 mg, 1.5 mmol, 1.0 equiv.), (tri-
phenylphosphoranylidene)acetonitrile (11) (497 mg, 1.65 mmol,
1.1 equiv.), toluene (4.5 mL) and petroleum ether (b.p. 40–60 °C)/
diethyl ether (30:1 Ǟ 7:3) as the eluent afforded the title compound
13c as a yellow solid, which was subsequently recrystallised from
dichloromethane/hexane to afford pale yellow microcrystals
(199 mg, 60% conversion, 52% yield). M.p. 132–134 °C; Rf = 0.23
(petroleum ether (b.p. 40–60 °C)/diethyl ether (7:3), det. with p-ani-
saldehyde [mauve]). 1H NMR (400 MHz, CDCl3, 25 °C, ppm): δ =
8.16 (d, J = 7.4 Hz, 2 H, ArH), 8.10 (m, 3 H, ArH), 8.02 (s, 1 H,
ArH), 8.01 (s, 1 H, ArH), 7.98 (d, J = 7.5 Hz, 1 H, ArH), 7.85 (d,
J = 7.8 Hz, 1 H, ArH), 3.64 (t, J = 7.4 Hz, 2 H, ArCH2), 2.80 (t,
J = 7.4 Hz, 2 H, CH2CN). 13C NMR (75.4 MHz, CDCl3, 25 °C,
ppm): δ = 132.0, 131.7, 131.2, 131.1, 128.8, 128.6, 127.8, 127.7,
127.5, 126.5, 125.9, 125.6, 125.5, 125.4, 125.2, 122.4, 119.6 (CϵN),
7.5 Hz, 2 H, PhCH2), 2.54 (t, J = 7.5 Hz, 2 H, CH2CONMe2). 13
C
NMR (75.4 MHz, CDCl3, 25 °C, ppm): δ = 172.3 (C=O), 141.6,
128.6, 128.5, 126.1, 37.3 [N(CH3)2], 35.5 [N(CH3)2], 35.4
(CH CONMe ), 31.5 (PhCH ). IR (liquid film): ν = 3060, 3026,
˜
2
2
2
2931, 1646 (C=O), 1495, 1453, 1398, 1345, 1266, 1141, 1076, 753,
700 cm–1. MS (EI+ = 70 eV): m/z (%) = 177 (100) [M·+], 133 (8),
105 (46), 91 (64), 77 (24), 72 (42), 45 (39). HRMS (EI+, 70eV):
C11H15NO requires 177.1153, found 177.1157.
N-Methoxy-N-methyl-3-phenylpropionamide (8f): According to Ge-
neral procedure 2 using N-methoxy-N-methyl(triphenylphosphor-
anylidene)acetamide (799 mg, 2.2 mmol, 1.1 equiv.) and petroleum
ether (b.p. 40–60 °C)/diethyl ether (9:1 Ǟ 7:3) as the eluent afforded
the title compound 8f as a pale yellow oil (164 mg, 75% conversion,
47% yield). Rf = 0.13 (petroleum ether (b.p. 40–60 °C)/diethyl ether
(7:3), det. with p-anisaldehyde [orange]). 1H NMR (300 MHz,
CDCl3, 25 °C, ppm): δ = 7.20–7.32 (s, 5 H, ArH), 3.61 [s, 3 H,
NMe(OCH3)], 3.18 [s, 3 H, NCH3(OMe)], 2.97 (t, J = 7.5 Hz, 2 H,
PhCH2), 2.74 [t, J = 7.5 Hz, 2 H, CH2CONMe(OMe)]. 13C NMR
(300 MHz, CDCl3, 25 °C, ppm): δ = 173.8 (C=O), 141.4, 128.6,
128.5, 126.2, 61.3 (OCH3), 33.9 [CH2CONMe(OMe)], 32.3
29.6 (ArCH ), 19.7 (CH CN). IR (KBr): ν = 3041, 2927, 2241
˜
2
2
(CϵN), 1604, 1586, 1466, 1457, 1431, 1418, 1318, 1248, 1187, 1098,
841, 755, 712 cm–1. MS (EI+ = 70 eV): m/z (%) = 255 (27) [M·+],
220 (27), 215 (97), 213 (74), 205 (100), 105 (36), 71 (33), 57 (74),
43 (54). HRMS (EI+ = 70 eV): C13H13N requires 255.1048, found
255.1051.
3-(p-Chlorophenyl)propionitrile (13d): According to general pro-
cedure
3 using p-chlorobenzyl alcohol (285 mg, 2.0 mmol,
1.0 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl ether (30:1
Ǟ 9:1) as the eluent afforded the title compound 13d as a pale
yellow oil (217 mg, 83% conversion, 66% yield). Rf = 0.28 (petro-
leum ether (b.p. 40–60 °C)/diethyl ether (7:3), det. with p-anisal-
(NCH OMe), 30.8 (PhCH ). IR (liquid film): ν = 3061, 3026, 2936,
˜
3
2
1690, 1603, 1587, 1495, 1454, 1385, 1340, 1178, 1103, 1075, 1029,
989, 940, 751, 695 cm–1. MS (EI+ = 70 eV): m/z (%) = 193 (44)
[M·+], 133 (21), 105 (95), 91 (100), 77 (20), 61 (29). HRMS (EI+ =
70 eV): C11H15NO2 requires 193.1103, found 193.1104.
1
dehyde [weak yellow]). H NMR (300 MHz, CDCl3, 25 °C, ppm):
δ = 7.31 (d, J = 8.3 Hz, 2 H, ortho-ArH), 7.17 (d, J = 8.3 Hz, 2 H,
meta-ArH), 2.92 (t, J = 7.3 Hz, 2 H, ArCH2), 2.60 (t, J = 7.3 Hz,
2 H, CH2CN). 13C NMR (75.4 MHz, CDCl3, 25 °C, ppm): δ =
136.5 (Cipso), 133.3 (Cpara), 130.0 (Cortho), 129.2 (Cmeta), 119.0
Dihydrocinnamonitrile (13a): According to general procedure 3
using benzyl alcohol (5) (216 mg, 2.0 mmol, 197 µL, 1.0 equiv.) and
petroleum ether (b.p. 40–60 °C)/diethyl ether (30:1) as the eluent
afforded the title compound 13a as a colourless oil (141 mg, 71%
conversion, 56% yield). 13a: Rf = 0.36 (petroleum ether (b.p. 40–
60 °C)/diethyl ether (7:3), det. with p-anisaldehyde [orange]) 1H
NMR (300 MHz, CDCl3, 25 °C, ppm): δ = 2.64 (t, J = 7.4 Hz, 2
H, CH2CN), 2.98 (t, J = 7.4 Hz, 2 H, PhCH2), 7.22–7.39 (m, 5 H,
ArH). 13C NMR (75.4 MHz, CDCl3, 25 °C, ppm): δ = 138.5, 129.3,
128.7, 127.6, 119.6 (CϵN), 32.0 (PhCH2), 19.8 (CH2CN). IR (li-
(CϵN), 31.0 (ArCH ), 19.5 (CH CN). IR (liquid film): ν = 3029,
˜
2
2
2932, 2869, 2246 (CϵN), 1598, 1493, 1427, 1424, 1408, 1092, 1015,
921, 832, 807, 776, 714 cm–1. MS (EI+ = 70 eV): m/z (%) = 167/165
(7, 21) [M·+], 127/125 (33, 100). HRMS (EI+ = 70 eV): C9H8N37Cl
requires 165.0345, found 165.0349.
3-(Furan-2-yl)propionitrile (13e): According to General procedure 3
using furfuryl alcohol (196 mg, 2.0 mmol, 174 µL, 1.0 equiv.) and
petroleum ether (b.p. 40–60 °C)/diethyl ether (30:1) as the eluent
afforded the title compound 13e as a pale yellow liquid (76 mg,
61% conversion, 31% yield). Rf = 0.44 (petroleum ether (b.p. 40–
quid film): ν = 3063, 3029, 2931, 2246 (CϵN), 1603, 1496, 1454,
˜
1079, 1030, 748, 699 cm–1. MS (EI+ = 70 eV): m/z (%) = 131 (22)
[M·+], 91 (100). HRMS (EI+ = 70 eV): C9H9N requires 131.0735,
found 131.0733.
1
60 °C)/diethyl ether (7:3), det. with p-anisaldehyde [turquoise]). H
NMR (400 MHz, CDCl3, 25 °C, ppm): δ = 7.26 (d, J = 1.5 Hz, 1
H, ArH), 6.23 (app. t, J = 2.7 Hz, 1 H, ArH), 6.09 (d, J = 3.1 Hz,
1 H, ArH), 2.93 (t, J = 7.4 Hz, 2 H), 2.60 (t, J = 7.4 Hz, 2 H). 13C
NMR (75.4 MHz, CDCl3, 25 °C, ppm): δ = 151.4, 142.2, 118.8
3-(Naphthalen-2-yl)propionitrile (13b): According to general pro-
cedure
3 using naphthalen-2-ylmethanol (316 mg, 2.0 mmol,
1.0 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl ether (19:1
Ǟ 9:1) as the eluent afforded the title compound 13b as a colourless
oil which solidified as a dense cream solid upon standing (168 mg,
83% conversion, 46% yield). M.p. 72–73 °C; Rf = 0.32 (petroleum
ether (b.p. 40–60 °C)/diethyl ether (7:3), det. with p-anisaldehyde
[red]). 1H NMR (CDCl3, 300 MHz, 25 °C, ppm): δ = 7.83–7.88 (m,
3 H, ArH), 7.72 (br. s, 1 H, ArH), 7.47–7.55 (m, 2 H, ArH), 7.37
(CϵN), 110.6, 106.9, 24.4 (ArCH ), 16.7 (CH CN). IR (CDCl ): ν
˜
2
2
3
= 2954, 2926, 2850, 2255 (CϵN), 1601, 1507, 1440, 1426, 1342,
1261, 1160, 1145, 1077, 1016, 927, 804 cm–1. MS (EI+ = 70 eV):
m/z (%) = 121 (22) [M·+]; 81 (100), 53 (26). HRMS (EI+ = 70 eV):
C7H7NO requires 121.0527, found 121.0528.
(dd, J = 1.5, 8.4 Hz, 1 H, ArH), 3.16 (t, J = 7.4 Hz, 2 H, ArCH2), 3-(Indol-3-yl)propionitrile (13f): According to General procedure 3
2.75 (t, J = 7.4 Hz, 2 H, CH2CN). 13C NMR (75.4 MHz, CDCl3, using indol-3-ylmethanol (294 mg, 2.0 mmol, 1.0 equiv.) and petro-
Eur. J. Org. Chem. 2006, 4367–4378
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4375

P. J. Black, M. G. Edwards, J. M. J. Williams
FULL PAPER
leum ether (b.p. 40–60 °C)/diethyl ether (9:1 Ǟ 1:2) as the eluent
afforded the title compound 13f as a pale brown oil which solidified
as a glassy pale brown solid on standing (170 mg, 82% yield, 50%
yield). M.p. 66–68 °C; Rf = 0.22 (petroleum ether (b.p. 40–60 °C)/
diethyl ether (1:1), det. with p-anisaldehyde [purple]). 1H NMR
(400 MHz, CDCl3, 25 °C, ppm): δ = 8.13 (br. s, 1 H, NH), 7.59 (m,
1 H, ArH), 7.41 (dt, J = 1.1, 8.0 Hz, 1 H, ArH), 7.27 (dt, J = 1.1,
(718 mg, 2.2 mmol, 1.1 equiv.) and petroleum ether (b.p. 40–60 °C)/
diethyl ether (40:1) as the eluent afforded the title compound, 26d,
as a colourless liquid (177 mg, 62% conversion, 46% yield). Rf =
0.57 (petroleum ether (b.p. 40–60 °C)/diethyl ether (7:3), det. with
1
p-anisaldehyde [grey]). H NMR (300 MHz, CDCl3, 25 °C, ppm):
δ = 7.06–7.20 (m, 5 H, ArH), 3.99 (q, J = 7.1 Hz, 2 H, OCH2CH3),
2.85–2.97 [m, 1 H, PhCH2CH(CH3)CO2Et], 2.53–2.68 (m, 2 H,
7.0 Hz, 1 H, ArH), 7.22 (dt, J = 1.1, 7.0 Hz, 1 H, ArH), 7.15 (br. PhCH2), 1.04–1.11 (m, 6 H, CH3). 13C NMR (300 MHz, CDCl3,
d, J = 2.4 Hz, 1 H, ArH), 3.17 (t, J = 7.1 Hz, 2 H, ArCH2), 2.73
(t, J = 7.1 Hz, 2 H, CH2CN). 13C NMR (75.4 MHz, CDCl3, 25 °C, (OCH2CH3), 42.1 [PhCH2CH(CH3)CO2Et], 40.3 (PhCH2), 17.3
ppm): δ = 136.4, 126.7, 122.5, 122.3, 119.9 (CϵN), 119.8, 119.3, [PhCH CH(CH )CO Et], 14.7 (OCH CH ). IR (liquid film): ν =
112.7, 111.6, 21.8 (ArCH ), 18.8 (CH CN). IR (CDCl ): ν = 3478,
25 °C, ppm): δ = 176.7 (C=O), 140.0, 129.0, 128.6, 126.8, 60.8
˜
2
3
2
2
3
3063, 3028, 2977, 2935, 2876, 1732, 1604, 1495, 1454, 1376, 1350,
3061, 2928, 2861, 2248 (CϵN), 1620, 1489, 1457, 1420, 1369, 1336, 1282, 1173, 1117, 1094, 1028, 745, 700 cm–1. MS (EI+ = 70 eV):
1251, 1229, 1094, 1070, 1012 cm–1. MS (EI+ = 70 eV): m/z (%) = m/z (%) = 191 (19) [M·+], 147 (7), 119 (22), 118 (61), 91 (100), 77
170 (24) [M·+]; 135 (27), 130 (100), 92 (91), 77 (15), 65 (15). HRMS (7). HRMS (EI+ = 70 eV): C12H16O2 requires 192.1150, found
˜
2
2
3
(EI+ = 70 eV): C11H10N2O requires 170.0844, found 170.0846.
119.1154.
Tridecanenitrile (26a): According to General procedure 3 using un-
decan-1-ol (345 mg, 2.0 mmol, 416 µL, 1.0 equiv.) and petroleum
(b.p. 40–60 °C)/diethyl ether (40:1) as the eluent afforded the title
compound as a colourless liquid (126 mg, 43% conversion, 32%
yield). Rf = 0.74 (petroleum (b.p. 40–60 °C)/diethyl ether (7:3), det.
with p-anisaldehyde [weak green]). 1H NMR (300 MHz, CDCl3,
25 °C, ppm): δ = 2.32 (t, J = 7.1 Hz, 2 H, CH2CN), 1.64 (quint, J
= 7.1 Hz, 2 H, CH2CH2CN), 1.38–1.45 (m, 2 H, CH2CH2CH2CN),
1.26 (br. s, 16 H), 0.87 (t, J = 7.0 Hz, 3 H, CH3). 13C NMR
(75.4 MHz, CDCl3, 25 °C, ppm): δ = 14.3 (CH3), 17.2, 22.8, 25.5,
28.8, 28.9, 29.4, 29.5, 29.6, 29.7, 32.0 (CH2CN), 119.9 (CϵN). IR
4-Phenylbutan-2-one (26e): In analogy to general procedure 2 using
alcohol 5 (216 mg, 2.0 mmol, 197 µL, 1.0 equiv.), (triphenylphos-
phoranylidene)acetone (700 mg, 2.2 mmol, 1.1 equiv.) and petro-
leum ether (b.p. 40–60 °C)/diethyl ether (40:1 Ǟ 19:1) as the eluent
afforded the title compound 26e as a colourless liquid (46 mg, 87%
conversion, 16% yield). Rf = 0.33 (petroleum ether (b.p. 40–60 °C)/
diethyl ether (7:3), det. with p-anisaldehyde [mauve]). 1H NMR
(300 MHz, CDCl3, 25 °C, ppm): δ = 7.16–7.22 (m, 2 H, ArH),
7.08–7.12 (m, 3 H, ArH), 2.81 (t, J = 7.5 Hz, 2 H, PhCH2), 2.67
[t, J = 7.5 Hz, 2 H, CH2C(O)Me], 2.05 (s, 3 H, CH3). 13C NMR
(74.5 MHz, CDCl3, 25 °C, ppm): δ = 206.9 (C=O), 140.0, 127.5,
127.3, 125.0, 44.1 [CH2C(O)Me], 29.0 (PhCH2), 28.7 (CH3). IR
(liquid film): ν = 292, 2864, 2246 (CϵN), 1465, 1432, 1377,
˜
722 cm–1. MS (EI+ = 70 eV): m/z (%) 166 (17) [(M–Et·)+], 152 (34),
138 (34), 124 (50), 110 (74), 97 (100), 96 (65), 83 (46), 82 (45), 69
(41), 57 (55), 55 (54), 43 (67), 41 (75). MS (FAB+): m/z = 196
[[M+H]+]. HRMS (FAB+): C13H26N [[M+H]+] requires 196.2065,
found 196.2067.
(liquid film): ν = 3061, 3027, 2924, 1717, 1602, 1579, 1496, 1453,
˜
1409, 1356, 1282, 1216, 1161, 1080, 1030, 965, 750, 699 cm–1. MS
(EI+ = 70 eV): m/z (%) = 148 (77) [M·+], 133 (15), 105 (92), 91 (71),
77 (24), 43 (100). HRMS (EI+ = 70 eV): C10H12O requires
148.0888, found 148.0892.
(E)-5-Phenyl-4-pentenenitrile (26b): According to General pro-
cedure 3 using cinnamyl alcohol (268 mg, 2.0 mmol, 1.0 equiv.) and
petroleum ether (b.p. 40–60 °C)/diethyl ether (50:1 Ǟ 40:1 Ǟ 30:1
Ǟ 19:1 Ǟ 9:1 Ǟ 7:3) as the eluent a very complex mixture of
fractions was isolated. From this mixture the title compound 26b
(47 mg, 15%) was isolated as a colourless liquid. Rf = 0.33 [petro-
leum ether (b.p. 40–60 °C)/diethyl ether (7:3), det. with UV]. 1H
NMR (300 MHz, CDCl3, 25 °C, ppm): δ = 7.15–7.30 (m, 5 H,
ArH), 6.44 (d, J = 15.8 Hz, 1 H, PhCH=), 6.12 (dt, J = 6.5,
15.8 Hz, 1 H, PhCH=CHCH2), 2.38–2.50 (m, 4 H). 13C NMR
(75.4 MHz, CDCl3, 25 °C, ppm): δ = 137.0, 133.0, 128.7, 128.7,
126.4, 125.6, 119.6 (CϵN), 29.2 (CH2CH2CN), 18.0 (CH2CN).
3-Phenylbutyronitrile (26c): According to General procedure 3
using ( )-sec-phenethyl alcohol (244 mg, 2.0 mmol, 242 µL,
1.0 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl ether (40:1)
as the eluent afforded the title compound 26c as a colourless oil
(38 mg, 19% conversion, 14% yield). Rf = 0.25 [petroleum ether
(b.p. 40–60 °C)/diethyl ether (7:3), det. with UV]. 1H NMR
(300 MHz, CDCl3, 25 °C, ppm): δ = 7.37–7.42 (m, 2 H, ArH),
7.27–7.34 (m, 3 H, ArH), 3.20 [app. sextet, J = 7.0 Hz, 1 H,
PhCH(CH3)CH2CN], 2.67 (dd, J = 6.4, 16.6 Hz, 1 H, CHaHbCN),
2.58 (dd, J = 7.5, 16.6 Hz, 1 H, CHaHbCN), 1.49 (d, J = 7.0 Hz, 3
H, CH3). 13C NMR (75.4 MHz, CDCl3, 25 C, ppm): δ = 143.2,
128.9, 126.6, 127.4, 118.7 (CϵN), 36.6 [PhCH(CH3)CH2CN], 26.4
1,2-Diphenylethane (26f): According to general procedure 4 using
benzyltriphenylphosphonium chloride (856 mg, 2.20 mmol,
1.1 equiv.) and petroleum ether (b.p. 40–60 °C)/diethyl ether (100:1)
as the eluent afforded the title compound 26f as a colourless oil
that solidified as a cream solid upon standing (73 mg, 20% yield,
40% conversion). M.p. 47–49 °C, Rf = 0.67 [petroleum ether (b.p.
1
40–60 °C)/diethyl ether (7:3), det. with UV]: H NMR (300 MHz,
CDCl3, 25 °C, ppm): δ = 7.33–7.38 (m, 4 H, ArH), 7.23–7.30 (m,
6 H, ArH), 3.00 (s, 4 H, PhCH2). 13C NMR (75.4 MHz, CDCl3,
25 °C, ppm): δ = 141.9, 128.6, 128.4, 126.0, 38.1 (PhCH2). IR
(KBr): ν = 3056, 3025, 2917, 2853, 1599, 1490, 1450, 1184, 1120,
˜
1063, 1026, 750, 721, 697 cm–1. MS (EI+ = 70 eV): m/z (%) = 182
(36) [M·+], 91 (100), 65 (13). HRMS (EI+ = 70 eV): C10H14 requires
182.1096, found 182.1096.
Pentylbenzene (26g): According to general procedure 4 using n-bu-
tyltriphenylphosphonium bromide (876 mg, 2.20 mmol, 1.1 equiv.)
and petroleum ether (b.p. 40–60 °C)/diethyl ether (100:1) as the elu-
ent afforded the a mixture of the title compound (40 mg, 14% yield,
40% conversion) and intermediate pent-1-enylbenzene following
chromatography. Oxidative removal of the alkene by-product also
leads to the destruction of the majority of the alkane product, al-
though small amounts of alkane could be isolated. In order to con-
firm the identity of the product an alternative workup involving
hydrogenation of the alkene (10% w/w Pd-C, 1 atm H2, 2 hours)
was employed. This furnished only the title compound 26g, a
colourless liquid. Rf = 0.76 [petroleum ether (b.p. 40–60 °C)/diethyl
ether (7:3), det. with UV/weak KMnO4]. 1H NMR (300 MHz,
CDCl3, 25 °C, ppm): δ = 7.17–7.22 (m, 2 H, ArH), 7.06–7.11 (m,
3 H, ArH), 2.51 (t, J = 7.6 Hz, 2 H, PhCH2), 1.54 (app. quint, J =
(CH CN), 20.7 (CH ). IR (CDCl ): ν = 3065, 3031, 2968, 2931,
˜
2
3
3
2253 (CϵN), 1602, 1495, 1453, 1422, 1383, 1358, 1015 cm–1. MS
(EI+ = 70 eV): m/z (%) = 145 (17) [M·+], 105 (100), 77 (16). HRMS
(EI+ = 70 eV): C10H11N requires 145.0891, found 145.0891.
Ethyl 2-Methyl-3-phenylpropionate (26d): According to General
procedure 2 using ethyl 2-(triphenylphosphoranylidene)propionate
4376
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4367–4378

Borrowing Hydrogen: Indirect “Wittig” Olefination for C–C Bonds from Alcohols
FULL PAPER
Soc. 2004, 126, 73; d) K. Fujita, T. Fujii, R. Yamaguchi, Org.
Lett. 2004, 6, 3525–3528; e) C. S. Cho, B. T. Kim, H.-S. Kim,
T.-J. Kim, S. C. Shim, Organometallics 2003, 22, 3608–3610; f)
C. S. Cho, B. T. Kim, T.-J. Kim, S. C. Shim, J. Org. Chem. 2001,
66, 9020; g) R. Grigg, T. R. B. Mitchell, S. Sutthivaiyakit, N.
Tongpenyai, Tetrahedron Lett. 1981, 22, 4107.
7.0 Hz, 2 H, PhCH2CH2), 1.23 (m, 4 H, CH2CH2CH3), 0.81 (t, J
= 7.0 Hz, 3 H, CH3). 13C NMR (75.4 MHz, CDCl3, 25 °C, ppm):
δ = 141.1, 128.5, 128.3, 125.7 36.1 (PhCH2), 31.7, 31.4, 22.7
(CH CH ), 14.2 (CH ). IR (liquid film): ν = 3062, 3026, 2956, 2927,
˜
3
2
3
2856, 1604, 1496, 1453, 1377, 1111, 1030, 1026, 744, 697 cm–1. MS
(EI+ = 70 eV): m/z (%) = 148 (54) [M·+], 105 (16), 93 (65), 91 (100).
HRMS (EI+ = 70 eV): C11H16 requires 148.1252, found 148.1249.
[11] See for example: a) K. Fujita, T. Fujii, R. Yamaguchi, Org.
Lett. 2004, 6, 3525–3528; b) K. Fujita, Z. Li, N. Ozeki, R.
Yamaguchi, Tetrahedron Lett. 2003, 44, 2687–2690; c)
R. A. T. M. Abbenhuis, J. Boersma, G. van Koten, J. Org.
Chem. 1998, 63, 4282–4290; d) K. Felfoldi, M. S. Klyavlin, M.
Bartok, J. Organomet. Chem. 1989, 362, 193–195; e) Y. Watan-
abe, Y. Tsuji, H. Ige, Y. Ohsugi, T. Ohta, J. Org. Chem. 1984,
49, 3359; f) S.-I. Murahashi, K. Kondo, T. Hakata, Tetrahedron
Lett. 1982, 23, 229.
[12] a) G. Cami-Kobeci, P. A. Slatford, M. K. Whittlesey, J. M. J.
Williams, Bioorg. Med. Chem. Lett. 2005, 15, 535; b) G. Cami-
Kobeci, J. M. J. Williams, Chem. Commun. 2004, 1072.
[13] P. J. Black, G. Cami Kobeci, M. G. Edwards, P. A. Slatford,
M. K. Whittlesey, J. M. J. Williams, Org. Biomol. Chem. 2006,
4, 116.
Supporting Information (see also the footnote on the first page of
this article): Full characterisation data for all hydrogen/substrate
loss compounds and isolated by-products.
Acknowledgments
The EPSRC (M. G. E), University of Bath and Roche Discovery
(P. J. B.) are thanked for support of this study.
[14] S. Sakaguchi, T. Yamaga, Y. Ishii, J. Org. Chem. 2001, 66, 4710.
[15] Z. Mouloungui, R. Elmestour, M. Delmas, A. Gaset, Tetrahe-
dron 1992, 48, 1219.
[16] G. Facchin, R. Bertani, M. Calligaris, G. Nardin, M. Mari, J.
Chem. Soc. Dalton Trans. 1987, 1381.
[17] P. Lidstrom, J. Tierney, B. Wathey, J. Westman, Tetrahedron
2001, 57, 10229.
[18] X. C. Yu, X. Huang, Synlett 2002, 1895.
[1] See for example: a) L. F. Tietze, Chem. Rev. 1996, 96, 115; b)
L. F. Tietze, J. Heterocycl. Chem. 1990, 27, 47; c) F. Ziegler,
Chem. Rev. 1988, 88, 1423; d) G. H. Posner, Chem. Rev. 1986,
86, 831; e) L. F. Tietze, in: Selectivity A Goal for Synthetic Effi-
ciency (Eds.: W. Bartmann, B. M. Trost), VCH, Weinheim,
1984, p. 299.
[2] a) P. J. Black, W. Harris, J. M. J. Williams, Angew. Chem. 2001,
113, 4607; b) P. J. Black, W. Harris, J. M. J. Williams, Angew.
Chem. Int. Ed. 2001, 40, 4475; c) P. J. Black, M. G. Edwards,
J. M. J. Williams, Tetrahedron 2004, 61, 1363.
[3] G. Cami-Kobeci, J. M. J. Williams, Synlett 2003, 124.
[4] We have also applied the Catalytic Electronic Activation con-
cept to the stereoselective epoxidation of enals: P. J. Black, K.
Jenkins, J. M. J. Williams, Tetrahedron: Asymmetry 2002, 13,
317.
[5] a) C. F. de Graauw, J. A. Peters, H. van Bekkum, J. Huskens,
Synthesis 1994, 1007; b) C. Djerassi, in: Org. React., vol. 6
(Eds.: R. Adams, W. E. Bachmann, L. F. Fieser, J. R. Johnson,
H. R. Snyder), John Wiley, New York, 1953, pp. 207; c) A. L.
Wilds, in: Org. React., vol. 2 (Eds.: R. Adams, W. E. Bach-
mann, L. F. Fieser, J. R. Johnson, H. R. Snyder), John Wiley &
Sons, New York, 1944, pp. 178.
[6] The literature contains very few reports of stoichiometric cross-
over transfer hydrogenation reactions between alcohols and al-
kenes. A few examples of crossover transfer hydrogenation
using a small excess of alkene acceptor have been reported. See
for example: M. E. Krafft, B. Zorc, J. Org. Chem. 1986, 51,
5482.
[7] a) M. G. Edwards, J. M. J. Williams, Angew. Chem. 2002, 114,
4934; b) M. G. Edwards, J. M. J. Williams, Angew. Chem. Int.
Ed. 2002, 41, 4740.
[8] There are numerous reports of tandem oxidation-olefination
reactions (2 of the 3 steps involved in our domino reaction
sequence). See for example: a) L. Blackburn, R. J. K. Taylor,
Synlett 2002, 215; b) G. Kim, D. G. Lee, S. Chang, Bull. Korean
Chem. Soc. 2001, 22, 943; c) L. Blackburn, X. Wei, R. J. K.
Taylor, Chem. Commun. 1999, 1337; d) D. Crich, X. Mo, Syn-
lett 1999, 67; e) X. Wei, R. J. K. Taylor, Tetrahedron Lett. 1998,
39, 3815; f) S. Shuto, S. Niizuma, A. Matsuda, J. Org. Chem.
1998, 63, 4489; g) A. G. M. Barrett, D. Hamprecht, M.
Ohkuba, J. Org. Chem. 1997, 62, 9376.
[19] E. von Rudloff, Can. J. Chem. 1956, 34, 1413.
[20] Control experiments in the presence of excess alcohol exhibited
no hydrogenolysis of the benzyl ester.
[21]
T. W. Greene, P. G. Wuts, Protective groups in Organic synthesis,
Wiley-Interscience, New York, 1991.
[22]
G. R. A. Adair, M. G. Edwards, J. M. J. Williams, Tetrahedron
Lett. 2003, 44, 5523.
[23]
[24]
[25]
[26]
M. Grayson, P. T. Keough, J. Am. Chem. Soc. 1960, 82, 3919.
F. C. Schaffer, G. A. Peters, J. Org. Chem. 1961, 26, 412.
L. I. Krimen, D. J. Cota, Org. React. 1969, 17, 213.
See for example: a) K. Hattori, H. Sajiki, K. Hirota, Tetrahe-
dron 2001, 57, 4817; b) H. Sajiki, K. Hattori, K. Hirota, J.
Chem. Soc. Perkin Trans. 1 1998, 4043; c) R. Mozingo, C. Spen-
cer, K. Folkers, J. Am. Chem. Soc. 1944, 66, 1859.
The ultimate fate of carbon monoxide is unknown. Iridium
carbonyl complexes are stable species thus it is plausible that
this step represents a catalyst termination event.
See for example: a) C. H. Jun, C. W. Huh, S. J. Na, Angew.
Chem. 1998, 110, 150; b) C. H. Jun, C. W. Huh, S. J. Na, An-
gew. Chem. Int. Ed. 1998, 37, 145.
[27]
[28]
[29]
[30]
H. Adkins, R. M. Elofson, A. G. Rossow, C. C. Robinson, J.
Am. Chem. Soc. 1949, 71, 3622.
a) R. Umaga, C. Crevisy, R. Gree, Chem. Rev. 2003, 103, 27;
b) D. V. McGrath, R. H. Grubbs, Organometallics 1994, 13,
224; c) B. M. Trost, R. J. Kulawiec, J. Am. Chem. Soc. 1993,
115, 2027; d) Y. Sasson, G. L. Rempel, Tetrahedron Lett. 1974,
47, 4133.
[31]
3-methyl-2-buten-1-ol (not shown) afforded a complex reaction
mixture from which none of the desired product could be iso-
lated.
[32]
[33]
Reactions with the cyano ylide 19 yielded mixtures of E/Z α,β-
unsaturated nitriles, even under the thermodynamic conditions.
See for example: a) E. J. Corey, J. Kang, J. Am. Chem. Soc.
1982, 104, 4724; b) K. Utimoto, M. Tamura, K. Sisido, Tetra-
hedron 1973, 29, 1169; c) C. Ruechardt, S. Eichler, P. Panse,
Angew. Chem. 1963, 75, 85; d) S. Trippett, D. M. Walker, J.
Chem. Soc. 1961, 1266.
[9] M. G. Edwards, R. F. R. Jazzar, B. M. Paine, D. J. Shermer,
M. K. Whittlesey, J. M. J. Williams, D. D. Edney, Chem. Com-
mun. 2004, 90.
[10] See for example: a) M. S. Kwan, N. Kim, S. H. Seo, I. S. Park,
R. K. Cheedrala, J. Park, Angew. Chem. 2005, 117, 7073; b)
M. S. Kwan, N. Kim, S. H. Seo, I. S. Park, R. K. Cheedrala, J.
Park, Angew. Chem. Int. Ed. 2005, 44, 6913; c) K. Taguchi, H.
Nakagawa, T. Hirabayashi, S. Sakaguchi, Y. Ishii, J. Am. Chem.
[34]
[35]
O. I. Kolodiazhnyi, Phosphorus Ylides, Wiley-VCH, Weinheim,
1999.
This particular substrate proved troublesome to isolate both
because of its volatility and the tendency for it to be destroyed
Eur. J. Org. Chem. 2006, 4367–4378
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4377

P. J. Black, M. G. Edwards, J. M. J. Williams
FULL PAPER
under the oxidative workup conditions; identification of the
product was therefore achieved by hydrogenation of the alkene
present in the alkene/alkane mixture.
[38] R. G. Shea, J. N. Fitzner, J. E. Fankhauser, A. Spaltenstein,
P. A. Carpino, R. M. Peevey, D. V. Pratt, B. J. Tenge, P. B. Hop-
kins, J. Org. Chem. 1986, 51, 5243.
[36] Under the domino reaction conditions the olefination reaction
is rapid.
Received: January 27, 2006
Published Online: July 27, 2006
[37] J. Suffert, J. Org. Chem. 1989, 54, 509.
4378
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4367–4378