Hydrogen borrowing catalysis using 1° and 2° alcohols: Investigation and scope leading to α and β branched products

Article abstract of DOI:10.1016/j.tet.2021.132051

The alkylation of a variety of ketones using 1° or 2° alcohols under hydrogen borrowing catalysis is described. Initial research focused on the α-alkylation of cyclopropyl ketones with higher 1° alcohols (i.e. larger than MeOH), leading to the formation of α-branched products. Our search for additional substrates with which to explore this chemistry led us to discover that di-ortho-substituted aryl ketones were also privileged scaffolds, with Ph? (C₆Me₅) ketones being the optimal choice. Further investigations revealed that this motif was crucial for alkylation with 2° alcohols forming β-branched products, which also provided an opportunity to study diastereoselective and intramolecular hydrogen borrowing processes.

Full text of DOI:10.1016/j.tet.2021.132051

Tetrahedron 86 (2021) 132051
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Hydrogen borrowing catalysis using 1ꢀ and 2ꢀ alcohols: Investigation
and scope leading to
a and b branched products
James R. Frost a, Choon Boon Cheong a, Wasim M. Akhtar a, Dimitri F.J. Caputo a,
,
*
Kirsten E. Christensen a, Neil G. Stevenson b, Timothy J. Donohoe a
a Chemistry Research Laboratory, University of Oxford, Oxford, OX1 3TA, United Kingdom
b GlaxoSmithKline, Medicines Research Centre, Stevenage, SG1 2NY, United Kingdom
a r t i c l e i n f o
a b s t r a c t
The alkylation of a variety of ketones using 1ꢀ or 2ꢀ alcohols under hydrogen borrowing catalysis is
described. Initial research focused on the
-alkylation of cyclopropyl ketones with higher 1ꢀ alcohols (i.e.
larger than MeOH), leading to the formation of -branched products. Our search for additional substrates
Article history:
Received 19 January 2021
Received in revised form
15 February 2021
Accepted 19 February 2021
Available online 5 March 2021
a
a
with which to explore this chemistry led us to discover that di-ortho-substituted aryl ketones were also
privileged scaffolds, with Ph* (C6Me5) ketones being the optimal choice. Further investigations revealed
that this motif was crucial for alkylation with 2ꢀ alcohols forming
b-branched products, which also
Dedicated to Professor J. M. J. Williams.
provided an opportunity to study diastereoselective and intramolecular hydrogen borrowing processes.
© 2021 Elsevier Ltd. All rights reserved.
Keywords:
Hydrogen borrowing catalysis
Iridium
Alkylation
Cyclopentane
retro-FriedeleCrafts
1. Introduction
[4], Andersson [5], and others [6] also cemented MeOH as an
appropriate alcohol with which to form
products. In order to increase the usefulness of this chemistry, we
also employed BaeyereVilliger reaction that provided
a-branched (methylated)
Hydrogen borrowing chemistry has emerged as a powerful
method for the rapid construction of new CeC and CeN bonds
under catalytic conditions [1]. In a typical one-pot reaction mani-
fold, an alcohol functional group is initially oxidized to the corre-
sponding carbonyl by an appropriate metal catalyst, generating a
metal hydride species. The carbonyl formed in situ then condenses
with a nucleophile (e.g. enolate or amine), and the subsequently
formed compound (e.g. enone or imine) is reduced by the metal
hydride to complete the catalytic cycle. The entire process involves
no net change in oxidation state, and avoids laborious chemical
manipulations and toxic reagents[2].
a
a
straightforward synthesis of the corresponding
[7].
a-branched esters
However, initial attempts at expanding this methodology to
encompass higher 1ꢀ alcohols (i.e. larger than MeOH) as alkylating
agents had limited success. A subsequent literature search revealed
very few examples whereby a-branching had been accomplished,
with most processes pertaining to Guerbet-type reactions [8].
Furthermore, it was apparent that 2ꢀ alcohols were also not
generally employed for alkylation under hydrogen borrowing
conditions. A rare example of this was reported by Obora, who
showed that MeCN (10 equiv.) could be alkylated with 2ꢀ alcohols,
albeit at high temperatures (130e200 ꢀC) [9]. Initially, we hy-
pothesized that the additional steric bulk imparted by higher al-
cohols (both 1ꢀ and 2ꢀ) destabilized the intermediate aldol adduct,
favoring the retro-aldol reaction over E1cB elimination (Scheme
1B). Our strategy therefore focused on the use of substrates that
were less sterically hindered such that aldol condensation could
occur profitably in situ.
We have recently sought to expand the synthetic utility of
hydrogen borrowing catalysis, and in the process develop new CeC
bond forming reactions. Our initial research revealed that
a-
methylation of ketones could be achieved using MeOH under
rhodium-catalyzed conditions (Scheme 1A) [3]. This also resulted
in the formation of a-branched ketones, which at that time had
little literature precedent. Further contributions by the groups of Li
* Corresponding author.
E-mail address: timothy.donohoe@chem.ox.ac.uk (T.J. Donohoe).
https://doi.org/10.1016/j.tet.2021.132051
0040-4020/© 2021 Elsevier Ltd. All rights reserved.

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
Table 1
Optimization of hydrogen borrowing conditions with cyclopropyl ketone 1 and
BnOH.
observed otherwise (entries 7 and 8). Whilst we replaced KOH with
several different bases (K2CO3 and NaOMe, not shown), only KOtBu
gave complete conversion, albeit in inferior isolated yield (entry 9,
50%). The amount of product 2 could be marginally improved
(~10%) by increasing the loading of [Ir(cod)Cl]2 and ( )-BINAP
(maintained in a 1:2 ratio). However, we considered this gain
insufficient to justify committing additional reagents (entries 11
and 12). With this in mind, several other catalysts that were known
to facilitate hydrogen borrowing chemistry without an additional
ligand were also screened. Both [Cp*IrCl2]2 and [Cp*RhCl2]2 were
found to be effective, with the former providing 2 in near identical
yield (entry 13) to the [Ir(cod)Cl]2/( )-BINAP system (entry 6) in
only 24 h. This considerable reduction in reaction time was highly
desirable, and so the conditions in entry 13 were chosen to evaluate
the substrate scope.
Scheme 1. Hydrogen borrowing reactions with higher 1ꢀ and 2ꢀ alcohols.
Herein we disclose our full research effort towards the devel-
opment of the aforementioned a-alkylation of ketones with higher
1ꢀ alcohols (Scheme 1C). Further development of this chemistry
thereafter led to a general iridium catalyzed alkylation using 2ꢀ
alcohols to form b-branched products (Scheme 1D) [10,11]. Subse-
quent work in the literature has reported related hydrogen
borrowing alkylation using a variety of metal (and earth abundant
metal) catalysts [12]. This account will also highlight our pre-
liminary studies towards the development of intramolecular
hydrogen borrowing reactions (Scheme 1E) [10b].
Next, non-benzylic alcohols were assessed under the optimized
conditions. Pleasingly, nBuOH gave a-branched compound 5 in 86%
2. Results and discussion
isolated yield, whilst more sterically hindered aliphatic alcohols
(cyclopropylmethanol and 3,3-dimethyl-1-butanol) also worked in
good to excellent yield (Scheme 2, 86% and 68% respectively).
Replacing the phenyl group on the substrate sidechain with nPr or
iBu was also tolerated in combination with these alcohols. A diox-
olane ring was then incorporated into the substrate backbone,
leading to compounds 14e18, and thereby providing useful func-
tionality for further synthetic manipulation. Of particular note was
the use of iBuOH as a coupling partner (to give 18 in 71% yield),
which represents one of the most sterically hindered alcohols
2.1.
ketones
a-Alkylation of cyclopropyl ketones and related cycloalkyl
Our study began with cyclopropyl ketone 1, employing [Ir(cod)
Cl]2 (2 mol%), ( )-BINAP (4 mol%), KOH (3 equiv.) and BnOH (5
equiv.) at 65 ꢀC under an Ar atmosphere for 48 h (Table 1, entry 1).
Pleasingly, analysis of the crude 1H NMR spectrum showed the
desired
a-alkylated product 2 (12%), as well as enone (7%, as a
mixture of E/Z-3 isomers) and unreacted starting material (47%)
[13]. While ligand variation did very little to change the product
distribution (select examples, entries 2e4), a breakthrough came
with an increase in reaction temperature. Heating to 85 ꢀC
improved the product ratio to 41:8:20 (2:(E/Z)-3:1, entry 5), whilst
employed to afford
a-branched compounds under hydrogen
borrowing conditions. Several cyclopropyl ketones bearing het-
erocyclic sidechains were also processed with nBuOH. The furan
moiety was found to be the least well-tolerated substrate (Scheme
2, 19, 22% yield), whilst thiophene and pyridine were more robust
(20 and 21, both 45% yield).
105 ꢀC gave desired
a-alkylated compound 2 in 74% isolated yield
(entry 6). With this result in hand, we attempted to lower the
stoichiometry of KOH from 3 equiv. but incomplete conversion was
2

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
Scheme 4. Ring opening of a-alkylated cyclopropyl ketones.
2.3.
a-Alkylation of ortho-substituted aromatic ketones
Scheme 2. Cycloalkyl ketone substrate scope.
During our study of cycloalkyl ketones we also chose to assess
whether -branched aryl ketones could be prepared by hydrogen
borrowing alkylation. An initial lead came from 3-phenyl-1-(ortho-
tolyl)propan-1-one, which underwent -benzylation using similar
a
a
starting conditions to those previously described for cyclopropyl
ketones [10a]. This system was then optimized independently, with
our final preferred conditions being [Ir(cod)Cl]2 (1 mol%), dppBz
(2 mol%), KOH (2 equiv.), BnOH (10 equiv.) at 85 ꢀC under an Ar
atmosphere for 24 h (Scheme 5). Pleasingly, this afforded com-
pound 32 in 70% yield, with the remaining mass comprised of
reduced starting material (3-phenyl-1-(ortho-tolyl)propan-1-ol).
The ortho-methyl group was found to be critical for the success of
the reaction, as its omission returned mostly unreacted substrate or
its reduced form (see 31). Since it is to be expected that ortho-
methyl substitution on an aryl ketone results in a twisting effect
whereby the aromatic ring is forced out of conjugation with the
ketone, we hypothesized that this might be the reason for differing
reactivity compared to unsubstituted arene 31.
Scheme 3. Ir catalyst-free control experiments.
Having assessed cyclopropyl ketones we were also curious to
determine whether other cycloalkyl ketones could be used in this
reaction manifold. Thus, cyclobutyl, cyclopentyl and cyclohexyl
ketones were synthesized and underwent the desired a-butylation
reaction accordingly, but were lower yielding than the parent
cyclopropyl ketone (22, 23 and 24, 27e45% c.f. 5, 86%).
Control experiments omitting [Cp*IrCl2]2 from the reaction
mixture were also performed, and to our surprise product 2 was
obtained in 49% yield when BnOH was employed as a reaction
partner (Scheme 3). However, this catalyst-free protocol does not
appear to be a general process, since nBuOH did not react under
these conditions. Autocatalytic hydrogen borrowing alkylation re-
actions are precedented [14] and are thought to proceed through
MeerweinePonndorfeVerleyeOppenauer (MPV-O) type processes.
Benzylic alcohols are presumably well-suited for this process since
they undergo facile oxidation compared to their alkyl counterparts
(Scheme 3).
2.2. Ring opening of the cyclopropyl ketone products
With the substrate scope complete, we also saw an opportunity
to further diversify our products by ring opening of the cyclopro-
pane. This was firstly achieved by treatment of compound 5 with
either aq. HBr or H2SO4/H2O [15], affording bromide and alcohol
functionalities respectively. Modification of a literature procedure
enabled homoconjugate addition of a Lewis acid complexed cya-
nocuprate, providing 27, 28 and 29 in moderate to good yield [16].
Finally, Lewis acid-assisted ring opening with thiophenol was also
successful, this time forming a CeS bond (Scheme 4).
Scheme 5. ortho-Substituted aryl ketone substrate scope.
3

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
Variation of either the alcohol used (nBuOH and cyclo-
propylmethanol) or the substrate backbone was once again well-
tolerated, although differences between various aryl ortho-sub-
stituents were noted. Whilst substitution with an ortho-CF3 group
gave similar yields to the parent ortho-CH3 containing compounds
(e.g. 33 and 35), the introduction of a bulkier ortho-iPr group
improved yields and provided a cleaner reaction profile (see 37 and
38, 81% and 79% respectively). Interestingly, no reduced starting
material was detected during these reactions, with the ortho-iPr
group seeming to protect the ketone from reduction by steric
shielding. In contrast, substitution with an ortho-methoxy group
resulted in a much poorer yield of 21%, presumably since the
carbonyl group is more exposed to reduction (Scheme 5) [17].
chlorine was found to be a competing process (54). Having initially
observed that the introduction of an ortho-iPr group improved the
reaction yield (Scheme 5), we noted that an additional ortho-iPr
substituent had the reverse effect, delivering only 43% of 55. This
example illustrates the upper limit of steric bulk necessary to
obtain high yields of alkylated product. Finally, 1-anthracenyl ke-
tone was also found to be effective under the reaction conditions
(56). In total, this study suggested that di-ortho-methyl substitution
was optimal, providing high yields of
6).
a-alkylated product (Scheme
2.5. Aryl release strategy
Whilst we had succeeded in developing
a high yielding
approach to -branched aryl ketones under catalytic conditions, the
a
2.4.
a-Alkylation of di-ortho-substituted aromatic ketones
products obtained appeared to represent a limited set of com-
pounds. With this in mind, we attempted to cleave the mesityl ring
by BaeyereVilliger reaction in order to deliver the corresponding
ester (Scheme 7A). However, despite several attempts, only
unreacted starting material 40 was observed. Since our study of the
Since steric shielding of the carbonyl appeared to be beneficial,
we chose to synthesize a mesityl ketone in order to further study
this feature and increase the desired twisting effect. As expected,
this change delivered the a-branched products (40 and 41, 89% and
a
-alkylation reaction had shown that di-orthoemethyl substitution
94% respectively, Scheme 6), with no reduced material being
observed. In this instance application of 3,3-dimethyl-1-butanol
under hydrogen borrowing conditions resulted in a lower yield
(42: 48%) relative to the cyclopropyl ketone counterpart (7). The
majority of the remaining mass balance comprised of the inter-
mediate enone (45%), suggesting that the combination of the
mesityl and tBu groups hinders the final reduction step in this
system (Scheme 6). Incorporation of a nitrogen atom in the aro-
matic ring was not detrimental (43 and 44), whilst addition of a
further ortho-methoxy substituent gave satisfactory yields
compared to the monosubstituted case (see 45). Modification of the
substrate backbone was again well-tolerated, with furan-
containing compounds also synthesized straightforwardly under
these conditions.
of the aromatic ring was beneficial in preventing reduction of the
ketone (by virtue of the aforementioned twisting effect), it seemed
likely that these methyl groups were also preventing nucleophilic
attack (Scheme 7A). In our search for alternative methods, we noted
that the esterification of mesitoic acid under ordinary acid catalysis
conditions is similarly challenging [18]. It is well-established that in
order to overcome this problem, mesitoic acid is initially treated
with conc. H2SO4 to form an intermediate acylium ion, then the
reaction mixture is poured onto cold MeOH to trap this species and
form the desired methyl ester (Scheme 7B) [18]. Our a-alkylated
systems bore a resemblance to this literature example, and so we
hypothesized that mesityl cleavage might be possible under similar
conditions but via a retro-FriedeleCrafts reaction involving ipso
addition to the arene (Scheme 7C). A thorough literature search
revealed a small set of papers whereby such a process had occurred
[19], and to our delight, treatment of ketone 40 with aq. H2SO4 at
a
-Alkylation could also be achieved even when the aromatic
ring contained di-ortho-chloro substitution, although loss of
Scheme 7. Aryl release strategy and application to ketone 40.
Scheme 6. Di-ortho-substituted aryl ketone substrate scope.
4

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
65 ꢀC successfully converted this compound to the corresponding
acid in 80% yield (58). Similarly, transacylation could also be ach-
ieved using TfOH (1.2 equiv.) in anisole at 100 ꢀC to afford 59
(Scheme 7D) [20].
Although the twisting effect was key to the retro-FriedeleCrafts
process, we recognized that the electron donating properties of the
methyl groups also played an important role. Mechanistically, it
seemed reasonable to assume that the ipso-position of the aromatic
ring (with respect to the carbonyl) was protonated under the
strongly acidic conditions, enabling an acylium ion to form and
release mesitylene. To make this process more functional group
tolerant and hence synthetically useful, we sought to develop non-
acidic conditions for aryl cleavage, and supposed that alternative
electrophiles could be employed [21]. As a proof of concept, neat
bromine was quickly established as a reagent with which to trigger
the retro-FriedeleCrafts reaction (unoptimized, not shown). This
represented an important breakthrough as aryl cleavage formed a
useful acid bromide in situ (see Section 2.6, Scheme 9). The acid
bromide could then be intercepted with nBuOH to form the cor-
responding nBu ester. In our hands, however, bromination of the
mesityl ring was also a competing process [12f], leading to di- and
tri-brominated mesitylene side-products. Thus, we decided to
further refine the design of the aromatic ring.
Scheme 9. Br2-mediated cleavage of Ph* and subsequent derivatization of the acid
bromide with various nucleophiles.
acid bromide generated (observed by 13C NMR, see ref 10a) could
then be easily intercepted in situ with different nucleophiles,
providing a range of different carboxylic acid derivatives in a one-
pot process (Scheme 9).
2.6. Design of pentamethylphenyl (Ph*) ketone
The most straightforward method to prevent unwanted
bromination of the aryl ring was to block the unsubstituted posi-
tions. Therefore, in order to ensure that the aryl ring remained
electron rich and amenable to cleavage by treatment with an
electrophile, we chose to add additional methyl groups. Since the
primary focus of our work was hydrogen borrowing catalysis,
several substrates containing this pentamethylphenyl (Ph*) group
were synthesized and treated under the optimized conditions.
These compounds were compatible with our well-established
range of alcohols, while sidechains bearing OBn, SBn and NBn2
functionality were also tolerated (Scheme 8).
We were also pleased to note that the compounds shown in
Scheme 8 were all crystalline. This proved to be a general property,
with almost all of the compounds synthesized that bear the Ph*
motif displaying this characteristic. This was advantageous, not
only for large scale purification without chromatography, but also
for straightforward determination of relative stereochemistry by
single crystal X-ray diffraction (see Scheme 10 for examples).
As expected, the Ph* group underwent clean retro-Friedele-
Crafts reaction when treated with bromine, giving 1-bromo-
2,3,4,5,6-pentamethylbenzene as the major byproduct (Scheme 9).
This reaction was subsequently optimized to reduce the amount of
bromine and concentration required, whilst also performing the
reaction at low temperature [Br2 (2 equiv.), CH2Cl2, e17 ꢀC]. The
2.7. Further analysis of the role of the Ph* group
In order to gain additional insight into the role of Ph* in the
hydrogen borrowing reaction, we synthesized separately the syn-
and anti-aldol adducts (76 and 77) that would form as in-
termediates under the reaction conditions (using compound 61 as a
reference). As expected, both aldol adducts were crystalline,
enabling single crystal X-ray diffraction studies to be performed
which supported our hypothesis that the Ph* motif was twisted out
of conjugation with the carbonyl group (Scheme 10A). Subjecting
either compound to the hydrogen borrowing reaction conditions
then provided the anticipated product 61. Next, we synthesized the
analogous syn- and anti-aldol adducts 78 and 79 without any ortho-
methyl substitution; the latter being crystalline with the resulting
structure determined by single crystal X-ray diffraction showing
the phenyl ring to be in conjugation with the ketone. This time,
subjecting 78 and 79 to the hydrogen borrowing conditions resul-
ted in only a small quantity of the desired a-alkylated product 31
being formed, with alcohol 81 being the major component of the
reaction mixture (together with a small amount of ketone 80).
Taken together, several conclusions can be drawn from these re-
sults. Firstly, the product distributions obtained for phenyl-
containing (i.e. non-Ph*) adducts 78 and 79 suggests that retro-
aldol reaction is favored over E1cB elimination, with the ketone
(80) then being reduced to the corresponding alcohol 81. Whilst it
is possible that alcohol 81 could then be re-oxidized in situ, the
large quantity of isolated material suggests that it does not readily
re-engage in the aldol process.
With these observations in mind, alcohol 81 was subjected to
the hydrogen borrowing alkylation conditions (Scheme 10B). This
returned 81 unchanged in 97% isolated yield. However, deuterated
substrate d-81 (ꢁ95% d, prepared separately) returned compound
81 (11% d by 1H NMR), thus suggesting that ketone 80 is formed by
reversible oxidation/reduction, but does not engage in hydrogen
borrowing chemistry. Interestingly, Ph*-containing alcohol 82 also
only gave starting material under these conditions, without the
observation of alkylated material. Since it had been well-
established throughout that di-ortho-methyl substitution of the
Scheme 8. Ph* ketone substrate scope.
5

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
coupled material 86 in 12% isolated yield, with 61% unreacted 83
recovered (Table 2, entry 1). Increasing the reaction temperature to
105 ꢀC resulted in a significantly improved yield of 86 (80%),
although 115 ꢀC was detrimental in comparison (entries 2 and 3).
Analysis of the reaction mixture suggested that the solubility of
KOH in PhMe may be a complicating factor, and so we chose to
assess different bases at lower temperature (85 ꢀC) (entries 4e6).
Whilst NaOH was found to outperform KOH at 85 ꢀC, KOtBu and
NaOtBu were superior, affording 86 in 86% and 97% yield respec-
tively. Further optimization was carried out with NaOtBu, and it
was quickly established that 85 ꢀC was the ideal temperature
(incomplete reaction at 65 ꢀC, entry 7), with 2 equiv. of base
required for complete conversion (entry 8). The amount of
[Cp*IrCl2]2 and 3-pentanol could also be reduced without affecting
the isolated yield (entries 10 and 12). Our final conditions therefore
involved treating starting material 83 with [Cp*IrCl2]2 (0.5 mol%),
3-pentanol (1.5 equiv.), NaOtBu (2 equiv.) and PhMe (4 M) at 85 ꢀC
under Ar for 24 h (entry 12). The reaction was successfully per-
formed on 1 g scale, affording 86 in 93% isolated yield (entry 14).
Finally, control experiments were run in the absence of NaOtBu and
[Cp*IrCl2]2. Whilst the omission of base resulted in only recovered
starting material 83 (entry 15), removing [Cp*IrCl2]2 gave partial
conversion to E/Z enones 85 (entry 16) and no reduced material
(86) [10b].
Similar to our earlier work involving
a-alkylation to form
branched products, we initially studied variations of the aryl ring
(Scheme 11). Compounds bearing di-ortho-methyl substitution
once again gave the desired coupled material in good to excellent
yield (see 86, 88, 89, 92, 93 and 95). As expected, ortho-tolyl per-
formed poorly (87, 15%), although interestingly the trimethoxyaryl
system gave 90 in only 32% yield. This is in contrast to previous
observations (see Scheme 6), suggesting that protection of the
carbonyl using this group is suboptimal when the substrate back-
bone itself does not contain a longer chain (i.e. CH2CH2Ph, com-
pounds 45e47). Similarly, anthracenyl ketone produced 91 in poor
yield. Substitution with iPr groups delivered the desired coupled
material in good yield (94, 62%), presumably since removal of the
substrate backbone compensates for the increased steric bulk of
these two ortho groups (c.f. 55). Double alkylation could also be
achieved, albeit with an increased loading of reagents (95). Lastly,
we also attempted to alkylate ethyl ketone 96 with 3-pentanol.
However, only unreacted starting material was returned (97%),
which was in line with our earlier observations that sterics play a
significant role in this chemistry. Overall, these results clearly
indicated that Ph* was the preferred aryl ring for this reaction,
providing the desired balance of substrate reactivity and steric
protection of the carbonyl group.
Scheme 10. Subjection of various
hydrogen borrowing conditions.
a-branching intermediates to the optimized
aryl ring prevents carbonyl reduction, this result suggests that
alcohol 82 does not undergo oxidation readily.
2.9. Alkylation with acyclic 2ꢀ alcohols
Taken together, these results indicate that the Ph* group enables
hydrogen borrowing alkylation by protecting the adjacent carbonyl
group against reduction/nucleophilic attack in situ. It is also
possible that the twisting effect of the Ph* group expedites for-
mation of the corresponding aldol and enone adducts, relative to
their Ph counterparts.
Next, we assessed the compatibility of other acyclic alcohols
with Ph*COMe 83. Alcohols bearing linear chains were readily
processed (98e100), as were those containing saturated ring sys-
tems (3-membered to 6-membered rings, 101e104). Phenyl sub-
stitution was also well-tolerated (105), whilst pyridine and
thiophene-containing alcohols behaved similarly (to give 106 and
107), albeit requiring an increased loading of reagents [[Cp*IrCl2]2
(2 mol%), NaOtBu (3 equiv.), alcohol (3 equiv.) and PhMe (4 M) at
85 ꢀC for 24 h under Ar]. Furans could also once again be incor-
porated (see 52 and 53 for earlier examples), although compound
108 was only isolated in 26% yield under these conditions. Finally,
long chain alcohols with distal OBn and NBn2 groups could be
combined with 83 to afford 109 and 111. A further increase in the
steric size of the alcohol (Me to Et) was also not detrimental to yield
(109 to 110, Scheme 12). It is worth noting at this stage that no self-
2.8. Alkylation with pentan-3-ol
Having investigated the role of di-ortho-substituted aryl ketones
in hydrogen borrowing chemistry, we next chose to assess 2ꢀ al-
cohols as alkylating agents. An initial hit was uncovered when
pentamethylacetophenone (83) was reacted with 3-pentanol (2
equiv.) in the presence of [Cp*IrCl2]2 (2 mol%), KOH (3 equiv.) and
PhMe (4 M) at 85 ꢀC for 24 h under Ar. This delivered the desired
6

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
Table 2
Optimization of hydrogen borrowing conditions with Ph* ketone 83 and 3-pentanol.
Scheme 12. Acyclic 2ꢀ alcohol scope.
condensed product (derived from 83) was detected in any of these
reactions. This suggests again that Ph* (specifically di-ortho-methyl
substitution of the aryl ring) is key to preventing this undesired side
reaction and promoting the cross-aldol process [10b].
2.10. Alkylation with cyclic 2ꢀ alcohols
Expansion of this methodology to include cyclic 2ꢀ alcohols was
also possible, but this time improved yields were obtained when
KOtBu was used in combination with an increased loading of re-
agents (Scheme 13). Simple cyclopentanol and cyclohexanol were
employed straightforwardly to give 112 and 113 in 82% and 86%
yield respectively. As shown previously, a dioxolane ring, NBn
substitution (to give the protected piperidine) and a benzylic
alcohol were found to be robust under hydrogen borrowing con-
ditions (114, 115 and 116) [10b].
Substituted cyclic 2ꢀ alcohols presented an exciting opportunity
to study diastereoselective hydrogen borrowing reactions, which
had little literature precedent [22]. Interestingly, reaction of cis-
decahydro-1-naphthol with 83 gave compound 117 in 66% isolated
yield in dr > 95:5 (the relative stereochemistry was determined by
X-ray crystallography), illustrating an initial proof of concept [23].
4-Substituted cyclohexyl alcohols were also found to couple in high
diastereoselectivity, with dr improving in line with an increase in
the size of the 4-substituent (tBu > CF3>CH3>OBn, 119e122, major
diastereomer shown in each case). Methyl substitution at the 2- or
3-position of the cyclohexyl ring also resulted in high diaster-
eoselectivity (123 and 124). X-Ray structures were obtained for
compounds 119 (major diastereomer), 121 (minor diastereomer)
and 122 (tBu), revealing that the CH2COPh* group in each major
diastereomer adopted an axial position (this was also true for
decalin 117). Therefore, it is reasonable to assume that the dia-
stereoselectivity can be rationalized by equatorial attack of an IreH
species onto an exocyclic enone, although reduction of an endo-
cyclic (migrated) alkene cannot be ruled out [24]. In contrast to
cyclohexyl-containing compounds, methyl-substituted cyclopentyl
alcohols coupled in poor diastereoselectivity (see compounds 125
Scheme 11. Di-ortho-substituted aryl ketone substrate scope.
7

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
either case, with only unreacted starting material present. Despite
extensive screening, compound 127 could only be produced in 12%
isolated yield (Scheme 14) [25].
In contrast, benzylic ketone 128 smoothly underwent a-alkyl-
ation to give 129 when treated with our standard set of conditions
(82% yield). Once again, steric hindrance appears to be playing a
significant role, with sp3-substitution at the
b-position adversely
affecting reactivity.
In order to overcome the lack of reactivity at the
decided to investigate an intramolecular hydrogen borrowing re-
action. -Substituted compounds 109 and 110 were ideally suited
a-position, we
b
for such an approach, as removal of the Bn-protecting group
revealed the requisite 1ꢀ alcohol (Scheme 15). Treatment of both
132 and 133 under [Cp*IrCl2]2 conditions resulted in smooth
cyclization to the corresponding cyclopentanes in good yield (136:
75% and 137: 72%). Of particular note was that both compounds
were formed in high diastereoselectivity (dr > 95:5), with the 1,2-
trans relationship confirmed by X-ray crystallography [10f]. This is
presumably due to reversible
thermodynamically favored product. To further investigate this
intramolecular process, we attempted to synthesize additional
a-deprotonation in situ, leading to the
b-
alkylated substrates via 2ꢀ alcohols. The chain length was fixed (in
order to form a cyclopentane) while the R substituent was varied. In
practice Ph (140) and benzyl (141) substitution were not well-
tolerated, providing messy reaction profiles and no coupled mate-
rial. Incorporation of a cyclopropyl group resulted in the desired
product (130) in 60% yield, although an increase in the ring size to
cyclopentyl gave skipped enone 143 as the major product. Dioxane-
containing compound 131 could also be prepared but in only
moderate yield (47%), likely due to the increased steric bulk of this
alcohol (Scheme 15). Deprotection of the cyclopropyl and dioxane
substituted compounds proceeded without incident to give the
corresponding 1ꢀ alcohols. Subjection of these substrates to
hydrogen borrowing conditions then delivered the desired cyclo-
pentanes in excellent diastereomeric ratio (138 and 139). Whilst an
Scheme 13. Cyclic 2ꢀ alcohol scope.
and 126). This likely reflects the relative conformational stability of
cyclohexanes over their cyclopentane counterparts.
2.11.
a-Alkylation of b-substituted ketones
intramolecular
the poor reactivity profiles associated with intermolecular pro-
cesses, the size of the
-sp3 substituent still exerts a considerable
a-alkylation approach can successfully overcome
Since alkylation with 2ꢀ alcohols had provided a range of
substituted ketones, it seemed appropriate to then test these
compounds as substrates for -alkylation. Thus, ketone 113 was
treated under both of our optimized reaction conditions (Scheme
14). Disappointingly no -butylated product was detected in
b-
b
influence on the reaction outcome. Nevertheless, these preliminary
studies led us to consider whether diols would be appropriate
partners for ring synthesis. This avenue of research proved highly
successful and has been published elsewhere, so will not be dis-
cussed here [10c-e].
a
a
2.12. Aryl release of Ph*
A number of structurally diverse b-substituted esters and am-
ides were synthesized from the alkylation products following
bromine-mediated release of Ph* (Scheme 16). Pleasingly, stereo-
defined compounds 117, 119, 121 and 136 underwent clean reaction
without erosion of diastereoselectivity (dr > 95:5 by 1H NMR).
Substrates bearing pyridine (148) and bicyclic (150) backbones
were also well-tolerated, but removal of Ph* in the presence of a
thiophene ring was initially problematic. Further investigation of
Ph* cleavage and subsequent optimization led us to develop con-
ditions employing TMSCl/HFIP/nBuOH at 40 ꢀC, giving ester 149 in
60% yield [10b,26]. Interestingly, treatment of benzyl alcohol 109
with bromine resulted in debenzylative lactonization to afford
compound 153 in 80% yield. Finally, a-amino esters were also found
to be suitable nucleophiles when added in situ to the acid bromide,
providing 156 and 157 in 75% and 87% yield respectively.
Scheme 14.
a-Alkylation of b-substituted substrates.
8

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
3. Conclusion
In conclusion, we have presented here in full our research effort
that has led to the formation of a- and b-substituted ketones under
hydrogen borrowing conditions. The initial approach focused on
the alkylation of cyclopropyl ketones with higher alcohols, leading
to a-branched products. Our search for additional substrates led us
to recognize that ortho-substituted aromatic ketones were also
privileged scaffolds, with di-ortho-substitution proving optimal.
We believe di-ortho-substitution is important for the following
reasons: a) the ortho-substituents twist the aromatic ring out of
conjugation with the ketone functionality, b) these substituents are
perfectly placed to protect the parent ketone from nucleophilic
attack. This latter design feature was shown to prevent competing
carbonyl reduction and self-condensation, providing a clean reac-
tion profile. To further expand the utility of our chemistry we
designed Ph* (C6Me5) as the substituted aromatic ring of choice
since this could be removed straightforwardly by a retro-Frie-
deleCrafts process using bromine. The acid bromide generated in
situ from an intermediate acylium ion could then be intercepted in
one pot with a range of nucleophiles. The Ph* group was found to
be vital to further research, facilitating alkylation reactions with 2ꢀ
alcohols (to form b-branched products) as well as an intramolecular
ring forming process. These latter advances provided an exciting
opportunity to study stereoselective hydrogen borrowing re-
actions, which to date remains an active area of research in our
laboratory.
Scheme 16. Br2-mediated cleavage of Ph* to give the corresponding Bu ester or amide
variants.
gradient BBFO “SMART” probe, Bruker AVII 500 equipped with a
5 mm BBFO “SMART” probe or 5 mm z-gradient TFI probe, or Bruker
AVIII HD equipped with a 5 mm triple resonance TBO probe, with
the deuterated solvent acting as an internal deuterium lock. 1H
NMR experiments were recorded at 400 or 500 MHz with 19F
decoupling where appropriate, 13C NMR at 101 or 126 MHz with
broadband proton decoupling, and 19F NMR at 377 MHz. Residual
protic solvent signal acted as an internal reference for 1H NMR, and
deuterated solvent carbon signal acted as an internal reference for
13C NMR (CDCl3: 1H NMR ¼ 7.26 ppm; 13C NMR ¼ 77.16 ppm;
DMSO‑d6: 1H NMR ¼ 2.50 ppm, 13C NMR ¼ 39.52 ppm. Chemical
shifts are quoted in ppm with the multiplicity of a signal reported as
such: s e singlet, d e doublet, t e triplet, q e quartet, quint. e
quintet, sext. e sextet, sept. e septet, qAB e AB quartet, m e
multiplet, app. e apparent or approximate, br. e broad, v. e very, or
4. Experimental section
4.1. General experimental
Unless otherwise stated, all reagents and solvents were pur-
chased and used as supplied from Sigma-Aldrich (now Merck
KGaA), Thermo Fisher Scientific (including Alfa Aesar, Fisher Sci-
entific and Acros Organics), Fluorochem, Honeywell, Tokyo Chem-
ical Industry, Apollo Scientific, Manchester Organics (part of Navin
Fluorine Int. Ltd.), Santa Cruz Biotechnology and Strem Chemicals.
All commercial reagents and solvents were used without additional
purification unless indicated in the text. NMR spectroscopy was
carried out using a Bruker AVIII HD 400 equipped with a 5 mm z-
gradient BBFO probe, Bruker AVIII HD 400 equipped with a 5 mm z-
Scheme 15. Intramolecular hydrogen borrowing reaction for the synthesis of 1,2-disubstitued cyclopentanes.
9

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
combinations thereof. Coupling constants and DAB are quoted in Hz
to the nearest 0.1 Hz. Fourier-transform infrared (FT-IR) spectra
were recorded from evaporated films on a Bruker Tensor 27 spec-
trometer equipped with a Pike Miracle Attenuated Total Reflec-
tance (ATR) sampling accessory. Absorption maxima are quoted in
wavenumbers (nmax) with units of cmꢂ1 and for the range of
3600e600 cmꢂ1. High resolution mass spectrometry (HRMS) under
ESI conditions were recorded on a Thermo Exactive Orbitrap mass
spectrometer equipped with a Waters Equity LC system, a Bruker
MicroToF mass spectrometer equipped with an Agilent 1100 HPLC
pump and autosampler, or on a Waters Xevo Quadrupole Time of
Flight (Q-ToF) mass spectrometer. The Thermo Exactive system
employs a flow rate of 0.2 mL minꢂ1 using H2O:MeOH:HCOOH
(10:89.9:0.1) as eluent, with a heated electrospray ionisation (HESI-
II) probe and has a resolution of 50,000 FWHM. The Bruker system
uses the built-in electrospray source, while the Waters system runs
on a lock-mass mode with ESI performed by a secondary electro-
spray source, both using conditions identical to the Thermo Exac-
tive system. Instrument control and data processing were
performed using the softwares Thermo Xcalibur for the Thermo
Exactive system, Compass DataAnalysis 4.0 for the Bruker system,
and MassLynx for the Waters system. Unless otherwise specified,
the mass reported for HRMS is the mass-to-charge ratio containing
the most abundant isotopes, with each value to 4 or 5 decimal
places and within 5 ppm of the calculated mass. Single crystal X-ray
diffraction was performed with a (Rigaku) Oxford Diffraction
2 ꢃ CH2CHCO), 29.9 (2C, 2 ꢃ CH2CH2CH3), 23.0 (2C, 2 ꢃ CH2CH3),
19.5 (cyclopropyl CH), 14.1 (2C, 2 ꢃ CH3), 10.8 (2C, cyclopropyl
CAH2CH and cyclopropyl CBH2CH); HRMS (ESIþ) Found
[MþH]þ ¼ 197.1901; C13H25O requires 197.1900,
D 0.75 ppm.
4.3. Representative procedure for a-alkylation of aromatic ketones:
synthesis of 2-(cyclopropylmethyl)-1-mesitylhexan-1-one (49)
To a 2e5 mL Biotage® microwave vial equipped with a stirrer
bar was added 1-mesitylhexan-1-one (50.5 mg, 0.23 mmol),
[Ir(cod)Cl]2 (1.54 mg, 1 mol%), dppBz (2.1 mg, 2 mol%), KOH
(25.8 mg, 0.46 mmol) and cyclopropylmethanol (0.19 mL,
2.30 mmol) sequentially in the open atmosphere. The reaction
vessel was sealed with a microwave vial cap (containing a Reseal™
septum) and purged with Ar for 5 min using a balloon. Following
this, the vial (complete with an Ar balloon) was heated to 85 ꢀC in a
preheated oil bath for 24 h. The mixture was cooled to RT, filtered
through a SiO2 plug (eluting with Et2O) and concentrated in vacuo.
Purification by column chromatography (SiO2, eluent load, Penta-
ne:Et2O, 99:1 / 98:2) afforded the title compound 49 (54.1 mg,
86%) as a colorless oil. Rf ¼ 0.59 (Pentane:Et2O, 95:5), [UV, KMnO4];
IR (film) nmax/cmꢂ1 2955, 2929, 2859, 1689, 1611, 1571, 1457, 1428,
1378, 1355, 1305, 1297, 1274, 1240, 1231, 1212, 1160, 1133, 1103, 1077,
1034, 1016, 981; 1H NMR (CDCl3, 400 MHz)
d
¼ 6.83 (2H, s, 2 ꢃ Ar-
CH), 2.99 (1H, app. quint., J ¼ 6.4 Hz, CHCO), 2.27 (3H, s, Ar-CH3),
2.23 (6H, s, 2 ꢃ Ar-CH3), 1.76e1.66 (2H, m, CHAHBCHCO and
CHCHAHBCHCO), 1.53e1.43 (1H, m, CHAHBCHCO), 1.37e1.23 (5H, m,
CHCHAHBCHCO, CH2CH2CH3 and CH2CH3), 0.86 (3H, t, J ¼ 7.2 Hz,
CH3), 0.83e0.71 (1H, m, cyclopropyl CH), 0.47e0.41 (2H, m,
2 ꢃ cyclopropyl CHAHB), 0.08e0.02 (2H, m, 2 ꢃ cyclopropyl CHAHB);
Supernovae A diffractometer using Cu-K
a
radiation (
l
¼ 1.54184 Å)
and graphite monochromator. Samples were mounted in
a
perfluoropoly-ethyl ether oil and cooled by a Cryostream N2 open-
flow cooling device to 150 K throughout the data collection process.
The diffraction patterns were integrated and reduced using the
software CrysAlisPro. The software CRYSTALS for Microsoft Win-
dows was used to obtain ab initio solutions using SuperFlip [27]
embedded within CRYSTALS [28] and to carry out structure
refinement. Melting points (m.p.) were obtained using a Leica
VMTG heated-stage microscope equipped with a Testo 720 ther-
mometer and are uncorrected.
13C NMR (CDCl3, 101 MHz)
d
¼ 212.7 (C]O), 138.9 (AreC), 138.5
(AreC), 133.9 (2C, 2 ꢃ AreC), 130.0 (2C, 2 ꢃ Ar-CH), 53.6 (CHCO),
34.7 (CHCH2CHCO), 29.9 (CH2CHCO), 29.7 (CH2CH2CH3), 23.0
(CH2CH3), 21.2 (Ar-CH3), 20.0 (2C, 2 ꢃ Ar-CH3), 14.1 (CH3), 9.7
(cyclopropyl CH), 5.4 (cyclopropyl CAH2), 5.3 (cyclopropyl CBH2);
HRMS (ESIþ) Found [MþH]þ ¼ 273.22125; C19H29O requires
273.22129,
D 0.15 ppm.
Detailed experimental procedures, characterization data and
NMR spectra for all novel alcohols, substrates and alkylated prod-
ucts are provided in the Supporting Information.
4.4. Representative procedure for alkylation with 2ꢀ alcohols:
synthesis of 3-cyclopropyl-1-(2,3,4,5,6-pentamethylphenyl)butan-
1-one (101)
4.2. Representative procedure for a-alkylation of cyclopropyl
ketones: synthesis of 2-butyl-1-cyclopropylhexan-1-one (8)
To a 2e5 mL Biotage® microwave vial equipped with a stirrer
bar was added 1-(2,3,4,5,6-pentamethylphenyl)ethan-1-one 83
(114 mg, 0.60 mmol), [Cp*IrCl2]2 (9.6 mg, 2.0 mol%), 1-
cyclopropylethan-1-ol (120 mg, 1.20 mmol), PhMe (0.15 mL) and
NaOtBu (173 mg, 1.80 mmol) sequentially in the open atmosphere.
The reaction vessel was sealed with a microwave vial cap (con-
taining a Reseal™ septum) and an Ar balloon fitted. The vial was
heated to 85 ꢀC in a preheated oil bath for 24 h. The mixture was
cooled to RT, filtered through a SiO2 plug (eluting with Et2O) and
concentrated in vacuo. Purification by column chromatography
(SiO2, eluent load, Pentane:Et2O, 98:2) afforded the title compound
101 (149 mg, 96%) as a colourless solid. Rf ¼ 0.38 (Pentane:Et2O,
95:5), [UV, KMnO4]; m.p. ¼ 46e47 ꢀC; IR (film) nmax/cmꢂ1 3077,
2998, 2988, 2955, 2928, 2910, 2873, 1700, 1574, 1459, 1427, 1401,
1383,1370,1350,1314,1300,1270, 1129,1095,1070,1045,1016, 999;
To a 2e5 mL Biotage® microwave vial equipped with a stirrer
bar was added 1-cyclopropylhexan-1-one (42.0 mg, 0.30 mmol),
[Cp*IrCl2]2 (4.8 mg, 2 mol%), KOH (50.5 mg, 0.90 mmol) and nBuOH
(0.14 mL, 1.50 mmol) sequentially in the open atmosphere. The
reaction vessel was sealed with a microwave vial cap (containing a
Reseal™ septum) and purged with Ar for 5 min using a balloon.
Following this, the vial (complete with an Ar balloon) was heated to
105 ꢀC in a preheated oil bath for 24 h. The mixture was cooled to
RT, filtered through a SiO2 plug (eluting with Et2O) and concen-
trated in vacuo. Purification by column chromatography (SiO2,
eluent load, Pentane:Et2O, 99:1) afforded the title compound 8
(38.0 mg, 65%) as a colorless oil. Rf ¼ 0.56 (Pentane:Et2O, 90:10),
[vanillin]; IR (film) nmax/cmꢂ1 3009, 2957, 2930, 2873, 2859, 1695,
1467, 1458, 1417, 1382, 1196, 1159, 1063, 1022, 997; 1H NMR (CDCl3,
1H NMR (CDCl3, 400 MHz)
d
¼ 2.88 (1H, dd, J ¼ 19.1, 4.2 Hz,
400 MHz)
propyl CH), 1.70e1.59 (2H, m, 2 ꢃ CHAHBCHCO), 1.49e1.39 (2H, m,
CHAHBCHCO), 1.35e1.17 (8H, m, CH2CH2CH3 and
d
¼ 2.59e2.50 (1H, m, CHCO), 1.99e1.92 (1H, m, cyclo-
CHAHBCO), 2.65 (1H, dd, J ¼ 19.1, 8.4 Hz, CHAHBCO), 2.23 (3H, s, Ar-
CH3), 2.19 (6H, s, 2 ꢃ Ar-CH3), 2.12 (6H, s, 2 ꢃ Ar-CH3), 1.54e1.41
(1H, m, CHCH2CO), 1.13 (3H, d, J ¼ 6.7 Hz, CH3), 0.69e0.58 (1H, m,
cyclopropyl CH), 0.49e0.36 (2H, m, cyclopropyl CAHAHB and
cyclopropyl CBHAHB), 0.23e0.12 (2H, m, cyclopropyl CAHAHB and
2
ꢃ
2
ꢃ
2 ꢃ CH2CH2CH3), 1.02e0.97 (2H, m, cyclopropyl CAHAHBCH and
cyclopropyl CBHAHBCH), 0.88 (6H, t, J ¼ 7.1 Hz, 2 ꢃ CH3), 0.86e0.81
(2H, m, cyclopropyl CAHAHBCH and cyclopropyl CBHAHBCH); 13C
cyclopropyl CBHAHB); 13C NMR (CDCl3, 101 MHz)
140.9 (AreC), 135.4 (AreC), 133.2 (2C, 2 ꢃ AreC), 127.4 (2C,
d
¼ 211.3 (C]O),
NMR (CDCl3, 101 MHz)
d
¼ 215.0 (C]O), 53.6 (CHCO), 31.7 (2C,
10

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
2 ꢃ AreC), 53.1 (CH2CO), 33.5 (CHCH2CO), 20.1 (CH3), 18.2 (cyclo-
propyl CH), 17.1 (2C, 2 ꢃ Ar-CH3), 16.8 (Ar-CH3), 16.1 (2C, 2 ꢃ Ar-
CH3), 4.4 (cyclopropyl CAH2), 4.1 (cyclopropyl CBH2); HRMS (ESIþ)
References
[1] For representative reviews on Hydrogen Borrowing Catalysis, see:. (a)
M.H.S.A. Hamid, P.A. Slatford, J.M. Williams, J. Adv. Synth. Catal. 349 (2007)
1555;
Found [MþH]þ ¼ 259.20575; C18H27O requires 259.20564,
D
0.41 ppm.
(b) G.E. Dobereiner, R.H. Crabtree, Chem. Rev. 110 (2010) 681;
€
(c) S. Bahn, S. Imm, L. Neubert, M. Zhang, H. Neumann, M. Beller, Chem-
CatChem 3 (2011) 1853;
(d) S. Pan, T. Shibata, ACS Catal. 3 (2013) 704;
4.5. Representative procedure for Br2-mediated cleavage: synthesis
(e) C. Gunanathan, D. Milstein, Science 341 (2013) 1229712;
(f) J.M. Ketcham, I. Shin, T.P. Montgomery, M.J. Krische, Angew. Chem. Int. Ed.
53 (2014) 9142;
of butyl cis-2-(4-(benzyloxy)cyclohexyl)acetate (145)
To a 2e5 mL Biotage® microwave vial equipped with a stirrer
bar was added cis-2-(4-(benzyloxy)cyclohexyl)-1-(2,3,4,5,6-
pentamethylphenyl)ethan-1-one (119) (50.0 mg, 0.13 mmol) and
CH2Cl2 (0.64 mL) sequentially in the open atmosphere. The reaction
vessel was sealed with a microwave vial cap (containing a Reseal™
septum) and cooled to ꢂ17 ꢀC (ice/NaCl bath). Following this, Br2
(g) Y. Obora, ACS Catal. 4 (2014) 3972;
(h) Q. Yang, Q. Wang, Z. Yu, Chem. Soc. Rev. 44 (2015) 2305;
(i) A. Nandakumar, S.P. Midya, V.G. Landge, E. Balaraman, Angew. Chem. Int.
Ed. 54 (2015) 11022;
(j) J. Leonard, A.J. Blacker, S.P. Marsden, M.F. Jones, K.R. Mulholland,
R. Newton, Org. Process Res. Dev. 19 (2015) 1400;
(k) A. Corma, J. Navas, M. Sabater, Chem. Rev. 4 (2018) 1410;
(l) T. Kwok, O. Hoff, R.J. Armstrong, T.J. Donohoe, Chem. Eur J. 26 (2020) 12912.
[2] For examples, see:. (a) K. Taguchi, H. Nakagawa, T. Hirabayashi, S. Sakaguchi,
Y. Ishii, J. Am. Chem. Soc. 126 (2004) 72;
(13.0
until TLC analysis indicated complete consumption of the substrate
(typically 15 min). To this, was added nBuOH (35.0 L, 0.38 mmol)
mL, 0.26 mmol) was added dropwise and the mixture stirred
(b) M.S. Kwon, N. Kim, S.H. Seo, I.S. Park, R.K. Cheedrala, J. Park, Angew. Chem.
Int. Ed. 44 (2005) 6913;
m
dropwise at ꢂ17 ꢀC, and the reaction warmed to RT and stirred for
1 h. The reaction was diluted with Et2O (3 mL) and H2O (3 mL). The
layers were separated and the aqueous layer extracted with Et2O (ꢃ
3). The combined organics were washed with sat. aq. Na2S2O3, sat.
aq. NaHCO3, brine, dried (MgSO4) and concentrated in vacuo. Pu-
rification by column chromatography (SiO2, eluent load, Penta-
ne:Et2O, 95:5 / 90:10) afforded the title compound 145 (27.0 mg,
67%, dr > 95:5) as a colorless oil. Rf ¼ 0.40 (Pentane:Et2O, 90:10),
[KMnO4]; IR (film) nmax/cmꢂ1 2957, 2928, 2859, 1731, 1496, 1454,
1444,1389,1373,1356,1331,1310,1286,1256,1215,1156,1108,1091,
(c) C. Lofberg, R. Grigg, M. Whittaker, A. Keep, A. Derrick, J. Org. Chem. 71
(2006) 8023;
(d) S.J. Pridmore, J.M.J. Williams, Tetrahedron Lett. 49 (2008) 7413;
(e) Y. Iuchi, Y. Obora, Y. Ishii, J. Am. Chem. Soc. 132 (2010) 2536;
(f) L. Guo, Y. Liu, X. Yao, Z. Huang, Org. Lett. 15 (2013) 1144.
[3] (a) L.M.K. Chan, D.L. Poole, D. Shen, M.P. Healy, T.J. Donohoe, Angew. Chem.
Int. Ed. 53 (2014) 761;
(b) D. Shen, D.L. Poole, C.C. Shotton, A.F. Kornahrens, M.P. Healy, T.J. Donohoe,
Angew. Chem. Int. Ed. 54 (2015) 1642.
[4] F. Li, J. Ma, N. Wang, J. Org. Chem. 79 (2014) 10447.
[5] X. Quan, S. Kerdphon, P.G. Andersson, Chem. Eur J. 21 (2015) 3576.
[6] (a) Y. Iuchi, M. Hyotanishi, B.E. Miller, K. Maeda, Y. Obora, Y. Ishii, J. Org. Chem.
75 (2010) 1803;
1064, 1028, 990; 1H NMR (CDCl3, 400 MHz)
d
¼ 7.38e7.31 (4H, m,
(b) S. Ogawa, Y. Obora, Chem. Commun. 50 (2014) 2491;
(c) F.-X. Yan, M. Zhang, X.-T. Wang, F. Xie, M.-M. Chen, H. Jiang, Tetrahedron
70 (2014) 1193;
(d) K. Chakrabarti, M. Maji, D. Panja, B. Paul, S. Shee, G.K. Das, S. Kundu, Org.
Lett. 19 (2017) 4750.
Ph), 7.29e7.23 (1H, m, Ph), 4.50 (2H, s, CH2Ph), 4.07 (2H, t, J ¼
6.7 Hz, CH2O), 3.64e3.59 (1H, m, OCH), 2.24 (2H, d, J ¼ 7.2 Hz,
CH2COO), 1.96e1.82 (3H, m, CHCH2COO and 2 ꢃ OCHCHAHB),
1.65e1.56 (2H, m, CH2CH2O), 1.55e1.44 (6H, m, 2 ꢃ OCHCHAHB and
2 ꢃ CHAHBCHCH2COO), 1.43e1.32 (2H, m, CH2CH3), 0.93 (3H, t, J ¼
[7] A. Baeyer, V. Villiger, Ber. Dtsch. Chem. Ges. 32 (1899) 3625.
[8] (a) S. Musa, L. Ackermann, D. Gelman, Adv. Synth. Catal. 355 (2013) 3077;
(b) R. Madsen, I.S. Makarov, J. Org. Chem. 78 (2013) 6593;
(c) C. Chaudhari, S.M.A. Hakim Siddiki, K. Shimizu, Top. Catal. 57 (2014) 1042;
(d) K. Wang, L. Zhang, W. Tang, H. Sun, D. Xue, M. Lei, J. Xiao, C. Wang, Angew.
Chem. Int. Ed. 59 (2020) 11408;
(e) T.W. Ng, G. Liao, K.K. Lau, H.-J. Pan, Y. Zhao, Angew. Chem. Int. Ed. 59
(2020) 11384.
[9] T. Sawaguchi, Y. Obora, Chem. Lett. 40 (2011) 1055.
[10] (a) J.R. Frost, C.B. Cheong, W.M. Akhtar, D.F.J. Caputo, N.G. Stevenson,
T.J. Donohoe, J. Am. Chem. Soc. 137 (2015) 15664;
7.3 Hz, CH2CH3); 13C NMR (CDCl3, 101 MHz)
d
¼ 173.4 (C]O), 139.4
(AreC), 128.4 (2C, 2 ꢃ Ar-CH), 127.4 (3C, 3 ꢃ Ar-CH), 73.0 (OCH),
69.7 (CH2Ph),64.2 (CH2O), 41.5 (CH2COO), 34.0 (CHCH2COO), 30.8
(CH2CH2O), 29.3 (2C, 2 ꢃ OCHCH2), 27.3 (2C, 2 ꢃ CH2CHCH2COO),
19.3
(CH2CH3), 13.9
(CH2CH3);
HRMS
(ESIþ)
1.20 ppm.
Found
[MþH]þ ¼ 305.21149; C19H29O3 requires 305.21112,
D
(b) W.M. Akhtar, C.B. Cheong, J.R. Frost, K.E. Christensen, N.G. Stevenson,
T.J. Donohoe, J. Am. Chem. Soc. 139 (2017) 2577;
(c) W.M. Akhtar, R.J. Armstrong, J.R. Frost, N.G. Stevenson, T.J. Donohoe, J. Am.
Chem. Soc. 140 (2018) 11916;
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
(d) R.J. Armstrong, W.A. Akhtar, T.A. Young, F. Duarte, T. Donohoe, J. Angew.
Chem. Int. Ed. 58 (2019) 12558;
(e) R.J. Armstrong, W.M. Akhtar, J.R. Frost, K.E. Christensen, N.G. Stevenson,
T.J. Donohoe, Tetrahedron 75 (2019) 130680;
(f) S. Wübbolt, C.B. Cheong, J.R. Frost, K.E. Christensen, T. Donohoe, J. Angew.
Chem. Int. Ed. 59 (2020) 11339;
(g) A.E.R. Chamberlain, K.J. Paterson, R.J. Armstrong, H.C. Twin, T. Donohoe,
J. Chem. Commun. 56 (2020) 3563;
Acknowledgments
(h) D.M.J. Cheang, R.J. Armstrong, W.M. Akhtar, T.J. Donohoe, Chem. Commun.
56 (2020) 3543;
(i) For related annulation reactions involving earth abundant catalysts, see:
A. Kaithal, L.-L. Gracia, C. Camp, E.A. Quadrelli, W. Leitner J. Am. Chem. Soc.
141 (2019) 17487;
(j) A. Jana, K. Das, A. Kundu, P.R. Thorve, D. Adhikari, B. Maji, ACS Catal. 10
(2020) 2615;
(k) L. Bettoni, S. Gaillard, J.-L. Renaud, Chem. Commun. 56 (2020) 12909.
We are grateful to the following for financial support EPSRC
[J.R.F. and T.J.D., Established Career Fellowship (EP/L023121/1)],
A*STAR, Singapore (C.B.C.), GlaxoSmithKline (W.M.A.) and EPSRC
Centre for Doctoral Training in Synthesis for Biology and Medicine,
generously supported by AstraZeneca, Diamond Light Source,
Defence Science and Technology Laboratory, Evotec, Glax-
oSmithKline, Janssen, Novartis, Pfizer, Syngenta, Takeda, UCB and
Vertex (D.F.J.C., EP/L015838/1). Sean Guggiari is thanked for syn-
thesizing compounds 128 and 129. Dr. Roly J. Armstrong is thanked
for helpful comments.
[11] For other recent examples of
a-alkylation of ketones or similar, see:. (a)
C. Schlepphorst, B. Maji, F. Glorius, ACS Catal. 6 (2016) 4184;
(b) Y. Iuchi, M. Hyotanishi, B.E. Miller, K. Maeda, Y. Obora, Y. Ishii, J. Org. Chem.
75 (2010) 1803;
(c) S. Ogawa, Y. Obora, Chem. Commun. 50 (2014) 2491;
(d) F.-X. Yan, M. Zhang, X.-T. Wang, F. Xie, M.-M. Chen, H. Jiang, Tetrahedron
70 (2014) 1193;
Appendix A. Supplementary data
(e) F. Li, X. Zou, N. Wang, Adv. Synth. Catal. 357 (2015) 1405.
[12] (a) X. Cao, X. Wan, F. Yang, K. Li, X. Hao, T. Shao, X. Zhu, M. Song, J. Org. Chem.
83 (2018) 3657;
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tet.2021.132051.
(b) L. Kabadwal, J. Das, D. Banerjee, Chem. Commun. 54 (2018) 14069;
11

J.R. Frost, C.B. Cheong, W.M. Akhtar et al.
Tetrahedron 86 (2021) 132051
€
(c) A. Kaithal, M. Schmitz, M. Holscher, W. Leitner, ChemCatChem 11 (2019)
5287;
[19] Previous conditions for the cleavage of related (mesityl) functional groups
required the use of strong protic acids. For selected examples, see:. (a)
W.M. Schubert, H.K. Latourette, J. Am. Chem. Soc. 74 (1952) 1829;
(b) M.L. Bender, H. Ladenheim, M.C. Chen, J. Am. Chem. Soc. 83 (1961) 123;
(c) M.L. Bender, M.C. Chen, J. Am. Chem. Soc. 85 (1963) 37.
[20] A. El-Khawaga, R. Roberts, K.M. Sweeney, J. Org. Chem. 50 (1985) 2055.
[21] (a) A.D. Andreou, R.V. Bulbulian, P.H. Gore, D.F.C. Morris, E.L. Short, J. Chem.
Soc. Perkin Trans. 2 (1981) 830;
€
(d) A. Kaithal, M. Schmitz, M. Holscher, W. Leitner, ChemCatChem 12 (2020)
781;
(e) S. Thiyagarajan, C. Gunanathan, J. Am. Chem. Soc. 141 (2019) 3822;
(f) L. Bettoni, S. Gaillard, J.-L. Renaud, Org. Lett. 22 (2020) 2064;
(g) P. Chakraborty, M.K. Gangwar, B. Emayavaramban, E. Manoury, R. Poli,
B. Sundararaju, ChemSusChem 12 (2019) 3463;
(h) S. Waiba, S.K. Jana, A. Jati, A. Jana, B. Maji, Chem. Commun. 56 (2020) 8376;
(b) M. Kazunori, K. Yasuo, T. Hisakazu, M. Toshio, T. Ichiro, K. Hidehiko,
K. Takashi, Bull. Chem. Soc. Jpn. 67 (1994) 757.
(i) For an example without
a metal catalyst M.B. Dambatta, J. Santos,
R.R.A. Bolt, L.C. Morrill, Tetrahedron 76 (2020) 131571.
[22] For examples of diastereoselective amination under hydrogen borrowing
conditions, see:. (a) X. Xi, Y. Li, G. Wang, G. Xu, L. Shang, Y. Zhang, L. Xia, Org.
Biomol. Chem. 17 (2019) 7651;
(b) K. Fujita, T. Fujii, R. Yamaguchi, Org. Lett. 6 (2004) 3525.
[23] We assume that in situ deprotonation and isomerization from cis- to trans-
decalone occurred under the reaction conditions.
[24] (a) J. Sauvage, R.H. Baker, A.S. Hussey, J. Am. Chem. Soc. 82 (1960) 6090;
(b) B.E. Huff, V.V. Khau, M.E. LeTourneau, M.J. Martinelli, N.K. Nayyar,
B.C. Peterson, Tetrahedron Lett. 38 (1997) 8627.
[13] Purification of the crude mixture did not allow for these three components to
be easily separated, and therefore they were isolated as a mixture, with the
product ratio determined based on the mass obtained and analysis of the 1H
NMR spectrum. We also identified alcohol 4 in the crude mixture, but this was
not quantified as it was difficult to separate from excess BnOH.
[14] (a) L.J. Allen, R.H. Crabtree, Green Chem. 12 (2010) 1362;
(b) Q. Xu, J. Chen, H. Tian, X. Yuan, S. Li, C. Zhou, J. Liu, Angew. Chem. Int. Ed. 53
(2013) 225. See also ref 12i.
[15] (a) T. Nakai, E. Wada, M. Okawara, Tetrahedron Lett. 16 (1975) 1531;
(b) C.U. Pittman, S.P. McManus, J. Am. Chem. Soc. 91 (1969) 5915.
[16] C. Mioskowski, S. Manna, J.R. Falck, Tetrahedron Lett. 24 (1983) 5521.
[25] Substrate 113 was also treated under our previously reported a-methylation
conditions. In this instance only unreacted starting material was isolated.
[26] (a) For a review on HFIP, see: I. Colomer, A.E.R. Chamberlain, M.B. Haughey,
T.J. Donohoe Nat. Rev. Chem. 1 (2017) 88;
[17] In similar fashion to cyclopropyl ketones, we also noted that
a-branched
product formed in the absence of catalyst when BnOH was used. Once again
this occurred in lower yield compared to the optimized system, with alkyl
alcohols again being unreactive in the absence of catalyst.
(b) C.B. Cheong, J.R. Frost, T. Donohoe, J. Synlett. 31 (2020) 1828.
[27] L. Palatinus, G.J. Chapuis, Appl. Crystallogr. 40 (2007) 786.
[28] (a) P. Parois, R.I. Cooper, A.L. Thompson, Chem. Cent. J. 9 (2015) 30;
(b) R.I. Cooper, A.L. Thompson, D.J. Watkin, J. Appl. Crystallogr. 43 (2010)
1100.
[18] P. Sykes, In A Guidebook to Mechanism in Organic Chemistry, John Wiley &
Sons, New York, 1986.
12