Cyclopropenium-activated cyclodehydration of diols

Article abstract of DOI:10.1021/ol102980t

The dehydrative cyclization of diols to cyclic ethers via cyclopropenium activation is described. Using 2,3-diphenylcyclopropene and methanesulfonic anhydride, a series of 1,4-and 1,5-diols are rapidly cyclized to furnish tetrahydrofurans and tetrahydropyrans in high yield. Eleven total substrates are shown, including a gram scale cyclization of a diterpene derivative.

Full text of DOI:10.1021/ol102980t

ORGANIC
LETTERS
2011
Vol. 13, No. 4
740–743
Cyclopropenium-Activated
Cyclodehydration of Diols
Brendan D. Kelly and Tristan H. Lambert*
Department of Chemistry, Columbia University, New York, New York 10027,
United States
tl2240@columbia.edu
Received December 9, 2010
ABSTRACT
The dehydrative cyclization of diols to cyclic ethers via cyclopropenium activation is described. Using 2,3-diphenylcyclopropene and
methanesulfonic anhydride, a series of 1,4- and 1,5-diols are rapidly cyclized to furnish tetrahydrofurans and tetrahydropyrans in high yield.
Eleven total substrates are shown, including a gram scale cyclization of a diterpene derivative.
Chemical processes that involve the direct or formal loss
of water, collectively termed dehydration reactions, repre-
sent perhaps the most widely utilized and important class
of reactions in organic synthesis. Included in this category
are such ubiquitous transformations as amidation (e.g.,
peptide coupling), esterification, glycosylation, ether for-
mation, and Mitsunobu inversion, to name a few. Despite
this importance, the traditional reagents used to effect sub-
stitution reactions often suffer from numerous limitations,
which may include reagent harshness, lack of substrate
scope, poor reactivity, offensive byproduct formation,
and, most notably, a profound lack of amenability to
enantioselective catalysis.
In an effort to address these limitations, we recently
disclosed a novel strategy for the promotion of dehydra-
tion reactions using cyclopropenium ions, a process we
termed “cyclopropenium activation” (Figure 1a).1 With
this approach, hydroxyl-bearing substrates such as
alcohols,1a carboxylic acids,1b or oximes1c are treated with
a cyclopropene of the type 1 bearing geminal leaving
groups, typically generated in situ from the corresponding
cyclopropenone 2. The propensity of cyclopropenes 1 to
ionize due to aromatic stabilization of the resulting 2π-
electron cyclopropenium ions2,3 leads to a rapid combina-
tion with hydroxylic substrates to produce intermediates 3.
As electron-deficient carbocations, these “cycloprope-
nium-activated” intermediates are highly susceptible to
nucleophilic displacement of the neutral cyclopropenone
to produce substitution or rearrangement products. Our
initial reports1 of this concept described the potency and
synthetic utility of cyclopropenium activation in the con-
text of alcohol chlorodehydration, carboxylic acid chloro-
dehydration, and Beckmann rearrangement (Figure 1b).
Each of these processes ultimately relies on the chloride ion
to serve as both the cyclopropenyl leaving group and the
nucleophilic species.
In this Letter, we report the advancement of our cyclo-
propenium activation strategy with the invention of a
powerful and convenient new strategy to achieve the
cyclodehydration of diols 4 to produce cyclic ethers 5. This
(2) (a) Breslow, R. J. Am. Chem. Soc. 1957, 79, 5318. (b) Breslow, R.;
Yuan, C. J. Am. Chem. Soc. 1958, 80, 5991.
(3) For a review on cyclopropenium ions, see: Komatsu, K.; Kitagawa,
(1) (a) Kelly, B. D.; Lambert, T. H. J. Am. Chem. Soc. 2009, 131,
13930. (b) Hardee, D. J.; Kovalchuke, L.; Lambert, T. H. J. Am. Chem.
Soc. 2010, 132, 5002. (c) Vanos, C. M.; Lambert, T. H. Chem. Sci. 2010,
1, 705.
T. Chem. Rev. 2003, 103, 1371.
r
10.1021/ol102980t
Published on Web 01/19/2011
2011 American Chemical Society

Figure 2. Mechanistic design for cyclopropenium ion mediated
cyclization of diols to form cyclic ethers.
with extreme convenience through the use of reagents
readily amenable to steric and electronic modulation.
In the design of a cyclopropenium activated cyclodehy-
dration reaction, we expected that a diol substrate 6 would
react with an activated cyclopropene reagent 7 to form an
equilibrium mixture of cyclopropene acetal6 8 and the
corresponding cyclopropenium ether 9 (Figure 2). Intra-
molecular nucleophilic closure of this intermediate would
then furnish cyclic ether 10 with concomitant production
of cyclopropenone 11. Key to the success of this design
would of course be the ability to readily access monoacti-
vated intermediate 9 and the ability of the cyclopropenium
ion to sufficiently activate a hydroxyl group for displace-
ment by the relatively poor nucleophilic alcohol moiety.
To test the feasibility of this proposed design, we first
examined the reaction of commercially available (S,S)-2,5-
hexanediol (12) with 1,1-dichloro-2,3-diphenylcyclopro-
pene (15), prepared in situ from commercially available
2,3-diphenylcyclopropenone 13 by treatment with oxalyl
chloride. In fact, when 12 and was added to a mixture of 13
and oxalyl chloride in acetonitrile at room temperature,
(R,S)-2,5-dimethyltetrahydrofuran (14) was produced in
93% yield (1H NMR) as a 12:1 diastereomeric mixture
after 2 h (eq 1). This experiment thus demonstrates the
power of cyclopropenium activation to effect dehydration
in the context of ether linkages and is notable for its
relatively high rate of reaction, the complete absence of
elimination or chlorination side products, and the lack of
need for base to effect ring closure.
Figure 1. Cyclopropenium activation.
work encompasses two key developments in our cyclopro-
penium program, namely, (1) demonstration of the effec-
tiveness of alternative cyclopropenyl leaving groups in the
form of methanesulfonate ions and (2) extension of the
nucleophilic species to include alcohols.
Cyclic ethers are a prevalent motif in biologically active
compounds. Accordingly, numerous synthetic methods
have been developed to access this important class of
heterocycles.4 Although the dehydrative cyclization of
diols is one of the most conceptually straightforward
strategies by which to construct cyclic ethers, from a
practical standpoint methods to achieve these transforma-
tions often suffer from lack of selectivity, the need for
forcing conditions, or the use of reagents that are incon-
venient to prepare, leave undesirable byproducts, and lack
broad tunability.5 In contrast, we reasoned that cyclopro-
penium activation could offer a powerful alternative strat-
egy for diol cyclodehydration by combining high reactivity
(4) Heitmann, W.; Strehlke, G.; Mayer, D. Ethers, Aliphatic. In
Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim,
Germany, 2002.
(5) See the following selected papers for diol cyclodehydrations
using: Platinium: (a) Shibata, T.; Fujiwara, R.; Ueno, Y. Synlett,
2005, 152. Triphenylphosphine: (b) Barry, C. N.; Evans, S. A., Jr. J.
Org. Chem. 1981, 46, 3361. (c) Robinson, P. L.; Barry, C. N.; Kelly,
J. W.; Evans, S. A., Jr. J. Am. Chem. Soc. 1985, 107, 5210. DMSO:(d)
Gillis, B. T.; Beck, P. E. J. Org. Chem. 1963, 28, 1388. Silver(I) oxide: (e)
Bouzide, A.; Sauve, G. Org. Lett. 2002, 4, 2329. Sulfuranes:(f) Martin,
J. C.; Franz, J. A.; Arhart, R. J. J. Am. Chem. Soc. 1974, 96, 4604.
Triphosphonitrilic chloride:(g) Matuszko, A. J.; Chang, M. S. J. Org.
On the other hand, the incomplete stereospecificity
observed in this transformation suggested that simple
SN2 closure is not the exclusive reaction pathway. In
addition, we found that certain other substrates, especi-
ally those possessing primary alcohols, were prone to
ꢀ
€ €
ꢀ
Chem. 1966, 31, 2004. (h) Torok, B.; Bucsi, I.; Beregszaszi, T.; Kapocsi, I.;
ꢀ
Molnar, A. J. Mol. Catal. A: Chem. 1996, 107, 305. Heteropoly acids:(i)
Duffy, M. G.; Grayson, D. H. J. Chem. Soc., Perkin Trans. I 2002, 1555.
Methanesulfonyl chloride:(j) Mihailovic, M. L.; Gojkovic, S.; Cekovic,
Z. J. Chem. Soc., Perkin Trans. I 1972, 2460.
(6) For a review on cyclopropenone acetals, see: Nakamura, M.;
Isobe, H.; Nakamura, E. Chem. Rev. 2003, 103, 1295.
Org. Lett., Vol. 13, No. 4, 2011
741

competitive chlorination. Thus for example diol 16 was
found to react to produce a mixture of 2-benzyltetrahy-
drofuran 17 and chloride 18 in 76% and 20% yields,
respectively (eq 2). This finding was not surprising, given
our previous description of 15 as a potent chlorodehydrat-
ing reagent. To solve this problem, we thus decided to
investigate the use of alternative cyclopropenes with the
intention of identifying a superior reagent.
leads to more stable (higher pKRþ) cyclopropenium ions,
significantly improved the rate of cyclic ether formation
(cf. 19, entry 3). Not surprisingly, we found that cyclopro-
penes bearing methanesulfonate (mesylate) groups were
significantly more effective than those bearing trifluoro-
acetoxy substituents. For example, methanesulfonic anhy-
dride (Ms2O) and 2,3-diphenylcyclopropenone effected
the cyclization of diol 12 in 95% yield after 2.5 h (entry
4), a reaction time only slightly longer than that observed
with the dichlorocyclopropene 15. Interestingly, changing
from phenyl to isopropyl substituents did not offer any
benefit in this case (entry 5). It is important to note that, in
contrast to the dichlorocyclopropene example noted
In this regard, we took note of the fact that the reaction
of cyclopropenones with trifluoroacetic anhydride (TF-
AA) had previously been shown7 to produce the corre-
sponding 1,1-bistrifluoroacetoxycyclopropenes, a rather
unusual transformation for ketone functionality. We thus
decided to investigate whether various acid anhydrides
could take the place of oxalyl chloride in our activation
strategy, and thereby minimize or eliminate the problems
plaguing the chloride-based reagents.
Table 2. Substrate Scope Studies for Dehydrative Cyclization of
Diolsa,b
Table 1. Optimization Studies for Dehydrative Cyclization of
Diolsa,b
activating
agent
syn:
antic
entry
R
time (h) yield (%)
1
2
3
4
5
6
7
8
Ph
13
13
19
13
19
(COCl)2
TFAA
TFAA
Ms2O
2
24
24
93
7
12:1
>20:1
>20:1
>20:1
>20:1
Ph
i-Pr
Ph
38
95
91
0
2.5
2.5
24
i-Pr
none
none
none
Ms2O
(COCl)2
TFAA
Ms2O
24
24
0
0
a Reactions were performed by the addition of “activating agent”
(1.5 equiv) to a solution of cyclopropenone (1.6 equiv) in CD3CN. After
30 min, diol 12 was added to the solution. b Yields determined by 1H
NMR analysis using internal standard (Bn2O). c Diastereomeric ratio
determined by 1H NMR.
As before, the following series of experiments were
conducted by in situ activation of cyclopropenone by
treatment with a given anhydride, followed by addition
of the diol 12 to the resulting cyclopropene solution. Thus
the use of TFAA and diphenylcyclopropenone 137 resulted
in low yet appreciable conversion of diol 12 to ether 14
(7%) after 24 h (Table 1, entry 2). Notably, changing the
cyclopropene substituents from phenyl to isopropyl, which
a General reaction conditions: Reactions were performed by the
addition of methanesulfonic anhydride (1.1 equiv) to a solution of
cyclopropenone 13 (1.2 equiv) in CH2Cl2. After 30 min, the diol
substrate was added to the solution. b Yields determined on isolated
and purified products unless otherwise noted. c Yield determined by 1H
NMR analysis using internal standard (Bn2O). d 1.4 equiv of 13 and 1.3
equiv of Ms2O were used.
(7) Oda, M.; Breslow, R.; Pecoraro, J. Tetrahedron Lett. 1972, 13,
4415.
742
Org. Lett., Vol. 13, No. 4, 2011

above, diastereomeric integrity was retained in all cases
with the use of anhydride activating agents. It should also
be stressed that no cyclization products were observed with
the use of oxalyl chloride, trifluoroacetic anhydride, or
mesyl anhydride in the absence of cyclopropenone (entries
6-8), or with the use of HCl, TFA, or methanesulfonic
acid (see the Supporting Information).
(entry 10), suggesting that this strategy can also be effective
for the preparation of benzocyclic ethers.
Finally, in order to demonstrate the potential prepara-
tive utility of this strategy, we examined the cyclization of
diol 38, which is the hydroboration/oxidation adduct of
(-)-isopulegol (eq 3). In the presence of 2,3-diphenylcy-
clopropenone (13) and methanesulfonic anhydride in
CH2Cl2 at room temperature, this substrate underwent
dehydrative cyclization to produce the bicyclic ether 39 in
95% yield over 8 h with complete stereospecificity on a
gram scale. For comparison, treatment of 38 under the
same conditions with methanesulfonic anhydride itself
resulted in no product, and in the presence of triethylamine
led to only 18% product over 12 h. Clearly the cyclopro-
penone provides a powerful enhancement of the dehydrat-
ing ability of methanesulfonic anhydride.
Substrate scope studies have shown 2,3-diphenylcyclo-
propenone (13) and Ms2O to be an effective reagent
combination for the dehydrative cyclization of 1,4- and
1,5-diols (Table 2).8 For example, the conversion of diol 16
to 2-benzyltetrahydrofuran (17) was readily achieved in
91% yield after 4 h at room temperature (entry 1). Similar
efficiency was found for the one-carbon homologated
substrate 20 for the production of the corresponding
tetrahydropyran 21, albeit with an expectedly longer reac-
tion time (entry 2). We note that, although dichlorocyclo-
propene 13 was only poorly effective for these substrates
due to competitive chlorination (see eq 2), the use of
mesylate counterions resulted in no observable sulfonate
substitution products. Indeed, even the bis-primary alco-
hol substrate 22 could be converted to tetrahydropyran 23
in high yield without complication (entry 3). In addition,
benzylic alcohol 24 (entry 4), R-hydroxyester 26 (entry 5),
and protected 1,2-diol 28 (entry 6) were all found to be
productive for dehydrative cyclization. Importantly, diol
28 was optically enriched, and stereochemical analysis of
the product 29 revealed that cyclization occurred exclu-
sively via nucleophilic substitution of the primary alcohol
by the secondary hydroxyl. Furthermore, despite its po-
tential sensitivity nitroalcohol 30 underwent cyclodehy-
dration without any observable β-elimination (entry 7). As
further proof of the SN2 nature of this reaction, we found
that the stereochemically complex tetra-O-benzylmannitol
(32) cleanly converted to tetrahydrofuran 33 in good yield
as a single diastereomer (entry 8). Cyclization in an SN20
sense was also facile, producing 2-vinyltetrahydrofuran
(35) in 95% yield (entry 9). Interestingly, a phenolic
hydroxyl was found to be a capable nucleophile for the
preparation of 2-methylchroman (31) with good efficiency
In conclusion, we have extended the concept of cyclo-
propenium activation with the demonstration of new acti-
vating agents and an alternative nucleophile in the context
of a novel dehydrative cyclization of 1,4- and 1,5-diols. This
method offers a potent and convenient means to construct
stereodefined tetrahydrofurans and pyrans from their cor-
responding acyclic diol precursors. Further development of
this strategy and application of cyclopropenium activation
to other reaction manifolds is currently underway.
Acknowledgment. T.H.L. is grateful for an Alfred P.
Sloan Research Fellowship and Young Investigator
Awards from Abbott and Amgen.
Supporting Information Available. Full experimental
details and characterization data. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
(8) Efforts to extend the current protocol to the formation of ring
sizes smaller than 5 or greater than 6 have not yet been successful.
Org. Lett., Vol. 13, No. 4, 2011
743