Mechanistic insights into the vanadium-catalyzed achmatowicz rearrangement of furfurol

Article abstract of DOI:10.1021/jo502641d

The Achmatowicz rearrangement is a powerful method for the construction of pyranones from simple furan derivatives. Here, we describe the development of improved reaction conditions and an interrogation into the fate of the metal center during this interesting transformation. The reaction to form the synthetically important lactol, 6-hydroxy-2H-pyran-3(6H)-one (3), proceeds cleanly in the presence of tert-butyl hydroperoxide (TBHP, 2) using low loadings of VO(OⁱPr)₃ as catalyst. The nonaqueous conditions developed herein allow for easy isolation of product 3 and synthetically important derivatives, a key advantage of this new protocol. Detailed experimental, spectroscopic, and kinetic studies along with kinetic modeling of the catalytic cycle support a positive-order dependence in both furfurol and TBHP concentrations, first-order dependence in catalyst (VO(OⁱPr)₃), and a negative dependence on the 2-methyl-2-propanol (4) concentration. ⁵¹V-NMR spectroscopic studies revealed that 2-methyl-2-propanol (4) competes with substrates for binding to the metal center, rationalizing its inhibitory effect.

Full text of DOI:10.1021/jo502641d

Article
pubs.acs.org/joc
Mechanistic Insights into the Vanadium-Catalyzed Achmatowicz
Rearrangement of Furfurol
†
,‡
‡
‡
‡
Yining Ji, Tamas Benkovics,* Gregory L. Beutner, Chris Sfouggatakis, Martin D. Eastgate,
,†
and Donna G. Blackmond*
†Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, United Sates
‡Chemical Development, Bristol-Myers Squibb, One Squibb Drive, New Brunswick, New Jersey 08903, United States
*
S Supporting Information
ABSTRACT: The Achmatowicz rearrangement is a powerful
method for the construction of pyranones from simple furan
derivatives. Here, we describe the development of improved
reaction conditions and an interrogation into the fate of the
metal center during this interesting transformation. The
reaction to form the synthetically important lactol, 6-
hydroxy-2H-pyran-3(6H)-one (3), proceeds cleanly in the
presence of tert-butyl hydroperoxide (TBHP, 2) using low
i
loadings of VO(O Pr) as catalyst. The nonaqueous conditions
3
developed herein allow for easy isolation of product 3 and
synthetically important derivatives, a key advantage of this new protocol. Detailed experimental, spectroscopic, and kinetic studies
along with kinetic modeling of the catalytic cycle support a positive-order dependence in both furfurol and TBHP
i
concentrations, first-order dependence in catalyst (VO(O Pr) ), and a negative dependence on the 2-methyl-2-propanol (4)
3
concentration. 51V-NMR spectroscopic studies revealed that 2-methyl-2-propanol (4) competes with substrates for binding to
the metal center, rationalizing its inhibitory effect.
INTRODUCTION
enable a facile isolation of the product, which meant eliminating
water both in the reaction mixture and during the workup
protocol. On the basis of literature reports, it was clear that
■
The Achmatowicz rearrangement of furans to provide
substituted pyranones is a powerful method for the preparation
of these densely functionalized building blocks. The trans-
formation is widely used in the synthesis of carbohydrates as
well as other natural products. Reagents such as m-
chloroperoxybenzoic acid or bromine−MeOH have been
routinely used for this transformation but suffer from significant
drawbacks from the standpoint of atom economy, process
efficiency, and ease of product isolation. More environmentally
benign oxidants such as hydrogen peroxide have been utilized,
but the presence of water in these systems often results in
serious challenges during isolation since many of these
pyranone derivatives have significant water solubility. This is
particularly true for the simplest product of this transformation
Scheme 1), 6-hydroxy-2H-pyran-3(6H)-one (3), generated
from furfurol (1).
In the context of a recent program, we wished to find a more
atom-economical and efficient method for the preparation of 3
from furfurol. Importantly, we desired conditions that would
8
1
tert-butyl hydroperoxide (TBHP, 2) was a competent oxidant
for this transformation, functioning in an anhydrous medium
that would allow for either the direct isolation of product 3 or
its direct use in a subsequent transformation. In this report, we
describe the development of a robust reaction and isolation
method for the Achmatowicz rearrangement of furfurol using
2
3
4
5
i
catalytic VO(O Pr)3 (Scheme 1), as well as mechanistic
6
51
investigations based on a combination of kinetic analysis, V-
7
NMR spectroscopic identification of catalytic intermediates,
and kinetic modeling to provide a more detailed understanding
of this transformation.
RESULTS AND DISCUSSION
■
(
In order to evaluate the performance of the Achmatowicz
reaction, we decided to investigate a range of oxidants and
vanadium catalysts. After optimization of the reaction, varying
i
vanadium source (VO(acac) vs VO(O Pr) ), oxidant (TBHP
2
3
vs cumene hydroperoxide), solvent, temperature, and addition
order, we were delighted to find that the oxidative rearrange-
ment gave the desired product in good in-process yield
Scheme 1. Achmatowicz Rearrangement of Furfurol
(
Scheme 2). Although detailed studies of temperature effects
were not carried out, the reaction proceeds cleanly at
Received: November 19, 2014
©
XXXX American Chemical Society
A
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Scheme 2. Preparation of Compounds 3, 5, and 6
Table 1. Reaction Conditions
reaction
[1] (M)
[2] (M)
comment
standard
0
0
a
b
c
d
e
f
1.0
0.5
0.5
0.5
0.5
1.0
1.0
1.0
1.5
1.0
1.5
1.0
1.0
1.5
1.5
1.5
same excess
different excess
0.4 M product 3 addition
0.4 M product 4 addition
i
0.5 M PrOH addition
i
g
h
1.0 M PrOH addition
t
VO(O Bu) catalyst
3
a
All reactions run in DCM at room temperature using 0.02 M
i
t
VO(O Pr) (VO(O Bu) for reaction h) as catalyst.
3
3
tration. Both protocols employ a simple graphical manipulation
of the reaction data to extract this information.
Figure 2 plots the concentration profiles for two “same
excess” experiments (reactions a and b, Table 1) designed to
temperatures between 10 °C and room temperature. Addition-
ally, lactol 3 could be isolated after quenching the residual
oxidant with trimethylphosphite followed by a direct
t
crystallization from heptanes/ BuOH. Alternatively, the crude
reaction mixture of 3 could be carried directly into an acylation
step with various anhydrides to provide improved access to
these useful synthetic building blocks as crystalline solids.
Although the vanadium-catalyzed furfurol oxidation pro-
duced 3 in a reproducible fashion, the reaction appeared to stall
despite an initial rapid reaction rate. In order to understand this
phenomenon in more detail, we undertook a detailed kinetic
i
study of the VO(O Pr) /TBHP system. FTIR spectroscopy was
3
employed for continuous monitoring of the reaction. The
formation of product 3 was monitored using the height of the
−1
peak at ca. 1708 cm , and consumption of furfurol (1) was
followed by a multicomponent method (concIRT). Validation
of this in situ method was carried out by comparison to an
alternative method; in this case, H NMR spectroscopy, where
an excellent correlation was observed (Figure 1). To character-
Figure 2. Concentration of substrate 1 as a function of time for
reaction a compared with “same excess” reaction b as noted in Table 1.
The time-adjusted profile for reaction b has the same [1] and [2] as
reaction a has at the point marked by the dotted lines.
1
test catalyst stability. As the dotted lines show, the
concentrations of furfurol (1) and TBHP (2) at the beginning
of reaction b (red circles) are identical to those of reaction a
(
blue circles) after ca. 4.5 min of reaction; i.e., reaction b is
started with a composition mimicking reaction a after 4.5 min.
Comparison of the profiles of these two reactions is then
possible by shifting the curve from reaction b to the appropriate
time on the x-axis. After this transformation, identical reaction
profiles should be expected if the activity of the catalyst remains
constant throughout the reaction. However, it is clear that
reaction b proceeds significantly faster than reaction a. The lack
of overlay of these “time adjusted” kinetic profiles suggests that
some process, other than the intrinsic kinetics related to the
temporal substrate concentrations of [1] and [2], contributes
to the observed rate changes.
Figure 1. Comparison of temporal kinetic profiles monitored by FTIR
and NMR spectroscopy of reactant 1 and product 3 for the reaction of
Scheme 1.
ize the kinetic behavior of the system, a series of experiments
was designed and performed (Table 1). Both the concentration
dependence and the stability of the catalyst system could then
be probed using the “different excess” and “same excess”
protocols through reaction progress kinetic analysis (RPKA),
Two key differences exist between the reactions a and b at
the point marked by the dotted arrows shown in Figure 2; (i)
the catalyst in reaction a has undergone several turnovers, while
reaction b employs fresh catalyst, and (ii) reaction a is at ca.
9
respectively.
t
Comparison of kinetic profiles under standard and “same
excess” conditions is a technique that has been employed
previously to probe catalyst robustness. Comparison of two or
more “different excess” experiments can also provide
information about the reaction order in substrate concen-
50% conversion and contains 0.4 M 3 and 4 BuOH while
reaction b contains only substrate and catalyst. Thus, two main
possibilities exist for the lack of overlay between curves a and b,
from the time-adjusted point onward: either the concentration
of catalyst in the productive cycle is decreasing through
B
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
irreversible deactivation or a reaction product is inhibiting
catalyst turnover.
a function of concentration. The “power-law” form of the rate
expression can be represented by eq 1:
Product inhibition can be independently tested by carrying
out a reaction under the conditions of reaction b while adding
the appropriate quantity of products 3 and/or 4 to mimic the
time-adjusted point. When the conditions of reaction b were
repeated in the presence of 0.4 M 3 (Figure 3, reaction d),
excellent superposition was observed, suggesting that lactol 3
does not interfer with the active catalysts turnover.
m
n
p
rate = k [1] [2] [4]
(1)
obs
n t
p
Normalization of the rate by [TBHP] [ BuOH] is shown as eq
2:
rate
2] [4]
m
=
k [1]
n
p
obs
[
(2)
The power−law values for reaction orders m, n, and p are
thus obtained from RPKA plots of eq 2 for multiple data sets,
with the function to the left of the equal sign plotted as the y-
axis and the function [1]m on the x-axis. Correctly chosen
values for m, n, and p would therefore produce straight line
correlations exhibiting “overlay” for a set of kinetic data from
reactions carried out under “same excess”, “different excess”,
and product added conditions. As illustrated in Figure 5, a
Figure 3. Effect of product 3 concentration on the temporal
concentration profile of reaction b. Reaction d includes product 3
added to match its concentration of reaction a at the point of the time
adjustment. Reaction d monitored using NMR spectroscopy.
In contrast, when the conditions of reaction b were repeated
in the presence of 0.4 M 4 (Figure 4, reaction e), a slower
Figure 5. RPKA plot of the function shown in eq 2 with m = 1, n = 1,
p = −1 for reaction entries a, b, c, and e of Table 1.
straight line with overlay between the four experimental profiles
is obtained with m = 1, n = 1, and p = −1. The linearity
confirms that the reaction is approximately first order in [1] (m
=
1), while the overlay between the profiles suggests that the
reaction exhibits an approximate first-order kinetic response (n
=
1) to substrate [2] and a negative first-order response (p =
1) to [4]. A demonstration of the effect on “overlay” by other
−
choice of reaction orders m, n, and p is given in the Supporting
Information.
Figure 4. Concentration of substrate 1 as a function of time for
reaction as compared to “same excess” conditions reaction e, with
product 4 added to a concentration close to that of reaction a at the
point of the time adjustment.
With an understanding of the orders of the different species
in solution, we wished to gain more knowledge on the
inhibitory effect of BuOH and therefore sought an alternative
t
technique that could provide more direct information on the
catalytic species. Thus, 51V NMR spectroscopy was employed
t
reaction was observed. Good superposition of reaction e was
observed with standard reaction (profile a), demonstrating that
the two reactions exhibit similar rates and suggesting that the
to probe the structural basis of the inihibitory effect of BuOH
on the reaction. The 51V isotope is perfectly suited to this task
due to the high natural abundance of this nucleus (N = 99.76%,
t
1
catalyst system is sensitive to BuOH inhibition. In addition, the
I = 7/2) and its high relative receptivity compared to H
10
reasonable overlay of these two profiles suggests that other
catalyst deactivation processes are likely not as significant as
alcohol inhibition under these conditions. The kinetic behavior
appears robust, consistent with the vanadium catalyst remaining
viable in the reaction medium throughout the reaction.
Having confirmed the inhibitory effect of the alcohol
byproduct, 2-methyl-2-propanol (4), as well as the absence of
any significant irreversible catalyst deactivation, we employed
the “different excess” protocol of RPKA to probe the reaction
orders in substrates [1] and [2]. The standard RPKA analysis
involves consideration of the data in the form of reaction rate as
(0.381). This technique has been widely used to study ligand-
exchange processes occurring at vanadium(V) atoms and is
sensitive to subtle changes in ligand structure due to the wide
11
chemical shift range for this nucleus (2000 ppm). We
hypothesized that this technique could allow for direct
observation of the catalytic species in solution and provide
further insight into the inhibition process as well as the overall
reaction mechanism.
First, the complexes formed from the interaction between
i
t
VO(O Pr) and BuOH (4) were investigated (Figure 6). A
3
i
t
mixture of VO(O Pr) and 4 equiv of BuOH (4) led to the
3
C
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
hydroperoxides were found to be insoluble under reaction
conditions.
Given the large number of alkoxyvanadium(V) complexes
that could be formed between the vanadium catalyst and the
various alcohols in solution, we sought to understand this
system in more detail. Therefore, the possible combinations of
t
furfurol (1), TBHP (2), lactol 3, and BuOH (4) were explored
i
in pairwise combination with VO(O Pr) . Similar studies have
3
been performed on other homogeneous vanadium-catalyzed
15
oxidations in order to characterize key reaction intermediates.
Selected examples from our studies are shown in Figure 9, and
full details of these structure elucidation experiments can be
found in the Supporting Information. For easy reference, the
values have been compiled in Table 2.
5
1
i
Figure 6. 131 MHz V NMR spectra of a solution of (a) VO(O Pr) ,
3
Having independently identified numerous vanadium com-
i
t
t
(
b) VO(O Pr) : BuOH (1:4), and (c) VO(O Bu) . All of the spectra
were recorded in CD Cl .
51
3
3
plexes potentially formed in the reaction mixture, in situ
V
2
2
NMR spectra were recorded over the course of the reaction
under the standard conditions (Table 1, reaction h). For these
formation of three new complexes which were assigned as
t
51
studies, VO(O Bu) was chosen as the catalyst since simpler V
i
t
i
t
t
3
VO(O Pr) (O Bu), VO(O Pr)(O Bu) , and VO(O Bu) (−639,
2
2
3
NMR spectra would be obtained in the absence of isopropoxide
−
654, and −671 ppm, respectively; Figure 6b). An authentic
i
t
t
12
groups. Since the reactions of VO(O Pr) and VO(O Bu)
3
3
sample of VO(O Bu) was prepared confirming its resonance
3
proceeded at similar rates (see Figure 7), the conclusions of the
study should prove general. By monitoring the reaction mixture
under reaction conditions h by 51V NMR over time, we were
able to identify almost all observable species formed during the
reaction mixture (Figure 8).
at −671 ppm (Figure 6c). These experiments demonstrated a
i
t
facile interchange between PrOH and BuOH at the vanadium-
(
V) center, an observation fully consistent with previous
13,14
studies.
t
Thus, we hypothesized that the inhibitory effect of BuOH
The kinetic profiles of these different vanadium species as a
could be caused by one of two factors: (1) the new species
i
t
i
t
t
function of time are plotted in Figure 9 along with fractional
[
VO(O Pr) (O Bu), VO(O Pr)(O Bu) , and VO(O Bu) ] could
2 2 3
t
conversion. At low conversions, the precatalyst VO(O Bu) (I,
3
possess different catalytic activities, or (2) binding of the
substrate 1 with the catalyst is competitive with the other
Table 2 and blue circles, Figure 9) is a minor species with the
t
t
t
dominant species being the complex VO(O Bu)(OO Bu)-
furfurol) (III, Table 2, and red circles). As the reaction
alcohols (e.g., IPA and BuOH) in solution.
(
To assess the reactivity of 2-methyl-2-propanol containing
catalysts against the triisopropoxide complex, we compared the
t
t
proceeds, III rapidly gives way to VO(O Bu) (OO Bu) (II,
2
t
i
Table 2, and green circles, Figure 9). The concentration of a
activity of VO(O Bu) with VO(O Pr) under the standard
3
3
product-bound complex (IV, Table 2, and beige circles) and
conditions (Figure 7, Table 1, reactions a and h). The fact that
t
the BuOH-bound complex I (blue circles) increases at higher
conversion. The formation of the oxidant byproduct 4 is known
16
to inhibit catalyst activity and is consistent with these results.
Computational studies have also described the role of oxidant
byproducts in the complex equilibria involved in the binding of
17
substrate to the catalyst’s metal center.
These results led us to propose the following reaction
mechanism (Scheme 3) which best fits the observed species
and kinetic data. Catalyst activation occurs via displacement of
one alcohol group on precatalyst species I by TBHP (2) to
1
8
provide entry into the catalytic cycle at species II.
Displacement of a second alcohol by binding of furfurol (1)
then occurs, leading to formation of III. Complex III undergoes
rate-determining, intramolecular oxidation in analogy to the
t
Figure 7. Comparison of the catalytic activity of VO(O Bu) (red
3
i
circles) vs VO(O Pr) (blue circles).
3
19
epoxidation of allylic alcohols. Rearrangement then occurs to
give the product complex IV, though the exact nature of the
bond rearrangements between III and IV is unclear. Several
proposals have been put forward in the literature, often
involving the intermediacy of 2,5-dihydrofuran intermediates,
but no definitive studies on the exact mechanism of the
the reaction rates were similar, regardless of catalyst structure,
strongly argues against the possibility that the vanadium
alkoxide complexes have different activities. Therefore, it
i
t
appears that perhaps competitive binding of PrOH, BuOH
4), and substrate 1 to the metal center was the cause of the
(
8b,20
apparent loss in catalyst activity. If this is true, the addition of
any alcohol should lead to a suppression of the reaction rate.
This was confirmed experimentally by the addition of PrOH,
rearrangement itself exist
Regardless, catalyst turnover is
completed by displacement of the product 3 by another
molecule of TBHP (2). The off-cycle species I is in equilibrium
with II, and both the reversible binding of furfurol (1) and
product 3 are likely quasiequilibrium steps.
A rate expression for this mechanism derived using the
quasiequilibrium assumption and containing only the kinetically
meaningful, observed intermediate species may be written as
i
t
which produced a similar inhibitory effect to BuOH (Table 1,
reactions f and g, see the Supporting Information). Attempts to
counter this inhibitory effect by using other hydroperoxides
were unsuccessful. Cumyl alcohol from cumene hydroperoxide
inhibited the reaction similarly to TBHP (2), while bulkier
D
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 2. Summary of 51V NMR Chemical Shifts
Figure 8. Identification of catalytic species under reaction conditions h
t
(
Table 1) using VO(O Bu) as catalyst.
3
Figure 9. Temporal kinetic profile of reaction catalytic species (%)
t
against time for the reaction of Scheme 1 using VO( BuO) as the
3
catalyst. Comparison of experimentally observed (symbols) to the
simulated (lines) profiles from the mechanism proposed in Scheme 3.
Scheme 3. Plausible Mechanism for the Achmatowicz
Rearrangement of Furfurol
shown in eq 3. The magnitude of the equilibrium constants
Keq,1 and Keq,4 may be estimated from their relative
concentrations in Figure 9 at any point during the reaction as
shown in eqs 4 and 5. Using these values, the global profile can
be fitted to either eq 3 or to the series of elementary steps
represented by Scheme 3. This procedure was carried out using
produces a computational prediction of the temporal
concentrations for 1 and 3 and the intermediates I, II III,
and IV and allows for a comparison with the experimentally
determined profiles. The model predictions for these
intermediates is given by the solid lines of Figure 9. The
2
1
the Copasi program and the global kinetic profiles for
reaction entries a−c as shown in Table 1. The global fit also
E
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
good qualitative prediction of the trends for 1, 3, and all catalyst
intermediates supports this model as a realistic approximation
of the catalytic cycle under the reaction conditions. This study
offers a rare example of the use of global kinetic data to predict,
with fair accuracy, the concentrations of species proposed as
catalytic intermediates and observed experimentally. While the
reasonable fit to the experimental data shown in Figure 9
cannot preclude other mechanistic possibilities, or even the
possibility that these species are not active catalytic
intermediates, these combined experimental and kinetic
modeling results offer a reasonable rationalization for the
catalytic behavior of this system.
aqueous layer was discarded. The resulting DCM solution of TBHP
was used without further purification. To a 250 mL flask under N was
2
added 5.1 g (52 mmol) of furfurol (1) along with 80 mL of DCM and
i
2
50 μL of VO(O Pr) (1 mmol, 0.02 equiv). The solution was cooled
3
in a room temperature water bath, and the solution of TBHP in DCM
was added dropwise over 10 min. Quantitative HPLC analysis
indicated 93% conversion, 81% yield after 3 h, at which point residual
TBHP was quenched by addition of 0.4 mL of trimethyl phosphite at a
rate to maintain the temperature below 30 °C. The reaction mixture
was then concentrated under reduced pressure, and 5 mL of n-heptane
was added followed by 50 mg of pure 3. After a suspension had
formed, 10 mL of n-heptanes was added and the suspension aged. The
product was isolated via filtration. The cake was washed with 20 mL of
1
:1 n-heptane/IPA to obtain 3.74 g of 6-hydroxy-2H-pyran-3(6H)-one
[
1][2]
4]
1
krdsKeq,1Keq,2 [ [cat.]total
as an off-white solid (33 mmol, 63% yield). H NMR (500 MHz,
CDCl ): 4.14 (d, J = 17.0 Hz, 1 H), 4.36 (d, J = 5.7 Hz, 1 H), 4.57 (d,
2
rate =
3
⎛
⎜
[2]
[4]
[1][2]
[4]
Keq,1 [3] ⎞
⎟
J = 17.0 Hz, 1 H), 5.63 (dd, J = 5.2, 3.0 Hz, 1 H), 6.17 (d, J = 10.4 Hz,
1 + Keq,1
+ Keq,1Keq,2
+
13
2
⎝
Keq,4 [4] ⎠
(3)
(4)
(5)
1 H), 6.98 (dd, J = 10.4, 3.0 Hz, 1 H); C NMR (125 MHz, CDCl ):
3
6
1
6.7, 88.3, 127.9, 146.6, 195.3. IR (film): 3308, 1663, 1624, 1278,
−1
+
[2]0
[2]
090, 1008, 980, 847, 686 cm . HRMS (ESI+): [C H O ] calcd
5 7 3
Keq,1 = [[ x′ − 1 where x′ =
II]
I]
115.0390, measured 115.0380. Mp: 58−60 °C. Anal. Calcd for
C H O : C, 52.63; H, 5.30. Found: C, 52.66; H, 5.31.
5
6
3
Preparation of tert-Butyl 5-Oxo-2H-pyran-2-yl Carbonate
5). In a 250 mL flask, 2.00 g (20.4 mmol, 1 equiv) of furfuryl alcohol
Keq,4 = Keq,1 [[ where [3] = [4]
I]
IV]
(
was diluted with 40 mL of dichloromethane, followed by the addition
i
of 100 μL (0.408 mmol, 0.02 equiv) of VO(O Pr) . The resulting
3
solution was cooled to 10 °C with an ice bath, and 4.5 mL (∼5.5 M, 13
mmol, 1.25 equiv) of TBHP in decane was added dropwise over 5
min. The resulting solution was stirred at room temperature. After 4 h,
the solution was cooled to 5 °C, and 1.0 mL (8.4 mmol, 0.4 equiv) of
trimethyl phosphite was charged dropwise. After completion of the
quench, the temperature of the solution was cooled back to 5 °C, and
CONCLUSIONS
■
A robust and high-yielding method for the Achmatowicz
rearrangement of furfurol (1) is described which allowed for
simple isolation of lactol 3 and its crystalline derivatives. The
mechanism of the reaction has been investigated using a
combination of ReactIR and H and V NMR spectroscopy.
This has given an increased understanding of the mechanism
and elucidation of several important catalytic intermediates.
These studies differentiate between irreversible catalyst
5
.5 g (25.2 mmol, 1.2 equiv) of Boc anhydride was added in one
1
51
1
portion, followed by 80 mg (0.67 mmol, 0.03 equiv) of DMAP. H
NMR analysis at 15 min indicated full conversion. The crude solution
was concentrated and purified using column chromatography with
EtOAc in hexanes as the eluent. After evaporation of the clean
t
deactivation and reversible rate suppression by BuOH (4),
fractions, tert-butyl 5-oxo-2H-pyran-2-yl carbonate (3.48 g, 15.3 mmol,
1
which acts as a competitive ligand impacting the rate of
substrate activation and therefore turnover of the productive
transformation. From this analysis, it is apparent that the
reaction occurs without significant irreversible deactivation of
the catalytic species. The increase in concentration of BuOH
4) is a natural outcome of the productive reaction but is also
responsible for the decrease in reaction rate at high conversion.
Elucidation of this effect demonstrates that this is fundamen-
tally a robust transformation to generate lactol 3 under
conditions applicable to further synthetic manipulations.
75% yield) was isolated as a white solid. H NMR (500 MHz, CDCl
3
):
δ 6.87 (dd, J = 10.4, 3.6 Hz, 1H), 6.28 (d, J = 3.6 Hz, 1H), 6.20 (d, J =
1
1
0.3 Hz, 1H), 4.50 (d, J = 17.0 Hz, 1H), 4.16 (d, J = 17.0 Hz, 1H),
.46 (s, 9H). 13C NMR (125 MHz, CDCl ): δ 193.2, 151.8, 141.6,
3
+
t
128.9, 88.8, 83.8, 67.2, 27.7. HRMS (DCI): [C10
H
15
O
5
] calcd
2
15.0919, measured 215.0921. Mp: 83−85 °C.
(
Preparation of 5-Oxo-2H-pyran-2-yl 2-Phenylacetate (6). In a
50 mL flask, 1.00 g (10.2 mmol, 1 equiv) of furfuryl alcohol was
2
diluted with 20 mL of dichloromethane, followed by the addition of 50
i
μL (0.204 mmol, 0.02 equiv) of VO(O Pr) . The resulting solution
3
was cooled to 10 °C with an ice bath, and 2.3 mL (∼5.5 M, 13 mmol,
1
.25 equiv) of TBHP in decane was added dropwise over 5 min. The
EXPERIMENTAL SECTION
General Experimental Methods. VO(O Pr) was stored under
nitrogen gas and used as received. For the kinetic studies, furfurol was
distilled from potassium bicarbonate and stored over molecular sieves
resulting solution was stirred at room temperature. After 4 h, the
solution was cooled to 5 °C, and 0.5 mL (4.24 mmol, 0.4 equiv) of
trimethyl phosphite was charged dropwise. After the completion of the
quench, the temperature of the solution was cooled back to 5 °C, and
■
i
3
3
.20 g (12.6 mmol, 1.2 equiv) of 2-phenylacetyl 2-phenylacetate was
(
4 Å) under inert atmosphere. TBHP/decane (∼5.5 M) used for
22
added in one portion, followed by 40 mg (0.33 mmol, 0.03 equiv) of
kinetic experiments was carefully titrated (×3) with Na S O /NaI,
2
2
3
1
DMAP. H NMR analysis at 15 min indicated full conversion. The
and the concentration was found to be 5.42 M. Tri-tert-butoxyvanadate
2
3
crude solution was concentrated and purified using column
chromatography with EtOAc in hexanes as the eluent. After
evaporation of the clean fractions, 2.7 g of desired material
contaminated with phenylacetic acid was isolated. The material was
diluted with 15 mL of ethyl acetate, washed twice with saturated
sodium bicarbonate solution, and dried with sodium sulfate to receive
was synthesized according to previously reported procedures. 1,3,5-
Trimethoxybenzene was employed as the internal standard. In situ
FTIR reaction analysis was performed using a ReactIR 45m instrument
fitted with a ATR probe. NMR spectra were calibrated using residual
proteo-solvent as an internal reference. VOCl (neat) was used as an
3
5
1
external reference for V NMR analysis. In order to obtain
quantitative data we measured the 90 deg flip angle (P1/4) and the
longitudinal relaxation time (T1) for furfurol.
5-oxo-2H-pyran-2-yl 2-phenylacetate (1.66 g, 7.15 mmol, 70% yield)
1
as white solid after evaporation. H NMR (500 MHz, CDCl
): δ
3
Preparation of 6-Hydroxy-2H-pyran-3(6H)-one (3). Anhy-
drous TBHP was prepared by mixing 8 g (62 mmol, 1.2 equiv) of
0% aqueous tert-butyl hydroperoxide (TBHP) together with 20 mL
7.40−7.19 (m, 5H), 6.91 (dd, J = 10.4, 3.6 Hz, 1H), 6.50 (d, J = 3.6
Hz, 1H), 6.26 (d, J = 10.4 Hz, 1H), 4.37 (d, J = 17.0 Hz, 1H), 4.17 (d,
J = 17.0 Hz, 1H), 3.69 (s, 2H). 13C NMR (125 MHz, CDCl ): δ 193.3,
7
3
of DCM. The phases were allowed to settle over 1 h, and then the
170.1, 142.1, 133.1, 129.2, 128.7, 128.7, 127.4, 86.8, 67.2, 41.2. HRMS
F
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
−
(
5
ESI-): [C H O ] calcd 231.06519, measured 231.06600. Mp: 55−
1973, 12, 337−342. (c) Gould, E. S.; Hiatt, R. R.; Irwin, K. C. J. Am.
Chem. Soc. 1968, 90, 4573−4579.
13
11
4
7 °C.
(
14) Curci, R.; Di Furia, F.; Testi, R.; Modena, G. J. Chem. Soc.,
Perkin Trans. 2 1974, 752−757.
ASSOCIATED CONTENT
Supporting Information
Full experimental details and supplementary experiments. This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
(
15) (a) Adao, P.; Kuznetsov, M. L.; Barroso, S.; Martins, A. M.;
̃
*
S
Avecilla, F.; Pessoa, J. C. Inorg. Chem. 2012, 51, 11430−11449.
(
(
b) Figiel, P. J.; Sobczak, J. M. J. Catal. 2009, 263, 167−172.
c) Hartung, J. Pure Appl. Chem. 2005, 77, 1559−1574. (d) Bryliakov,
K. P.; Talsi, E. P. Kinet. Catal. 2003, 44, 334−346. (e) Smith, T. S.;
Pecoraro, V. L. Inorg. Chem. 2002, 41, 6754−6760.
AUTHOR INFORMATION
Corresponding Authors
E-mail: tamas.benkovics@bms.com.
E-mail: blackmond@scripps.edu
(16) (a) Bryliakov, K. P.; Talsi, E. P.; Stas’ko, S. N.; Kholdeeva, O. A.;
Popov, S. A.; Tkachev, A. V. J. Mol. Catal. A 2003, 194, 79−88.
■
*
*
(
b) Talsi, E. P.; Chinakov, V. D.; Babenko, V. P.; Zamaraev, K. I. J.
Mol. Catal. 1993, 81, 235−254.
17) Vandichel, M.; Leus, K.; Van Der Voort, P.; Waroquier; Van
Speybroeck, V. J. Catal. 2012, 294, 1−18.
18) (a) Greb, M.; Harung, J.; Koehler, F.; Spehar, K.; Kluge, R.;
(
Notes
̌
(
The authors declare no competing financial interest.
Csuk, R. Eur. J. Org. Chem. 2004, 18, 3799−3812. (b) Di Furia, F.;
Modena, G.; Curci, R.; Bachofer; Edwards, J. O.; Pomerantz, M. J. Mol.
Catal. 1982, 14, 219−229.
ACKNOWLEDGMENTS
■
This work was supported by collaboration between Bristol-
Myers Squibb Company and The Scripps Research Institute.
We thank Drs. David Kronenthal and Joel Barrish for
supporting this work.
(19) Itoh, T.; Jitsukawa, K.; Kaneda, K.; Teranishi, S. J. Am. Chem.
Soc. 1979, 101, 159−169.
(20) (a) Achmatowicz, O.; Burzynska, M. H. Carbohydr. Res. 1985,
141, 67−76. (b) Sammes, P. G.; Thetford, D. J. Chem. Soc., Chem.
Commun. 1985, 352−353. (c) Pelter, A.; Ward, R. S.; James, D. C.;
Kamakshi, C. Tetrahedron Lett. 1983, 24, 3133−3136. (d) Torii, S.;
Tanaka, H.; Anoda, T.; Simizu, Y. Chem. Lett. 1976, 495−498.
REFERENCES
■
(
1) Georgiadis, M. P.; Albizati, K. E.; Georgiadis, T. M. Org. Prep.
Proc. 1992, 24, 95−118.
2) (a) Cuccarese, M. F., O’Doherty, G. A. In Asymmetric Synthesis II:
More Methods and Applications; Christmann, M., Braese, S., Eds.;
Wiley−VCH: Weinheim, 2012; pp 249−259. (b) Guo, H. B.;
O’Doherty, G. A. J. Org. Chem. 2008, 73, 5211−5220. (c) Harris, J.
M.; Li, M.; Scott, J. G.; O’Doherty, G. A. Strategies Tactics Org. Synth.
(21) Hopps, S.; Sahle, S.; Gauges, R.; Lee, C.; Pahle, J.; Simus, N.;
Singhal, M.; Xu, L.; Mendes, P.; Kummer, U. COPASI − A Complex
(
Pathway Simulator. Bioinformatics 2006, 22 (24), 3067−3074.
(22) Verhoeven, T. R.; Sharpless, K. B. Aldrichimica Acta 1979, 12,
6
(
3
3−74.
23) Mittal, R. K.; Mehrotra, R. C. Z. Anorg. Allg. Chem. 1964, 327,
11−314.
2
1
004, 5, 221−253. (d) Achmatowicz, O.; Bielski, R. Carbohydr. Res.
977, 55, 165−176. (e) Achmatowicz, O.; Bukowski, P.; Szechner, B.;
Zwierzchowska, Z. Tetrahedron 1971, 27, 1973−1996.
(
3) For representative examples, see: (a) Nicolaou, K. C.; Aversa, R.
J.; Jin, J.; Rivas, F. J. Am. Chem. Soc. 2010, 132, 6855−6861. (b) Zhou,
X. F.; Wu, W. Q.; Liu, X. Z.; Lee, C. S. Org. Lett. 2008, 10, 5525−5528.
(
(
c) Jones, R. A.; Krische, M. J. Org. Lett. 2009, 11, 1849−1851.
d) Abrams, J. N.; Babu, R. S.; Guo, H. B.; Le, D.; Le, J.; Osbourn, J.
M.; O’Doherty, G. A. J. Org. Chem. 2008, 73, 1935−1940. (e) Wender,
P. A.; Bi, F. C.; Buschmann, N.; Gosselin, F.; Kan, C.; Kee, J. M.;
Ohmura, H. Org. Lett. 2006, 8, 5373−5376.
(
4) (a) Shimokawa, J.; Harada, T.; Yokoshima, S.; Fukuyama, T. J.
Am. Chem. Soc. 2011, 133, 17634−17637. (b) Benkovics, T.; Ortiz, A.;
Guo, Z.; Goswami, A.; Deshpande, P. Org. Synth. 2014, 91, 293−306.
(
5) Kolb, H. C.; Hoffmann, H. M. R. Tetrahedron: Asymmetry 1990,
1
(
, 237−250.
6) See further discussion on p 106 of ref 1.
(
7) (a) Wahlen, J.; Moens, B.; De Vos, D. E.; Alsters, P. L.; Jacobs, P.
A. Adv. Synth. Catal. 2004, 346, 333−338. (b) Finlay, J.; McKervey, M.
A.; Gunaratne, H. Q. N. Tetrahedron Lett. 1998, 39, 5651−5654.
(
8) (a) Jackson, K. L.; Henderson, J. A.; Morris, J. C.; Motoyoshi, H.;
Phillips, A. J. Tetrahedron Lett. 2008, 49, 2939−2941. (b) Ho, T. L.;
Sapp, S. G. Synth. Commun. 1983, 13, 207−211.
(
9) (a) Blackmond, D. G. Angew. Chem., Int. Ed. 2005, 44, 4302−
4
320. (b) Mathew, J. S.; Klussmann, M.; Iwamura, H.; Valera, F.;
Futran, A.; Emanuelsson, E. A. C.; Blackmond, D. G. J. Org. Chem.
006, 71, 4711−4722.
10) Rehder, D. In Multinuclear NMR; Mason, J., Ed.; Plenum Press:
New York, 1987; pp 488−493.
11) (a) Crans, D. C.; Chen, H.; Felty, R. A. J. Am. Chem. Soc. 1992,
2
(
(
1
14, 4543−4550. (b) Hillerns, F.; Rehder, D. Chem. Ber. 1991, 124,
2
249−2254.
(
12) Bryant, B. E.; Fernelius, W. C.; Busch, D. H.; Stouffer, R. C.;
Stratton, W. Inorganic Synthesis 1957, 5, 113−116.
13) (a) Bortolini, O.; Di Furia, F.; Modena, G. J. Mol. Catal. 1983,
9, 319−329. (b) Su, C.-C.; Reed, J. W.; Gould, E. S. Inorg. Chem.
(
1
G
DOI: 10.1021/jo502641d
J. Org. Chem. XXXX, XXX, XXX−XXX