Highly efficient indium(III)-mediated cyclisation of 5-hydroxy-1,3- diketones to 2,3-dihydro-4H-pyran-4-ones; Mechanistic insights from in situ Fourier transform infrared spectroscopy

Article abstract of DOI:10.1039/b919402a

5-Hydroxy-1,3-diketones have been synthesised in a facile one-pot reaction from the treatment of acid chlorides with non-substituted ketones and LiHMDS. Subsequent cyclisation to 2,3-dihydro-4H-pyran-4-ones occurs rapidly and in high yield (89-99%) when mediated by anhydrous indium(iii) chloride. A spectroscopic study of the reaction using in situ Fourier transform infrared (FTIR) spectroscopy has shown the reaction to be highly dependent on temperature, metal complex formation and InCl₃ concentration. Since the reaction is deactivated by the precipitation of [InCl₃·(H ₂O)₃], the concurrent use of a stronger drying agent, such as molecular sieves 4 A or anhydrous MgSO₄, allows the reaction to be successfully carried out at relatively low loadings of InCl ₃ (1-10%). In their absence, the optimum reaction conditions were found to be a diketone:InCl₃ ratio of 3:1 in toluene, and a reaction temperature of 80 °C.

Full text of DOI:10.1039/b919402a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Highly efficient indium(III)-mediated cyclisation of 5-hydroxy-1,3-diketones
to 2,3-dihydro-4H-pyran-4-ones; mechanistic insights from in situ Fourier
transform infrared spectroscopy†
Philip C. Andrews,*a William J. Gee,a Peter C. Junka and Harald Krautscheidb
Received 17th September 2009, Accepted 6th November 2009
First published as an Advance Article on the web 10th December 2009
DOI: 10.1039/b919402a
5-Hydroxy-1,3-diketones have been synthesised in a facile one-pot reaction from the treatment
of acid chlorides with non-substituted ketones and LiHMDS. Subsequent cyclisation to
2,3-dihydro-4H-pyran-4-ones occurs rapidly and in high yield (89–99%) when mediated by anhydrous
indium(III) chloride. A spectroscopic study of the reaction using in situ Fourier transform infrared
(FTIR) spectroscopy has shown the reaction to be highly dependent on temperature, metal complex
formation and InCl3 concentration. Since the reaction is deactivated by the precipitation of
˚
[InCl3·(H2O)3], the concurrent use of a stronger drying agent, such as molecular sieves 4 A or
anhydrous MgSO4, allows the reaction to be successfully carried out at relatively low loadings of InCl3
(1–10%). In their absence, the optimum reaction conditions were found to be a diketone : InCl3 ratio of
3 : 1 in toluene, and a reaction temperature of 80 ◦C.
constantly monitored and evaluated, eliminating the potential
for transfer contamination and sample degradation between the
source and the spectrometer.
Introduction
2,3-Dihydro-4H-pyran-4-ones are important biological scaffolds.
They are incorporated in several natural products including
In this paper, we report the rapid, one-pot synthesis of a series
Prelog–Djerassi lactone and Compactin, a competitive inhibitor
of 5-hydroxy-1,3-diketones, and describe their subsequent conver-
in cholesterol biosynthesis,1,2 as well as serving as intermediates
in the synthesis of many other biologically active compounds.3
sion to 2,3-dihydro-4H-pyran-4-ones mediated by InCl3. The 2,3-
dihydro-4H-pyran-4-ones are generated from acid chlorides and
Synthetic routes to this class of compound from 5-hydroxy-1,3-
non-substituted ketones, borrowing from a previously described
diketones are well established, utilising strong acids including
method for generating diketones from acid chlorides and lithium
hydrochloric, sulfuric, and p-toluenesulfonic acids,4–6 as well as
Lewis acids, including TiCl4 and AlCl3.7,8 However, each of
these methods suffers from either by-product formation4–8 or
inconsistent yields.4–6,9 Recently, indium(III) chloride has been
described as an efficient and high yielding Lewis acid for the
formation of flavones from 2¢-hydroxychalcones8 and phenol-
substituted 1,3-diketones,9 albeit with a loading requirement of
20–50% InCl3. The mechanism for the indium mediated cyclisation
remains unclear, though the latter report implicates an ultimate
dehydration pathway.9
In situ Fourier transform infrared (FTIR) spectroscopy is a
powerful analytical tool for probing reaction mechanisms and
is proving invaluable to both the chemical and pharmaceutical
industries.10,11 In recent years, the application of in-line spec-
troscopy has proved effective in quality control,12 in under-
standing kinetic studies,13–15 in postulating reaction mechanisms,15
in reaction optimisation, and in the identification of reaction
intermediates.16,17 In situ measurements allow reactions to be
enolates.18 This constitutes the first synthesis of 2,3-dihydro-4H-
pyran-4-ones from acid chlorides in a single step, improving
efficiency and allowing the rapid isolation of products.
In addition, we describe the optimum reaction conditions,
in terms of rate and yield, and a plausible reaction pathway
deduced from in situ FTIR spectroscopy. We report the need for
participation in the reaction mixture of an efficient dehydrating
agent for a low level of InCl3 catalyst loading to be effective.
Results and discussion
Synthesis of 2,3-dihydropyranones
Precursor 5-hydroxy-1,3-diketones 1–6 (Table 1), required for
cyclisation to 2,3-dihydro-4H-pyran-4-ones, were synthesised by a
one-pot condensation reaction of a 1,3-dicarbonyl dianion with a
ketone ◦(Scheme 1). A nitrogen-flushed reaction vessel was cooled
to -78 C and charged with dry THF. One equivalent of freshly
formed lithium propen-2-olate was generated by the addition
of lithium hexamethyldisilazide (LiHMDS) to acetone. A single
aSchool of Chemistry, Monash University, Clayton, Melbourne, Vic 3800,
Australia. E-mail: phil.andrews@sci.monash.edu.au; Fax: +61 3 99054597;
Tel: +61 3 99055509
bInstitut fu¨r Anorganische Chemie, Universita¨t Leipzig, 04103, Leipzig,
Germany
† Electronic supplementary information (ESI) available: Generated in
situ IR spectra and waterfall representations. Experimental procedure
for assigning conversion percentage for drying agent study. See DOI:
10.1039/b919402a
Scheme 1 Reagents and conditions: (i) 1 eq. lithium propen-2-olate,
-78 ◦C, 5 min; (ii) 2 eq. LiHMDS, -78 ◦C, 5 min; (iii) 1 eq. acetone,
-78 → 0 ◦C, 2 h; (iv) HCl(aq).
698 | Org. Biomol. Chem., 2010, 8, 698–705
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 5-Hydroxy-1,3-diketones and 2,3-dihydro-4H-pyran-4-ones synthesised
Acid chloride
Intermediate
Diketone yield (%)
77
Product
Pyranone yield (%)
99
66
80
97
97
68
72
46
95
98
89
equivalent of acid chloride was then added with rapid stirring to
generate the diketone. Addition of a further two equivalents of
LiHMDS to the solution gave the dianion, which then reacted
with a final equivalent of acetone. The more nucleophilic terminal
enolate reacts preferentially, giving the 5-hydroxy-1,3-diketone
after acidic workup.
achieve total conversion of hydroxydiketone to dihydropyranone.
Hydroxydiketone 3 was chosen as the representative substrate
since it is obtained in high yield and, being solid, is easily purified
through simple recrystallisation. Scanning from 4000 to 650 cm-1
at one minute intervals commenced from the time indium(III)
chloride was added to the reaction mixture. This approach,
rather than focusing on one absorption band, allows us to plot
continual and cumulative changes in the spectra against time,
thereby encompassing rate and conversion changes arising from
intermediate species en route to the final product. If each new
spectrum differs from its immediate predecessor and the change is
significant then a steep incline occurs when this change is plotted
relative to time. The converse is also therefore true for spectra that
show no or only slight cumulative changes per unit time.
Optimisation reactions were undertaken by varying the ratio
of InCl3 relative to a 0.055 M toluene solution of 3, heated at
80 ◦C. The most rapid conversion was observed for a molar ratio of
3 : 1 (diketone : InCl3), with the reaction reaching completion after
13 min (Fig. 1). Surprisingly, increasing the amount of InCl3 was
found to negatively impact on the conversion rate of cyclisation,
with equimolar equivalents more than doubling the time required
to effect complete conversion. A reduction in the amount of InCl3
relative to the diketone resulted not only in a drop in conversion
rate, but also in incomplete conversion to the pyranone. This
corresponds with previous reports in the literature.8 When a ratio
Performing the reaction at -78 ◦C and using a 1 : 1 ratio
of enolate to acid chloride inhibited the possible formation
of triketone by-products, improving overall reagent economy.
Conducting the reaction at 0 ◦C results in a sharp increase in
triketone species. Table 1 displays the 5-hydroxy-1,3-diketones and
2,3-dihydro-4H-pyran-4-ones synthesised in this study, alongside
their corresponding isolated yields.
Cyclisation of 1,3-diketo-5-hydroxy moieties was achieved
through the addition of 0.33 equivalents of anhydrous InCl3 to
a pre-dried toluene solution of each respective hydroxy diketone
(1–6) at 80 ◦C. Optimisation of the reaction conditions is described
below. After stirring for 30 min the solution was filtered and all
volatiles removed in vacuo yielding the 2,3-dihydro-4H-pyranones
(7–12).
Optimisation of reaction
An optimisation study was undertaken using in situ FTIR spec-
troscopy to determine the minimum loading of InCl3 required to
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 698–705 | 699
©

View Article Online
Fig. 1 Rate of change in IR spectra for transformation of diketone 3
into pyran-4-one dependant on diketone : InCl3 ratio. Reaction conditions:
0.055 M toluene solution of 3 maintained at 80 ◦C.
Fig. 2 Rate of change in IR spectra for transformation of diketone 3 into
pyran-4-one dependant on reaction temperature.
ate species with variations in rate signifying the transition between
species. Identification of the intermediate species is discussed
below.
of 6 : 1 (diketone : InCl3) was employed, the yield of pyran-4-one
was only 75%. Extending the reaction time by five hours led to no
improvement in the overall yield.
In situ FTIR analyses
A temperature dependence conversion study was undertaken
next. Varying the reaction temperature from 80 ◦C to 25, 60
and 120 ◦C, whilst retaining the optimised 3 : 1 diketone : InCl3
ratio, led to interesting variations in conversion rate over the
reaction lifetime (Fig. 2). Increasing the temperature to 120 ◦C
initially resulted in an increased rate of conversion relative to
the reaction conducted at 80 ◦C. However, after approximately
75% conversion, a sharp decrease in rate was observed, such that
complete conversion was achieved 15 min later than the 80 ◦C
experiment. Analysis of the reaction at 25 ◦C indicates that the
reaction proceeds to completion, but with total conversion only
achieved after approximately 12 h (not shown). The kinetic profile
To aid in the elucidation of the reaction mechanism for the
cyclisation of diketone 3 to dihydropyranone 9, a peak library
was generated for the starting material and product using in situ
FTIR spectroscopy. Spectra of 0.055M solutions of analyte in
toluene were collected. Metal interaction can be expected to have
a marked effect on carbonyl stretches; hence, assignment focused
on the region of 1700–1200 cm-1. Assigned bands for both 3 and
9 are shown in Table 2.
1H NMR spectral analysis of 3 identified the conjugated vinylic
and enolic protons at 6.26 and 15.82 ppm, respectively, indicating
the dominant tautomer in solution to be the enolized form. This
is consistent with the IR spectrum, which shows a characteristic
very strong and broad band from 1550–1680 cm-1. The location
and extreme intensity of this band is attributed to the large
◦
of the 80 C experiment is suggestive of Michaelis–Menton-type
kinetics; however, this assumption was invalidated by variations
in rate observed for the 60 and 120 ◦C experiments, implying a
multistep reaction mechanism containing multiple species.19
Based on these observations it is postulated that the conversion
of diketone 3 to dihydropyranone 9 occurs via transient intermedi-
=
increase in charge polarisation of the C O bond, shifting the
absorption to a lower wavenumber and increasing the charge on
the carbonylic oxygen.20 The overall resolution of bands in the IR
Table 2 Assignment of observed wavenumbers for diketone 3, dihydropyranone 9, and intermediates 13 and 14
Wavenumber/cm-1
Assignment
Wavenumber/cm-1
Assignment
Diketone 3
1577
1532
1349
1320
Intermediate 13
1727
1693
1657
1637
1618
1586
1538
1521
Intermediate 14
1674
1616
1606
1547
1506
1491
1340
=
=
=
C
N
N
O, C C stretch
O asym. stretch
O sym. stretch
C
C
C
C
C
C
C
N
O stretch trans
=
=
=
=
=
=
=
O stretch cis
=
O stretch cis
C–H in-plane bend
C–H in-plane bend
C stretch cis
1300
C stretch cis
C stretch trans
O stretch diketone
=
O asym. stretch
2,3-Dihydropyranone 9
=
=
=
=
1596
1570
1533
1505
1484
1353
1310
C
C
N
O stretch
C stretch
C
C
C
N
=
C
=
C
N
O stretch
C aromatic stretch
C aromatic stretch
O asym. stretch
C aromatic stretch
C aromatic stretch
=
=
O asym. stretch
=
=
C
C
N
C aromatic stretch
C aromatic stretch
=
=
O sym. stretch
=
C–H in-plane bend
O sym. stretch
700 | Org. Biomol. Chem., 2010, 8, 698–705
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
spectrum of diketone 3 was poor, attributed to various rotameric
isomers giving broadened, overlapping vibrations. The infrared
spectrum of dihydropyranone 9 displays greater resolution of
bands owing to greater rigidity afforded by the pyranone ring.
=
=
Both the u(C C) and u(C O) bands are evident at 1570 and
1596 cm , respectively, with well defined aromatic C C stretching
-1
=
bands observed between 1505 and 1484 cm-1. The appearance of
a trisubstituted vinylic C–H out of plane bend at 852 cm-1 further
confirms the presence of the pyranone ring. Twin nitro asymmetric
=
and symmetric stretches (N O) are prominent for both 3 and 9 at
approximately 1530 and 1349 cm-1.
Intermediate species in the conversion of diketone 3 to pyranone
9 were sought using real-time infrared scanning. Species differenti-
ation by the software ConcIRTTM was improved through reducing
the conversion rate by using toluene at 50 ◦C with the optimised
diketone : InCl3 ratio of 3 : 1. Concentration profiles were identified
for diketone 3, dihydropyranone 9 as well as two intermediate
species across the 30 min reaction period (Fig. 3).
Scheme 2 Species observed by in situ IR spectroscopy.
14 had occurred. The increase in concentration of 14 was seen
to mirror the decline of indium-complexed diketone 13. The
=
carbonyl region of 14 shows a single dominant u(C O) band
at 1674 cm . Sharp aromatic u(C C) pairs are seen at 1606–
-1
=
1616 cm-1 and at 1491–1506 cm-1 overlapping the asymmetric
nitro stretch. Ring closure was determined by the appearance of
an intense u(C–O) stretch at 1061 cm-1. At t = 20 min, 11 min after
the addition of indium(III) chloride, evidence for the formation of
dihydropyranone 9 is obtained. Occurrence of the product again
mirrors the decline in intermediate 14 with complete conversion
effected after a further 10 min. The component spectrum of
9 correlates well with that of the solid state IR spectrum and
peak library determination. The final detectable species is water,
identified by an intense hydroxyl band located at 1321 cm-1.
Exact correlation with an in situ IR reference of deionised water
was observed. The appearance of water was seen to coincide
with 9, and existed for approximately 10 min before sharply
subsiding. This disappearance is attributed to the coordination
of water molecules to indium(III) chloride, which results in loss of
water from the solution phase and precipitation of the insoluble
hydrate [InCl3·(H2O)3]. A comparison by Raman spectroscopy
of indium(III) chloride hydrate, generated in situ from a sample
of indium(III) chloride exposed to moist air, yielded an identical
spectrum, containing a single u(In–Cl) band at 297 cm-1. Both
species also gave a broad hydroxyl band at 3315 cm-1.
Based on these observations, the proposed mechanism initiates
with activation of a keto group by coordination to indium(III) chlo-
ride. Interaction with the metal disrupts intramolecular hydrogen
bonding through binding of the diketone to the indium centre,
most probably in the more thermodynamically stable chelate mode
but in equilibrium with a proportion bound only in a monodentate
fashion. A range of conformers are therefore observed by FTIR
as multiple vibrations in the carbonyl region. Significantly, only
when the diketone is acting as a monodentate ligand, as epitomised
by conformer 13, can the hydroxyl group be in close proximity
to the indium-activated carbonyl group, thus enabling cyclisation
through an intramolecular SN2 attack resulting in an ether linkage
and an anionic indate species. Performing the reaction at 120 ◦C
gave an increased conversion rate (Fig. 2), most likely resulting
Fig. 3 Identified reaction components (compounds 3, 9, 13 and 14) as a
function of time.
The reaction commenced with the addition of InCl3 to a solution
of 3 at T = 9 min. Delaying the addition of InCl3 was required
to allow ConcIRTTM to generate reference spectra and allow the
concentration of diketone to equilibrate prior to the reaction
commencing. Post addition, free diketone 3 is consumed as the
concentration of an indium coordinated intermediate 13 increases,
reaching a maximum after 5 min (Scheme 2).
Chelation to the metal by the diketone represents the most stable
-1
=
conformer, seen as an intense u(C O) stretch at 1538 cm , yet
a large number of conformational rotamers are observed in the
component IR spectrum as six bands in the double bond stretching
region located at 1586, 1618, 1637, 1657, 1693 and 1727 cm-1
(Table 2). These bands are characteristic of cis and trans rotamers
of keto enol ethers,12 arising in this case through disruption
of intramolecular hydrogen bonding by indium coordination.
These bands can be assigned as an equilibrium of kinetically
favoured cis b-keto enol ethers and thermodynamically stable
trans forms. The lifetime of intermediate 13 was 10 min, after
which complete conversion to a second pyran-4-one intermediate
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 698–705 | 701
©

View Article Online
Scheme 3 Proposed mechanism of pyran-4-one formation.
Table 3 Assignment of observed wavenumbers for diketone 3, dihydropy-
ranone 9 and intermediates 13 and 14
from a shift in the complex equilibrium favouring monodentate
binding of the ligand and an increase in kinetic energy promoting
the intramolecular SN2 attack.
Conversion ratio
Protonation results in cleavage of the unstable indium–oxygen
bond, regenerating indium(III) chloride and releasing the hy-
droxypyranone, 14. The IR spectrum of intermediate 14 shows
retention of the more stable keto form. Elimination of water
requires a tautomeric shift to the enol form, constituting the
second rate step observed in Fig. 2. At higher temperatures the
equilibrium shifts to favour the keto form,21–23 and as such,
the decrease in the rate of cyclisation observed at 120 ◦C can
be justified by the fact that in the keto form, elimination of water
is not energetically favourable. Only in the enol form does proton
transfer occur, leading to the elimination of water and formation
of the conjugated 2,3-dihydro-4H-pyranone. Scheme 3 displays
the observed species 3, 9, 13 and 14 with the proposed mechanism
overlayed.
Variation in the rate of conversion induced by differing ratios
of indium(III) chloride to 5-hydroxy-1,3-diketone (Fig. 1) can also
be understood in light of the mechanism. Incomplete conversion
is observed when a molar ratio of less than 3 : 1 (diketone : InCl3)
is employed, attributed to the rapid capture of eliminated water
by the indium cation to generate hydrated indium(III) chloride.
The coordination of three equivalents of water saturates the
coordination sphere of indium(III) chloride, binding with greater
strength than the diketone. As a result the diketone is denied access
to the metal center, resulting in termination of reaction.
Drying agent
—
InCl3 loading (%)
(diketone : pyranone)
10
0
10
0
10
1
67 : 43
100 : 0
18 : 82
100 : 0
0 : 100
28 : 72
˚
4 A molecular sieves
˚
4 A molecular sieves
Anhydrous MgSO4
Anhydrous MgSO4
Anhydrous MgSO4
It would be expected that an increase in the concentration of
InCl3 would lead to both a rate enhancement and an overall
yield improvement since a greater proportion of the diketone is
complexed and activated, and excess anhydrous InCl3 would act
as a dehydrating agent. One plausible explanation why this is not
the case could be that as the InCl3 concentration increases then so
does the amount of diketone coordinated to two InCl3 molecules;
one on each carbonyl functionality, with the hydroxy group also
binding to give a stable six-membered chelate, as illustrated by 15
in Fig. 4. This would inhibit the diketone adopting configurations
capable of undergoing intramolecular SN2 attack. An increase
in concentration of InCl3 would also increase the possibility of
stabilising the keto tautomer of intermediate 16, further resulting
in a rate decrease.
˚
In light of this, the addition of molecular 4 A sieves to a
solution loaded with only 10% InCl3 (i.e. 10 : 1 molar ratio)
led to a near doubling in the observed yield when compared
with the analogous reaction conducted in the absence of sieves
(43 vs. 82%; Table 3). That the conversion is not quantitative may
be indicative of competitive binding of water between the sieves
and InCl3. Subsequent substitution of the sieves with an evenly
distributed suspension of anhydrous magnesium sulfate (MgSO4)
did produce a quantitative conversion of diketone to pyranone.
However, reducing the catalyst loading to 1% InCl3 resulted in
only 72% conversion to the pyranone product, presumably due to
slow poisoning of the catalyst by water. The hydrated indium(III)
chloride can be easily recovered by filtration and dehydrated
through sublimation.
Fig. 4 Possible intermediates 15 and 16 implicated in rate reduction.
Conclusions
We have developed a new one-pot synthetic protocol for the
formation of 5-hydroxy-1,3-diketones from acid chlorides, non-
substituted ketones and the bulky lithium base, LiHMDS. These
can be rapidly and efficiently converted to 2,3-dihydro-4H-pyran-
4-ones through a cyclisation reaction mediated by anhydrous
Increased concentrations of indium (i.e. greater than 3 : 1
diketone:indium) were found to cause a decline in rate (Fig. 1).
702 | Org. Biomol. Chem., 2010, 8, 698–705
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
indium(III) chloride. For this latter process, examination and
analysis of the reaction by in situ FTIR spectroscopy has
indicated that the optimum reaction conditions in the absence
of a dehydrating agent involve an InCl3 : diketone ratio of 1 : 3,
heated at 80 ◦C. The reaction rate and yield are affected favourably
by shifting the keto–enol equilibrium in favour of the enol form
(lower temperature), and by monodentate binding of the diketone
to InCl3 rather than by chelation. The nucleophilic cyclisation and
final dehydration step to give the pyranone ring are also favoured
by higher temperatures, indicating that a fine balance is required
to maximise product yield. A higher concentration of InCl3 leads
to a reduced reaction rate, most likely due to formation of an
[(InCl3)2·(diketone)] complex with a configuration which prohibits
cyclisation.
organic washes dried (MgSO4) and concentrated to dryness under
reduced pressure. The crude product was then purified by column
chromatography yielding the 5-hydroxy-1,3-diketone.
General method for the synthesis of 2,3-dihydro-4H-pyran-4-ones
The 5-hydroxy-1,3-diketone (1.00 mmol) and indium chloride
(22.1 mg, 0.100 mmol) were suspended in dry toluene in a nitrogen-
flushed, oven-dried flask. The mixture was stirred at 80 ◦C for
40 min. After this time, the solution was filtered and evaporated
under reduced pressure yielding the 2,3-dihydro-4H-pyran-4-one
product.
(Z)-1,5-dihydroxy-5-methyl-1-phenylhex-1-en-3-one (1). Vis-
cous yellow oil (0.340 g, 77%); IR (KBr) n 3431 s, 2974 s, 1603 s,
1461 s, 1379 s, 1187 s, 1159 s, 1079 s, 1028 m, 1001 m, 981 m,
950 m, 909 s, 859 w, 813w, 765 s, 699 s, 648 m, 571 m, 498 m cm-1;
1H NMR (500 MHz, CDCl3, dH) d 1.33 (s, 6H, CH3), 2.62 (s, 2H,
H-4), 3.43 (s, 1H, OH), 6.18 (s, 1H, H-2), 7.46 (t, 2H, J = 7.2 Hz,
Ar), 7.54 (t, 1H, J = 7.2 Hz, Ar), 7.88 (d, 2H, J = 7.2 Hz, Ar),
16.14 (s, 1H, OH); 13C NMR (125 MHz, CDCl3, dC) d 29.5 (CH3),
51.3 (C-4), 70.4 (C-5), 98.0 (C-2), 127.1 (Ar), 128.7 (Ar), 132.6
(Ar), 134.3 (Ar), 182.6 (C-1), 196.8 (C-3); HRMS (ESI) calcd for
(M + H) C13H17O3: 221.1178. Found: 221.1172.
The discovery that the reaction is deactivated by the precip-
itation of [InCl3·(H2O)3] led to the reaction being successfully
conducted at low InCl3 loadings (1–10%) when a more competitive
˚
drying agent, molecular sieves 4 A or anhydrous MgSO4, was used
in the reaction medium.
Experimental
General
(Z)-4-(1,5-dihydroxy-5-methyl-3-oxohex-1-enyl)benzonitrile (2).
Colourless oil (0.322 g, 66%); IR (KBr) n 3445 brm, 2974 m, 2231
m, 1615 s, 1598 s, 1562 s, 1501 m, 1463 m, 1437 m, 1379 m, 1288 s,
1161 m, 1114 m, 909 m, 855 m, 806 m, 776 m, 545 m cm-1; 1H NMR
(500 MHz, CDCl3, dH) d 1.34 (s, 6H, CH3), 2.67 (s, 2H, H-4), 3.23
(s, 1H, OH), 6.23 (s, 1H, H-2), 7.75 (d, 2H, J = 8.4 Hz, Ar), 7.97
(d, 2H, J = 8.4 Hz, Ar), 15.87 (s, 1H, OH); 13C NMR (125 MHz,
CDCl3, dC) d 29.6 (CH3), 51.8 (C-4), 70.5 (C-5), 99.1 (C-2), 115.5
(Ar), 118.0 (CN), 127.5 (Ar), 132.2 (Ar), 138.3 (Ar), 179.0 (C-1),
198.6 (C-3); HRMS (ESI) calcd for (M - H) C14H14NO3: 244.0974.
Found: 244.0980.
All chemicals were obtained from commercial sources and were
used as received unless otherwise stated. Anhydrous solvents were
obtained from a MBRAUN MB SPS-800 solvent purification
system, and all reaction glassware was oven-dried prior to use.
1H NMR spectra and 13C NMR spectra were recorded on a
Varian Unity Nova 500 MHz spectrometer. Chemical shifts were
recorded on the d scale and referenced to deuterated solvent.
13C NMR spectra were recorded at 125 MHz, with resonances
referenced to deuterated solvent. Solid-state IR spectra were
recorded on a Bruker Equinox 55 Infrared Spectrometer fitted
with a Specac Diamond ATR source. Solution RTIR scanning
measurements were recorded using a Mettler Toledo ReactIR
10 spectrometer fitted with a DiComp probe and connected
to an MCT detector by a K6 Conduit. Scanning was per-
formed in the region of 4000–650 wavenumbers at 8 wavenumber
resolution. Elucidation of reaction components was performed
using ConcIRTTM software. All positive and negative electrospray
ionization mass spectroscopy (ESI( )-MS) was performed on a
Micromass Platform II QMS spectrometer using cone voltages
varying from 35–50 V. Melting points were measured on a Stuart
Scientific Melting Point Apparatus in an open capillary.
(Z)-1,5-dihydroxy-5-methyl-1-(4-nitrophenyl)hex-1-en◦-3-one
(3). Red crystalline solid (0.426 g, 80%); Mp 56.1–56.6 C; IR
(KBr) n 3276 m, 2979 m, 2929 m, 1584 s, 1520 s, 1382 m, 1344 s,
1319 s, 1297 m, 1157 m, 1110 m, 1074 m, 1010 m, 989 m, 948 w,
912 m, 895 m, 866 w, 814 w, 804 m, 783 m, 764 m, 750 w, 729 m,
707 m, 683 m, 651 w, 583 w, 486 w, 444 w, 416 w cm-1; 1H NMR
(500 MHz, CDCl3, dH) d 1.35 (s, 6H, CH3), 2.69 (s, 2H, H-4), 3.19
(s, 1H, OH), 6.26 (s, 1H, H-2), 8.04 (d, 2H, J = 8.4 Hz, Ar), 8.30
(d, 2H, J = 8.4 Hz, Ar), 15.82 (s, 1H, OH); 13C NMR (125 MHz,
CDCl3, dC) d 29.7 (CH3), 51.9 (C-4), 70.5 (C-5), 99.4 (C-2), 123.9
(Ar), 128.0 (Ar), 139.9 (Ar), 149.9 (Ar), 178.4 (C-1), 199.1 (C-3);
HRMS (ESI) calcd for (M - H) C13H14NO5: 264.0872. Found:
264.0876.
General method for the synthesis of 5-hydroxy-1,3-diketones
To a solution of freshly prepared LiHMDS (1.25 mL, 1.6 M in
hexanes, 2.0 mmol) in dry THF (5 mL) at -78 ◦C was added one
equivalent of acetone (142 mL, 2.00 mmol) with rapid stirring for
approximately 1 min. After this time, an equimolar quantity of
acid chloride was added in one portion with rapid stirring. After
approximately 5 min two further equivalents of LiHMDS were
added and the solution was stirred for an additional 5 min. A
second equivalent of acetone was added and the solution allowed
to warm to room temperature. The reaction was quenched with
water (~10 mL) and acidified to pH 5 with 1 M HCl. The
solution was then extracted with EtOAc (3 ¥ 25 mL) and the
(Z)-1,5-dihydroxy-5-methyl-1-(4-(trifluoromethyl)phenyl)hex-1-
en-3-one (4). White powder (0.390 g, 68%); Mp 76.8–77.2 ◦C;
IR (KBr) n 3515 m, 3099 m, 2987 m, 2927 m, 1617 s, 1576 s, 1518
m, 1444 m, 1390 m, 1379 m, 1329 s, 1261 m, 1162 s, 1129 s, 1113 s,
1080 m, 1064 s, 1016 m, 984 w, 948 w, 913 w, 887 w, 855 m, 822 s,
1
718 m, 685 w, 598 w, 519 w, 428 w cm-1; H NMR (500 MHz,
CDCl3, dH) d 1.34 (s, 6H, CH3), 2.66 (s, 2H, H-4), 3.24 (s, 1H,
OH), 6.21 (s, 1H, H-2), 7.72 (d, 2H, J = 8.0 Hz, Ar), 7.98 (d,
2H, J = 8.0 Hz, Ar), 15.91 (s, 1H, OH); 13C NMR (125 MHz,
CDCl3, dC) d 29.6 (CH3), 51.9 (C-4), 70.4 (C-5), 98.7 (C-2), 125.6
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 698–705 | 703
©

View Article Online
(Ar), 125.7 (Ar), 127.3 (Ar), 137.5 (Ar), 180.1 (C-1), 198.1 (C-3);
HRMS (ESI) calcd for (M - H) C14H14F3O3: 287.0895. Found:
287.0900.
CDCl3, dC) d 26.2 (CH3), 46.4 (C-3), 83.5 (C-2), 102.4 (C-5), 124.0
(Ar), 128.3 (Ar), 138.3 (Ar), 150.1 (Ar), 170.3 (C-6), 197.8 (C-4);
HRMS (ESI) calcd for (M + H) C13H14NO4: 248.0923. Found:
248.0915.
(Z)-1,5-dihydroxy-1-(4-methoxyphenyl)-5-methylhex-1-en-3-
one (5). Colourless oil (0.361 g, 72%); IR (KBr) n 3436 m, 2972
m, 2841 m, 1602 m, 1258 s, 1175 s, 1114 m, 1077 m, 1029 s, 982 m,
951 m, 909 m, 844 m, 806 m, 774 m, 682 w, 633 w cm-1; 1H NMR
(500 MHz, CDCl3, dH) d 1.31 (s, 6H, CH3), 2.58 (s, 2H, H-4), 3.50
(s, 1H, OH), 3.86 (s, 3H, CH3O), 6.11 (s, 1H, H-2), 6.94 (d, 2H,
J = 8.5 Hz, Ar), 7.86 (d, 2H, J = 8.5 Hz, Ar), 16.18 (s, 1H, OH);
13C NMR (125 MHz, CDCl3, dC) d 29.6 (CH3), 51.9 (C-4), 55.9
(OCH3), 70.5 (C-5), 99.3 (C-2), 109.1 (Ar), 114.1 (Ar), 126.8 (Ar),
130.5 (Ar), 178.3 (C-1), 199.0 (C-3); HRMS (ESI) calcd for (M +
H) C14H19O4: 251.1283. Found: 251.1280.
2,2-Dimethyl-6-p-((trifluoromethyl)phenyl)-2H -pyran-4(3H)-
one (10). Colourless oil (0.255 g, 95%); IR (KBr) n 1633 s, 1417 s,
1
1324 s, 1115 m, 1069 m, 1015 m, 983 m, 936 m, 809 m cm-1; H
NMR (500 MHz, CDCl3, dH) d 1.58 (s, 6H, CH3), 2.76 (s, 2H,
H-3), 6.25 (s, 1H, H-5), 7.69 (d, 2H, J = 8.0 Hz, Ar), 7.86 (d, 2H,
J = 8.0 Hz, Ar); 13C NMR (125 MHz, CDCl3, dC) d 26.3 (CH3),
47.1 (C-3), 82.6 (C-2), 101.9 (C-5), 125.9 (Ar), 127.4 (Ar), 133.6
(CF3), 133.8 (Ar), 136.6 (Ar), 169.1 (C-6), 195.5 (C-4); HRMS
(ESI) calcd for (M + H) C14H14F3O3: 271.0946. Found: 271.0942.
2,2-Dimethyl-6-p-(methoxyphenyl)-2H-pyran-4(3H)-one (11).
Colourless oil (0.227 g, 98%); IR (KBr) n 1633 s, 1249 s, 1169 s,
5-Hydroxy-5-methyl-1-((3aS,5R,5aR,8aS,8bS)-2,2,7,7-tetra-
methyltetrahydro-3aH -bis[1,3]dioxolo[4,5-b:4¢,5¢-d]pyran-5-yl)-
hexane-1,3-dione (6). Colourless oil (0.344 g, 46%); IR (KBr) n
3478 s, 2979 s, 1704 s, 1603 s, 1257 s, 901 m, 841 m, 787 m, 678 w,
633 s cm-1; 1H NMR (500 MHz, CDCl3, dH) d 1.23 (s, 6H, CH3),
1.28 (s, 3H, (O)2C(CH3)2), 1.33 (s, 3H, (O)2C(CH3)2), 1.41 (s, 3H,
(O)2C(CH3)2), 1.52 (s, 3H, (O)2C(CH3)2), 2.17 (s, 2H, H-2), 2.51
(s, 2H, H-4), 2.63 (s, 1H, (OH), 4.33-4.38 (m, 2H, H-3¢, H-5¢),
4.62 (dd, 1H, J = 7.5 Hz, J2 = 2.0 Hz, H-2¢), 4.67 (dd, 1H, J =
7.5 Hz, J2 = 2.0 Hz, H-4¢), 5.61 (d, 1H, J = 4.5 Hz, H-1¢); 13C
NMR (125 MHz, CDCl3, dC) d 24.7 (C(CH3)2), 25.1 (C(CH3)2),
26.1 (C(CH3)2), 26.3 (C(CH3)2), 29.5 (CH3), 29.7 (CH3), 50.6 (C-
4), 70.0 (C-4¢), 70.6 (C-5), 70.8 (C-2¢), 70.9 (C-3¢), 77.0 (C-2), 96.67
(C-5¢), 100.01 (C-1¢), 109.32 (C(CH3)2), 110.02 (C(CH3)2), 191.42
(C-3), 192.40 (C-1); HRMS (ESI) calcd for (M + H) C18H29O8:
373.1862. Found: 373.1862
1
1108 s, 983 s, 936 s, 892 m, 842 m, 771 m, 733 s, 679 m cm-1; H
NMR (500 MHz, CDCl3, dH) d 1.56 (s, 6H, CH3), 2.83 (s, 2H,
H-3), 3.86 (s, 3H, CH3O), 6.34 (s, 1H, H-5), 6.93 (d, 2H, J =
9.0 Hz, Ar), 7.86 (d, 2H, J = 9.0 Hz, Ar); 13C NMR (125 MHz,
CDCl3, dC) d 26.3 (CH3), 46.2 (OCH3), 55.7 (C-3), 82.2 (C-2),
99.2 (C-5), 114.4 (Ar), 124.7 (Ar), 129.5 (Ar), 163.6 (Ar), 173.3
(C-6), 196.2 (C-4); HRMS (ESI) calcd for (M + H) C14H17O3:
233.1178. Found: 233.1170.
2,2-Dimethyl-6-((3aS,5R,5aR,8aS,8bS)-2,2,7,7-tetramethyl-
tetrahydro-3aH-bis[1,3]dioxolo[4,5-b:4¢,5¢-d]pyran-5-yl)-2H-py-
ran-4(3H)-one (12). Colourless oil (0.315 g, 89%); IR (KBr) n
1652 s, 1455 s, 1372 s, 1259 s, 1055 m, 933 w, 898 w, 802 m,
733 w, 692 w cm-1; 1H NMR (500 MHz, CDCl3, dH) d 1.30
(s, 3H, (O)2C(CH3)2), 1.33 (s, 3H, (O)2C(CH3)2), 1.37 (s, 3H,
(O)2C(CH3)2), 1.38 (s, 3H, (O)2C(CH3)2), 1.45 (s, 3H, C(CH3)2),
1.53 (s, 3H, C(CH3)2), 2.48 (d, 1H, J = 16.5 Hz, H-3a), 2.57 (d,
1H, J = 16.5 Hz, H-3b), 4.27 (m, 1H, H-5¢), 4.36 (dd, 1H, J =
5.0 Hz, J2 = 2.5 Hz, H-3¢), 4.42 (dd, 1H, J = 8.0 Hz, J2 = 2.5 Hz,
H-2¢), 4.64 (dd, 1H, J = 8.0 Hz, J2 = 2.5 Hz, H-4¢), 5.59 (d,
1H, J = 5.5 Hz, H-1¢), 5.66 (s, 1H, H-5); 13C NMR (125 MHz,
CDCl3, dC) d 24.5 (C(CH3)2), 24.9 (C(CH3)2), 25.0 (C(CH3)2),
26.1 (C(CH3)2), 26.2 (CH3), 27.9 (CH3), 47.9 (C-3), 67.6 (C-4¢),
70.7 (C-2¢), 70.9 (C-3¢), 71.7 (C-5¢), 82.0 (C-2), 96.6 (C-1¢), 102.7
(C-5), 109.1 (C(CH3)2), 109.9 (C(CH3)2), 170.5 (C-6), 192.8 (C-
4); HRMS (ESI) calcd for (M + H) C18H27O7: 355.1757. Found:
355.1751.
2,2-Dimethyl-6-phenyl-2H-pyran-4(3H)-one (7). Yellow oil
(0.200 g, 99%); IR (KBr) n 1633 s, 1493 m, 1451 s, 1372 s, 1255 s,
1171 s, 1056 s, 1028 m, 983 m, 935 m, 891 m, 772 m, 692 m cm-1;
1H NMR (500 MHz, CDCl3, dH) d 1.55 (s, 6H, CH3), 2.67 (s,
2H, H-3), 6.12 (s, 1H, H-5), 7.43 (t, 2H, J = 8.0 Hz, Ar), 7.48
(t, 1H, J = 8.0 Hz, Ar), 7.74 (d, 2H, J = 8.0 Hz, Ar); 13C NMR
(125 MHz, CDCl3, dC) d 26.3 (CH3), 47.2 (C-3), 81.6 (C-2), 100.8
(C-5), 126.9 (Ar), 128.8 (Ar), 131.9 (Ar), 133.3 (Ar), 169.8 (C-6),
194.5 (C-4); HRMS (ESI) calcd for (M + H) C13H15O2: 203.1072.
Found: 203.1067.
2,2-Dimethyl-6-p-(cyanophenyl)-2H-pyran-4(3H)-one
(8).
Colourless oil (0.219 g, 97%); IR (KBr) n 2230 m, 1630 s, 1417 m,
1256 m, 1170 m, 1108 m, 1057 m, 1017 m, 983 m, 935 m cm-1; 1H
NMR (500 MHz, CDCl3, dH) d 1.59 (s, 6H, CH3), 2.85 (s, 2H,
H-3), 6.40 (s, 1H, H-5), 7.74 (d, 2H, J = 8.0 Hz, Ar), 7.87 (d, 2H,
J = 8.0 Hz, Ar); 13C NMR (125 MHz, CDCl3, dC) d 26.2 (CH3),
46.6 (C-3), 83.1 (C-2), 102.1 (C-5), 115.5 (Ar), 118.0 (CN), 127.7
(Ar), 132.6 (Ar), 136.8 (Ar), 169.8 (C-6), 197.0 (C-4); HRMS
(ESI) calcd for (M + H) C14H14NO2: 228.1025. Found: 228.1015.
Acknowledgements
We gratefully acknowledge financial support from the Australian
Research Council, Monash University and Universita¨t Leipzig.
References
1 S. Danishefsky, N. Kato, D. Askin and J. F. Kerwin, J. Am. Chem. Soc.,
1982, 104, 360.
2 S. Danishefsky, J. F. Kerwin Jr. and S. Kobayashi, J. Am. Chem. Soc.,
1982, 104, 358.
3 S. Danishefsky, E. R. Larson and D. Askin, J. Am. Chem. Soc., 1982,
104, 6457.
4 R. J. Light and C. R. Hauser, J. Org. Chem., 1961, 26, 1716.
5 J. R. Peterson, T. J. Winter and C. P. Miller, Synth. Commun., 1988, 18,
949.
2,2-Dimethyl-6-p-(nitrophenyl)-2H-pyran-4(3H)-one (9). Red
crystalline solid (0.240 g, 97%); Mp 154–156 ◦C; IR (KBr) n 1659 s,
1603 s, 1580 s, 1514 m, 1471 m, 1416 m, 1343 s, 1324 s, 1300 m,
1249 m, 1105 s, 1044 m, 1012 w, 981 m, 862 s, 849 s, 834 s, 815
1
m, 756 s, 693 s cm-1; H NMR (500 MHz, CDCl3, dH) d 1.56 (s,
6H, CH3), 2.62 (s, 2H, H-3), 6.06 (s, 1H, H-5), 7.88 (d, 2H, J =
6 D. Obrecht, Helv. Chim. Acta, 1991, 74, 27.
7 J. R. Peterson and E. W. Kirchhoff, Synlett, 1990, 394.
9.0 Hz, Ar), 8.27 (d, 2H, J = 9.0 Hz, Ar); 13C NMR (125 MHz,
704 | Org. Biomol. Chem., 2010, 8, 698–705
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
8 Y. R. Lee and K. Y. Kang, Lett. Org. Chem., 2007, 4, 440.
9 N. Ahmed, H. Ali and J. E. Van Lier, Tetrahedron Lett., 2005, 46, 253.
10 S. Tummala, J. W. Shabaker and S. S. W. Leung, Curr. Opin. Drug
Discovery Dev., 2005, 8, 789.
11 R. N. Landau, S. M. Penix, S. M. Donahue and A. J. Rein, Proc.
SPIE–Int. Soc. Opt. Eng., 1992, 1681, 356.
12 R. N. Landau, P. F. McKenzie, A. L. Forman, R. R. Dauer, M. Futran
and A. D. Epstein, Process Control Qual., 1995, 7, 133.
13 A. Pintar, J. Batista and J. Levec, Analyst, 2002, 127, 1535.
14 K. Zajsek and A. Gorsek, Rev. Chim, 2008, 59, 1308.
15 I. Poljansˇek, B. Likozar and M. Krajnc, J. Appl. Polym. Sci., 2007, 106,
878.
16 G. Keglevich, I. Csontos, N. Szila´gyi and I. Greiner, Chem. Eng.
Technol., 2008, 31(3), 421.
17 G. Keglevich, I. Csontos and N. Szila´gyi, Spectrosc. Lett., 2009, 42(2),
67.
18 S. T. Heller and S. R. Natarajan, Org. Lett., 2006, 8, 2675.
19 Comprehensive Chemical Kinetics, ed. N. Green, 2001, 38, p. 17.
20 R. S. Rasmussen, D. D. Tunnicliff and R. R. Brattain, J. Am. Chem.
Soc., 1949, 71, 1068.
21 J. G. Dawber and M. M. Crane, J. Chem. Educ., 1967, 44, 150.
22 A. Scho¨nberg, A. Mustafa and W. Asker, Nature, 1953, 171, 222.
23 J. Dubois, M. El-Alaoui and J. Toullec, J. Am. Chem. Soc., 1981, 103,
5393.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 698–705 | 705
©