One-pot synthesis of 1,3,5-tribenzoylbenzenes by three consecutive michael addition reactions of 1-phenyl-2-propyn-1-ones in pressurized hot water in the absence of added catalysts

Article abstract of DOI:10.1002/chem.201001946

Cyclotrimerization of 1-phenyl-2-propyn-1-one in pressurized hot water gave 1,3,5-tribenzoylbenzene in one pot in 65% yield after 7min at 200°C, or in 74% yield after 60min at 150°C. The reaction did not take place in the absence of water, and added base promoted the reaction at 250°C, suggesting a mechanism of three-consecutive Michael addition reactions. The reaction rates increased with temperature, but the yield of 1,3,5-tribenzoylbenzene decreased at the expense of formation of acetophenone as a side product at higher temperatures. p-Methyl and p-chloro-substituents on the phenyl ring retarded and enhanced the reaction, respectively. A mechanism involving the enol of benzoylacetaldehyde at a branching point of the pathway leading to 1,3,5-tribenzoylbenzene and acetophenone was suggested.

Full text of DOI:10.1002/chem.201001946

DOI: 10.1002/chem.201001946
One-Pot Synthesis of 1,3,5-Tribenzoylbenzenes by Three Consecutive
Michael Addition Reactions of 1-Phenyl-2-propyn-1-ones in Pressurized Hot
Water in the Absence of Added Catalysts
Makoto Tanaka,[b] Kazuya Nakamura,[b] Tatsuya Iwado,[b] Toshiyuki Sato,[b]
Masaki Okada,[b] Kiwamu Sue,[c] Hiizu Iwamura,*[a] and Toshihiko Hiaki*[b]
Abstract: Cyclotrimerization of 1-
phenyl-2-propyn-1-one in pressurized
hot water gave 1,3,5-tribenzoylbenzene
in one pot in 65% yield after 7 min at
2008C, or in 74% yield after 60 min at
1508C. The reaction did not take place
in the absence of water, and added
base promoted the reaction at 2508C,
suggesting a mechanism of three-con-
secutive Michael addition reactions.
The reaction rates increased with tem-
perature, but the yield of 1,3,5-triben-
zoylbenzene decreased at the expense
of formation of acetophenone as a side
product at higher temperatures. p-
Methyl and p-chloro-substituents on
the phenyl ring retarded and enhanced
the reaction, respectively. A mecha-
nism involving the enol of benzoylace-
taldehyde at a branching point of the
pathway leading to 1,3,5-tribenzoylben-
zene and acetophenone was suggested.
Keywords: cycloaddition
·
high-
pressure chemistry · Michael addi-
tion · supercritical fluids · water
chemistry
Introduction
the catalysts and removal of the solvents, which can cause
deterioration of waste water quality and emission of volatile
Typical organic reactions are often performed by the action
of acid or base catalysts in organic solvents. These reactions
usually require workup procedures, such as neutralization of
organic compoundsACTHUGNTERN(UNG VOC) into the atmosphere, thus in-
creasing the burden on the environment. Pressurized hot
water near the critical point (TC =374.158C, PC =
22.12 MPa) has various merits as an alternative reaction
medium. Its high specific dielectric constant (er) of 78 at
room temperature drops in parallel with the decrease in its
density to about 28 at 2508C (cf. er =28 for ethanol and 33
for methanol at 258C) and down to approximately 2 at
4508C (cf. er =2.3 for benzene at 258C),[1] enabling higher
solubility/miscibility of organic compounds in this medium.
Because the ionic dissociation of water is endothermic, its
ion product of 10À14 at 258C increases with temperature and,
because its er drops at higher temperature, it reaches a maxi-
mum three orders of magnitude greater at temperatures of
around 2508C under the saturated vapor pressure.[2] These
conditions facilitate various acid- or base-catalyzed organic
reactions in water in the absence of added catalysts.[3,4,5]
Herein, we report a one-pot cyclotrimerization reaction
(Scheme 1) of 1-phenyl-2-propyn-1-one to produce 1,3,5-tri-
benzoylbenzene in the absence of any added catalyst. It is
[a] Prof. Dr. H. Iwamura
College of Science and Technology
Nihon University
1-8-14 Kanda-Surugadai
Chiyoda-ku, Tokyo 101-8308 (Japan)
Fax : (+3)-3259-0468
E-mail: iwamura.hiizu@nihon-u.ac.jp
[b] M. Tanaka, K. Nakamura, T. Iwado, Dr. T. Sato, Dr. M. Okada,
Prof. Dr. T. Hiaki
College of Industrial Technology
Nihon University
1-2-1 Izumi-cho
Narashino, Chiba 275-8575 (Japan)
Fax : : (+47)-474-2579
E-mail: hiaki.toshihiko@nihon-u.ac.jp
[c] Dr. K. Sue
Nanosystem Research Institute
National Institute of
Advanced Industrial Science and Technology
Tsukuba Central 5, 1-1-1 Higashi
Tsukuba, Ibaraki 305-8565 (Japan)
ꢀ
well known that monosubstituted acetylenes RC CH under-
go cyclotrimerization reactions to give substituted benzenes
under the action of various transition-metal catalysts.[6]
Reppeꢀs catalysts, consisting of nickel or cobalt carbonyls,
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem201001946.
606
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 606 – 612

FULL PAPER
tion polymer,[14] and tri(p-octyloxybenzoyl)benzene has been
fabricated into a monolayer.[15] When attached with poly-
AHCTUNGTREG(NNUN oxyethylene) units, the resulting crownophanes crystallized
as inclusion solids.[16] Moreover, the propiolyl group has
been introduced into a calixarene moiety, which was cyclo-
trimerized to form tri- and pentacalixarenes that served as
host molecules for amines, polyamines, and C60.[17] 1,3,5-Tri-
benzoylbenzene serves, among other things, as key synthetic
intermediates for the synthesis of dendrimers of various
functions, including superparamagnetic polycarbenes and
1,3,3-trisubstituted 2-oxindoles.[12b,18,19] Hyperbranched poly-
mers with a high degree of branching, such as poly(aroxycar-
bonylphenylene)s and ferrocene-containing hyperbranched
poly(aroylarylene)s have been prepared.[20a,b] The physico-
chemical properties of disc-like molecules of medium size
thus warrants further investigation, and herein we report a
synthetic study of 1,3,5-triaroylbenzene in pressurized hot
water.[21]
Scheme 1. One-pot cyclotrimerization of 1-phenyl-2-propyn-1-one.
give mainly 1,2,4-trisubstituted benzenes, with lower
amounts of the 1,3,5-isomers also generally obtained, and
little, if any, of the 1,2,3-isomers.[7] The Ziegler-type catalyst,
a tri-isobutylaluminum-TiCl4 system, has been used to cyclo-
trimerize 2-ethynylfuran.[8] More recently, a (4-hydroxybu-
tyrylcyclopentadienyl)Co-h4-COD catalyst (COD=1,5-cy-
clooctadiene) and a ruthenium(IV) precatalyst have been
successfully used in the cyclotrimerization of acetylenes in
aqueous solvents.[9a,b] Furthermore, [Co(CO)2(Cp)]-cata-
lyzed (Cp=cyclopentadienyl) reactions have even been car-
ried out in supercritical water.[9c,d] All of the transition-
metal-catalyzed reactions, including the latter approach,
have the same problem regarding the regioselective forma-
tion of 1,3,5-isomers from terminal acetylenes.
Whereas Sasaki and Suzuki had obtained 1,3,5-tri(2-furan-
carbonyl)benzene in 20% yield during the oxidation of 1-(2-
furyl)-2-propyn-1-ol with active manganese oxide,[10] it was
Balasubramanian et al. who first reported the unique forma-
tion of 1,3,5-tribenzoylbenzene by three consecutive Michael
addition reactions of 1-phenyl-2-propyn-1-one in the pres-
ence of diethylamine as a catalyst in organic solvents such
as dichloromethane, N,N-dimethylformamide (DMF), and
toluene.[11] Mechanistic studies of the cyclotrimerization re-
action by Matsuda et al. revealed that the amine catalyst ini-
tially reacts with an alkynyl moiety to form an enaminoke-
tone. Sequential condensation of this enamine with two ad-
ditional molecules of aryl ethynyl ketone results in the for-
mation of a cyclohexadiene, from which the amine is elimi-
nated, thereby generating the central benzene ring.[12] Our
conjecture is based on the high OHÀ concentration in pres-
surized hot water, which might be able to replace NH-
Results and Discussion
Formation of 1,3,5-tribenzyolbenzenes: All reactions gave
colorless powders in various amounts depending on the re-
action conditions. The solid products were shown to be
1,3,5-tribenzyolbenzene on the following grounds: 1) Its
melting point (118–1198C) was in good agreement with the
reported melting point (1198C).[11,12a,22] 2) A mass spectral
parent peak at m/z 390, and a base peak at m/z 105 due to
the fragment ion C6H5CO, and other fragmentation peaks
agreed well with those of an authentic sample of 1,3,5-tri-
benzyolbenzene. 3) The FTIR spectra of a sample in a KBr
disc showed peaks that were characteristic of conjugated
1
carbonyl absorptions at 1660 cmÀ1.[22] 4) The H NMR spec-
trum showed a singlet at d=8.47 ppm due to three protons
at the 2-, 4-, and 6-positions on the inner benzene ring, mul-
tiplets (d=7.91–7.96 ppm) due to six protons ortho to the
carbonyls and remaining multiplets (9 H, d=7.57–7.71 ppm)
due to the three outer benzene rings.[22] The identities of
1,3,5-tri(p-methylbenzoyl)benzene and 1,3,5-tri(p-chloroben-
1
zoyl)benzene were confirmed by comparing their H NMR
G
spectra with those reported in the literature.[22]
through hydrogen bonding and other weak interactions,
1,3,5-triaroylbenzene itself serves as a molecular host not
only in solution but also in the solid phase.[13] The tri(p-
cyanophenyl) derivative was used as a ligand for a coordina-
The effect of the amount of water on the yield of 1,3,5-tri-
benzoylbenzene: A series of experiments were performed at
2508C (10min) in which the amount of water was adjusted
within the molar ratio 0–600 relative to one mole of 1-
phenyl-2-propyn-1-one as a starting material. The results are
given in the Supporting Information (Table S1) and are sum-
marized in Figure 1.
It is significant that 1,3,5-tribenzoylbenzene was not
formed in the absence of water. Its yield increased sharply
with increasing relative amounts of water, leveling off at a
yield of approximately 55% upon introducing water in a
300:1 ratio with phenylpropynone. The results are interpret-
ed in terms of the highest ion product of water at 2508C
under saturated vapor pressure and catalysis of the reactions
Scheme 2. X=NHACHTUNGTRENNUNG
(C2H5)2 in the previous work[11,12] and OHÀ due to
pressurized hot water in this work.
Chem. Eur. J. 2011, 17, 606 – 612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
607

H. Iwamura, T. Hiaki et al.
uct is thermally stable in hot pressurized water. The highest
yield of 1,3,5-tribenzoylbenzene was 65% at 2008C after
7 min or 74% at 1508C after 60 min. The initial rates, which
were estimated from the less reliable data after 1 min of re-
action and the slopes of the rising curves in Figure 2, were
greater as the temperature increased. However, the saturat-
ed yields of 1,3,5-tribenzoylbenzene became lower at higher
temperatures. We also noted that most of the phenylpropy-
none starting material was lost during the early stages of the
reactions. We determined the amount of unreacted starting
materials by HPLC analysis and found that the molar
amount was less than 1% after 1 min at 2008C. The obser-
vation strongly implies the presence of branching in the
mechanism, in which product formation is accompanied by
a loss of entropy (see below). This is not surprising given
the nature of the cyclotrimerization reaction.
Figure 1. Changes in the amount of 1,3,5-tribenzoylbenzene with the rela-
tive amount of water for the reactions fixed at 2508C for 10 min.
through latent H+ or OHÀ ions available from the large
excess of water.
As described in the Experimental Section, possible cata-
lytic metal sites on the inner wall of the reaction vessel were
quenched by “aging”. The inner wall may be regarded as
being coated with oxide films of Fe3O4 (sometimes Fe2O3),
NiO, and CrOOH. Even if these sites were crucial for the
cyclotrimerization reaction, however, it was concluded that
they would not provide a suitable reaction environment in
the absence of water. According to the thermodynamic eval-
uation by Ziemniak,[23] the concentration of the solubilized
oxides are estimated to be 2.5, 0.6, and 0.01 ppb for Fe3O4,
NiO, and CrOOH, respectively, at temperatures of approxi-
mately 2508C, and these values decrease as the temperature
increases.
Because the reaction vessel was sealed at ambient temper-
ature and pressure, the water in the reactor was considered
to be in equilibrium between the vapor and liquid phases
below the critical point. At 150, 200, 250, 300, and 3708C,
the density of neat water is estimated to decrease in the
order: 0.92, 0.86, 0.80, 0.71, and 0.45 gcmÀ3, and the internal
pressure of the vessel increases in the order 0.48, 1.6, 4.0,
8.6, and 21 MPa, respectively. By definition of the supercriti-
cal state, only one phase, with a density of 0.35 gcmÀ3 and a
pressure of 30 MPa, should be present at 4008C. We found
no drastic change in the temperature dependence of the for-
mation of 1,3,5-tribenzoylbenzene on passing through the
critical point (Figure 2).
The effects of reaction time and temperature: Temporal var-
iation of the amounts of 1,3,5-tribenzoylbenzene formed
when the reaction was performed in a 400-fold molar excess
water was examined at temperatures of between 150–4008C.
The results are given in the Supporting Information
(Table S2) and summarized in Figure 2.
The effect of the pH of water on the rates of the reaction:
To provide support for the mechanism of the three consecu-
tive Michael addition reactions for the formation of 1,3,5-tri-
benzoylbenzene, the pH of the aqueous solutions were ad-
justed to 3.0, 4.0, 5.0, 6.0, 7.0, and 8.0 at 258C by prior addi-
tion of measured amounts of aqueous hydrochloric acid
(0.1m) or sodium hydroxide (0.01m) to pure water to make
10À3, 10À4, 10À5, and 10À6 moldmÀ3 solutions of HCl and
NaOH. The effective pH values at 2508C were calculated by
using the relevant dissociation constants of water, hydro-
chloric acid, and sodium hydroxide, and the ionic activity
coefficients reported in the literature,[24] as described in the
previous paper.[5] The results are listed in Table 1.
Prolonged heating did not decrease the amount of 1,3,5-
tribenzoylbenzene once formed, showing that the final prod-
The product yields of 1,3,5-tribenzoylbenzene in 400-
molar excess water after 1 min at 2508C were less than 10%
at pH 3–5, it rose gradually, and reached 48% at pH 7.2, as
shown in Figure 3. No improvement in the accuracy of the
Table 1. Calculated pH values of the reaction media at 2508C.
pH at 258C
pH at 2508C
3.0
4.0
5.0
6.0
7.0
8.0
3.0
4.0
5.0
5.3
5.6
7.2
Figure 2. Temporal variation in the amounts of product 1,3,5-tribenzoyl-
~
*
benzene formed in 400-fold excess water at 150 ( ), 200 ( ), 250 (&), 300
~
&
*
(
), 370 ( ), and 4008C ( ).
608
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 606 – 612

Synthesis of 1,3,5-Tribenzoylbenzenes
FULL PAPER
invoked to explain the lack of reactivity in the hot pressur-
ized water system used in this study. The lower solubility of
the starting material due to the naphthyl moiety in the sub-
critical water could, however, be responsible for the lack of
reactivity.
Formation of acetophenone as a byproduct and reaction
mechanism: Acetophenone, as a byproduct, always accom-
panied the formation of tribenzoylbenzene. A mass spectral
parent peak at m/z 120, a base peak at m/z 105 due to frag-
ment ion C6H5CO, and other fragmentation peaks agreed
well with those of an authentic sample of acetophenone.
The FTIR spectra contained peaks that were characteristic
of conjugated carbonyl absorptions at 1690 cmÀ1. The
1H NMR spectrum showed a singlet at d=2.50 ppm (3H,
CH3), together with protons in the aromatic region d=7.35–
7.38 (2H), 7.44–7.47 (1H), and 7.85–7.90 ppm (2H).
Figure 3. Effect of pH on the yield of 1,3,5-tribenzoylbenzene in 400-
molar excess water at the earliest reaction time of 1 min at 2508C.
Temporal variation in the production of acetophenone in
a 400-fold molar excess of water was determined at 150–
4008C, and the results are given in the Supporting Informa-
tion (Table S3) and Figure 4. The amount of acetophenone
gradually increased with reaction time. At approximately
3008C, the molar yield of acetophenone became comparable
to that of 1,3,5-tribenzoylbenzene, and the former became
the more predominant product at higher temperatures.
measurements or in the analysis could establish a linear re-
lationship between logarithm of the initial rate R of the
product formation and pH (logRꢁpH), which would be ex-
pected for a simple base catalysis in the rate-determining
step of the reaction sequence. This would suggest that the
Michael addition reaction involves stronger base catalysis
and weaker acid catalysis. In the crossed aldol condensation
in pressurized hot water, both V-shaped and linear plots are
reported for the formation of benzalacetone and pentacene-
quinone,[4a,5] respectively.
1-(p-Substituted phenyl)-2-propyn-1-ones: To extend the ap-
plicability of the cyclotrimerization reaction and to obtain
experimental support for the mechanism, three 1-phenyl-2-
propyn-1-ones carrying a para substituent on the 1-phenyl
ring, were studied. The results obtained by using a 400-
molar excess of water after 5 min at 2008C are summarized
in Table 2 and show that, during the earlier parts of the re-
actions, the p-chloro and p-methyl substituents seem to en-
hance and retard the reactions, respectively. The electron-
withdrawing and -donating substituents differ in the Ham-
+
mett sp constants by 0.425, revealing the buildup of nega-
tive charges at the side chains of the phenyl ring in the tran-
sition state of the reaction. The 1-(2-naphthyl) derivative re-
mained unreacted even after 6 min at 2508C. Because even
1-(1-naphthyl)-2-propyn-1-one is reported to undergo the
cyclotrimerization reaction to give the corresponding 1,3,5-
tri(1-naphthoyl)benzene under classical amine-catalyzed
conditions in organic solvent,[22] a steric effect could not be
Figure 4. Temporal variation in the formation of acetophenone in 400-
~
~
*
*
fold excess water at 150 ( ), 200 ( ), 250 (&), 300 ( ), 370 ( ), and
&
4008C ( ).
Although comparison of the GC-MS minor peaks to refer-
1
ence peaks held in the database, and the H NMR spectrum
of the crude reaction mixture both suggested the additional
formation of 1-indenone and 4-(hydroxymethylene)-1,5-di-
phenyl-2-pentene-1,5-dione (dimer in Scheme 3), the
amount of these byproducts was small. A precedent for the
formation of 1-indenone is found in the formation of 3-
arylindenone from 1,3-diarylpropynones in superacids.[26]
In addition to the observation that the starting 1-phenyl-
propynone disappeared sooner than the triketone formed,
Table 2. Reactions of 1-(substituted phenyl)-2-propyn-1-ones after 5 min
at 2008C.
ꢀ
Ar in ArCOC CH
Yield [%]
1-phenyl
37.1
33.6
49.0
0
1-(p-methylphenyl)
1-(p-chlorophenyl)
1-(2-naphthyl)
Chem. Eur. J. 2011, 17, 606 – 612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
609

H. Iwamura, T. Hiaki et al.
vored at higher temperature at
the expense of the cyclization
reaction.
Conclusion
It has been established that the
threefold Michael addition of 1-
phenyl-2-propyn-1-one gives cy-
clotrimerized 1,3,5-tribenzoyl-
benzene in good yield in pres-
surized hot water, in the ab-
sence of any organic solvent
and added catalyst. In contrast
to the use of triethylamine as a
catalyst, which is a hazardous
organic base, if not carcinogen-
ic,[27] and an organic solvent
such as dichloromethane, tolu-
ene, or DMF in the convention-
al organic preparative methods,
the present approach of using
pure water has the merit of re-
ducing the burden on the envi-
ronment. The size of the reac-
tion vessel used in this work
was miniaturized for rapid tem-
perature equilibration in the
metal salt bath and to conserve
Scheme 3. A plausible reaction pathway.
the starting material. It could,
however, readily be scaled up
HPLC analysis of the material balance during the course of
the reaction under various conditions revealed that, at tem-
peratures higher than approximately 3008C, the molar
amounts of 1,3,5-tribenzoylbenzene and acetophenone ac-
counted for all the starting materials consumed. At lower
temperatures, the material balance was not satisfactory; the
total amount of both of the products was considerably lower
than that of the starting material consumed during the earli-
er part of the reaction.
All the results can be interpreted in terms of a plausible
reaction mechanism given in Scheme 3. The nucleophilic ad-
dition of OHÀ at the terminal ethynyl carbon of the phenyl-
propynone must be a rate-determining step, which would be
followed by protonation of the vinylcarbanion to give the
enol of benzoylacetaldehyde. This intermediate serves as an
important branching point in the reactions. The reaction
with a second molecule of 1-phenyl-2-propyn-1-one, al-
though unfavorable in terms of reaction entropy, leads to
the formation of 1,3,5-tribenzoylbenzene. The preparation
of benzoylacetaldehyde by base-catalyzed aldol condensa-
tion of acetophenone with alkyl formate in nonhydrolytic
solvents has been described in the literature,[25a,b] and under-
goes a hydroxide ion catalyzed reverse reaction to give ace-
tophenone and a formate ion.[25c] The latter reaction should
be favorable in terms of reaction entropy and, therefore, fa-
to a cylinder accommodating 100 cm3 of water. A continu-
ous flow reactor used for the production of nanoparticles of
various oxides[28] could also be modified to produce 1,3,5-tri-
benzoyolbenzene at a rate of 1 kg per day.
An analysis of the byproducts formed provided hints for
the mechanism of the reaction. A series of experiments
showed unequivocally that the increased solubility of the
starting carbonyl compounds, as a result of lowered dielec-
tric constant and the higher ion product under saturated
water vapor pressure of pressurized hot water at 2508C, are
responsible for the successful addition reaction. It is clear
that another merit of the pressurized hot water system as an
alternative reaction media is the high temperature and con-
sequent short reaction time, although we have to note that
the selectivity of the parallel reactions at high temperatures,
if present, could be lower than in lower temperature reac-
tions. Finally, we would like to suggest that the process
might be extended to sets of two or three different 1-aryl-2-
propyn-1-ones that should lead to unsymmetrically substitut-
ed 1,3,5-triaroylbenzenes, which are precursors of a number
of interesting dendrimers.[18,19] Such reactions have been suc-
cessfully performed in organic solvents with added amine
catalysts. Furthermore, the phenyl ring of 1-phenyl-2-
propyn-1-one can include a protected propiolyl group as a
substituent. Deprotection of the formed 1,3,5-triaroylben-
610
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 606 – 612

Synthesis of 1,3,5-Tribenzoylbenzenes
FULL PAPER
ing the reactor from the heat bath and immersing it promptly in cold
water. The reactor heat-up time was on the order of 30 s, which was gen-
erally short compared with the reaction times employed.
zene, followed by cyclotrimerization, is expected to produce
unsymmetrically extended dendrimers, as reported by Mat-
suda and co-workers under classical conditions for nona-
and dodecaketones. Such compounds, after conversion into
the corresponding polydiazo derivatives followed by photol-
ysis at cryogenic temperature, gave strongly paramagnetic
super-high-spin organic compounds.[18]
Analyses of the reaction mixture: GC–MS analyses were carried out with
a Shimadzu GC–MS-QP2010 spectrometer. Into a column of Agilent
DB-5MS, 1.5 mL aliquots of the solution of the products (dried over
MgSO4) in dichloromethane were injected with a split ratio of 100 with a
flow rate of helium carrier gas of 28 cmsÀ1. After being kept at 608C for
4 min, the column was heated to 2808C at a rate of 208CminÀ1 and kept
at 2808C for 45 min. NMR spectra were recorded with a Bruker AXS
NMR Avance-400 spectrometer operating at 400 MHz, with samples dis-
solved in CDCl3 and TMS as an internal standard. FTIR spectra were
measured with a Shimadzu FTIR-8400S spectrometer, with solid samples
in KBr discs.
Experimental Section
Chemicals: 1-Phenyl-2-propyn-1-ol of chemical purity greater than 98%
was purchased from Tokyo Chemical Industry and used without further
purification for the synthesis of 1-phenyl-2-propyn-1-one after oxidation
with potassium permanganate supported on active manganese dioxide.[29]
The 1H NMR and GC-MS spectra of the product, which was obtained in
quantitative yield, agreed with those reported in the literature.[30] Ultra-
pure water with a specific resistivity greater than 18.2 MW and TOC less
than 20 ppb was obtained by treating tap water through a Millipore
Milli-RX75 ultrapure water purification system. Other chemicals and sol-
vents were commercially available and used as received. Aqueous hydro-
chloric acid (0.1m) and sodium hydroxide (0.01m) from Wako Pure
Chemical Industries were used directly for pH-adjusted reactions.
For quantitative analyses, methanol was added to the reaction mixture in
the reactor to make a homogeneous solution of a given amount in the
range 15–20 cm3. An aliquot (1 mL) was injected into a Shimadzu high-
pressure liquid chromatograph analyzer (Prominence HPLC20) and de-
veloped with 50% aqueous acetonitrile under the flow rate of
1 mLminÀ1. The amounts of 1,3,5-tribenzoylbenzene and acetophenone
produced were determined accurately by reference to added acetone as
an internal standard and on the basis of calibration curves obtained inde-
pendently for the pure samples. The variance of the yield data in these
studies were dominated by the uncertainty of these manual processes and
was estimated to be 5%.
The amount of substituted 1,3,5-tribenzoylbenzene was determined by
NMR spectroscopic analysis of the crude reaction products with a given
amount of 1,3,5-triacetylbenzene added as an internal standard.
Reaction vessel: A batch reactor (Figure 5 and Figure S1 in the Support-
ing Information) made of a SUS316 1/16-inch plug (Swagelok, No. SS-
100-P), a 1/16-inch to 1/8-inch reducing connector (Swagelok, No. SS-
Acknowledgements
This research was supported in part by a Grant from the Ministry of Edu-
cation, Culture, Sports, Science and Technology to promote multi-disci-
plinary research projects. We thank Professor Yoshiaki Nishibayashi and
Dr. Yoshihiro Miyake of the University of Tokyo for the generous offer
of samples of variously substituted 1-phenylpropynone and their spectral
data. We are also grateful to Professor Sakae Uemura and Professor
Kenji Matsuda for valuable advice given at the earliest stage of this
work.
[1] a) M. Uematsu, E. U. Franck, Ber. Bunsen-Ges. 1980, 95, 1586–
1593; b) M. Uematsu, E. U. Franck, J. Phys. Chem. Ref. Data 1980,
9, 1291–1306; c) D. P. Fernandez, A. R. H. Goodwin, E. W.
Lemmon, J. M. H. L. Sengers, R. C. Williams, J. Phys. Chem. Ref.
Data 1997, 26, 1125–1166.
[2] a) G. M. Schneider, Ber. Bunsen-Ges. 1972 76, 325–331; b) W. L.
Marshall, E. U. Franck, J. Phys. Chem. Ref. Data 1981, 10, 295–304.
[3] Reviewed in: a) P. E. Savage, S. Gopalan, T. I. Mizan, C. J. Martino,
E. E. Brock, AIChE J. 1995, 41, 1723–1778; b) A. R. Katritzky,
S. M. Allin, M. Siskin, Acc. Chem. Res. 1996, 29, 399–406; c) P. E.
Savage, Chem. Rev. 1999, 99, 603–621; d) A. Kruse, E. Dinjus, J. Su-
percrit. Fluids 2007, 39, 362–380.
[4] For representative papers, see: a) C. M. Comisar, P. E. Savage,
Green Chem. 2004, 6, 227–231; b) R. Mestres, Green Chem. 2004, 6,
583–603; c) Y. Nagai, N. Matsubayashi, M. Nakahara, Bull. Chem.
Soc. Jpn. 2004, 77, 691–697; d) N. Akiya, E. Savage, Ind. Eng.
Chem. Res. 2001, 40, 1822–1831; e) Y. Harano, H. Sato, F. Hirata, J.
Am. Chem. Soc. 2000, 122, 2289–2293; f) M. Sasaki, Ind. Eng.
Chem. Res. 2002, 41, 6642–6649; g) Y. Ikushima, K. Hatakeda, O.
Sato, T. Yokoyama, M. Arai, J. Am. Chem. Soc. 2000, 122, 1908–
1918; h) Y. Ikushima, O. Satoa, M. Sato, K. Hatakeda, M. Arai,
Chem. Eng. Sci. 2003, 58, 935–941.
Figure 5. A SUS316 reaction vessel.
200-R-8), and a 1/8-inch to 1/2-inch reducing connector (Swagelok, No.
SS-100-R-2) were fitted with a 1/2-inch cap (Swagelok, No. SS-810-C),
and had an effective inner volume of 2.1 cm3.[5] Prior to its use in experi-
ments, the reactor was loaded with 2% aqueous hydrogen peroxide and
conditioned for 1 h at 3708C to remove any lubricants/oils that might
remain from the manufacture of the Swagelok parts and possible catalyt-
ic metal sites on the inner wall, unless otherwise stated. The reactor was
cleaned with acetone and dried prior to repeated use of about 10 and 5
times for the reactions carried out at 250 and 4008C, respectively.
Heating, temperature control, and a typical run of the cyclotrimerization
reaction: 1-Phenyl-2-propyn-1-one (0.0136 g) and water (0.751 g), making
a molar ratio of 1:400, were introduced into the reaction vessel and the
contents were purged with a stream of argon. The capped vessel was im-
mersed in a metal salt bath (Taiatsu Techno, Model TBC-B600) preheat-
ed at 200, 250, 300, 370, or (400Æ1)8C. The pressure in the reaction
vessel was calculated to be 1.6, 4.0, 8.6, 21, or 30 MPa, respectively. After
the given reaction time (1–10 min), heating was discontinued by remov-
[5] T. Sawada, S. Nakayama, A. Kawai-Nakamura, K. Sue, H. Iwamura,
T. Hiaki, Green Chem. 2009, 11, 1675–1680.
[6] M. Lautens, W. Klute, W. Tam, Chem. Rev. 1996, 96, 49–92.
Chem. Eur. J. 2011, 17, 606 – 612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
611

H. Iwamura, T. Hiaki et al.
[7] a) W. Reppe, O. Schlichtig, K. Klager, T. Topel, Liebigs Ann. 1948,
560, 1–92; b) W. Reppe, N. von Kutepow, A. Magin, Angew. Chem.
1969, 81, 717–723; Angew. Chem. Int. Ed. Engl. 1969, 8, 727–733.
[8] a) L. S. Meriwether, E. C. Clothup, G. W. Kennerly, R. N. Reusch, J.
Org. Chem. 1961, 26, 5155; b) V. O. Reikhstelꢀd, K. L. Makovetskii,
Russ. Chem. Rev. 1966, 35, 510.
[9] a) M. S. Sigman, A. W. Fatland, B. E. Eaton, J. Am. Chem. Soc.
1998, 120, 5130–5131; b) V. Cadierno, S. E. Garcꢂa-Garrido, J.
Gimeno, J. Am. Chem. Soc. 2006, 128, 15094–15095; c) K. S.
Jerome, E. J. Parsons, Organometallics 1993, 12, 2991–2993; d) H.
Borwieck, O. Walter, E. Dinjus, J. Rebizant, J. Organomet. Chem.
1998, 570, 121–127.
[10] For a review of the chemistry of ethynyl ketones as of February,
1971, see: R. L. Bolꢀshedvorskaya, L. I. Vereshchagin, Russ. Chem.
Rev. 1973, 42, 225, however, no mention was made of the work on
possible cyclotrimerization of 1-(2-furyl)-2-propyn-1-one by T.
Sasaki, Y. Suzuki, Tetrahedron Lett. 1967, 8, 3137–3140.
[19] A. Vandendriessche, J. Thomas, C. Van Oosterwijck, J. Huybrechts,
B. Dervaux, S. Dꢀhollander, F. Du Prez, W. Dehaen, M. Smet, Eur.
Polym. J. 2009, 45, 3196–3209.
[20] a) C. K. W. Jim, A. Qin, J. W. Y. Lam, J.-Z. Liu, M. Haeussler, B. Z.
Tang, Sci. China, Ser. B: Chem. 2008, 51, 705–708; b) H. Dong, A.
Qin, C. K. W. Jim, J. W. Y. Lam, M. Haussler, B. Z. Tang, J. Inorg.
Organomet. Polym. 2008, 18, 201–205.
[21] C. Tschierske, Nanoscale Stereochemistry in Liquid Crystals in Chir-
ality at the Nanoscale (Ed.: D. B. Amabilino), Wiley-VCH, Wein-
heim, 2009, ch. 9.
[22] I. Elghamry, Synthesis 2003, 2301–2303.
[23] We thank one of the referees for pointing out that high-temperature
water often leaches metals from 316 steel and the metals might be
catalyzing the reaction. It is a continuing problem to experimental
chemists in this field to exclude the effect of such artifacts. Although
poly(tetrafluoroethylene) has a melting point of 3278C, its proper-
ties degrade above 2608C, and so cannot be used to cover the inner
wall of the reaction vessel. It was difficult to find chelating agents
that can withstand pressurized hot water. For these reasons, we
relied on the reported amounts of metal oxides, see: S. E. Ziemniak,
J. Solution Chem. 1992, 21, 745–760.
[24] K. Sue, M. Uchida, T. Usami, T. Adschiri, K. Arai, J. Supercrit.
Fluids 2004, 28, 287–296.
[25] a) J. Rahil, R. F. Pratt, J. Am. Chem. Soc. 1977, 99, 2661–2665; b) N.
Talınlı, B. Karlıga, J. Heterocycl. Chem. 2004, 41, 205–209; c) L. B.
Fedor, B. S. R. Murty, N. C. De, J. Am. Chem. Soc. 1975, 97, 4308–
4312.
[26] A. V. Vasilyev, S. Walspurger, P. Pale, J. Sommer, Tetrahedron Lett.
2004, 45, 3379–3381.
[27] For example, see: Fine Chem. Trading—UK: http://www.
chemfinder.co.uk/diethylamine_MSDS_DEA.html.
[28] a) T. Adschiri, Chem. Lett. 2007, 36, 1188–1193; b) T. Sato, K. Sue,
W. Suzuki, M. Suzuki, K. Matsui, Y. Hakuta, H. Hayashi, K. Arai, S.
Kawasaki, A. Kawai-Nakamura, T. Hiaki, Ind. Eng. Chem. Res.
2008, 47, 1855–1860.
[11] K. Balasubramanian, P. S. Selvaraj, P. S. Venkataramani, Synthesis
1980, 29–30.
[12] a) K. Matsuda, N. Nakamura and H. Iwamura, Chem. Lett. 1994,
1765–1768; b) K. Matsuda, N. Nakamura, K. Takahashi, K. Inoue,
N. Koga, H. Iwamura, J. Am. Chem. Soc. 1995, 117, 5550–5560.
[13] a) F. C. Pigge, F. Ghasedi, N. P. Rath, Tetrahedron Lett. 1999, 40,
8045–8048; b) F. C. Pigge, Z. Zheng, N. P. Rath, New J. Chem. 2000,
24, 183–185; c) F. C. Pigge, F. Ghasedi, Tetrahedron Lett. 2000, 41,
6545–6549; d) F. C. Pigge, F. Ghasedi, Z. Zheng, N. P. Rath, G.
Nichols, J. Chickos, J. Chem. Soc. Perkin Trans. 2 2000, 2458–2464.
[14] F. C. Pigge, M. D. Burgard, N. P. Rath, Crystal Growth Des. 2003, 3,
331–337.
[15] M. K. Dighe, F. J. Dover, K. J. Stine, F. C. Pigge, Thin Solid Films
2008, 516, 3227–3238.
[16] a) F. C. Pigge, M. K. Dighe, J. C. D. Houtman, J. Org. Chem. 2008,
73, 2760–2767; b) F. C. Pigge, A. V. Schmitt, N. P. Rath, J. Inclusion
Phenom. Macrocyclic Chem. 2008, 62, 105–110.
[17] J. Wang, C. D. Gutsche, J. Org. Chem. 2002, 67, 4423–4429.
[18] a) H. Iwamura, K. Matsuda, in Modern Acetylene Chemistry (Eds.:
P. J. Stang, F. Diederich), VCH, Weinheim, 1995, pp. 385–414; b) K.
Matsuda, N. Nakamura, K. Inoue, N. Koga, H. Iwamura, Chem. Eur.
J. 1996, 2, 259–264; c) K. Matsuda, N. Nakamura, K. Inoue, N. Kog,
H. Iwamura, Bull. Chem. Soc. Jpn. 1996, 69, 1483–1494.
[29] A. Shaabani, P. Mirzaei, S. Naderi, D. G. Lee, Tetrahedron 2004, 60,
11415–11420.
[30] Y. Maeda, N. Kakiuchi, S. Matsumura, T. Nishimura, T. Kawamura,
S. Uemura, J. Org. Chem. 2002, 67, 6718–6724.
Received: September 16, 2010
Published online: November 12, 2010
612
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 606 – 612