A thermodynamic study of ketoreductase-catalyzed reactions 2. Reduction of cycloalkanones in non-aqueous solvents

Article abstract of DOI:10.1016/j.jct.2005.06.005

The equilibrium constants for the ketoreductase-catalyzed reactions (cycloalkanone + 2-propanol = cycloalkanol + acetone) have been measured in n-hexane, n-pentane, and supercritical carbon dioxide SCCO₂ (pressure = (8.0 to 12.0) MPa). The cycloalkanones included in this study were: cyclobutanone, cyclopentanone, cyclohexanone, cycloheptanone, and cyclooctanone. The equilibrium constants for the reactions involving cyclobutanone and cyclohexanone were measured in n-hexane over the range T = (288.35 to 308.05) K. The thermodynamic quantities at T = 298.15 K are: K = (0.763 ± 0.001); ΔrGm=(0.670±0.002)kJ·mol-1; ΔrHm=-(1.09±0.11) kJ·mol-1, and ΔrSm=-(5.9±0.4)J·K-1·mol-1 for the reaction involving cyclobutanone; and K = (15.7 ± 0.2); ΔrGm=-(6.82±0.02)kJ·mol-1; ΔrHm=-(4.6±1.0) kJ·mol-1, and ΔrSm=(7.4±3.3)J·K-1·mol-1 for the reaction involving cyclohexanone, respectively. An inspection of the equilibrium constants for these reactions in n-hexane, n-pentane, and SCCO ₂ shows that solvent dependence is not significant. The equilibrium constants of cycloalkanones decrease with increasing value of the number of carbons, N_C with the exception of cyclohexanone. The cyclohexanol, which adopts a nearly strainless, idealized tetrahedral conformation around each carbon, is thermodynamically favored and more stable compared to other cycloalkanol rings, and this is reflected in the significantly higher value of the equilibrium constant obtained for this reaction. Comparisons with results obtained by using two independent thermochemical routes are also made.

Full text of DOI:10.1016/j.jct.2005.06.005

J. Chem. Thermodynamics 38 (2006) 388–395
www.elsevier.com/locate/jct
A thermodynamic study of ketoreductase-catalyzed reactions
2. Reduction of cycloalkanones in non-aqueous solvents
a,
b
c
Yadu B. Tewari *, Karen W. Phinney , Joel F. Liebman
a
Biotechnology Division, National Institute of Standards and Technology, Gaithersburg, MD 20899-8312, USA
Analytical Chemistry Division, National Institute of Standards and Technology, Gaithersburg, MD 20899-8312, USA
b
c
Department of Chemistry, University of Maryland, Baltimore County, Baltimore, MD 21250, USA
Received 19 May 2005; received in revised form 8 June 2005; accepted 9 June 2005
Available online 1 August 2005
Abstract
The equilibrium constants for the ketoreductase-catalyzed reactions (cycloalkanone + 2-propanol = cycloalkanol + acetone)
have been measured in n-hexane, n-pentane, and supercritical carbon dioxide SCCO2 (pressure = (8.0 to 12.0) MPa). The
cycloalkanones included in this study were: cyclobutanone, cyclopentanone, cyclohexanone, cycloheptanone, and cyclooctanone.
The equilibrium constants for the reactions involving cyclobutanone and cyclohexanone were measured in n-hexane over the
range T = (288.35 to 308.05) K. The thermodynamic quantities at T = 298.15 K are: K = (0.763 0.001); DrGmꢀ
¼
ð0.670 ꢁ 0.002Þ kJ ꢂ molꢃ1; DrHꢀ ¼ ꢃð1.09 ꢁ 0.11Þ kJ ꢂ molꢃ1, and DrSꢀ ¼ ꢃð5.9 ꢁ 0.4Þ J ꢂ Kꢃ1 ꢂ molꢃ1 for the reaction involving
m
cyclobutanone; and K = (15.7 0.2); DrGꢀ ¼ ꢃð6.82 ꢁ 0.02Þ kJ ꢂmmolꢃ1
;
DrHꢀ ¼ ꢃð4.6 ꢁ 1.0Þ kJ ꢂ molꢃ1
,
and DrSꢀm
¼
m
m
ð7.4 ꢁ 3.3Þ J ꢂ Kꢃ1 ꢂ molꢃ1 for the reaction involving cyclohexanone, respectively. An inspection of the equilibrium constants
for these reactions in n-hexane, n-pentane, and SCCO2 shows that solvent dependence is not significant. The equilibrium con-
stants of cycloalkanones decrease with increasing value of the number of carbons, NC with the exception of cyclohexanone. The
cyclohexanol, which adopts a nearly strainless, idealized tetrahedral conformation around each carbon, is thermodynamically
favored and more stable compared to other cycloalkanol rings, and this is reflected in the significantly higher value of the equi-
librium constant obtained for this reaction. Comparisons with results obtained by using two independent thermochemical routes
are also made.
Published by Elsevier Ltd.
Keywords: Acetone; Cycloalkanol; Cycloalkanone; Equilibrium constant; Ketoreductase; 2-Propanol; Supercritical carbon dioxide; Thermody-
namics
1. Introduction
Ketoreductase-catalyzed redox reactions [9–14] have
been used for the synthesis of chiral secondary alcohols
that are useful intermediates in a number of industries.
New industrial applications of biocatalysis continue to
be realized as enzymes are customized through recombi-
nant and directed evolution techniques for a particular
set of operating conditions.
There have been several reports from our laboratory
dealing with the thermodynamics of enzyme-catalyzed
reactions in organic solvents [15–18]. In recent studies
[19,20], we have also used supercritical carbon dioxide
(SCCO2) as a solvent. It is an attractive solvent because
Biocatalysis in organic solvents is playing an increas-
ingly important role in the synthesis of rare chemicals
that are useful intermediates for the pharmaceutical
and agrochemical industries [1–4]. Lipase-catalyzed
reactions in non-aqueous solvents have been widely used
for stereoselective resolution of racemic mixtures [3,5–8].
*
Corresponding author.
E-mail addresses: yadu.tewari@nist.gov (Y.B. Tewari), karen.phin-
ney@nist.gov (K.W. Phinney), jliebman@umbc.edu (J.F. Liebman).
0021-9614/$ - see front matter. Published by Elsevier Ltd.
doi:10.1016/j.jct.2005.06.005

Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
389
of its low toxicity, non-flammability, abundant avail-
cyclooctanoneðsolnÞ þ 2-propanolðsolnÞ ¼
cyclooctanolðsolnÞ þ acetoneðsolnÞ.
Here ‘‘soln’’ denotes the organic or supercritical carbon
dioxide solvents used in this study.
The activity of the ketoreductase-catalyzed reactions
depends on the presence of a small amount of reduced
b-nicotinamide-adenine dinucleotide (NADPred). The
ketoreductase-catalyzed reaction proceeds in two steps.
In the first step, the cycloalkanone is reduced to the
corresponding cycloalkanol and NADPox, and in the
second step, NADPred is regenerated in the presence of
2-propanol:
ability, environmental friendliness, and recyclability.
The low viscosity and high diffusivity of SCCO2 also
help to provide favorable mass transfer properties which
can enhance the catalysis. The thermodynamic results
obtained in these studies are essential for understanding
of the energetics of these reactions and also for the prac-
tical utilization of enzyme-catalyzed reactions in these
solvents.
ð5Þ
In this study, we report results of equilibrium mea-
surements for the following ketoreductase-catalyzed
reactions (see figure 1):
cyclobutanoneðsolnÞ þ 2-propanolðsolnÞ ¼
2NADPredðaqÞ þ cycloalkanoneðsolnÞ ¼
cycloalkanolðsolnÞ þ 2NADPoxðaqÞ;
cyclobutanolðsolnÞ þ acetoneðsolnÞ;
ð1Þ
ð2Þ
ð3Þ
ð6Þ
ð7Þ
cyclopentanoneðsolnÞ þ 2-propanolðsolnÞ ¼
cyclopentanolðsolnÞ þ acetoneðsolnÞ;
2NADPoxðaqÞ þ 2-propanolðsolnÞ ¼
2NADPredðaqÞ þ acetoneðsolnÞ.
By combining equations (6) and (7), we have the follow-
ing overall reaction for the reduction of the cycloalka-
none to its corresponding cycloalkanol.
cyclohexanoneðsolnÞ þ 2-propanolðsolnÞ ¼
cyclohexanolðsolnÞ þ acetoneðsolnÞ;
cycloheptanoneðsolnÞ þ 2-propanolðsolnÞ ¼
cycloheptanolðsolnÞ þ acetoneðsolnÞ;
cycloalkanoneðsolnÞ þ 2-propanolðsolnÞ ¼
cycloalkanolðsolnÞ þ acetoneðsolnÞ.
ð4Þ
ð8Þ
OH
O
O
OH
+
+
cyclobutanol
acetone
cyclobutanone
2-propanol
O
OH
O
OH
+
+
cyclopentanol
cyclopentanone
acetone
2-propanol
O
OH
OH
O
+
+
cyclohexanone
2-propanol
cyclohexanol
acetone
OH
O
O
OH
OH
+
+
cycloheptanone
2-propanol
acetone
cycloheptanol
O
O
OH
+
+
acetone
2-propanol
cyclooctanone
cyclooctanol
FIGURE 1. Structures of the substances in reactions (1) to (5).

390
Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
TABLE 1
Principal substances used in this study with their Chemical Abstracts Service (CAS) registry numbers, empirical formulas, molecular masses Mr,
Supplier (A = Aldrich, B = Biocatalytics, and T = TCI America), mass fraction of moisture w determined by Karl Fischer, mole fraction purity x as
stated by the vendors, and method(s) used to determine the mole fraction purity
Substance
CAS number
Formula
Mr
Supplier
w
x
Method(s)
Aetone
67-64-1
C3H6O
C4H8O
C4H6O
58.08
72.11
70.09
116.20
112.17
100.16
98.15
128.22
126.20
86.13
A
A
A
A
A
A
A
T
A
A
A
A
A
B
0.010
0.007
0.007
0.004
0.006
0.003
0.004
0.006
0.002
0.007
0.005
0.99
0.99
0.99
0.97
0.99
0.99
0.998
0.995
0.98
0.99
0.99
0.99
0.99
g.c.
Cyclobutanol
Cyclobutanone
Cycloheptanol
Cycloheptanone
Cyclohexanol
Cyclohexanone
Cyclooctanol
Cyclooctanone
Cyclopentanol
Cyclopentanone
n-Hexane
2919-23-5
1191-95-3
502-41-0
502-42-1
108-93-0
108-94-1
696-71-9
502-49-8
98-41-3
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
FT-n.m.r., FT-i.r., g.c.
g.c.
C7H16
C7H14
C6H14
C6H12
C8H18
C8H16
C5H12
C5H10
C6H14
C6H14
O
O
O
O
O
O
O
O
120-92-3
110-54-3
111-27-3
84.12
86.18
102.18
1-Hexanol
Ketoreductase
n-Pentane
O
0.001
0.001
g.c.
109-66-0
67-63-0
C5H12
C3H8O
72.15
60.10
A
A
0.99
0.995
g.c.
2-Propanol
FT-n.m.r., FT-i.r.
A principal aim of our study is to determine how the
equilibrium constant for reaction (8) varies with the
number of carbon atoms in the cycloalkanone. In our
previous study [20], it was found that the solvent depen-
dency of ketoreductase-catalyzed reactions is not signif-
icant. In this study, the equilibrium constants of
reactions (1) to (3) have been measured by using n-hex-
ane, n-pentane, and supercritical carbon dioxide
(SCCO2). The equilibrium constants of reactions (4)
and (5) have been determined only in n-hexane and
n-pentane. The temperature dependencies of the equilib-
rium constants for reactions (1) and (3) were also mea-
sured in n-hexane. Reactions (6) and (7) require a
small amount of water to dissolve the NADPred and
NADPox. In fact, it is highly likely that the actual catal-
ysis occurs in the aqueous phase. However, since the
organic reactants and products are only slightly soluble
in water, the organic or SCCO2 solvents are necessary
for the overall reaction (8) to proceed to any substantial
degree. Since the concentrations of the reactants and
products are measured in the bulk non-aqueous phase,
the measured equilibrium constants pertain to these
phases.
water given in table 1 for the compounds involved in
these reactions and the internal standard, 1-hexanol,
were determined by using Karl Fischer titration [21].
The ketoreductase (EC 1.1.1.2) used in this study was
from Biocatalytics, Pasadena, CA.1 This enzyme was
prepared from a recombinant bacterial source and gene
expressed in Escherichia coli. NADPred was embedded in
the ketoreductase preparation by the vendor.
2.2. Chromatography and quantitative analysis
For quantitative analysis of reactants and products,
we used a Hewlett-Packard (HP) 5890 gas chromato-
graph (g.c. Agilent Technologies, Wilmington, DE,
USA), equipped with a flame ionization detector (FID)
and a fused silica Phenomenex ZB-FFAP capillary col-
umn (30 m long, 0.53 mm id, 0.53 lm thick film coating).
The injector and detector temperatures were T = (523
and 543) K, respectively; the head pressure of the helium
carrier gas was P = 283 kPa. The initial column temper-
ature of T = 313 K was held for 3 min and then raised to
T = 513 K at a rate of 0.333 K Æ sꢃ1 and held at
T = 513 K for 5 min. The substance 1-hexanol was used
as an internal standard for the quantitative determina-
tion of the concentrations of the substances in reactions
(1) to (5) in n-hexane, n-pentane, and SCCO2. Under the
chromatographic conditions described above, the peaks
of all compounds involved in these reactions and the
internal standard were well separated.
2. Experimental
2.1. Materials
Relevant information about the substances used in
this study is given in table 1. The reported purities and
the methods used for their determinations are those sta-
ted by the vendors. The purities of these substances were
further checked by using the g.c. method described be-
low and found to have purities greater than or equal
to those stated by the vendors. The mass fractions of
1
Certain commercial equipment, instruments, or materials are
identified in this paper to specify the experimental procedures
adequately. Such identification is not intended to imply recommenda-
tion or endorsement by the National Institute of Standards and
Technology, nor is it intended to imply that the materials or equipment
identified are necessarily the best available for the purpose.

Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
391
2.3. Response factor ratios
Wilmington, DE, USA) with a modified three-port
valve. Stock solutions of approximately equimolar mix-
tures of cyclohexanone and 2-propanol{concentration
c(cyclohexanone) = 6.25 mol (kg soln)ꢃ1 and c(2-propa-
nol) = 6.42 mol (kg soln)ꢃ1} for the forward reaction,
and of cyclohexanol and acetone {c(cyclohexa-
nol) = 6.22 mol (kg Æ soln)ꢃ1 and c(acetone) = 6.48 mol
(kg Æ soln)ꢃ1} for the reverse reaction were prepared
and used in all studies involving this system. Similar
stock solutions were also prepared for the cyclobuta-
none and cyclopentanone systems involving reactions
in SCCO2 solvent.
A standard solution of acetone, cyclobutanone,
cyclobutanol, 2-propanol, and 1-hexanol in n-hexane
was prepared gravimetrically. Using this solution the re-
sponse factors (ratio of concentration to peak area) with
reference to 1-hexanol were determined for acetone,
cyclobutanone, cyclobutanol, and 2-propanol. For reac-
tions (2) to (5) standard solutions of cycloalkanone, its
cycloalkanol, and 1-hexanol were prepared in n-hexane
and their response factor ratios with respect to 1-hexa-
nol were also determined. Similarly, for the reactions
in n-pentane, a solution containing acetone, 2-propanol,
and 1-hexanol was also prepared and the response factor
ratios of acetone and 2-propanol with respect to 1-hex-
anol were measured. The response factor ratios of
acetone and 2-propanol in n-pentane with respect to
1-hexanol were required due to their overlap with n-pen-
tane. However, the response factor ratios of cycloalka-
nones and cycloalkanols involved in these reactions
were not influenced by the change of solvent.
For an equilibrium measurement, ꢄ0.20 g of the
stock solution, 0.10 cm3 of a solution containing the
enzyme [0.010 g of ketoreductase dissolved in
0.10 cm3 of phosphate buffer {K2HPO4 (c = 0.10 mo-
l Æ dm3), adjusted to pH 7.3 with H3PO4}] was added
to the vessel. Following assembly, the reaction vessel
was connected to the supercritical fluid chromato-
graph (SFC) system. The system was then filled with
SCCO2 to the desired pressure, as controlled by the
back-pressure regulator of the SFC system. The
assembly and pressurization of the reaction vessel re-
quired ꢄ20 min and the reaction was then allowed
to proceed for a specified period of time. During the
course of the reaction, the vessel was periodically sha-
ken gently at ꢄ15 min intervals. After completion of
the reaction, the vessel was depressurized by changing
the position on the three-way valve to open the needle
valve and bubbling the CO2 into a 25 mL volumetric
flask that contained ꢄ7 mL of n-hexane to collect
any products that were contained in the CO2. When
the pressure was ꢄ0.1 MPa, the vessel was discon-
nected from the SFC system and the cover of the
reaction vessel was removed. This depressurization
process also required ꢄ20 min. The n-hexane in the
25 mL volumetric flask was poured into the reaction
vessel. Additional n-hexane was then added to the
reaction vessel to fill it completely. This reaction mix-
ture in the vessel was then analyzed by g.c. for the
concentrations of reactants and products using 1-hex-
anol as an internal standard as described below.
For the quantitative analysis of the reactants and
products in the equilibrated reaction mixtures in n-
hexane and n-pentane, ꢄ4 cm3 of reaction mixture
and ꢄ0.100 cm3 of 1-hexanol internal standard solu-
tion were gravimetrically added to a vial, and tightly
capped. Then, ꢄ6 Æ 10ꢃ4 cm3 of this solution was in-
jected into the g.c., and the concentrations of reac-
2.4. Equilibrium studies
The equilibrium measurements were carried out by
approaching equilibrium from both directions of the
reaction. The forward direction solutions consisted of
(cycloalkanone + 2-propanol) and the reverse direction
solutions were (cycloalkanol + acetone). The substrates
were first dissolved in the solvent, n-hexane or n-pen-
tane. The total volume of the solutions used was
ꢄ10 cm3. Then, 0.10 cm3 of a solution containing the en-
zyme [0.010 g of ketoreductase dissolved in 0.10 cm3 of
phosphate buffer {K2HPO4 (c = 0.10 mol Æ dm3), ad-
justed to pH 7.3 with H3PO4}] was added to each reac-
tion mixture. These reaction mixtures were then placed
in a constant temperature bath ( 0.01 K) and shaken
laterally at ꢄ30 shakes Æ minꢃ1. The solutions were peri-
odically analyzed by g.c. to determine the extent of reac-
tion. A reaction was considered to be at equilibrium
when the ratios of the g.c. peak areas of products/reac-
tants were essentially identical for the forward and the
reverse reaction mixtures. An additional amount of en-
zyme solution was added to the reaction mixtures if it
was found necessary to speed up the reactions to reach
equilibrium. Reactions (1) and (2) equilibrated in 4 h
in both solvents and reaction (3) reached equilibrium
in 12 h. Reactions (4) and (5) required additional en-
zyme solution and needed ꢄ5 d to reach equilibrium.
For the equilibrium study in SCCO2, a ꢄ 15 mL vol-
ume reactor vessel fabricated from stainless steel [19]
was used. The reaction vessel was sealed via a neoprene
O-ring to a cover that had h.p.l.c. connections (outer
diameter = 0.159 cm) in it. As described previously
[19], the reaction vessel was connected to a HP supercrit-
ical fluid chromatograph (SFC, Agilent Technologies,
tants
and
products
were
measured.
The
concentration c{expressed as mol (kg Æ soln)ꢃ1} of
each substance involved in a reaction was determined
from its respective chromatographic peak area, its re-
sponse factor ratio with respect to 1-hexanol, the con-
centration of 1-hexanol, and the chromatographic
peak area of 1-hexanol.

392
Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
3. Results and discussion
obtained from the forward and the reverse directions.
It is important to note that the values of the equilibrium
constants obtained from the forward and the reverse
directions of a reaction are generally in agreement with-
in their statistical uncertainties. It is judged that reason-
able estimates of the standard uncertainties [22] due to
possible systematic errors in the values of the equilib-
rium constants for these reactions are: 0.03K in the
g.c. measurements of the concentrations of the reactants
and products (this includes possible errors in the values
of the response factors) and 0.01K due to possible
sample impurities. These estimated uncertainties have
been combined in quadrature together with the statisti-
cal uncertainties, expressed as one estimated standard
deviation of the mean, to obtain combined standard
uncertainties [22]. These reported uncertainties are the
expanded uncertainties assigned to the average values
of K (see column 9 in table 2).
An examination of the equilibrium constants of the
cycloalkanones shows that, with the exception of reac-
tion (3) involving cyclohexanol, they follow a similar
pattern as observed in our earlier studies of 2-alkanones
[20]. Specifically, the values of the equilibrium constants
decrease with increasing value of carbon number NC of
the cycloalkanols. The large value of the equilibrium
constant for the reaction involving cyclohexanol {reac-
tion (3)} implies that cyclohexanol is significantly more
stable vis a vis cyclohexanone than the other alkanols
are to their corresponding cycloalkanone. A likely
explanation is the fact that cyclohexanol has a nearly
idealized tetrahedral conformation (1.911 rad = 109.5ꢀ)
around each carbon atom and thus has virtually no ste-
ric strain, whereas the keto-carbon of cyclohexanone
precludes tetrahedral symmetry. The other cyclic
alkanols have similar conformational strain as their cor-
responding cyclic ketones. Because of this, these reac-
tions are not as favorable for cyclic alkanol formation.
Inspection of the equilibrium constants in table 2
shows that the values of the equilibrium constants are
essentially comparable in the different solvent systems
used in this study. This is consistent with the results of
our previous study [20] on the ketoreductase-catalyzed
reduction of 2-alkanones in which it was also found that
the values of the equilibrium constants were essentially
independent of the solvent used to carry out the reac-
tion. The effect of pressure on the equilibrium constant
for the cyclohexanone reaction in SCCO2 was examined
by measuring the value of the equilibrium constant for
this reaction at three different pressures (8, 10 and
12 MPa). As seen in table 2, no significant change in
the value of the equilibrium constant was observed with-
in this pressure range.
3.1. Thermodynamic formalism
The equilibrium constants for reactions (1) to (5) are
K ¼ cfcyclobutanolðsolnÞg ꢂ cfacetoneðsolnÞg=
½cfcyclobutanoneðsolnÞg ꢂ cf2-propanolðsolnÞgꢅ; ð9Þ
K ¼ cfcyclopentanolðsolnÞg ꢂ cfacetoneðsolnÞg=
½cfcyclopentanoneðsolnÞg ꢂ cf2-propanolðsolnÞgꢅ;
ð10Þ
K ¼ cfcyclohexanolðsolnÞg ꢂ cfacetoneðsolnÞg=
½cfcyclohexanoneðsolnÞg ꢂ cf2-propanolðsolnÞgꢅ;
ð11Þ
K¼cfcycloheptanolðsolnÞgꢂcfacetoneðsolnÞg=
fcycloheptanoneðsolnÞgꢂcf2-propanolðsolnÞgꢅ; ð12Þ
K¼cfcyclooctanolðsolnÞgꢂcfacetoneðsolnÞg=
½cfcyclooctanoneðsolnÞgꢂcf2-propanolðsolnÞgꢅ. ð13Þ
Here, c is the concentration of the indicated substance
{expressed as mol (kg Æ soln)ꢃ1}. Since these reactions
are symmetrical with respect to reactants and products,
the equilibrium constants of these reactions are dimen-
sionless quantities and are independent of the units cho-
sen to express the concentrations. In organic solvents,
the substrates involved in these reactions are in a non-
ionized form and their concentrations are small. Hence,
we have assumed that their activity coefficients are unity
and that the measured equilibrium constants can be
identified as thermodynamic equilibrium constants de-
fined in terms of activities of the reactants and the
products.
3.2. Results of equilibrium measurements
The results of the equilibrium measurements for the
ketoreductase-catalyzed reduction of cycloalkanone to
corresponding cycloalkanol are given in table 2. This ta-
ble includes equilibrium data for reactions (1)–(3) in
SCCO2, n-hexane, and n-pentane, and also equilibrium
data for reactions (4) and (5) in n-hexane and n-pentane.
The rates of reaction for reactions (4) and (5) in SCCO2
were too slow to allow for equilibrium measurements.
The equilibrium constants K of the reactions (column
8) are calculated from the concentrations of reactants
and products (columns 4 to 7). The uncertainties given
in this column are equal to two estimated standard devi-
ations of the mean based only on the random errors in
the measurements and do not include possible system-
atic errors. The equilibrium constants ÆKæ reported in
column 9 are the averages of the equilibrium constants
The standard molar Gibbs free energy change DrGꢀm,
the standard molar enthalpy change DrHꢀm, and the
standard molar entropy change DrSꢀm for reactions (1)
and (3) in n-hexane have been calculated from the

Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
393
TABLE 2
Results of equilibrium measurements of the ketoreductase-catalyzed reduction of cyclobutanone, cyclopentanone, cyclohexanone, cycloheptanone, and cyclooctanone to
the corresponding cycloalkanols in several solvents {reactions (1) to (5)}a
Solvent
Direction
T/K
c(c-alkanone)
c(2-propanol)
c(c-alkanol)
c(acetone)
K
ÆKæ
mol Æ (kg Æ soln)ꢃ1
mol Æ (kg Æ soln)ꢃ1
mol Æ (kg Æ soln)ꢃ1
mol Æ (kg Æ soln)ꢃ1
Cyclobutanone(soln) + 2-propanol(soln) = cyclobutanol(soln) + acetone(soln) (1)
n-Hexane
Forward
Reverse
Forward
Reverse
Forward
Reverse
Forward
Reverse
Forward
Reverse
288.35
288.35
292.99
292.99
298.02
298.02
303.04
303.04
308.05
308.05
0.0246
0.0269
0.0253
0.0246
0.0154
0.0160
0.0247
0.0269
0.0141
0.0153
0.0113
0.0133
0.0154
0.0145
0.0127
0.0125
0.0169
0.0181
0.0118
0.0126
0.0123
0.0156
0.0162
0.0144
0.0113
0.0110
0.0175
0.0178
0.0112
0.0114
0.0175
0.0178
0.0184
0.0191
0.0132
0.0139
0.0180
0.0208
0.0111
0.0128
0.774 0.005
0.775 0.025
0.776 0.003
0.765 0.007
0.771 0.007
0.763 0.003
0.765 0.002
0.755 0.004
0.760 0.003
0.747 0.007
0.757 0.008
0.768 0.025
0.764 0.025
0.758 0.024
0.752 0.024
n-Pentane
Forward
Reverse
298.15
298.15
0.00505
0.00389
0.00628
0.00618
0.00265
0.00387
0.00905
0.00470
0.756 0.018
0.757 0.005
0.757 0.025
0.722 0.025
SCCO2
Forward
Reverse
303.15
303.15
0.0268
0.0204
0.00791
0.00678
0.0173
0.0133
0.00878
0.00755
0.717 0.011
0.726 0.008
(10.0 MPa)
Cyclopentanone(soln) + 2-propanol(soln) = cyclopentanol(soln) + acetone(soln) (2)
n-Pentane
n-Hexane
Forward
Reverse
298.15
298.15
0.00882
0.01247
0.0142
0.0105
0.00634
0.00558
0.00702
0.00842
0.355 0.004
0.359 0.002
0.357 0.011
0.342 0.011
0.339 0.011
Forward
Reverse
298.15
298.15
0.00908
0.00768
0.0134
0.0112
0.00649
0.00696
0.00641
0.00423
0.342 0.005
0.342 0.002
SCCO2
Forward
Reverse
303.15
303.15
0.0369
0.0139
0.0135
0.0197
0.00847
0.00316
0.335 0.003
0.343 0.010
(10.0 MPa)
0.00493
0.00744
Cyclohexanone(soln) + 2-propanol(soln) = cyclohexanol(soln) + acetone(soln) (3)
n-Hexane
Forward
Reverse
Forward
Reverse
Forward
Reverse
Forward
Reverse
Forward
Reverse
288.35
288.35
293.20
293.20
298.04
298.04
303.05
303.05
308.19
308.19
0.0142
0.00739
0.00769
0.00711
0.00735
0.00776
0.00661
0.01172
0.01145
0.00806
0.00818
0.0454
0.0484
0.0320
0.0337
0.0381
0.0332
0.0493
0.0514
0.0343
0.0356
0.0385
0.0395
0.0299
0.0310
0.0332
0.0281
0.0470
0.0472
0.0313
0.0321
16.7 0.2
17.1 0.3
16.0 0.2
16.1 0.1
15.7 0.1
15.6 0.1
15.0 0.2
15.5 0.3
14.7 0.2
14.9 0.2
16.9 0.6
16.0 0.5
15.6 0.5
15.3 0.5
14.8 0.5
0.0145
0.00842
0.00884
0.01039
0.00904
0.01316
0.01367
0.00906
0.00937
n-Pentane
Forward
Reverse
298.15
298.15
0.00194
0.00326
0.00682
0.00376
0.0135
0.0115
0.0139
0.0152
14.2 0.1
14.3 0.2
14.3 0.5
SCCO2
8.0 MPa
Forward
Reverse
303.15
303.15
0.00842
0.00626
0.00250
0.00267
0.0449
0.0583
0.00748
0.00462
16.0 0.2
16.1 0.2
16.0 0.2
16.0 0.4
16.5 0.2
10.0 MPa
12.0 MPa
Forward
Reverse
303.15
303.15
0.00715
0.01005
0.00326
0.00433
0.0444
0.0444
0.0872
0.0151
16.6 0.3
15.4 0.3
Forward
Reverse
303.15
303.15
0.00942
0.00915
0.00247
0.00469
0.0502
0.0819
0.00758
0.00865
16.4 0.2
16.5 0.2
Cycloheptanone(soln) + 2-propanol(soln) = cycloheptanol(soln) + acetone(soln) (4)
n-Hexane
n-Pentane
Forward
Reverse
298.15
298.15
0.0103
0.00756
0.00524
0.00355
0.00165
0.00323
0.00466
0.147 0.002
0.147 0.001
0.147 0.005
0.152 0.006
0.00996
Forward
Reverse
298.15
298.15
0.00672
0.00995
0.01131
0.00757
0.00323
0.00164
0.00346
0.00720
0.147 0.003
0.157 0.003
Cyclooctanone(soln) + 2-propanol(soln) = cyclooctanol(soln) + acetone(soln) (5)
8.86 Æ 10ꢃ5
n-Hexane
Forward
Reverse
298.15
298.15
0.0104
0.0108
0.00322
0.00300
0.0112
0.0111
0.0296 0.0002
0.0332 0.0005
0.0314 0.0028
0.0279 0.0009
9.70 Æ 10ꢃ5
n-Pentane
Forward
Reverse
298.15
298.15
0.01274
0.01048
0.01163
0.01122
9.41 Æ 10ꢃ4
3.25 Æ 10ꢃ4
0.00433
0.01019
0.0275 0.0002
0.0282 0.0005
a
Abbreviations are: cycloalkanone, c-alkanone; cycloalkanol, c-alkanol; and supercritical carbon dioxide, SCCO2. Unless specified otherwise, the pressure
P = 0.1 MPa. The concentrations of the substrates involved in the reaction {cycloalkanone(soln) + 2-propanol(soln) = cycloalkanol(soln) + acetone(soln)} are given in
columns 4 to 7. Each concentration is the average of five measurements. The values of the equilibrium constants K (column 8) are calculated from the concentrations
given in columns 4 to 7 by using equations (9) to (13), respectively. The quantity ÆKæ is the average of the equilibrium constants which were measured from the forward
and the reverse directions of reaction. The uncertainties in the values of K (column 8) are based on two estimated standard deviations of the mean. The uncertainties
reported in column 9 are the combined expanded uncertainties in the values of the equilibrium constants ÆKæ.

394
Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
temperature dependency of the equilibrium constant.
In this calculation, we used the Clark and Glew
equation [23] with the assumption that the standard
molar heat capacity change DrCꢀp;m for these reactions
is zero. The calculated thermodynamic quantities at
equilibrium constants reported by Muller and Blanc
¨
[24] are K = (10.9, 21.7, 40.9, 4.7, and 2.0) for reactions
(14) to (18), respectively. For the reactions involving
cyclohexanone {reactions (15) to (18)}, the calculated
values of K are approximately twice as large as their
measured values. However, there is excellent agreement
between the calculated and measured values of K for
reactions (17) and (18).
T = 298.15 K
are:
K = (0.763 0.001);
DrGꢀ ¼
ð0.670 ꢁ 0.002Þ kJ ꢂ molꢃ1
;
DrHꢀ ¼ ꢃð1.09 ꢁ 0.11ÞmkJꢂ
m
molꢃ1
,
and DrSꢀm ¼ ꢃð5.9 ꢁ 0.4Þ J ꢂ Kꢃ1 ꢂ molꢃ1 for
reaction (1); and K = (15.7 0.2); DrGꢀ ¼ ꢃð6.82ꢁ
The second thermochemical route uses values of the
standard molar formation properties for the condensed
phases of the reactants and products to calculate values
of DrHꢀm; DrGꢀm, and K for reactions (1) to (5) in the neat
solvents. To do this, we have used the standard molar
formation properties given in table 3. Many of these val-
ues are taken from Domalski and Hearing [25]. The ba-
sis of the remaining values is given in the footnotes to
table 3. It should be noted that estimates had to be made
for several property values. These formation propertiꢀes
have been used to calculate values of DrHꢀm, DrSm,
DrGꢀm, and K for reactions (1) to (5) in the neat solvents
(see table 4). Inspection of table 4 shows that there is a
fair agreement in regards to the value of K for reaction
(1), good agreement in the value of K for reaction (2),
and a fair agreement in the values of K and DrHꢀm reac-
tion (3). The differences in the remaining property values
are most likely explained by consideration of the uncer-
tainties in the formation properties in table 3 as well as
the fact that the values given in table 4 refer to the neat
0.02Þ kJ ꢂ molꢃ1; DrHꢀ ¼ ꢃð4.6 ꢁ 1.0Þ kJ ꢂmmolꢃ1, and
m
DrSmꢀ ¼ ð7.4 ꢁ 3.3Þ J ꢂ Kꢃ1 ꢂ molꢃ1 for reaction (3). The
uncertainties are equal to two estimated standard devi-
ations of the mean.
3.3. Comparison with literature data
While there are no measurements in the literature
that allow for a direct comparison with the results ob-
tained herein, there are two independent routes that al-
low for a degree of comparison. The first study is that of
Muller and Blanc [24] who reported equilibrium con-
¨
stants at T = 353.15 K for the oxidation of the following
cycloalkanols by cycloalkanones in benzene in the pres-
ence of aluminum isopropoxide acting as a catalyst:
cyclobutanolðbenzÞ þ cyclohexanoneðbenzÞ ¼
cyclohexanolðbenzÞ þ cyclobutanoneðbenzÞ;
ð14Þ
ð15Þ
ð16Þ
ð17Þ
ð18Þ
cyclopentanolðbenzÞ þ cyclohexanoneðbenzÞ ¼
cyclohexanolðbenzÞ þ cyclopentanoneðbenzÞ;
TABLE 3
Standard molar enthalpies of formation Df Hꢀm, standard molar Gibbs
free energies of formation Df Gꢀm, and standard molar entropies Sꢀm of
the compounds involved in reactions (1) to (5) at T = 298.15 K and
P = 0.1 MPa
cycloheptanolðbenzÞ þ cyclohexanoneðbenzÞ ¼
cyclohexanolðbenzÞ þ cycloheptanoneðbenzÞ;
Substance and
state
Formula Df Hꢀm
Df Gꢀm
Smꢀ
cyclooctanolðbenzÞ þ cycloheptanoneðbenzÞ ¼
cycloheptanolðbenzÞ þ cyclooctanoneðbenzÞ;
kJ Æ molꢃ1 kJ Æ molꢃ1 J Æ Kꢃ1 Æ molꢃ1
Acetone(l)
C3H6O ꢃ248.10a ꢃ155.25
200.41a
180.58a
187.0c
2-Propanol(l)
Cyclobutanol(l)
Cyclobutanone(l)
Cyclopentanol(l)
C3H8O ꢃ318.10a ꢃ180.37
cycloheptanolðbenzÞþcyclopentanoneðbenzÞ¼
cyclopentanolðbenzÞþcycloheptanoneðbenzÞ.
C4H8O ꢃ199.14b
ꢃ61.62
ꢃ39.85
ꢃ87.02
ꢃ63.88
ꢃ97.01
ꢃ64.19
ꢃ67.76
ꢃ54.29
ꢃ58.64
ꢃ45.14
C4H6O ꢃ139.49d
183.4c
C5H12
O
O
O
O
O
O
O
O
ꢃ298.43a
ꢃ237.40a
ꢃ349.81a
ꢃ272.63a
ꢃ349.98a
ꢃ297.65a
ꢃ375.76e
ꢃ323.42d
206.30a
202.7c
The values of DrGꢀm and then the equilibrium constants
for reactions (14) to (18) can be calculated by using
our results for reactions (1) to (5): DrGꢀmð14Þ ¼
DrGꢀmð3Þ ꢃ DrGꢀmð1Þ; DrGmꢀ ð1 5Þ ¼ DrGmꢀ ð3Þ ꢃ DrGꢀ ð2Þ;
DrGꢀmð16Þ ¼ DrGmꢀ ð3Þ ꢃ DrGmꢀ ð4Þ; DrGmꢀ ð17Þ ¼ DrGmꢀ mð4Þꢃ
DrGꢀmð5Þ; and DrGmꢀ ð18Þ ¼ DrGꢀmð2Þ ꢃ DrGꢀmð4Þ. These
calculations assume that the use of benzene as a solvent
will not significantly change the values of DrGꢀm and the
equilibrium constants. If we also assume that the values
of K for reactions (14) to (18) are independent of tem-
perature, we can compare the calculated values of K
for reactions (14) to (18) with the measured values re-
Cyclopentanone(l) C5H10
Cyclohexanol(l)
Cyclohexanone(l)
Cycloheptanol(l)
Cycloheptanone(l) C7H14
Cyclooctanol(l)
C6H14
C6H12
C7H16
203.87a
221.98a
241.63a
241.3a
261.0c
C8H18
C8H16
Cyclooctanone(l)
260.6c
a
Values are from Domalski and Hearing [25].
Based on the standard molar enthalpy of combustion DcHꢀm mea-
b
sured by Rocˇek and Radkowsky [26] and also discussed by Liebman
and Slayden [27].
c
Estimated values based on the differences of Sꢀm of cyclopentane(l),
cycloheptane(l), and cyclooctane(l) [25]; the average difference in the
values of Sꢀm per CH2 group is 19.3 J Æ Kꢃ1 Æ molꢃ1
.
ported by Muller and Blanc [24] at T = 353.15 K. Thus,
¨
d
From Wolf [28] and Weiberg et.al. [29].
The value of Df Hꢀm for cyclooctanol(l) was extrapolated from the
e
for reactions (14) to (18) the calculated values are,
respectively, K = (21, 44, 107, 4.7 and 2.4). The values
values of Df Hꢀm for cyclopentanol(l) and cycloheptanol(l) [25].

Y.B. Tewari et al. / J. Chem. Thermodynamics 38 (2006) 388–395
395
TABLE 4
Calculated values (columns 2, 4, 5, and 6) of the standard molar enthalpy changes DrHꢀm, standard molar entropy changes DrSmꢀ , standard molar
Gibbs free energy changes DrGꢀm, and the equilibrium constants K for five condensed phase reactions in the neat solvents at T = 298.15 K and
P = 0.1 MPa
a
b
a
a
Reaction
DrHꢀm
DrHꢀm
DrSꢀm
DrGꢀm
Ka
Kb
kJ Æ molꢃ1 kJ Æ molꢃ1
J Æ Kꢃ1 Æ molꢃ1 kJ Æ molꢃ1
Cyclobutanone(l) + 2-propanol(l) = cyclobutanol(l) + acetone(l)
Cyclopentanone(l) + 2-propanol(l) = cyclopentanol(l) + acetone(l)
Cyclohexanone(l) + 2-propanol(l) = cyclohexanol(l) + acetone(l)
Cycloheptanone(l) + 2-propanol(l) = cycloheptanol(l) + acetone(l)
Cyclooctanone(l) + 2-propanol(l) = cyclooctanol(l) + acetone(l)
a
10.35
8.97
ꢃ7.18
17.67
17.66
ꢃ1.1
ꢃ4.6
23.43
23.43
1.72
20.16
20.23
3.36
1.98
ꢃ7.69
11.66
11.63
0.26
0.45
22.3
0.0091
0.0092
0.763
0.344
15.6
0.147
0.0314
These calculated values are based on the standard molar formation property values given in table 3.
b
The experimental values of K for reactions (1) to (5) and DrHꢀm for reactions (1) and (3) were obtained by using n-hexane as the solvent.
[13] S. Kambourakis, J.D. Rozzell, Tetrahedron 60 (2004) 663–669.
[14] D. Zhu, B.E. Rios, J.D. Rozzell, L. Hua, Tetrahedron Asymm.
16 (2005) 1541–1546.
[15] Y.B. Tewari, M.M. Schantz, D.J. Vanderah, J. Chem. Eng. Data
44 (1999) 641–647.
solvents. Clearly, the property values obtained by the di-
rect measurements performed in this paper are more
accurate than the values estimated by using the forma-
tion property values.
[16] Y.B. Tewari, J. Mol. Catal. B: Enzymatic 9 (2000) 83–90.
[17] Y.B. Tewari, D.M. Bunk, J. Mol. Catal. B. Enzymatic 15 (2001)
135–145.
Acknowledgments
[18] Y.B. Tewari, D.J. Vanderah, J.D. Rozzell, J. Mol. Catal. B:
Enzymatic 21 (2003) 123–131.
[19] Y.B. Tewari, T. Ihara, K.W. Phinney, M.P. Mayhew, J. Mol.
Catal. B: Enzymatic 30 (2004) 131–136.
[20] Y.B. Tewari, M.M. Schantz, K.W. Phinney, J.D. Rozzell, J.
Chem. Thermodyn. 37 (2005) 89–96.
The authors thank Drs. D.T. Gallagher, R.N. Gold-
berg, and J.D. Rozzell for helpful discussions and
suggestions.
[21] Y.B. Tewari, N. Kishore, R. Bauerle, W.R. La Course,
R.N. Goldberg, J. Chem. Thermodyn. 33 (2001) 1791–
1805.
[22] B.N. Taylor, C.E. Kuyatt, Guidelines for Evaluating and
Expressing the Uncertainty of NIST Measurement Results, NIST
Technical Note 1297, U.S. Government Printing Office, Wash-
ington, DC, 1994.
References
[1] A.M.P. Koskinen, A.M. Klibanov, Enzymatic reactions in
organic media, Blackie, Glasgow, 1996.
[2] R.D. Schmidt, R. Verger, Angew. Chem. Ind., Ed. (Engl.) 37
(1998) 1608–1633.
[3] G. Carrea, S. Riva, Angew. Chem. Int., Ed., 39 (2000) 2226–2254.
[4] S.H. Krishna, N.G. Karanth, Catal. Rev. 44 (2002) 499–591.
[5] G. Hedstrom, M. Backlund, J.P. Slotte, Biotechnol. Bioeng. 42
(1993) 618–624.
[23] E.C.W. Clarke, D.N. Glew, Trans. Faraday Soc. 62 (1966) 539–
547.
[24] P. Muller, J. Blanc, Helv. Chim. 63 (1980) 1759–1766.
¨
[25] E.S. Domalski, E.D. Hearing, J. Phys. Chem. Ref. Data 25 (1996)
[6] A.R. Margolin, Enzyme Microb. Technol. 15 (1993) 266–280.
[7] A.M.P. Koskinen, A.M. Klibanov, Enzymatic reactions in
organic media, Blackie, Glasgow, 1996.
1;
E.S. Domalski, E.D. Hearing, J. Phys. Chem. Ref. Data 22 (1993)
4.
[8] R.N. Patel, A. Banerjee, V. Nanduri, A. Goswami, F.T.
Comezoglu, J. Am. Oil Chem. Soc. 77 (2000) 1015–1019.
[9] W. Kruse, W. Hummel, U. Kragl, Recl. Trav. Chim. Pays-Bas
115 (1996) 239–243.
[10] R.A. Devaux-Basseguy, A. Bergel, M. Comtat, Enzyme Microb.
Technol. 20 (1997).
[11] V.B. Nanduri, A. Banerjee, J.M. Howell, D.B. Brzozowski, R.F.
Eiring, R.N. Patel, J. Ind. Microb. Biotech 25 (2000) 171–175.
[12] S. Kambourakis, J.D. Rozzell, Adv. Synth. Catal. 345 (2003) 699–
705.
[26] J. Rocˇek, A.E. Radkowsky, J. Am. Chem. Soc. 95 (1973) 7123–
7132.
[27] J.F. Liebman, S.W. Slayden, in: Z. Rappoport, J.F. Liebman
(Eds.), The Chemistry of Cyclobutanes, Wiley, Chichester, 2005,
pp. 133–175.
[28] G. Wolf, Helv. Chim. Acta 55 (1972) 1446–1459.
[29] K.B. Wiberg, L.S. Crocker, K.M. Morgan, J. Am. Chem. Soc.
113 (1991) 3447–3450.
JCT 05-124