Fourier transform near-infrared absorption spectroscopic study of catalytic isomerization of quadricyclane to norbornadiene by copper(II) and tin(II) salts

Article abstract of DOI:10.1021/jp012199n

By using Fourier transform near-infrared (NIR) absorption spectroscopy, we have investigated the catalytic conversion of quadricyclane to norbornadiene. Either CuSO₄ in chloroform or SnCl₂ in benzene is used as catalyst. To avoid the effect of sample heterogeneity, the reaction mixture is kept still without stirring. The NIR light beam is guided to propagate through the solution right above the surface of metal salt. The NIR absorption spectra are acquired at 5-min intervals for a reaction period of 6 h. The related concentrations of quadricyclane and norbornadiene in the temporal evolution are determined with the method of partial least squares. A kinetic model for the pseudo-first-order reaction is derived considering the diffusion motion. Accordingly, the second-order rate constant for the isomerization catalyzed by CuSO₄ and SnCl₂, respectively, are determined to be (1.38 ?± 0.04) ?? 10^-3 and (4.62 ?± 0.09) ?? 10^-3 min^-1 g^-1. The norbornadiene is produced via a one-site coordination between the reactant and the catalyst. The product contribution from the intermediate of a two-site coordination is negligible in our system. The obtained result for CuSO₄ is comparable with that detected by using Raman spectroscopy.

Full text of DOI:10.1021/jp012199n

132
J. Phys. Chem. B 2002, 106, 132-136
Fourier Transform Near-Infrared Absorption Spectroscopic Study of Catalytic
Isomerization of Quadricyclane to Norbornadiene by Copper(II) and Tin(II) Salts
Eric Chau-Chin Chuang and King-Chuen Lin*
Department of Chemistry, National Taiwan UniVersity, and Institute of Atomic and Molecular Sciences,
Academia Sinica, Taipei 106, Taiwan, Republic of China
ReceiVed: June 11, 2001
By using Fourier transform near-infrared (NIR) absorption spectroscopy, we have investigated the catalytic
conversion of quadricyclane to norbornadiene. Either CuSO4 in chloroform or SnCl2 in benzene is used as
catalyst. To avoid the effect of sample heterogeneity, the reaction mixture is kept still without stirring. The
NIR light beam is guided to propagate through the solution right above the surface of metal salt. The NIR
absorption spectra are acquired at 5-min intervals for a reaction period of 6 h. The related concentrations of
quadricyclane and norbornadiene in the temporal evolution are determined with the method of partial least
squares. A kinetic model for the pseudo-first-order reaction is derived considering the diffusion motion.
Accordingly, the second-order rate constant for the isomerization catalyzed by CuSO4 and SnCl2, respectively,
are determined to be (1.38 ( 0.04) × 10-3 and (4.62 ( 0.09) × 10-3 min-1 g-1. The norbornadiene is
produced via a one-site coordination between the reactant and the catalyst. The product contribution from the
intermediate of a two-site coordination is negligible in our system. The obtained result for CuSO4 is comparable
with that detected by using Raman spectroscopy.
I. Introduction
Moore and co-workers have determined the quadricyclane and
norbornadiene concentrations by gas chromatography equipped
with both flame ionization and mass spectroscopy detection.6
Their procedure requires the removal and analysis of the aliquots
of the reaction mixture at intervals during the reaction. To offer
the potential for automated, nondestructive, continuous monitor-
ing of the isomerization reaction, Vickers and co-workers
demonstrated with fiber-optic Raman spectroscopy.7 Neverthe-
less, the suspended solids increased the sample opacity. Such
effect of sample heterogeneity did not change much the detection
limit, but decreased the measurement sensitivity.7
To avoid the effect of sample heterogeneity, in this work we
acquire Fourier transform near-infrared (NIR) absorption spectra
for the isomerization reaction of quadricyclane without stirring
the suspended catalysts. The CuSO4 and SnCl2 catalysts are
remaining still in the reaction medium. The reactants diffuse in
the solid stacks to undergo catalytic isomerization on the metal
surface, while the products are probed outside the solid stacks.
The quadricyclane and norbornadiene concentrations in the
reaction evolution may be determined with the aid of partial
least-squares method. A kinetic model involving the diffusion
motion is derived. On the basis of this model, the conversion
rate constants for these two catalysts are determined and
compared with previous measurements.
The interconversion of quadricyclane (Q) and norbornadiene
(N) has attracted considerable attention because of its potential
for solar energy storage.1,2 Energy is stored when norbornadiene
is photochemically converted to quadricyclane,3-5 which is
relatively stable at room temperature with a lifetime of 140 h
at 100 °C. The energy is then released when quadricyclane is
catalytically converted back to norbornadiene.6-8
In a practical design of a solar energy system based on
valence isomerization, the catalyst desired for the conversion
of quadricyclane should fit two conditions.1 First, the reaction
is rapid without byproducts, which may reduce the conversion
efficiency and probably precipitate to cause the cleaning
problem. Second, the photochemical reaction chamber and the
catalytic conversion chamber must be kept apart. Thus, the
catalyst should remain insoluble in the reaction medium to avoid
contamination of the other photochemical reaction chamber. In
this sense, transition metal salts such as Sn, Cu, W, Mo, Co,
Ni, Rh, and Pd are found to be potential candidates.1,6-12 Among
them, the Sn(II) and Cu(II) salts are cheap catalysts and
frequently used for study. Moore and co-workers reported that
the hydrated Cu(II) and Sn(II) salts cannot catalyze the
isomerization.6 A coordination site on the metal must be
available for the catalytic conversion. In addition, the copper
chloride and bromide may lead to the side products of C7H8X2
and CuX (X ) Cl and Br), such that the depletion rate of
quadricyclane becomes larger than the production rate of
norbornadiene.6 The catalytic surface also changes to white
color. To avoid the above effects, in this work we choose
anhydrous tin(II) chloride and copper(II) sulfate to catalyze
quadricyclane.
II. Experiment
A step-scan Fourier transform spectrometer (FTS, IFS-88,
Bruker) was implemented with a tungsten lamp as the NIR
radiation source and a Si diode detector available for the
wavelength range from 6000 to 25000 cm-1. The FTS contains
a classical Michelson interferometer, which can be operated in
either rapid-scan or step-scan mode.14-16 In this work a rapid-
scan mode was used. The obtained interferogram in each scan,
as completed within milliseconds, was Fourier transformed and
displayed. Each spectrum acquired for the reaction mixture was
the assemble average over 200 measurements.
The kinetics of the conversion of quadricyclane to norbor-
nadiene by copper and tin salts has been investigated.6-8,13
* Author to whom correspondence should be addressed. Fax: 886-2-
236 21483. E-mail: kclin@mail.ch.ntu.edu.tw.
10.1021/jp012199n CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/28/2001

Isomerization of Quadricyclane to Norbornadiene
J. Phys. Chem. B, Vol. 106, No. 1, 2002 133
A cylinder cell of 2.5-cm diameter and 5-cm length used as
the reactor was settled in the compartment of the FTS. A volume
of 15 mL of 0.1-0.14 M (mol/L) quadricyclane solution with
metal salt, prepared in the range of 0.2-1.2 g, was contained
in the cell. When anhydrous CuSO4 (Merck) was used as the
catalyst, quadricyclane solution was made in chloroform. When
anhydrous SnCl2 (Janssen) was used, the solvent was changed
to benzene. In either case, the solvent was kept still for 1 h to
ensure that the solid may precipitate completely before acquiring
the NIR spectra of the solvent. Then the appropriate amount of
quadricyclane was added and kept still for 10 min before
acquiring the NIR spectra of the reaction solution. The light
beam was guided to propagate through the solution right above
the solid surface. Since the suspended salts had settled down in
the cell bottom, the effect of radiation scattering and sample
opacity was minimized. The NIR spectrum was recorded at
5-min intervals over a period of 6 h.
One finds that C ) Cs when quadricyclane is on the catalyst
surface (i.e., x f 0), while C becomes zero when quadricyclane
is far from the catalyst surface. Thus the boundary conditions
are expressed as
Cd ) 1, as X ) 0
(5)
and
dC
Cd ) 0 and d ) 0, as X ) ∞
(6)
dX
Considering the boundary conditions, the solution to eq 3 yields
Cd ) e-X
(7)
Substitution into the second derivative of C with respect to x
gives rise to
Quadricyclane (quadricyclo[2.2.1.0.0] heptane) and norbor-
nadiene (bicyclo[2.2.1] hepta-2,5-diene) (99%, Aldrich) were
used as purchased without further purification. The concentration
of each compound was prepared in the range 0.005-0.11 M
for the partial least-squares analysis.
d2Cd
dx2
d2C
dx2
D
) DCs
) Cske-xxk/D
(8)
Equation 1 (n ) 1) can be solved exactly with substitution of
eq 8. Given the initial concentration, C ) Co at t ) 0, the
solution yields
III. Kinetic Modeling
The behavior of reactant motion in the solid stacks results in
different diffusion mechanisms.17 For the molecular mechanism,
the pore size of the catalyst is larger than the mean free path,
a distance that a molecule moves between collisions. Then the
molecular motion behaves as in the solvent, neglecting distur-
bance by the catalyst. On the other hand, when the pore size is
smaller than the mean free path, the wall collisions become
important and the diffusion of the reactant is dominated by the
Knudsen mechanism.17 The reactant diffusivity may depend on
the diameter of the pore size and the mean molecular velocity.
For simplicity, we assume the diffusivity of quadricyclane
inside the stacks of catalyst to be constant, D. Then the kinetic
rate equation including the diffusion motion can be expressed
as18
P2
C ) Coe-kt
+
(1 - e-kt
)
(9)
k
where
P2 ) Cske-xxk/D
(10)
Equations 9 and 10 describe the concentration change of
quadricyclane in the reaction evolution. Given the measured
data of quadricyclane concentration, the reaction rate constant
may be determined accordingly.
IV. Results and Discussion
A. Determination of Isomer Concentrations. The chloro-
form solvent absorbs NIR radiation from 8000 to 11500 cm-1
.
dC
dt
d2C
dx2
) D
- kCn
(1)
When the background spectra is subtracted, only the major band
at 8700 cm-1 leaves some noise in the baseline, as shown in
Figure 1a. Fortunately, the NIR spectra of quadricyclane and
where C is the concentration of quadricyclane, x is the depth of
the stacks of catalyst, n is the reaction order, and k is the n-order
rate constant for isomerization. In this work, the isomer
conversion belongs to the first-order reaction, n ) 1. We assume
that the catalyst surface may be covered steadily by the reactant
during the reaction. Therefore, the steady-state condition may
be applied, and eq 1 becomes
norbornadiene in chloroform peak at 8840 and 8900 cm-1
,
respectively, such that the solvent background may lead to the
least disturbance. Figure 1b shows the NIR spectra for conver-
sion of quadricyclane to norbornadiene catalyzed by CuSO4 at
reaction times of 10, 120, and 360 min, respectively. The slight
difference in the absorption bands of these two isomers allows
us to use partial least squares regression to model and determine
the individual concentration.19-21
d2C
dx2
D
) kC
(2)
When SnCl2 is used as the catalyst, the chloroform is changed
to benzene. By analogy with the background spectrum subtrac-
tion described previously, we measure the NIR spectra of
benzene and subtract from the reaction mixture. However, even
after the background subtraction, the quadricyclane and nor-
bornadiene are still buried in the large noise in the 8540-8850
cm-1 region. The NIR spectra for quadricyclane and norbor-
The concentration in eq 2 becomes dimensionless by substitution
of Cd ) C/Cs, in which Cs indicates the steady-state concentra-
tion of quadricyclane covering the catalyst surface. The expres-
sion yields
nadiene can only be acquired in the range of 8850-9050 cm-1
,
d2C
d ) Cd
(3)
(4)
in which their absorption intensities are stronger than the
background spectra, as shown in Figure 2. The NIR spectra
exhibit different features between quadricyclane and norborna-
diene. As shown in Figure 2b, NIR spectra of the reaction
mixture are recorded at the reaction times of 10, 60, 120, 240,
and 360 min, respectively. As the heterogeneous reaction
dX2
where
x
X ) x k/D

134 J. Phys. Chem. B, Vol. 106, No. 1, 2002
Chuang and Lin
Figure 1. (a) Upper trace: the NIR absorption spectrum of chloroform;
lower trace: the NIR spectrum of quandricyclane in chloroform with
the solvent background subtracted. A significant noise appears at 8700
cm-1 after the background subtraction. (b) The NIR spectra for
isomerization of quadricyclane to norbornadiene catalyzed by CuSO4
over a reaction period of (a) 10 min, (b) 120 min, and (c) 360 min,
respectively.
Figure 2. (a) Lower trace: the NIR spectrum of benzene; upper
trace: the NIR spectrum of quadricyclane in benzene with the solvent
background subtracted. A large portion of quadricyclane spectrum is
buried in the noise even after background subtraction. (b) The NIR
spectra for isomerization of quadricyclane to norbornadiene catalyzed
by SnCl2 in benzene over a reaction period of (a) 10 min, (b) 60 min,
(c) 120 min, (d) 240 min, and (e) 360 min, respectively.
evolves, the maximum absorption shifts to the position at 8900
cm-1, which indicates that the reaction mixture is predominated
by the product.
catalyst weight shows a straight line in Figure 4b, with a slope
yielding the second-order rate constant, kII ) 0.00138 ( 0.00004
min-1 g-1
.
In this work, more than 70 set of the NIR absorption spectra
for each catalytic reaction system was acquired. With the
analysis of partial least squares, either quadricyclane or nor-
bornadiene concentration in the reaction mixture could be
determined by a fit of the calibration set to the observed
spectra.19-21 For the CuSO4 catalyst, 84 calibration set of various
concentration combinations in the range 0.005-0.105 M of
quadricyclane and norbornadiene were made, while for the SnCl2
catalyst, 60 sets of concentration combinations in the same range
were made. As shown in Figure 3a and 3b, the fit between the
predicted and actual concentrations for samples are characterized
by a straight line with a regression coefficient > 0.98, thereby
ensuring reliability of the determined concentrations.
B. Determination of Rate Constants. Given the determined
quadricyclane concentrations in the isomerization evolution, the
fit of eq 9 gives rise to the pseudo-first-order rate constant, k,
and P2; the latter parameter is associated with the diffusion
contribution. The weight of CuSO4 is varied from 0.2 to 1.2 g.
For each weight, we may obtain fitted parameters of k and P2,
as listed in Table 1. Figure 4a gives an example for the decay
of the quadricyclane concentration catalyzed by CuSO4 and the
corresponding fit. The rate constants on the Table 1 appear to
increase with the catalyst weight. The plot of k as a function of
Analogously, when SnCl2 is used as the catalyst in benzene,
the corresponding rate constant and parameter P2 are fitted from
the decay of quadricyclane. The SnCl2 weight is varied from
0.2 to 1.2 g. The obtained parameters for each weight of catalyst
are listed in Table 2. The fit of the decay curve of quadricyclane
concentration is also demonstrated in Figure 5a. The dependence
of the pseudo-first-order rate constant on the SnCl2 weight is
plotted in Figure 5b, yielding a straight line from which the
second-order rate constant is determined to be kII ) 0.00462 (
0.00009 min-1 g-1
.
As the depth of catalyst stacks, x, in eq 10 is let to be zero,
the steady-state concentration, Cs, on the catalyst surface may
be determined from the parameter, P2. The results are also listed
in Tables 1 and 2 for both catalysts. The Cs values seem to
increase with the weight of catalyst and the initial concentration
of quadricyclane, if we ignore the data with the smaller weights,
which lead to a large relative error in the P2 determination.
However, the detailed relation between them is uncertain.
C. Comparison with Other Experiments. In the early work,
Moore and co-workers have proposed that in the heterogeneous
catalysis, the absorption of quadricyclane on the surface by
combination of a one-site and a two-site coordination may lead
to the production of norbornadiene.6 Therefore, the depletion

Isomerization of Quadricyclane to Norbornadiene
J. Phys. Chem. B, Vol. 106, No. 1, 2002 135
Figure 4. (a) Reaction time dependence of quadricyclane concentration
catalyzed by 1.2 g of CuSO4 in chloroform. (b) Pseudo-first-order rate
constant as a function of the weight of CuSO4. The second-order rate
constant determined from the slope is (1.38 ( 0.04) × 10-3 min-1
g-1
.
Figure 3. (a) Calibration plot for the determination of quadricyclane
concentration from the NIR spectra of reaction mixture catalyzed by
CuSO4 in chloroform. (b) Calibration plot for the determination of
quadricyclane concentration from the NIR spectra of reaction mixture
catalyzed by SnCl2 in benzene.
TABLE 2: The Kinetic Data of Quadricyclane Catalyzed by
SnCl2
SnCl2(g) k (×103 min-1
)
P2 (×104 M min-1
)
Co (M)a Cs (M)b
1.2
1.0
0.8
0.6
0.4
0.2
5.54 ( 0.47
4.5 ( 0.4
3.93 ( 0.47
2.8 ( 0.3
1.6 ( 0.7
1.1 ( 0.5
2.3 ( 0.3
2 ( 0.3
0.09
0.1
0.087
0.085
0.097
0.081
0.0415
0.0444
0.0382
0.0286
0.0313
0.0182
TABLE 1: The Kinetic Data of Quadricyclane Catalyzed by
CuSO4
1.5 ( 0.3
0.8 ( 0.2
0.5 ( 0.6
0.2 ( 0.4
CuSO4(g) k (×103 min-1
)
P2 (×105 M min-1
)
Co (M)a Cs (M)b
1.2
1.0
0.8
0.6
0.4
0.2
1.57 ( 0.13
1.41 ( 0.15
1.10 ( 0.15
0.89 ( 0.15
0.61 ( 0.16
0.41 ( 0.16
7 ( 1
0.1348 0.0446
6.0 ( 1.6
3.0 ( 1.5
1.0 ( 1.6
1.0 ( 1.7
1.0 ( 1.7
0.132
0.121
0.126
0.121
0.122
0.0426
0.0273
0.0112
0.0164
0.0244
a
b
Initial concentration of quadricyclane. Steady-state concentration
of quadricyclane on the catalyst surface.
Raman spectroscopy.7 They demonstrated that the temporal
evolution of quadricyclane prepared initially at 0.0997 M
decreased almost exponentially to 0.03 M within 240 min, as
catalyzed by 0.375 g CuSO4 in chloroform. Given the above
a
b
Initial concentration of quadricyclane. Steady-state concentration
of quadricyclane on the catalyst surface.
rate of quadricyclane can be expressed as
dQ
experimental conditions and the rate constants, k1II and k2 , by
II
Moore and co-workers, eq 11 yields the quadricyclane concen-
tration to be 0.00058 M. The product concentration is about 2
orders of magnitude smaller than that obtained by Vickers and
co-workers.7 The rate constants reported by Moore and co-
workers are apparently much larger.
) [k1 W + k2 W2]Q
(11)
II
II
dt
where Q denotes the quadricyclane concentration, and W the
weight of catalyst. They have determined the one-site and two-
Given the rate constant, kII ) 0.00138 min-1 g-1, for the
CuSO4 catalyst in this work, eq 9 with substitution of the
experimental conditions provided by Vickers and co-workers
gives rise to a quadricyclane concentration to be 0.090 M. This
value has the same order of magnitude as that by the Raman
spectroscopy. The rate constant determined here seems to be
site coordinated rate constants, k1 and k2 , to be 2.9 × 10-2
min-1 g-1 and 7.3 × 10-2 min-1 g-2, respectively, for the
CuSO4 catalyst in benzene by the detection of GC-MS.6
Vickers and co-workers have measured the kinetic behavior
of quadricyclane and norbornadiene simultaneously by using
II
II

136 J. Phys. Chem. B, Vol. 106, No. 1, 2002
Chuang and Lin
of 2 × 10-7 (mol Sn/L)-1 s-1 with an effective polymer-
anchored SnCl2 catalyst.13 By using NMR spectroscopy, the
same group later obtained a rate constant of 1.82 × 10-4 M-1
s-1 by dissolving 50 mg of SnCl2 in 1 mL of CD3OD.8
Conclusion
By using Fourier transform NIR absorption spectroscopy, we
have determined the kinetic data for the catalytic reaction of
quadricyclane. For a heterogeneous reaction, the suspended solid
may increase the sample opacity and cause the problem of
radiation scattering. In this work, the reaction mixture is kept
still to prevent from the effect of sample heterogeneity. The
NIR radiation is guided to pass through the solution above the
solid surface. Accordingly, a kinetic model taking into account
the contribution of diffusion motion has been derived to give
rise to the related kinetic parameters. The obtained depletion
rate of quadricyclane is linearly proportional to the weight of
catalyst. That is, the norbornadiene is produced via a one-site
coordination (1:1 complex) between the reactant and the catalyst.
The product contribution from the intermediate of a two-site
coordination becomes negligible in our system. Despite a distinct
kinetic mechanism resulting from a still reaction medium, the
obtained rate constants are reliable. The depletion rates of
quadricyclane are found to be on the same order of magnitude
as that by using Raman spectroscopy in a continuously stirred
solution. Therefore, in this work we have successfully provided
an alternative technique for the kinetic measurement of hetero-
geneous catalysis.
Acknowledgment. This work is supported by the National
Science Council and Chinese Petroleum Company of the
Republic of China under Contract No. NSC89-2119-M-002-
007. We thank C. B. Ke for preparing the plots.
Figure 5. (a) Reaction time dependence of quadricyclane concentration
catalyzed by 1.2 g of SnCl2 in benzene. (b) Pseudo-first-order rate
constant as a function of the weight of SnCl2. The second-order rate
constant determined from the slope is (4.62 ( 0.09) × 10-3 min-1
References and Notes
(1) Hautala, R. R.; King, R. B.; Kutal, C. Solar Energy: Chemical
ConVersion and Storage; Humana Press: Clifton, NJ, 1979; p 333-344.
(2) Bren, V. A.; Dubonosov, A. D.; Minkin, V. I.; Chernoivanov, V.
A. Russ. Chem. ReV. 1991, 60, 451.
(3) Cuppoletti, A.; Dinnocenzo, J. P.; Goodman, J. L.; Gould, I. R. J.
Phys. Chem. A 1999, 103, 11253.
g-1
.
more reasonable than that by Moore and co-workers.6 It should
be noted that the latter group used benzene as the solvent. It is
uncertain whether the solvent effect may cause such great
difference in the isomerization rate.
(4) Sluggett, G. W.; Turro, N. J.; Roth, H. D. J. Phys. Chem. A 1997,
101, 8834.
(5) Franceschi, F.; Guardigli, M.; Solari, E.; Floriani, C.; Chiesi, V.
A.; Rizzoli, C. Inorg. Chem. 1997, 36, 4099.
(6) Fife, D. J.; Morse, K. W.; Moore, W. M. J. Am. Chem. Soc. 1983,
105, 7404.
(7) Ford, J. F.; Mann, C. K.; Vickers, T. J. Appl. Spectrosc. 1994, 48,
592.
(8) Patrick, T. B.; Bechtold, D. S. J. Org. Chem. 1984, 49, 1935.
(9) Noyori, R.; Umeda, I.; Kawauchi, H.; Takaya, H. J. Am. Chem.
Soc. 1975, 97, 812.
A distinct kinetic behavior for the heterogeneous catalysis
under either a still or a stirred condition is reflected in three
aspects. First, the depletion rates of quadricyclane in this work
are expected to be slower than the system in which the solution
is stirred to enhance the collision frequency of the reactants.
That consideration interprets why the estimated quadricyclane
concentration based on our rate constant is about three times
larger than that obtained by Vickers and co-workers. Second,
the measured decay of quadricyclane concentration in this work
cannot be well fitted by using eq 11 derived by Moore and co-
workers. Since our reaction system is kept still without stirring,
the formation of a two-site coordination (1:2 complex) between
reactant and catalyst should be negligible. That is, k2II is ignored,
and k1II is equivalent to our obtained kII. Third, the isomerization
rate should take into account the diffusion contribution as in
this work, which is ignored in the kinetics for a continuously
stirred heterogeneous solution.
(10) Manassen, J. J. Catal. 1970, 18, 38.
(11) Taylor, R. B.; Jennings, P. W. Inorg. Chem. 1981, 20, 3997.
(12) Sen, A.; Thomas, R. R. Organometallics 1982, 1, 1251.
(13) Landis, M. E.; Gremaud, D.; Patrick, T. B. Tetrahedron Lett. 1982,
23, 375.
(14) Wang, C. C.; Chin, T. L.; Lin, K. C. J. Chem. Phys. 1997, 107,
10348.
(15) Chin, T. L.; Lin, K. C. Appl. Spectrosc. 1999, 53, 22.
(16) Wang, C. C.; Chen, Y. P.; Chin, T. L.; Lin, K. C. J. Chem. Phys.
2000, 112, 10204.
(17) Karger, J.; Ruthven, D. M. Diffusion in Zeolities and Other
Microporous Solids; Wiley Press: New York, 1992.
(18) Crank, J. The Mathmatics of Diffusion; Oxford University Press:
Ely House, London, 1975.
The isomerization rate catalyzed by SnCl2 in benzene is found
to be three times faster than that by CuSO4 in chloroform.
However, to our knowledge, we cannot find appropriate
heterogeneous reactions reported for comparison. Some rate
constants have been measured under different conditions.8,13 For
instance, Patrick and co-workers have obtained a rate constant
(19) Lindberg, W.; Persson, J.-A. Anal. Chem. 1983, 55, 643.
(20) Phelan, M. K.; Barlow, C. H.; Kelly, J. J.; Jinguji, T. M.; Callis, J.
B. Anal. Chem. 1989, 61, 1419.
(21) Schrieve, G. D.; Ullman, A. H. Appl. Spectrosc. 1991, 45, 713.