Direct spectrophotometric measurement of demetalation kinetics of 5,10,15,20-tetraphenylporphyrinatozinc(II) at the liquid-liquid interface by a centrifugal liquid membrane method

Article abstract of DOI:10.1021/ac971353s

The equilibrium and kinetics of the protonation of 5,10, 15,20-tetraphenylporphyrin (H₂TPP) and the demetalation of 5,10,15,20-tetraphenylporphyrinatozinc(II) (ZnTPP) at a dodecane-aqueous acid interface were investigated by means of a new in situ spectrophotometric method, the centrifugal liquid membrane method, which can provide the ultrathin two-phase liquid membrane system in a rotating glass cell. The consumption of H₂-TPP in the bulk dodecane phase and the production of the diprotonated aggregate, (H₄TPP²⁺)_n, adsorbed at the liquid-liquid interface were directly measured from the spectral change. The equilibrium constants of the interfacial aggregation of H₄TPP²⁺ and the demetalation of ZnTPP were determined as log(K_e1/dm⁶ mol^-2) = 2.14 ± 0.07 and log(K_e2/dm⁹ mol^-3) = -6.05 ± 0.04 at 298 K, respectively. The observed rate constant of the demetalation of ZnTPP depended upon the first order of the acidity function, and it was suggested that the rate-determining step is the formation of the monoprotonated intermediate, [ZnTPPH]⁺, at the liquid-liquid interface. The demetalation rate constant of ZnTPP was determined as k₁ = (8.6 ± 1.3) × 10^-5 dm³ mol^-1 s^-1 at 298 K. In the aggregation of H₄TPP²⁺, the rate-determining step was controlled by molecular diffusion of H₂TPP in the bulk dodecane phase.

Full text of DOI:10.1021/ac971353s

Anal. Chem. 1998, 70, 2860-2865
Direct Spectrophotometric Measurement of
Demetalation Kinetics of
5,10,15,20-Tetraphenylporphyrinatozinc(II) at the
Liquid-Liquid Interface by a Centrifugal Liquid
Membrane Method
Hirohisa Nagatani and Hitoshi Watarai*
Department of Chemistry, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan
species and applicable to a fast interfacial reaction of submillisec-
onds, but the dispersed two-phase system formed was stable only
for ∼1000 ms. Thus, to establish the direct spectroscopic method
of the interfacial kinetics, several requirements have to be
resolved, including the stability of the two-phase system, the
discrimination of overlapped absorption spectra of the bulk phase
species, and the large specific interfacial area.
In this article, we propose a new in situ spectrophotometric
method to study liquid-liquid interfacial reaction kinetics. The
two ultrathin liquid membranes were made in a rotating optical
cell, and the spectral change was directly observed.
The kinetics of the demetalation of metalloporphyrin are well
studied in homogeneous systems.4-6 However, the kinetics and
liquid-liquid interfacial reaction mechanism of the metallopor-
phyrin have not been reported. Recently, we studied the adsorp-
tion behavior and reaction mechanism of several metalloporphy-
rins at the liquid-liquid interface.7,8 5,10,15,20-Tetraphenylpor-
phyrinatozinc(II) (ZnTPP) is specifically adsorbed at the liquid-
liquid interface, compared with other divalent metal complexes,
e.g., those of cobalt, nickel, copper, and vanadyl. In the present
study, the kinetics and mechanism of demetalation of ZnTPP at a
dodecane-aqueous acid interface are investigated to demonstrate
the advantages of the new method.
The equilibrium and kinetics of the protonation of 5 ,1 0 ,
1 5 ,2 0 -tetraphenylporphyrin (H2 TP P ) and the demetala-
tion
of 5 ,1 0 ,1 5 ,2 0 -tetraphenylporphyrinatozinc(II)
(ZnTP P ) at a dodecane-aqueous acid interface were
investigated by means of a new in situ spectrophotometric
method, the centrifugal liquid membrane method, which
can provide the ultrathin two-phase liquid membrane
system in a rotating glass cell. The consumption of H2 -
TP P in the bulk dodecane phase and the production of
the diprotonated aggregate, (H4 TP P 2 +)n , adsorbed at the
liquid-liquid interface were directly measured from the
spectral change. The equilibrium constants of the inter-
facial aggregation of H4 TP P 2 + and the demetalation of
ZnTP P were determined as log(Ke1 / dm6 mol-2 ) ) 2 .1 4
( 0 .0 7 and log(Ke2 / dm9 mol-3 ) ) -6 .0 5 ( 0 .0 4 at 2 9 8
K, respectively. The observed rate constant of the de-
metalation of ZnTP P depended upon the first order of the
acidity function, and it was suggested that the rate-
determining step is the formation of the monoprotonated
intermediate, [ZnTP P H]+, at the liquid-liquid interface.
The demetalation rate constant of ZnTP P was determined
as k1 ) (8 .6 ( 1 .3 ) × 1 0 -5 dm3 mol-1 s-1 at 2 9 8 K. In
the aggregation of H4TP P 2+, the rate-determining step was
controlled by molecular diffusion of H2 TP P in the bulk
dodecane phase.
EXPERIMENTAL SECTION
Reagents. 5,10,15,20-Tetraphenylporphyrinatozinc(II) (Zn-
TPP) was prepared from 5,10,15,20-tetraphenylporphyrin (Dojindo
Laboratories) and zinc(II) acetate dihydrate (Wako Chemicals,
>99.9%) in acetic acid by the conventional method.9 ZnTPP
prepared in acetic acid was extracted into dodecane and washed
several times with Milli-Q water. Dodecane as an organic solvent
was obtained from nacalai tesque, G.R., and purified by distillation
after being washed with a mixture of fuming sulfuric acid (nacalai
tesque, E.P., 25%) and sulfuric acid. The hydrochloric acid
Only a limited number of methods and devices have been
developed to study interfacial reactions and interfacial adsorption
at liquid-liquid interfaces. Most of those methods are indirect,
in that the interfacial phenomena must be estimated from
concentration changes in the bulk phases, i.e., the high-speed
stirring method, or from changes in interfacial energy, i.e., the
interfacial tension methods.1,2 The assignment of an interfacial
species by these methods, therefore, is rather ambiguous. Previ-
ously, we reported the direct stopped-flow spectroscopic measure-
ment of the diprotonation of 5,10,15,20-tetraphenylporphyrin
(H2TPP) at a liquid-liquid interface.3 The two-phase stopped-
flow spectrometry was very sensitive to detect the interfacial
(4) Lavallee, D. K. Coord. Chem. Rev. 1 9 8 5 , 61, 55-96.
(5) Hambright, P. In Porphyrins and Metalloporphyrins; Smith, K. M., Ed.;
Elsevier: New York, 1975; Chapter 6.
(6) Berezin, B. D. Russ. J. Inorg. Chem. 1 9 9 2 , 37, 634-648.
(7) Nagatani, H.; Watarai, H. Chem. Lett. 1 9 9 7 , 167-168.
(8) Nagatani, H.; Watarai, H. J. Chem. Soc., Faraday Trans. 1 9 9 8 , 94, 247-
252.
(1) Watarai, H. Trends Anal. Chem. 1 9 9 3 , 12, 313-318.
(2) Hanna, G. J.; Noble, R. D. Chem. Rev. 1 9 8 5 , 85, 583-598.
(3) Nagatani, H.; Watarai, H. Anal. Chem. 1 9 9 6 , 68, 1250-1253.
(9) Banks, C. V.; Bisque, R. E. Anal. Chem. 1 9 5 7 , 29, 522-526.
2860 Analytical Chemistry, Vol. 70, No. 14, July 15, 1998
S0003-2700(97)01353-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/04/1998

Figure 2. Typical transmission absorption spectra of the ultrathin
two-phase liquid membrane system. The solid and dotted lines refer
to the demetalation of ZnTPP and the protonation of H2TPP,
respectively. The absorption maximum wavelengths were 416 nm for
ZnTPP and H2TPP in dodecane (79 µm in thickness) and 466 nm for
(H4TPP2+
) at the interface, respectively. The initial concentrations
n
in dodecane of ZnTPP and H2TPP were 1.86 × 10-5 and 1.71 ×
10-5 mol dm-3, respectively. The concentrations of hydrochloric acid
were 2.0 mol dm-3 for the demetalation of ZnTPP and 0.17 mol dm-3
for the protonation of H2TPP.
tively.10 Thus, the dodecane was spread as an inner liquid
membrane of 79-µm film thickness and the aqueous phase as an
outer liquid membrane of 145-µm thickness, sandwiched between
the dodecane layer and the cell wall. The interfacial area between
the two phases, Si, was 17 cm2, and the specific interfacial area,
Figure 1. Schematic drawing of the centrifugal liquid membrane
apparatus. The ultrathin two-phase liquid membrane system was
stable at rotation speeds ranging from 6000 to 7500 rpm. In the
present system, the organic phase, 0.150 cm3, and the aqueous
phase, 0.250 cm3, were spread as an inner liquid membrane of 79-
µm film thickness and an outer liquid membrane of 145-µm thickness,
respectively. The transmission absorption spectrum was measured
from the direction perpendicular to the rotation axis by the diode array
spectrophotometer.
Si/ V , was calculated as 113 cm-1
. The summation of the
o
absorption spectra of both interfacial and bulk organic phase
species was measured from the direction perpendicular to the
rotation axis with Hewlett-Packard HP8452A diode array spectro-
photometer. Thus, the light beam passed twice through the
liquid-liquid interface and both bulk phases in the rotating optical
cell. The typical transmission absorption spectra of the ultrathin
two-phase liquid membrane systems are shown in Figure 2.
Equilibrium Measurements. The equilibrium constants of
the interfacial protonation of H2TPP and the demetalation of
ZnTPP in the dodecane-hydrochloric acid system were deter-
mined by the CLM method in a thermostated room at 298 ( 2 K.
The concentrations of H2TPP and ZnTPP were 1.71 × 10-5 and
9.7 × 10-6 mol dm-3, respectively.
solution of various concentrations was used as an aqueous phase,
and the ionic strength was kept at 2.0 by the addition of sodium
chloride. All of the acids and sodium chloride were of reagent
grade, nacalai tesque, G.R. The aqueous solutions used for the
experiments were all prepared in Milli-Q water purified with
Millipore Milli-Q SP.TOC.
Kinetic Measurements. A stable, two-phase liquid mem-
brane system in the rotating optical cell was established at ∼4 s
after the beginning of the rotation. The kinetic measurement of
protonation and aggregation of H2TPP in the dodecane-
hydrochloric acid system was carried out by using the dodecane
solution of H2TPP ranging in concentration from 3.23 × 10-5 to
5.9 × 10-6 mol dm-3, and the demetalation of ZnTPP was studied
by using 8.9 × 10-6 mol dm-3 ZnTPP in dodecane. The
absorbance changes at the absorption maximum wavelengths
were recorded at 1.0-s intervals, with an integration time of 1.0 s.
The typical kinetic profiles obtained in the CLM measurements
are shown in Figure 3. The kinetic measurements were carried
out in a thermostated room at 298 ( 2 K.
Centrifugal Liquid Membrane Method. The centrifugal
liquid membrane (CLM) method can form a stable, ultrathin two-
phase liquid membrane by the centrifugal force produced by the
rotation. The CLM apparatus is schematically shown in Figure
1. The organic phase, 0.150 cm3, and the aqueous phase, 0.250
cm3, were put into the cylindrical glass cell and stoppered by a
Teflon plug attached to the rotation shaft of the high-speed motor
(Hitachi GP 2SA). The rotation speed was regulated by a Toshiba
Slidac SD105 and monitored by a digital tachometer (Ono Sokki
HT-431). The inner diameter and inner height of the cylindrical
cell were 19 and 29 mm, respectively. When the densities of
organic and aqueous phases are different, the two-phase system
is spread out at the inner wall of the cylindrical glass cell by high-
speed rotation. The ultrathin two-phase liquid membrane system
produced was stable in the rotation speed range of ∼6000-7500
rpm and could be regarded just as coated liquid films. In the
present system, the densities of dodecane used as an organic
solvent and water at 298 K are 0.745 and 0.997 g cm-3, respec-
RESULTS AND DISCUSSION
P rotonation and Aggregation of H2 TP P . The kinetic
measurement of protonation and aggregation of tetraphenylpor-
(10) Riddick, J. A.; Bungel, W. B.; Sakano, T. K. Organic Solvents, 4th ed.; John
Wiley & Sons: New York, 1986; pp 74, 132.
Analytical Chemistry, Vol. 70, No. 14, July 15, 1998 2861

Figure 4. Logarithmic plot of [(H4TPP2+)n]iSi/[H2TPP]oVo and
[Zn2+][(H4TPP2+)n]i/[ZnTPP]i against the acidity function of hydrochlo-
ric acid, -H0. The closed and open circles refer to the diprotonation
of H2TPP and the demetalation of ZnTPP, respectively. The slopes
of the straight lines were 2.3 for the protonation and 4.1 for the
demetalation systems, respectively. The equilibrium measurements
were carried out in a thermostated room at 298 ( 2 K.
accompanied by the formation of (H4TPP2+
) at 466 nm. The
n
broad absorption peak at 698 nm is the Q-band of (H4TPP2+)n.
The molar absorptivities, ꢀ, at λmax were 4.33 × 105 dm3 mol-1
cm-1 for H2TPP in dodecane and 3.06 × 105 dm3 mol-1 cm-1 for
H4TPP2+ forming the aggregate at the interface, respectively.
Previously, we studied the interfacial diprotonation and aggrega-
tion of H2TPP by the two-phase stopped-flow method in the
dispersed dodecane-hydrochloric acid system. It was concluded
that the diprotonation process from H2TPP to H4TPP2+ is faster
than the aggregation process from H4TPP2+ to (H4TPP2+)n, and
Figure 3. First-order kinetic profiles observed from the CLM
measurements in the diprotonation of H2TPP (a) and the demetalation
of ZnTPP (b). The solid line is the fitting curve obtained by the first-
order analysis. The initial concentrations of H2TPP and ZnTPP were
1.71 × 10-5 and 8.9 × 10-6 mol dm-3, respectively. The concentration
of hydrochloric acid was 2.0 mol dm-3
.
phyrin (H2TPP) in the dodecane-hydrochloric acid system was
carried out by applying the centrifugal liquid membrane method.
When the dodecane solution of H2TPP was contacted with the
aqueous acidic solution, H2TPP, which was hardly adsorbed at
the interface, was protonated, and the diprotonated species
formed, H4TPP2+, was changed to the aggregate, (H4TPP2+)n, at
the interface,
λ
max is 440 nm for H4TPP2+ and 457 nm for (H4TPP2+)n. It is noted
that the λmax for (H4TPP2+
is not consistent between the two-
)
n
phase stopped-flow experiment, 457 nm, and the CLM experiment,
466 nm. This disagreement may be attributable to the different
conditions of the interface, i.e., the interface of droplets in the
stopped-flow system is appreciably perturbed during the measure-
ment, but the interface in the CLM system is not perturbed. Since
the larger red-shift of the Soret band in the aggregate of H4TPP2+
corresponds to the larger aggregation number,11,12 it can be
concluded that the aggregation of H4TPP2+ in the stopped-flow
system proceeds less than that in the CLM system.
H2TPPo + 2H+ f H4TPP2+
(1)
(2)
i
nH4TPP2+i f (H4TPP2+
)
ni
where the subscripts o and i refer to the bulk organic phase and
the interface, respectively. The overall reaction was represented
by
The equilibrium constant of the interfacial diprotonation of H2-
TPP, Ke1 (dm6 mol-2), was defined as
k0
H2TPPo + 2H+ h (H4TPP2+
[(H4TPP2+)n]iSi/ Vo
)
(4)
(5)
H2TPPo + 2H+ 98 (H4TPP2+
)
(3)
ni
ni
Ke1
)
where k0 (s-1) is the pseudo-first-order rate constant. This
aggregate existed only at the interface, and the counterions for
the diprotonated species were postulated as chloride ions. The
diprotonation and aggregation were indicated by the large red-
shift of the Soret band of H2TPP as shown in Figure 2, in which
the transmission absorption spectra of H2TPP in the dodecane
[H2TPP]o[H+]2
The value of log Ke1 was determined as 2.14 ( 0.07 from the
intercept of the logarithmic linear function shown in Figure 4,
and the slope, which means the reaction order of hydrogen ion
was 2.29 ( 0.08. Since the concentration of hydrogen ion in the
aqueous acid phase was in large excess over that of H2TPP, the
phase and (H4TPP2+
)
at the interface were depicted. The
n
isosbestic point was observed at ∼430 nm, and the absorption
maximum wavelength, λmax, was 416 nm for H2TPP in dodecane
and 466 nm for (H4TPP2+)n, respectively. The diprotonated
monomer, H4TPP2+, at the interface was obtained at 440 nm in
the initial stage of the aggregation; however, it disappeared rapidly,
(11) Barber, D. C.; Freitag, R. A.; Whitten, D. G. J. Phys. Chem. 1 9 9 1 , 95, 4074-
4086.
(12) Jin, R.-H.; Aoki, S.; Shima, K. J. Chem. Soc., Faraday Trans. 1 9 9 7 , 93, 3945-
3953.
2862 Analytical Chemistry, Vol. 70, No. 14, July 15, 1998

absorbance changes at λmax of H2TPP and (H4TPP2+
analyzed by a pseudo-first-order rate law written as follows:
) were
n
Table 1. Rate Constants for the Diprotonation of
H2TPP Determined at Several Vo Conditionsa
d[H2TPP]o d[(H4TPP2+)n]i Si
conditions
rate constants
Vo, 10-3
Si/ Vo,
do, 10-4
k0, 10-2
kcalc, 10-2
-
)
) k0[H2TPP]o (6)
dt
dt
Vo
cm3 b
cm-1 c
cmd
s-1 e
s-1 f
100
150
200
250
170
113
85
53
79
100
132
9.9 ( 3.4
4.4 ( 0.7
2.1 ( 0.1
1.5 ( 0.3
11.3
5.0
3.0
1.8
The typical first-order kinetic profiles are shown in Figure 3a and
the rate constant, k0 (s-1), were determined by least-squares curve-
fitting of the absorbance change at each λmax. The diprotonation
and aggregation attained equilibrium after ∼100 s, and the values
of k0 did not depend on the concentrations of H2TPP in the
dodecane phase and of hydrogen ion in the aqueous phase. The
values of the observed rate constants were (3.9 ( 0.4) × 10-2 s-1
for the decrease of H2TPP and (4.4 ( 0.4) × 10-2 s-1 for the
increase of (H4TPP2+)n, respectively, and the average value was
(4.1 ( 0.5) × 10-2 s-1. Thus, it was suggested that the interfacial
diprotonation of H2TPP in the present system proceeded by the
diffusion-controlled mechanism as concluded in the two-phase
stopped-flow experiments.3 The rate law for the diffusion-
controlled protonation of H2TPP at the interface is derived from
Fick’s first law as follows:13
68
a
The kinetic measurements were carried out at a constant aqueous
phase volume of 0.250 cm3. The concentration of HCl and the ionic
strength were 1.0 mol dm-3 and 2.0, respectively. b The organic phase
volume introduced into the rotating glass cell. c Si is a constant value,
17 cm2. d The thickness of an organic liquid membrane. e Experimental
rate constants at 298 K. f Calculated mass transport constants.
a function of V and δo ()do). When V is decreased, k0 should
be increased according to eq 9. The values of k0 and do were
o
o
determined in several conditions of V with a constant aqueous
o
phase volume, 0.250 cm3. The calculated mass transport constant,
k
calc (s-1), was obtained from eq 9 replacing δo by do. The values
of k0 and kcalc are summarized in Table 1. The values of k0 are
∼10-30% smaller than kcalc; however, it is thought that k0 agreed
with kcalc within experimental errors. Thus, it was confirmed that
the diprotonation of H2TPP in the CLM system is governed by
the diffusion-controlled mechanism, and k0 is mass transport
constant. It was also shown that the thickness of the diffusion
layer can be easily regulated by changing the organic phase
volume, and the values of k0 correspond to the limiting value of
mass transfer rate in the present two-phase liquid membrane
systems.
dCδ Do S
-
)
i (Cδ - Ci)
(7)
dt
δo Vo
where Cδ, Ci, Do, and δo are the concentration of H2TPP in
dodecane at a distance δo from the liquid-liquid interface, the
concentration of H2TPP at the interface, the diffusion coefficient
of H2TPP in dodecane3 (3.5 × 10-6 cm2 s-1), and the thickness of
the diffusion layer in dodecane with a linear concentration
gradient, respectively. The thickness of the diffusion layer is
assumed to be equal to the thickness of the organic liquid
membrane, do (cm), in the present system, because the two-phase
liquid membrane system formed by the CLM method is the
ultrathin unstirred system, and the mass transfer is expected to
be carried out only by molecular diffusion. For t ) 0 s, the initial
concentrations are all same, Cδ ) Ci. Since the intrinsic proto-
nation rate of H2TPP is much faster than the mass-transfer rate,
the value of Ci becomes to zero soon after the start of the reaction,
and the concentration of H2TPP in the organic layer decreases
from Cδ to zero with a linear gradient. In this situation, the
concentration of H2TPP observed in the CLM measurement is
the mean concentration of the organic layer,
Demetalation Kinetics of ZnTP P . In the acidic systems,
ZnTPP adsorbed at the interface was decomposed to zinc ion and
free base porphyrin by the attack of hydrogen ions.
ZnTPPo f ZnTPPi
(10)
(11)
ZnTPPi + 2H+ f Zn2+ + H2TPPi
Immediately, H2TPP was diprotonated, and the aggregate of H4-
TPP2+ was produced at the interface (cf. eqs 1 and 2). The overall
demetalation of ZnTPP can be written,
ZnTPPo + 4H+ f Zn2+ + (H4TPP2+
)
(12)
ni
Cδ - Ci Cδ
[H2TPP]o )
=
(8)
where (H4TPP2+
) refers to the unit porphyrin constructing the
ni
2
2
aggregate. The demetalation represented in eq 12 was observed
from the significant spectral change, since λmax at the Soret band
was changed from 416 nm for ZnTPP to 466 nm for (H4TPP2+)n.
The molar absorptivity of ZnTPP in dodecane was 4.98 × 105 dm3
mol-1 cm-1 at 416 nm. The equilibrium constant of the interfacial
demetalation of ZnTPP, Ke2 (dm9 mol-3), was defined as
From eqs 6-8, the rate constant, k0, is expressed as
Do Si
k0 )
(9)
δo Vo
ZnTPPi + 4H+ h Zn2+ + (H4TPP2+
)
(13)
(14)
where Do and Si are constant value. Equation 9 shows that k0 is
ni
[Zn2+][(H4TPP2+)n]i
(13) Danesi, P. R. In Principles and Practices of Solvent Extraction; Rydberg, J.,
Ke2
)
Musikas, C., Choppin, R. G., Eds.; Marcel Dekker: New York, 1992; Chapter
5.
[ZnTPP]i[H+]4
Analytical Chemistry, Vol. 70, No. 14, July 15, 1998 2863

The logarithmic values of [Zn2+][(H4TPP2+)n]i/ [ZnTPP]i were
plotted against the acidity function of hydrochloric acid, -H0,14
as shown in Figure 4. The intercept, log Ke2, and the slope of the
linear plots were determined as -6.05 ( 0.04 and 4.1 ( 0.3,
respectively. At initial concentrations of ZnTPP higher than ∼1.3
× 10-5 mol dm-3 in the acidic conditions, -H0 g 0, the final
amount of (H4TPP2+)n produced at the dodecane-water interface
was saturated, giving an absorbance value of 0.067 ( 0.004 at 466
nm. In this situation, the dodecane-water interface was com-
pletely occupied by the monolayer of the diprotonated aggregates,
and the remaining ZnTPP in the bulk organic phase cannot be
demetalated any more. Therefore, this limiting value refers to
the saturated interfacial concentration, a (mol cm-2), of H4TPP2+
forming the aggregate at the interface, and the value of a can be
evaluated from
Figure 5. Logarithmic plot of the observed demetalation rate
constants of ZnTPP, kobs (s-1), vs acidity function of hydrochloric acid,
-H0, at 298 K. The slope of the straight line is 0.95. The concentration
of ZnTPP was 8.9 × 10-6 mol dm-3, and the ionic strength was
maintained at 2.0.
Asat
1
a )
(15)
103ꢀ
2
function of hydrochloric acid, as shown in Figure 5. The slope
of the linear plots was calculated as 0.95, and a first-order reaction
for the hydrogen ion was suggested. Since kobs of the demetalation
of ZnTPP was much smaller than k0 of the protonation of H2TPP,
it was suggested that the rate-determining step of the demetalation
at the dodecane-water interface is the protonation of the pyrrole
ring, as shown in eq 18.
where Asat and ꢀ are the saturated absorbance at 466 nm and the
molar absorptivity of the monomer unit in (H4TPP2+)n at 466 nm
(3.06 × 105 dm3 mol-1 cm-1), respectively. The saturated
interfacial concentration and the occupied area per molecule were
determined as (1.10 ( 0.06) × 10-10 mol cm-2 and (1.51 ( 0.08)
nm2, respectively. Taking into account that the molecular area
occupied by H4TPP2+ is predicted to be ∼1.9 nm2 from the
molecular mechanics (MM2) calculation, the experimental value
is ∼20% smaller than the predicted value. We have already
observed that λmax of the aggregate was red-shifted from 440 nm
of H4TPP2+ to 466 nm. Therefore, it was suggested that H4TPP2+
is adsorbed as J-aggregates at the dodecane-hydrochloric acid
interface. The tilting angle of the pyrrole ring plane from the
interface and the overlapped area of the monomers were estimated
as less than 27° and 40%, respectively.
k1
ZnTPPi + H+ 98 [ZnTPPH]+
(18)
i
k2
[ZnTPPH]+i + H+ 98 Zn2+ + H2TPPo
(k2 . k1) (19)
Thus, it can be thought that the second attack of hydrogen ion
shown in eq 19 is a very fast process, and the rate constant of the
second step, k2 (dm3 mol-1 s-1), is much larger than the rate
constant of the first step, k1 (dm3 mol-1 s-1). In this case, the
rate law of the demetalation is described as follows:
The absorbance decrease at λmax of ZnTPP was analyzed as
pseudo-first-order kinetics, since the concentration of hydrogen
ion in the aqueous acid phase was in large excess. The rate law
of the demetalation of ZnTPP represented by eq 12 was described
as
d[ZnTPP]i Si
Si
i Vo
-
) k1[H+][ZnTPP]
(20)
dt
Vo
d[ZnTPP]
Si
i Vo
-
T ) kobs[ZnTPP]
(16)
dt
From eqs 16, 17, and 20, the value of k1 can be calculated by
kobs
where kobs (s-1) is the observed rate constant and [ZnTPP]T is
the total or initial concentrations of ZnTPP dissolved in the bulk
organic phase, which equals the sum of the ZnTPP concentrations
in the organic phase and at the interface.
k1 )
(21)
(1 + (1/ K′)(Vo/ Si))[H+]
where the adsorption constant of ZnTPP in the dodecane-water
system, K′, is defined as
Si
i Vo
[ZnTPP]T ) [ZnTPP]o + [ZnTPP]
(17)
[ZnTPP]i
K′ )
(22)
[ZnTPP]o
The rate constants were obtained by least-squares curve-fitting
of the absorbance change at λmax of ZnTPP. The values of kobs
depended on the concentration of hydrochloric acid, and the
logarithmic values of kobs were linearly correlated with the acidity
and the value of K′ in the dodecane-water system was determined
as 2.1 × 10-4 dm.7 Therefore, the rate law of the overall reaction
can be rewritten as
(14) Cox, R. A.; Yates, K. Can. J. Chem. 1 9 8 1 , 59, 2116-2124.
2864 Analytical Chemistry, Vol. 70, No. 14, July 15, 1998

Figure 6. Schematic drawing of the demetalation mechanism of ZnTPP in the dodecane-hydrochloric acid system. ZnTPP distributes between
the bulk dodecane phase and the interface with the adsorption constant defined in eq 22. At first, ZnTPP is monoprotonated at the interface and
changed to the intermediate, [ZnTPPH]+. Instantly, [ZnTPPH]+ is further protonated, and the free base H2TPP is produced. The rate-determining
step of the interfacial demetalation of ZnTPP is the first attacking step of a hydrogen ion with the rate constant, k1 ) (8.6 ( 1.3) × 10-5 dm3
mol-1 s-1 at 298 K. The diprotonation of H2TPP in the present conditions is much faster than the demetalation of ZnTPP, and at the end of
reaction, the diprotonated aggregate, (H4TPP2+
) is formed at the interface.
n
d[ZnTPP]T
Table 2. Parameters of Protonation and Demetalation
-
)
Determined in the Dodecane-Hydrochloric Acid
dt
System at 298 K
Si
i Vo
k1(1 + (1/ K′)(Vo/ Si))[H+][ZnTPP]
(23)
system
log Kca
rate constants
a, mol cm-2 b
interfacial protonation
of H2TPP
2.14 ( 0.07 (4.1 ( 0.5)
× 10-2 s-1
interfacial demetalation -6.05 ( 0.04 (8.6 ( 1.3) × 10-5 (1.10 ( 0.06)
The averaged value of k1 was (8.6 ( 1.3) × 10-5 dm3 mol-1 s-1
,
of ZnTPP
dm3 mol-1 s-1
× 10-10
and the parameters obtained in the present study are summarized
in Table 2.
a
The logarithmic values of equilibrium constants, Ke1 (dm6 mol-2
)
and Ke2 (dm9 mol-3), refer to the diprotonation and the demetalation
equilibria, respectively. b The saturated interfacial concentration de-
termined by direct spectrophotometric measurements.
The demetalation mechanism of ZnTPP in the dodecane-
hydrochloric acid system was schematically shown in Figure 6.
In homogeneous systems, a second-order hydrogen ion depen-
dence has been reported for the demetalation of various
metalloporphyrins.15-19 However, a first-order dependence of
hydrogen ion concentration was obtained in this work. This
discrepancy suggests the unique property of the liquid-liquid
interface as a reaction field. Since zinc atom is a relatively positive
part in the ZnTPP molecule and located out of the plane of the
pyrrole ring in its structure, it is expected that the zinc atom side
of the adsorbed complex faces the aqueous phase.7 It may be
difficult for the hydrogen ion to approach the nitrogen atom of
the pyrrole ring in the ZnTPP molecule, which is somewhat
solvated by dodecane. Once the zinc complex is monoprotonated
by a hydrogen ion, it will become more hydrophilic than the
neutral compound, ZnTPP, and the produced intermediate,
acid system was done, and it was confirmed that the rate-
determining step is the first attack of hydrogen ion on ZnTPP at
the interface. The specific interfacial area of the CLM method is
a constant value, regardless of the kind of solute species. The
value of 113 cm-1 in the present study is sufficiently large in
comparison with other direct spectrophotometric methods to study
the interfacial reaction in the liquid-liquid system, e.g., 20 cm-1
of the Teflon capillary plate method, 67 cm-1 of the optical stir
cell method,20 and 777 cm-1 of the two-phase stopped-flow
method.3 Thus, the CLM method is a sensitive and reliable
method for the study of the interfacial kinetics and the adsorption
equilibria. When the volume of the organic phase is reduced,
the liquid membrane becomes thinner, and the specific interfacial
area can readily be increased. Furthermore, the applicability of
the CLM method is not restricted by the presence of surfactant
and the properties of an organic solvent, e.g., polarity, viscosity,
and refractive index. A difference in the densities of both phases
is only necessary for the operation of the CLM method. This
method may possibly be combined with other spectroscopic
techniques, such as fluorescence and Raman spectrometries.
[ZnTPPH]+ , will be adsorbed at the interface closer to the
i
aqueous phase. Thus, the second protonation at the pyrrole ring
can easily proceed. Thus, the rate-determining step of the
demetalation of ZnTPP at the interface is ascribable to the first
attacking step of a hydrogen ion.
CONCLUSIONS
A new spectrophotometric method to study the interfacial
reaction, named the centrifugal liquid membrane (CLM) method,
was established in this study. By using this method, a kinetics
study of the demetalation of ZnTPP in a dodecane-hydrochloric
ACKNOWLEDGMENT
This work was supported by a Grant-in-Aid for General
Scientific Research (A) from the Ministry of Education, Science
and Culture, Japan (No. 07404042). The authors thank Prof. H.
Yokoyama of Yokohama City University for his valuable sugges-
tions.
(15) Sutter, T. P. G.; Hambright, P. Inorg. Chem. 1 9 9 2 , 31, 5089-5093.
(16) Nwaeme, J.; Hambright, P. Inorg. Chem. 1 9 8 4 , 23, 1990-1992.
(17) Thompson, A. N.; Krishnamurthy, M. J. Inorg. Nucl. Chem. 1 9 7 9, 41, 1251-
1255.
(18) Cheung, S. K.; Dixon, F. L.; Fleischer, E. B.; Jeter, D. Y.; Krishnamurthy,
Received for review December 15, 1997. Accepted April
27, 1998.
M. Bioinorg. Chem. 1 9 7 3 , 2, 281-294.
(19) Hambright, P.; Fleischer, E. B. Inorg. Chem. 1 9 7 0 , 9, 1757-1761.
(20) Watarai, H.; Chida, Y. Anal. Sci. 1 9 9 4 , 10, 105-107.
AC971353S
Analytical Chemistry, Vol. 70, No. 14, July 15, 1998 2865