Thermophysical and structural properties of crystalline solvates of tetraphenylporphyrin and their zinc, copper and cadmium metallo-complexes

Article abstract of DOI:10.1016/s0040-6031(97)00472-3

The results of a thermochemical and structural investigation of the process of desolvation of crystalline solvates of tetraphenylporphyrin H₂TPP with o-, m-xylene; zinc(II) tetraphenylporphyrin ZnTPP with benzene, toluene, o-, m-, p-xylene; cadmium(II) tetraphenylporphyrin CdTPP and copper(II) tetraphenylporphyrin CuTPP with o-, m-xylene are reported. Thermogravimetry and differential scanning calorimetry were used to determine the stoichiometry of crystalline solvates, the energy of activation and the heat effect of the desolvation process. The lattice parameters of the unsolvated crystalline CdTPP, ZnTPP and crystalline solvates of ZnTPP with p-xylene and CuTPP with m-xylene were determined using the X-ray powder diffraction method and the algorithm of Ito. The stoichiometry 1:1 for crystalline solvates [ZnTPP+(p-xylene)₁] and [CuTPP+(m-xylene)₁] was observed. The values of free volume per solvent molecules V^free defined as the difference between volume of unit cell of crystalline solvate and the sum of the volume of a unit cell of unsolvated phase of an investigated compound and Van der Waals volume of solvent molecules were calculated. The packing coefficients K_chan of the solvent molecules located in channels of the porphyrin matrix were also calculated. For ZnTPP crystalline solvates, we found the correlations between heat of desolvation and free volume per guest molecules of unit cell, as well as between the coefficient of channel packing and the temperature for which the half-lifetime of desolvation phase is equal to 60 min. A linear kinetic compensation effect for the crystalline solvates studied was found.

Full text of DOI:10.1016/s0040-6031(97)00472-3

Thermochimica Acta 311 (1998) 163±171
Thermophysical and structural properties of crystalline solvates of
tetraphenylporphyrin and their zinc, copper and cadmium
metallo-complexes
German L. Perlovichab, Wojciech Zielenkiewicza,*, Zbigniew Kaszkura,
Eva Utziga, Oleg A. Golubchikovb
a Institute of Physical Chemistry, Polish Academy of Sciences, Kasprzaka 44/55, 01-224 Warsaw, Poland
b Institute of Non-Aqueous Solution Chemistry, Russian Academy of Sciences, 153045 Ivanovo, Russian Federation
Received 25 July 1997; received in revised form 21 November 1997; accepted 2 December 1997
Abstract
The results of a thermochemical and structural investigation of the process of desolvation of crystalline solvates of
tetraphenylporphyrin H2TPP with o-, m-xylene; zinc(II) tetraphenylporphyrin ZnTPP with benzene, toluene, o-, m-, p-xylene;
cadmium(II) tetraphenylporphyrin CdTPP and copper(II) tetraphenylporphyrin CuTPP with o-, m-xylene are reported.
Thermogravimetry and differential scanning calorimetry were used to determine the stoichiometry of crystalline solvates, the
energy of activation and the heat effect of the desolvation process. The lattice parameters of the unsolvated crystalline CdTPP,
ZnTPP and crystalline solvates of ZnTPP with p-xylene and CuTPP with m-xylene were determined using the X-ray powder
diffraction method and the algorithm of Ito. The stoichiometry 1:1 for crystalline solvates [ZnTPP‡(p-xylene)1] and
[CuTPP‡(m-xylene)1] was observed. The values of free volume per solvent molecules Vfree de®ned as the difference between
volume of unit cell of crystalline solvate and the sum of the volume of a unit cell of unsolvated phase of an investigated
compound and Van der Waals volume of solvent molecules were calculated. The packing coef®cients Kchan of the solvent
molecules located in channels of the porphyrin matrix were also calculated. For ZnTPP crystalline solvates, we found the
correlations between heat of desolvation and free volume per guest molecules of unit cell, as well as between the coef®cient of
channel packing and the temperature for which the half-lifetime of desolvation phase is equal to 60 min. A linear kinetic
compensation effect for the crystalline solvates studied was found. # 1998 Elsevier Science B.V.
Keywords: Coef®cient of packing; Crystalline solvate; Desolvation; Porphyrin sponges; Tetraphenylporphyrin; Tetraphe-
nylporphyrin metallo-complexes; Thermal behaviour; X-ray
1. Introduction
molecular structure. Investigations of structural prop-
erties of porphyrin sponges have advanced appreci-
ably in recent years [1±4], as a result of advanced
developed techniques available of growing organic
crystals with programmable lattice architecture. The
crystalline solvates contain three-dimensional chan-
nels occupied by solvent molecules (Fig. 1). Sizes of
Porphyrins are cyclic aromatic polyamines based on
a macrocycle skeleton containing 16 atoms. The inter-
est in this type of compound arises from their unique
*Corresponding author. Fax: 48 22 6325 276.
0040-6031/98/$19.00 # 1998 Elsevier Science B.V. All rights reserved
P I I S 0 0 4 0 - 6 0 3 1 ( 9 7 ) 0 0 4 7 2 - 3

164
G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
spectra of these substances were found to conform to
the literature data.
Benzene, toluene, o-, m-, p-xylene (Fluka) were
dried by boiling with anhydrous CaCl2 for 3 h. These
solvents were twice distilled and stored in a protected
(moisture-free) container.
All crystalline solvates were grown by slow eva-
poration of saturated solutions of the compounds
followed by drying at room temperature. We did
not succeed in preparing the crystalline solvates of
H2TPP from benzene, toluene and p-xylene solutions.
Nevertheless, the X-ray data for [H2TPP‡(p-xylene)1]
are reported in the literature [1]. We believe that this
phase is stable in saturated solution but in an air
atmosphere at 298 K it quickly decomposes to the
unsolvated state. Crystalline solvates CdTPP obtained
from solution were very unstable and therefore the TG
measurements were performed immediately after the
preparation.
Fig. 1. A section parallel of the b and c axes for porphyrin sponges
[1±4].
the channels may vary depending on the parameters of
the crystal growing process and the geometry of guest
molecules (the topological structure of porphyrin
matrix does not change). Observation of the desolva-
tion process suggests the possibility of removing a
de®nite number of solvent molecules from the chan-
nels and therefore to control stoichiometry and the
packing density of the channels as well as orientation
of the guest molecules in the channels. The objective
of this work is to ®nd correlations between energetic
and structural properties which characterize the evo-
lution of selected crystalline phases. Presented further
are the results of systematic thermophysical and struc-
tural investigations undertaken on the crystalline sol-
vates based on tetraphenylporphyrin and its zinc(II),
cadmium(II), and copper(II) metallo-complexes. The
work is a continuation of our previous study [9]
enlarging our knowledge by investigations of new
groups of crystalline solvates.
X-Ray measurements were made using the Rigaku±
Denki powder diffractometer, CuKꢀ radiation
(ꢁˆ1.5418), and Bragg±Brentano geometry. The
powder diffraction pattern was scanned in steps of
0.058 (2ꢂ) up to 508 and a ®xed counting time (10 s)
was employed. The calculation of the unit cell para-
meters and the identi®cation of re¯ection planes were
performed using the Ito algorithm [8].
The following parameters were calculated: the
change of the unit cell volume of the crystalline
solvate during process of desolvation ÁV; the free
volume per guest molecules Vfree; coef®cients of
crystalline solvate packing Kcryst; and coef®cients of
packing of guest molecules in channels of porphyrin
matrix Kchan
.
The free volume per solvent molecules Vfree de®ned
as a difference between the volume of the unit cell of
the crystalline solvate V(Z[MTPP‡(solvent)n]) and
the sum of the volume of unit cell of the unsolvated
phase V(Z[MTPP]) of the compound and the Van der
Waals volume of the solvent Vsolvent located in the
channels of porphyrin matrix were calculated accord-
ing to the relations (1)±(2):
2. Materials and methods
The free-base tetraphenylporphyrin H2TPP and
four-coordinate metallo-complexes tetraphenylpor-
phyrin MTPP (where M=Cu, Zn and Cd) were synthe-
sized according to known methods [5,6]. All the
compounds were thoroughly puri®ed by repeated
preparative chromatography (using a mixture of
chloroform±benzene) and repeated recrystallization
procedures. The ®nal stages of puri®cation were
effected by carefully controlled sublimation techni-
ques using a ¯ow of inert gas (N2) [7]. The absorption
‰MTPP ‡ ꢀsolvent† Š ! ‰MTPPŠ ‡ n solvent
n
(1)
ÁV ˆVꢀZ‰MTPP‡ꢀsolvent† Š† VꢀZ‰MTPPŠ†
(2)
n
Vfree ˆ ÁV ZnVsolvent

G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
165
where Z is the number of structure units of crystalline
solvate in a unit cell; n is the number of molecules of a
solvent.
The packing coef®cient of crystalline solvate was
calculated according to the relation:
Kcryst ˆ ꢀZꢀnVsolvent ‡ Vhost††=
VꢀZ‰MTPP ‡ ꢀsolvent† Š†
(3)
n
where Vhost is the Van der Waals volume of porphyrin
molecule.
The packing coef®cient of a channel was calculated
according to the relation:
Kchan ˆ ꢀZnVsolvent†=ÁV
(4)
The thermal properties of crystalline compounds
were investigated in the range 298±623 K using a Du
Pont Thermal Analysis System 2100, equipped with
DSC 910 and TG 951 cells. The DSC device was
calibrated using an indium sample and the TG cell was
calibrated using of calcium oxalate monohydrate as
described previously [9]. The relative error in thermal
measurements was 1%. The DSC runs were performed
1
at a scan rate of 10 K min in an atmosphere of
¯owing dry argon (100 ml min 1) using open standard
aluminium sample pans. On the basis of the TG curves
obtained for various heating rates (0.5, 1, 2, 5, 10 and
20 K min 1), the activation energy of the thermal
processes Ea, the frequency or pre-exponential factor
A, and the temperature, for which half-lifetime t1/2, of
desolvation phase is equal to 60 min, were calculated
according to the model described in [10,11]:
dꢀ=dt ˆ fꢀꢀ†‰A expꢀ Ea=RT†Š
where ꢀ is the fractional extent of reaction.
Assume, that A,f(ꢀ) and E are independent of T and
that A and E are independent of ꢀ.
(5)
Fig. 2. The powder X-ray diffraction diagrams obtained for
[ZnTPP] (a); [ZnTPP‡(p-xylene)1] (b); [CuTPP‡(m-xylene)1] (c)
[CdTPP] (d).
The TG Decomposition Kinetics Data Analysis
Program, Version 4.0 Du Pont Analyst 2000/2100,
was applied for these calculations.
of the desolvation process noted on the DSC curves are
endothermic.
In this work we do not attempt to describe the
desolvation kinetic of the crystalline solvates using
more complicated models. These problems will be an
object of our subsequent investigations.
The X-ray data for unsolvated phases of [ZnTPP],
[CdTPP] and for crystalline solvates [ZnTPP‡(p-
xylene)1], [CuTPP‡(m-xylene)1] were measured by
powder diffraction. Fig. 2 shows the X-ray diagrams
obtained for the compounds investigated. The results
of identi®cation of the re¯ection planes of the ®rst 20
peaks are given in Tables 4. Table 5 reviews the unit
cells parameters of the crystalline phases. The X-ray
3. Results and discussion
The results of thermogravimetric and DSC inves-
tigations are given in Tables 1±3. The enthalpy effects

166
G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
Table 1
The results of the thermogravimetric experiments for the reaction of desolvation at different heating rates, (ꢃ in K min 1). The temperatures of
start, Ts in K, and finish, Tf in K; the stoichiometry of crystalline solvate (host:guest), S
ꢃ
ZnTPP
Benzene
Toluene
m-xylene
o-xylene
p-xylene
Ts
Tf
Ts
Tf
Ts
Tf
Ts
Tf
Ts
Tf
0.5
1
Ð
Ð
305.3
307.0
309.7
311.2
313.2
314.2
1:2
332.6
345.5
351.2
360.7
372.7
385.4
317.6
320.3
324.8
328.9
331.3
336.9
1:2
350.1
355.8
359.0
368.1
377.5
395.1
318.6
320.1
325.8
327.4
331.8
333.2
1:2
363.2
378.2
396.8
408.3
414.4
431.7
Ð
Ð
310.3
314.0
314.8
324.7
324.7
1:2
357.8
368.7
373.8
383.8
393.4
345.9
351.1
367.7
370.9
382.2
1:1
408.8
415.1
429.6
441.9
458.1
2
5
10
20
S
H2TPP
m-xylene
Ts
CdTPP
m-xylene
Ts
o-xylene
o-xylene
Tf
Ts
Tf
Tf
Ts
Tf
0.5
1
Ð
Ð
297.4
297.9
298.5
299.0
299.2
299.7
1:2
325.5
358.8
365.9
375.0
387.0
397.4
296.2
296.2
296.2
296.2
297.2
297.5
1:2
311.6
315.2
320.7
333.0
343.3
368.6
Ð
Ð
317.9
325.7
329.7
337.3
336.4
1:2
366.4
370.3
374.3
380.9
330.0
296.2
296.2
296.4
296.5
297.2
327.6
336.9
342.7
366.2
373.7
1:2
2
5
10
20
S
Table 2
The results of the thermogravimetric experiments for the reaction of desolvation at different heating rates, (ꢃ in K min 1). The temperatures of
start, Ts in K, and finish, Tf in K; Ám in %; AML ± the average mass losses (%); the stoichiometry of reaction, SR; the stoichiometry of
crystalline solvate (host:guest), S
ꢃ
CuTPP‡(m-xylene)
First stage
CuTPP‡(o-xylene)
First stage
Second stage
Second stage
Ts/Tf
Ám
Ts/Tf
Ám
Ts/Tf
Ám
Ts/Tf
Ám
1
322.9/355.7
327.9/361.9
329.8/365.1
330.8/369.7
336.0/379.9
8.1
8.1
7.9
7.9
8.3
8.1
355.4/452.4
363.3/452.0
364.7/453.2
370.2/514.2
378.9/559.2
4.7
4.8
4.9
5.1
5.1
4.9
Ð
Ð
Ð
Ð
2
303.2/438.8
303.2/446.2
303.2/473.2
Ð
18.1
18.2
18.3
Ð
438.8/503.2
446.2/522.2
473.2/540.2
Ð
2.9
3.1
3.0
Ð
5
10
20
AML
SR
S
18.2
3.0
(5:5)!(5:2)
(5:2)!(5:0)
(20:20)!(20:3)
(20:3)!(20:0)
1:1
1:2
literature data for the crystalline phases studied
by us [1±4,12,13,15,18] (Table 6) and the data
obtained here for the unknown phases were used
for calculations of the ÁV, Vfree, Kcryst and Kchan values
(Table 7).
m-xylene have stoichiometry (1:2), similar to those
observed in the case of crystalline solvates H2TPP,
CdTPP with o-, m-xylene. This is in good agreement
with the data obtained by X-ray investigations [1±4].
The process of desolvation of these compounds occurs
in one stage with loss of all solvent molecules from the
lattice:
The thermogravimetric and DSC data indicate that
crystalline solvates ZnTPP with benzene, toluene, o-,

G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
167
Table 3
The activation energy, Ea in kJ mol 1; heat effect, Q in kJ mol 1; pre-exponential factor, Aˆ10ꢄ in min 1, of desolvation process; the
temperature, t1/2 in K, for which half-lifetime of desolvation phase is equal 60 min
Crystalline solvates
Ea
Q a
ꢄ
t1/2
M b
First
Second
First
Second
First
Second
First
Second
ZnTPP‡(benzene)2
ZnTPP‡(toluene)2
ZnTPP‡(m-xylene)2
ZnTPP‡(o-xylene)2
ZnTPP‡(p-xylene)1
H2TPP‡(m-xylene)2
H2TPP‡(o-xylene)2
CdTPP‡(m-xylene)2
CdTPP‡(o-xylene)2
120Æ1
80Æ2
Ð
77.2
127
Ð
Ð
Ð
Ð
Ð
Ð
Ð
Ð
Ð
2.9
6.4
7.2
17.3
12.0
17.0
14.4
10.5
15.5
8.7
Ð
326
308
324
333
326
325
289
285
284
337
331
323
Ð
834.31
862.37
890.43
890.43
784.27
827.02
827.02
937.46
937.46
754.39
782.45
782.45
Ð
Ð
Ð
117Æ1
104Æ1
86.0Æ0.5
108Æ1
59Æ1
Ð
145
Ð
Ð
Ð
Ð
120
Ð
Ð
Ð
Ð
72.1
113.4
55.5
71.0
61.6
7.6
Ð
Ð
Ð
Ð
Ð
Ð
Ð
Ð
Ð
91Æ3
15.0
10.0
10.0
27.0
6.0
65Æ3
Ð
Ð
Ð
c
CuTPP‡(benzene)1
CuTPP‡(m-xylene)1
CuTPP‡(o-xylene)2
79Æ1
102Æ1
68.0Æ0.5
100Æ1
10.2
7.8
10.1
436
364
435
182Æ2
49Æ1
33.6
5.4
a The relative error in thermal measurements was 1%.
b Molecular mass of crystalline solvate.
c Ref. [9].
‰MTPP ‡ ꢀsolvent† Š ! ‰MTPPŠ ‡ 2 solvent
2
From Table 1 it follows, that ZnTPP forms numer-
ous groups of crystalline solvates, having different
solvent molecules and the same stoichiometry:
[ZnTPP‡(solvent)2] where the solvent is benzene,
toluene, o-, m-xylene. Therefore we have attempted
to analyse the in¯uence of the nature of guest mole-
cules on the thermal behaviour of the desolvation
process for this group of the compounds.
The value of Q is a summarized characteristic of all
heat effects occurring during desolvation. Avolume of
unit cell is very sensitive to all kinds of interactions of
molecules in a crystal. We have attempted to ®nd a
correlation
between
Q
and
V free
for
Fig. 3. The dependence of a heat effect of desolvation process, Q,
versus a free volume per solvent molecules in an unit cell, Vfree, for
the crystalline solvates by based ZnTPP.
[ZnTPP‡(solvent)2] (Fig. 3). From Fig. 3, we con-
clude that an increase of V free leads to an increase of
Q. Simultaneously it is known that the value V free for
crystalline solvates of analogous architecture is
related, mainly with topology of the guest molecule
and the geometrical conjugation of guest±guest and
guest±host in a channel of porphyrin matrix [16].
to 60 min, versus the packing coef®cient of channel
Kchan is presented in Fig. 4. The thermal stability of
the studied [ZnTPP‡(solvent)2] crystalline system
increases with increase of the Kchan. For example,
the crystal ZnTPP with o-xylene is more thermally
stable (Kchanˆ0.698 and t1/2ˆ333 K), whereas ZnTPP
with toluene is one of less stable (Kchanˆ0.655 and
t1/2ˆ308 K).
Presumably
for
crystalline
solvates
[ZnTPP‡(solvent)2], the heat effect of desolvation
is connected with the act of free volume activation
in the crystalline lattice.
The dependence of the temperature t1/2 for which
the half-lifetime for desolvation of the phase is equal

168
G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
Table 4
Ê
The experimental, (d/n)e in A, and theoretical, (d/n)t in A, values of distance between planes of reflection; the intensity, I in relative units, and
Ê
the indexes, (hkl), for first 20 reflections
No.
(d/n)e
(d/n)t
I
(hkl)
(d/n)e
(d/n)t
I
(hkl)
CuTPP‡(m-Xylene)1
CdTPP
1
2
14.49Æ0.15
11.06Æ0.15
10.92Æ0.15
10.79Æ0.15
10.31Æ0.15
10.11Æ0.15
8.70Æ0.08
8.03Æ0.08
7.73Æ0.08
7.56Æ0.08
7.22Æ0.08
6.99Æ0.08
6.89Æ0.08
6.44Æ0.08
6.09Æ0.08
5.94Æ0.08
5.75Æ0.05
5.53Æ0.02
5.41Æ0.02
5.33Æ0.02
14.49
11.06
10.93
10.79
10.31
10.07
8.74
8.10
7.71
7.56
7.25
6.96
6.93
6.44
6.08
5.94
5.76
5.53
5.40
5.35
0.31
0.34
0.31
0.38
0.36
0.31
0.27
0.28
0.27
0.28
0.27
0.23
0.26
0.31
0.26
0.26
0.29
0.23
0.22
0.23
(100)
12.94Æ0.15
12.20Æ0.15
9.77Æ0.08
9.61Æ0.08
8.79Æ0.08
8.50Æ0.08
8.31Æ0.08
7.48Æ0.08
7.22Æ0.08
7.11Æ0.08
6.81Æ0.08
6.57Æ0.08
6.49Æ0.08
6.095Æ0.02
6.09Æ0.02
5.93Æ0.02
5.82Æ0.02
5.57Æ0.02
5.49Æ0.02
5.40Æ0.02
12.94
12.20
9.77
9.59
8.80
8.52
8.30
7.45
7.23
7.12
6.82
6.59
6.47
6.095
6.09
5.89
5.84
5.59
5.44
5.40
1.00
0.66
0.81
0.74
0.80
0.59
0.59
0.55
0.65
0.54
0.52
0.51
0.51
0.58
0.58
0.64
0.49
0.49
0.49
0.49
(100)
(010)
(001)
ꢀ110†
ꢀ011†
ꢀ101†
(110)
ꢀ111†
(101)
ꢀ111†
(011)
ꢀ111†
(200)
(020)
ꢀ210†
ꢀ021†ꢀ201†
ꢀ120†
(111)
ꢀ121†
(210)
(010)
3
ꢀ110†
4
(001)
5
ꢀ011†
6
ꢀ101†
7
ꢀ111†
8
ꢀ111†
9
(101)
10
11
12
13
14
15
16
17
18
19
20
(110)
(200)
ꢀ201†
ꢀ111†
(011)
ꢀ211†
ꢀ012†ꢀ120†
ꢀ211†ꢀ112†
ꢀ020†ꢀ102†
ꢀ020†ꢀ221†
ꢀ112†
Zn(TPP)‡p-Xylene)1
ZnTPP
1
2
14.59Æ0.15
12.27Æ0.15
10.90Æ0.15
9.94Æ0.08
9.79Æ0.08
9.10Æ0.08
8.96Æ0.08
8.34Æ0.08
8.05Æ0.08
7.28Æ0.08
7.12Æ0.08
6.76Æ0.08
6.64Æ0.08
6.37Æ0.08
6.22Æ0.08
6.11Æ0.02
6.00Æ0.02
5.90Æ0.02
5.45Æ0.02
5.39Æ0.02
14.59
12.27
10.90
9.91
9.79
9.14
8.94
8.37
8.12
7.30
7.13
6.73
6.58
6.34
6.23
6.12
6.00
5.88
5.45
5.37
0.40
0.57
0.82
0.50
0.64
0.50
0.49
0.51
0.48
0.47
0.53
0.43
0.42
0.42
0.43
0.41
0.41
0.41
0.53
0.42
(100)
(010)
12.73Æ0.15
10.62Æ0.15
9.50Æ0.08
7.90Æ0.08
7.35Æ0.08
7.13Æ0.08
6.44Æ0.08
5.86Æ0.05
5.50Æ0.05
5.47Æ0.05
5.37Æ0.05
5.21Æ0.05
5.12Æ0.02
4.77Æ0.02
4.68Æ0.02
4.61Æ0.02
4.45Æ0.02
4.34Æ0.02
4.26Æ0.02
4.22Æ0.02
12.73
10.68
9.50
7.91
7.35
7.10
6.50
5.82
5.48
5.46
5.34
5.26
5.11
4.75
4.69
4.66
4.43
4.34
4.25
4.24
1.00
0.69
0.81
0.75
0.94
0.65
0.64
0.59
0.57
0.57
0.57
0.57
0.59
0.57
0.57
0.57
0.81
0.60
0.71
0.90
(100)(010)
ꢀ110†
3
(001)
(001)
4
ꢀ110†
ꢀ110†ꢀ011†ꢀ101†
(011)(101)
ꢀ111†ꢀ111†
ꢀ120†ꢀ210†
(111)
5
ꢀ011†
6
ꢀ101†
7
(110)
8
(101)
9
ꢀ111†
ꢀ021†ꢀ201†
ꢀ121†ꢀ211†
ꢀ220†
10
11
12
13
14
15
16
17
18
19
20
(200)
(011)
ꢀ111†
ꢀ211†ꢀ121†
(021)(210)(201)
(002)
ꢀ210†
ꢀ201†
ꢀ021†
ꢀ121†ꢀ211†
ꢀ221†ꢀ221†
ꢀ130†ꢀ310†
ꢀ012†ꢀ102†ꢀ112†
ꢀ112†
(020)(111)
(210)
ꢀ120†
ꢀ002†ꢀ120†ꢀ211†
ꢀ112†
(030)(300)
For the TG and DSC investigations from a p-xylene
different stoichiometry: [ZnTPP‡(p-xylene)2] and
[ZnTPP2‡(p-xylene)1]. The [ZnTPP‡(p-xylene)2] is
thermodynamically unstable and can be obtained as a
kinetic product by seeding a saturated solution with
crystals of [ZnTPP‡(o-xylene)2]. The [ZnTPP2‡
solution, we were able to obtain only the crystalline
solvate [ZnTPP‡(p-xylene)1] (Table 1). Neverthe-
less, in the literature [2] there are reported results
for crystalline solvates of ZnTPP with p-xylene of

G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
169
Table 5
Lattice parameters a for investigated compounds calculated following Ito method
Ê
A
Ê 3
Vol A
Compounds
deg
Z
a
b
c
ꢀ
ꢃ
ꢅ
ZnTPP
ZnTPP‡(p-xylene)1
13.32
12.94
13.97
18.03
15.71
13.20
13.32
11.48
12.39
16.94
12.84
12.68
9.54
93
94
93
107
108
108
115
107
96
1615
2067
1673
4185
2087
1626
2
2
2
4
2
2
14.70
11.56
15.10
12.12
10.20
95
102
90
CuTPPb
110
90
b
CuTPP‡(benzene)1
CuTPP‡(m-xylene)1
112
104
98
98
CdTPP
a Relative accuracy: a, b, c ± 1%; ꢀ, ꢃ, ꢅ ± 2%.
b Ref. [9].
Table 6
Lattice parameters for investigated compounds
Vol a
Ê 3
Z
Space group
Ref.
Ê
Compounds
A
deg
A
a
b
c
ꢀ
ꢃ
ꢅ
ZnTPP
14.8
17.2
14.6
90
90
90
3716.6
798.1
2120
4
1
2
2
2
2
2
2
1
2
2
2
2
2
P212121
P1
[12]
[13]
[4]
ZnTPP
10.382
18.386
11.349
18.212
13.465
13.677
13.566
6.44
12.421
15.819
11.404
15.229
21.709
21.975
22.435
10.42
6.443
98.30
90
101.15
137.87
103.87
134.23
51.98
49.16
46.72
99.14
63.11
43.34
51.55
48.52
63.87
96.47
90
ZnTPP‡(benzene)2
ZnTPP‡(toluene)2
ZnTPP‡(toluene)2
ZnTPP‡(m-xylene)2
ZnTPP‡(o-xylene)2
ZnTPP‡(p-xylene)2
H2TPP
10.867
10.502
10.430
11.174
10.975
11.296
12.41
P21/n
P1
110.48
90
107.65
90
2237
[18]
[4]
(2073)
2326
2292
P21/n
64.80
66.94
65.06
96.06
54.87
60.85
65.76
66.97
53.71
72.99
72.97
73.40
101.12
70.97
80.94
72.55
73.03
70.88
[1]
[1]
2269
[2]
801.9
2308
1981
P1
[17]
[1]
H2TPP‡(m-xylene)2
H2TPP‡(p-xylene)1
CdTPP‡(m-xylene)2
CdTPP‡(o-xylene)2
CuTPP‡(m-xylene)2
13.632
13.667
13.563
13.807
13.528
19.389
17.692
21.970
22.109
19.098
12.020
14.470
10.995
11.085
11.984
[1]
2335
[2]
2330
(2231)
[2]
[2]
a A volume in parentheses indicates a low-temperature structure determination.
(p-xylene)1] is directly crystallized from a saturated
solution. One molecule of p-xylene located in the
channel of crystalline lattice [ZnTPP‡(p-xylene)1]
has Vfree approximately 1.5 times greater than in
crystalline solvate with stoichiometry (1:2). Probably,
the p-xylene molecule in relation to its largest linear
dimension is located in the channel lengthwise. Ana-
logous behaviour was observed in the case of
[H2TPP‡(p-xylene)1] described in [1].
separated one from another on the temperature scale.
It allows characterization of each stage (Table 3). The
process of desolvation can be presented as follows:
‰CuTPP ‡ ꢀm-xylene† Š ! ‰CuTPPŠ‡m-xylene
1
First stage: (5:5)!(5:2)
Second stage: (5:2)!(5:0)
‰CuTPP ‡ ꢀbenzene† Š ! ‰CuTPPŠ ‡ benzene
1
The crystalline solvates based on CuTPP were
prepared only for o-, m-xylene. Their stoichiometry
was (1:2) and (1:1) respectively. The crystalline
[CuTPP‡(benzene)1] has been studied by us earlier
[9]. The kinetics of desolvation is very similar for this
group of crystalline solvates: (a) the process occurs in
two independent stages (Table 2); (b) these stages are
First stage: (5:5)!(5:3)
Second stage: (5:3)!(5:0)
It can be seen, that for crystalline solvate with m-
xylene, the major part of the solvent molecules leave
the crystal lattice during the ®rst stage of desolvation,
whereas for [CuTPP‡(benzene)1], this occurs during
the second stage.

170
G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
Table 7
Unit cell volume of starting, V1, and final, V2, crystal structure and the volume loss during desolvation, ÁV; free volume per guest molecules,
free; packing coefficient of guest molecules in channels of porphyrin matrix, Kchan; packing coefficient of molecules in unit cell of crystalline
V
solvate, Kcryst
Ê 3
Reactions of desolvation
A
Kchan
Kcryst
V1
V2
ÁV
Vfree
[ZnTPP‡(benzene)2]![ZnTPP]
[ZnTPP‡(toluene)2]![ZnTPP]
[ZnTPP‡(m-xylene)2]![ZnTPP]
[ZnTPP‡(o-xylene)2]![ZnTPP]
[ZnTPP‡(p-xylene)1]![ZnTPP]
[H2TPP‡(m-xylene)2]![H2TPP]
[CdTPP‡(m-xylene)2]![CdTPP]
[CdTPP‡(o-xylene)2]![CdTPP]
[CuTPP‡(benzene)1]![CuTPP]
[CuTPP‡(m-xylene)1]![CuTPP]
2120 a
2237 c
2326 d
2292 a
2067 b
2308 d
2335 e
2330 e
2093 f
2087 b
1615 b
1615 b
1615 b
1615 b
1615 b
1604 g
1626 b
1626 b
1673 f
1673 f
505
622
711
677
452
704
709
704
420
414
163
215
238
204
216
231
236
231
249
178
0.676
0.655
0.665
0.698
0.523
0.672
0.667
0.672
0.407
0.571
0.729
0.720
0.721
0.732
0.697
0.707
0.719
0.721
0.656
0.689
a Ref. [4].
b This work.
c Ref. [18].
d Ref. [1].
e Ref. [2].
f Ref. [9].
g Ref. [17]
We observed that for the crystalline solvates with
similar values of the coef®cient Kchan, there is a certain
regularity in absolute values of heat of desolvation for
different compounds with o-, m-xylene:
‰ZnTPP ‡ ꢀm-xylene† Š
2
> ‰H2TPP ‡ ꢀm-xylene† Š
2
> ‰CdTPP ‡ ꢀm-xylene† Š
2
‰ZnTPP ‡ ꢀo-xylene† Š
2
> ‰CdTPP ‡ ꢀo-xylene† Š
2
The examination of the data in Table 3 indicates that
the total heat effect of solvation Q is different for
different host molecules. We assumed that Q depends
on nonspecific guest±host interactions.
Fig. 4. The dependence of a temperature for which half-lifetime of
desolvation phase equals to 60 min, t1/2, versus a coefficient of
packing guest molecules in channels of porphyrin matrix, Kchan, for
the crystalline solvates by based ZnTPP.
As a measure of these interactions, we have chosen
values of partial molar volumes of the host V02 in
benzene solutions. Then, the relation between Q and
V02 was derived (Fig. 5). It was found, that interactions
with the solvent molecules are stronger in the case of
ZnTPP molecules than for H2TPP, CdTPP and CuTPP.
It is very likely due to the more favorable redistribu-
tion of electron density between metal atom and the
porphyrin macrocycle.
An in¯uence of the nature of porphyrin molecule
(host) of crystalline solvate on energetics of the de-
solvation process was analysed as an example of
crystal matrixes, which have the same solvate (guest)
molecules but different host molecules: [MTPP‡
(m-xylene)2] and [MTPP‡(o-xylene)2], where
MˆH2; Zn; Cd; Cu.

G.L. Perlovich et al. / Thermochimica Acta 311 (1998) 163±171
171
process of the investigated crystalline solvates with
stoichiometry 1:2.
4. Conclusion
The presented experimental results lead to the
conclusion that the thermal stability of crystalline
solvates [ZnTPP‡(solvent)n] is connected with such
structural parameter as the packing coef®cient Kchan of
the solvent molecules located in the channels of the
porphyrin matrix. The heat effect Q of desolvation
process of investigated compounds is sensitive to: (a)
the value of free volume per solvent (guest) molecule
V
(c) the nature of the solvent (guest) molecule.
free; (b) the nature of the metal atom of MTPP (host);
Fig. 5. The dependence of a heat effect of desolvation process, Q,
versus a MTPP partial molar volume, V02, in benzene. & ±
[MTPP‡(m-xylene)2]; ~ ± [MTPP‡(o-xylene)2].
References
[1] M.P. Byrn, C.I. Curtis, S.I. Khan, P.A. Sawin, R. Tsurumi,
C.E. Strouse, J. Am. Chem. Soc. 112 (1990) 1865.
[2] M.P. Byrn, C.E. Strouse, J. Am. Chem. Soc. 113 (1991)
2501.
[3] M.P. Byrn, C.I. Curtis, I. Goldberg, Y. Hsion, S.I. Khan, P.A.
Sawin, S.K. Tendick, C.E. Strouse, J. Am. Chem. Soc. 113
(1991) 6549.
[4] M.P. Byrn, C.I. Curtis, Y. Hsiou, S.I. Khan, P.A. Sawin, S.K.
Tendick, A. Terzis, C.E. Strouse, J. Am. Chem. Soc. 115
(1993) 9480.
[5] A.D. Adler, F.R. Longo, J.D. Finarelli, J. Goldmacher, Y.
Assour, L. Korsakoff, J. Org. Chem. 32 (1967) 476.
[6] J.A. Mullins, A.D. Adler, R.M. Hochstrasser, J. Chem. Phys.
43 (1965) 2548.
[7] O.A. Golubchikov, G.L. Perlovich, B.K. Naneishvili, Zh. Fiz.
Khim. (Russian) 68 (1994) 2122.
[8] A.A. Rusakov, X-Ray of metals, Atomizdat, 1977, p. 207.
[9] G.L. Perlovich, W. Zielenkiewicz, E. Utzig, Z. Kaszkur, O.A.
Golubchikov, Thermochim. Acta 279 (1996) 121.
[10] C.D. Doyle, J. Appl. Polym. Sci. 6 (1962) 639.
[11] J.H. Flynn, L.A. Wall, Polym. Lett. 4 (1966) 323.
[12] E.B. Fleischer, C.K. Miller, L.E. Webb, J. Am. Chem. Soc. 86
(1964) 2342.
Fig. 6. The dependence of Ln(A) versus an energy of activation,
Ea, of a crystalline solvates [MTPP‡(solvent)2].
The dependences of Ln(A) versus energy of activa-
tion of crystalline solvates: [MTPP‡(solvent)2],
where MˆH2; Zn; Cd are presented in Fig. 6. As
can be seen from these ®gures, the relation can be
approximated as a straight line (linear kinetic com-
pensation effect) [14]:
[13] W.R. Scheidt, J.U. Mondal, C.W. Eigenbrot, A. Adler,
L.J. Radonovich, J.L. Hoard, Inorg. Chem. 25 (1986) 795.
[14] J. Szako, J. Therm. Anal. 9 (1976) 101.
[15] W. Zielenkiewicz, G.L. Perlovich, G.E. Nikitina, A.S.
Semeykin, J. Solution Chem. 25 (1996) 135.
[16] A. Gavezzotti, J. Am. Chem. Soc. 105 (1983) 5220.
[17] S.J. Silvers, A.J. Tulinsky, J. Am. Chem. Soc. 89 (1967)
3331.
LnꢀA† ˆ ꢀ2:2 Æ 1:3† ‡ ꢀ0:31 Æ 0:01†Ea
s ˆ 0:814 r ˆ 0:995 n ˆ 7
The observed behaviour of the kinetical parameters
(6)
[18] W.R. Scheidt, M.E. Kastner, K. Hatano, Inorg. Chem. 17
(1978) 706.
suggests an identical mechanism of the desolvation