Effect of degree of β-chlorination on photocatalytic activity of meso-tetraphenylporphyrin under homogeneous and nanoscale heterogeneous conditions: Chlorination vs. bromination

Article abstract of DOI:10.1016/j.jcat.2020.04.018

β-Chlorination of meso-tetraphenylporphyrin, H₂TPP leads to the formation of H₂TPPCl_x (X = 2, 4, 6 and 8) with absorption bands in the range of 421–451 and 650–715 nm, for the Soret and Q(0,0) bands, respectively. H_2<

Full text of DOI:10.1016/j.jcat.2020.04.018

Journal Pre-proofs
Effect of degree of β-chlorination on photocatalytic activity of meso-tetraphe-
nylporphyrin under homogeneous and nanoscale heterogeneous conditions:
Chlorination vs. bromination
Saeed Zakavi, Hadi Yavari, Akram Heydari-turkmani, Leila Alghooneh
PII:
S0021-9517(20)30153-6
https://doi.org/10.1016/j.jcat.2020.04.018
YJCAT 13720
DOI:
Reference:
To appear in:
Journal of Catalysis
Received Date:
Revised Date:
Accepted Date:
2 February 2020
6 April 2020
19 April 2020
Please cite this article as: S. Zakavi, H. Yavari, A. Heydari-turkmani, L. Alghooneh, Effect of degree of β-
chlorination on photocatalytic activity of meso-tetraphenylporphyrin under homogeneous and nanoscale
heterogeneous conditions: Chlorination vs. bromination, Journal of Catalysis (2020), doi: https://doi.org/10.1016/
j.jcat.2020.04.018
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Elsevier Inc. All rights reserved.

Effect of degree of β-chlorination on photocatalytic activity of meso-tetraphenylporphyrin
under homogeneous and nanoscale heterogeneous conditions: Chlorination vs. bromination
Saeed Zakavi*, Hadi Yavari, Akram Heydari-turkmani, Leila Alghooneh
Department of Chemistry, Institute for Advanced Studies in Basic Sciences (IASBS), Zanjan
45137-66731, Iran.
*Corresponding author. Tel.: +98 24 33153202; Fax: +98 24 33153232. E-mail addresses:
zakavi@iasbs.ac.ir; szakavi@gmail.com(S. Zakavi).
Abstract
β-Chlorination of meso-tetraphenylporphyrin, H2TPP leads to the formation of H2TPPClx (X = 2,
4, 6 and 8) with absorption bands in the range of 421-451 and 650-715 nm, for the Soret and Q(0,0)
bands, respectively. H2TPPClx were used as photosensitizers for the aerobic photooxidation of
cyclooctene under mild conditions and compared with that of the non-chlorinated porphyrin. The
lowest degree of catalyst degradation was observed in the case of H2TPPCl2. H2TPPCl2 and
H2TPPCl4 showed the highest photocatalytic activity among the series. In spite of the electron-
withdrawing effects of the chlorine atoms, with the exception of H2TPPCl2, the chlorinated
porphyrins showed a comparable oxidative degradation to that of H2TPP towards oxidative
degradation. The preference of blue LED lamp over the red one shows that the absorption at the
Soret band region of the chlorinated porphyrins is mainly responsible for the photosensitizer ability
of the catalyst. In this regard, H2TPPCl2 and H2TPPCl4 which have the highest overlap with the
emission bands of the blue LED lamp showed the highest photocatalytic activity than the other
H2TPPClx and H2TPP.
Also, control reactions performed in the presence of 1,3-
diphenylisobenzofuran and 1,4-benzoquinone provided evidence for the involvement of singlet

oxygen as the main reactive oxygen species in this reaction. Singlet oxygen quantum yields (φF)
of H2TPPClx were smaller than that of the non-chlorinated porphyrin and decreased as H2TPPCl8>
H2TPPCl4 > H2TPPCl6 ≥ H2TPPCl2.H2TPPClx showed much higher photocatalytic activities than
their brominated counterparts. Also, the φF values and oxidative stability of the chlorinated
porphyrins were usually higher than those of the brominated porphyrins, with the same degree of
β-halogenation. Protonation of H2TPPClx with weak and strong acids (HCOOH, CF3COOH and
HCl), also their immobilization on a nano-structured acidic polymer (Amberlyst 15) were used to
study the influence of core protonation on photocatalytic activity and oxidative stability of the
chlorinated porphyrins.
Keywords: Heterogeneous and homogeneous photocatalysis, β-chlorinatedporphyrins, Singlet
oxygen, aerobic photooxidation, olefins, Near-IR photosensitizers
1. Introduction
Porphyrins and porphyrinoids have been widely used as photosensitizers in organic
transformations and photodynamic (PDT) therapy of cancer or photodynamic inactivation (PDI)
of microorganisms over the past century [1-9]. Photodynamic therapy largely involves the
intermediacy of singlet oxygen, a reactive oxygen species (ROS) which can carry out various
organic transformations and therefore novel photosensitizers are commonly designed for organic
transformation as well as medicinal medical and water treatment applications [10,11]. In this
regard, porphyrin and porphyrinoid-based photosensitizers with red shifted absorption bands and
enhanced oxidative stability are of great interest in photocatalysis and medicinal applications
[12,13]. The oxidative stability of porphyrin derivatives may be increased by introduction of
electron-withdrawing and/or bulky substituents at the porphyrin periphery [14]. In a recent work,

we have reported high oxidative stability of mono-ortho substituted manganese porphyrins with
simple substituents such as methyl and chlorine [15]. Also, molecular complexation between
meso-tetra(aryl)porphyrins and Lewis and protic acids was found to increase their oxidative
stability and improve their photocatalytic activity. Indeed, electron donation from central nitrogens
to the acid molecules was used to induce electron deficiency on the aromatic macrocycle.
Furthermore, the absorption region of the molecular complexes appeared in remarkably longer
wavelengths with respect to that of the corresponding free base porphyrins [16,17]. The latter is
one of the main features of efficient photosensitizers, from an application point of view [18]. In
spite of the synthesis of various porphyrin and phthalocyanine based photosensitizers, no attention
was paid to the photosensitizing ability of β halogenated porphyrins [19]. Due to the electron-
withdrawing effects of halogen atoms also the spin-orbit coupling induced by heavy halogen atoms
such as chlorine and bromine, the substitution of porphyrin periphery with halogens is expected to
increase the efficiency of this family of photosensitizers [20,21]. Furthermore, the presence of
halogens was found to significantly decrease the fluorescence quantum yields φF of halogenated
porphyrin with respect to that of meso-tetraphenylporphyrin (H2TPP) [22]. It is noteworthy that
radiative decay of the S1 and T1 excited states of porphyrin photosensitizers occurs through
fluorescence and phosphorescence emission, respectively. The radiative deactivation of the excited
states decreases the efficiency of formation of reactive oxygen species (ROS, mainly singlet
oxygen and superoxide radical anion) through energy and electron transfer mechanisms [23]. Very
recently, we have reported the first study of the photocatalytic activity of β-brominated porphyrins
and their dications with weak and strong acids [19]. In the present study, a series of β-chlorinated
porphyrins (Fig. 1), H2TPPClx (X = 2, 4, 6 and 8), were used as photosensitizers to investigate the

relationship between the degree of chlorination of porphyrins and their photocatalytic properties
in the oxidation of olefins.
Ar
meso

NH
N
Ar
Ar
HN
N
Ar
Ar
Porphyrin -chlorinated porphyrins
H2TPP
Dications
Phenyl
H2TPPClx ( X = 2, 4, 6, 8) H4TPPClx(X)2 (X = Cl-, CF3COO-, HCOO-)
Nano-structured Amberlyst 15 supported porphyrins, H4TPPClx@nanoAmb
Photosensitizer, O2, room temperature
O
OOH
LED lamp (White or blue),
H2O:CH3CN:Toluene (10:9:1)
Fig. 1. Porphyrin derivatives used for aerobic photooxidation of olefins in this study.
2. Experimental
2.1. Instrumentals and chemicals
The NMR spectra were obtained on a BrukerAvance DPX-400 MHz spectrometer. The absorption
spectra were recorded on a Pharmacia Biotech Ultrospec 4000 UV–Vis spectrophotometer. A
Varian-3800 gas chromatograph equipped with a HP-5 capillary column (phenylmethylsiloxane
30 m × 320 µm × 0.25 µm) and a flame-ionization detector was used for GC analysis. Porosimetry
analyses (nitrogen adsorption/desorption) were carried out on a Belsorp Max instrument (BEL-
Japan, Inc.). A Malvern Zetasizer Nano-ZS instrument (Malvern Instruments Ltd., Worcestershire,
UK) was used to determine the mean diameter and size distribution of the nanosized polymer
support before and after loading of porphyrin at 25 °C with a detection angle of 90°. Field-emission
scanning electron microscopy (FE-SEM; Carl Zeiss ULTRA PLUS) was utilized for determination

of the morphology and size of the nanoparticles. More details were explained elsewhere
[19,24].The DLS experiments were conducted using a Malvern ZEN3600 Zetasizer (Nano ZS).
The formation of cyclooct-1-en-3-yl hydroperoxide and 2-cyclohexen-1-one in the oxidation of
cyclooctene and cyclohexene, was confirmed by 13C NMR (See supplementary material of
[19,25]). 10 W blue, white and red LED lamps were used as the light source (8 mm diameter with
full width at half maximum less than ca. 20 nm for the most intense emission bands, see
supplementary material of [26] for the full emission spectra). N-chlorosuccinimide (NCS) was
recrystallized from hot water before use. 1,3-diphenylisobenzofuran (97%, Sigma-Aldrich) and
1,4-benzoquinone (reagent grade, ≥ 98%, Sigma-Aldrich) were used as received.
2.2. Synthesis and characterization of porphyrins
Chlorination of the porphyrin was conducted using freshly recrystallized NCS as the chlorine
source. The synthesis, purification and characterization of the chlorinated porphyrins (H2TPPClX;
X = 2, 4, 6 and 8) were performed according to the literature [27-29].
2.2.1. Synthesis H2TPP
The non-chlorinated porphyrin was prepared and purified according to the Adler method [30].
H2TPP was firstly recrystallized from dichloromethane by the gradual addition of methanol and
further purified by column chromatography on neutral alumina using dichloromethane as the
eluent.
H2TPP: UV-Vis in CH2Cl2, λmax/nm (logε): 416 (5.79), 513 (4.58), 542 (4.38), 584 (4.30), 644
(4.29) (Fig. S1). 1H NMR (400 MHz, CDCl3, TMS), δ/ppm: -2.71 (2H, s, NH), 7.77-7.83 (8Hm and
4Hp, m J= 7.8), 8.26-8.28 (8Ho, dd J= 8.3), 8.90 (8Hβ, s J= 8.9) (Fig. S2).

2.2.2. Synthesis and characterization of H2TPPClx (x = 2, 4, 6 or 8)
The β-chloroporphyrins were prepared from H2TPP by the following procedure.
2.2.2.1. Synthesis of NiTPP
NiTPP was prepared from the reaction of H2TPP (614 mg, 1 mmol) and Ni(OAc)2.4H2O (800 mg,
3 mmol) in refluxing DMF (10 ml) according to the Adler method [31]. UV-Vis in CH2Cl2
(λmax/nm): 414, 527 (Fig. S3). 1H NMR (400 MHz, CDCl3, TMS), δ/ppm: 8.7 (s, 8H, β-pyrrole J=
8.8), 7.95-8.10 (8H, o-phenyl J= 8.0), 7.6 (12H, m- and p-phenyl J= 7.7) (Fig. S4).
2.2.2.2. Synthesis NiTPPClx (x = 2, 4, 6 or 8)
NiTPPCl2, NiTPPCl4, NiTPPCl6 and NiTPPCl8 were prepared by chlorination of NiTPP with
different mole ratios of NCS in refluxing 1,2-dichlorobenzene following a previously reported
procedure [29]. The Soret band shifted to 419, 424, 430 and 439 nm in the case of NiTPPCl2,
NiTPPCl4, NiTPPCl6 and NiTPPCl8, respectively. The solid residue obtained by evaporation of
the solvent under vacuum was washed with CH2Cl2/MeOH (30 ml, 1:10 v/v) to remove the soluble
1
impurities. NiTPPCl2, UV-Vis in CH2Cl2, λmax (nm): 419, 535 (Fig. S5). H NMR (400 MHz,
CDCl3, TMS), δ/ppm: 8.55-8.85 (m, 6H, β-pyrrole J= 8.7), 7.80-8.05 (m, 8H, o-phenyl J= 7.9),
7.60–7.80 (m, 12H, m- and p-phenyl J= 7.7) (Fig. S6).
NiTPPCl4, UV-Vis in CH2Cl2, λmax (nm): 426, 544 (Fig. S7).1H NMR (400 MHz, CDCl3, TMS),
δ/ppm: 8.30 -8.87 (m, 4H, β-pyrrole), 7.80-8.05 (m, 8H, o-phenyl), 7.58–7.78 (m, 12H, m- and p-
phenyl) (Fig. S8).

NiTPPCl6, UV-Vis in CH2Cl2, λmax (nm): 435, 548. (Fig. S9).1H NMR (400 MHz, CDCl3, TMS),
δ/ppm:8.25-8.50 (m, 2H, β-pyrrole), 7.88-7.98 (m, 8H, o-phenyl), 7.63–7.78 (m, 12H, m- and p-
phenyl) (Fig. S10).
NiTPPCl8, UV-Vis in CH2Cl2, λmax (nm): 439, 553. (Fig. S11). 1H NMR (400 MHz, CDCl3, TMS),
δ/ppm: 7.90-7.95 (m, 8H, o-phenyl), 7.66-7.76 (broad, 12H, m- and p-phenyl) (Fig. S12).
2.2.2.3. H2TPPClx (X = 2, 4, 6 or 8)
H2TPPCl2, H2TPPCl4, H2TPPCl6 and H2TPPCl8 were prepared by demetallation of the Ni(II)
complexes with H2SO4 in CH2Cl2 [29].
H2TPPCl2, UV-Vis in CH2Cl2, λmax/nm (logε): 421 (5.48), 518 (4.19), 547 (3.69), 595 (3.68), 650
(3.58)(Fig. S13).1H NMR (400 MHz, CDCl3, TMS), δ/ppm: 8.60-8.95 (m, 6H, β-pyrrole), 8.05-
8.25 (m, 8H, o-phenyl), 7.72–7.85 (m, 12H, m- and p-phenyl), -2.7, -3.0 (2H, NH). (Fig. S14).
Anal. Calcd. For C44H28Cl2N2.0.5H2O.0.25C6H14: C, 76.52; H, 4.59; N 7.84%. Found: C, 73.40; H
4.50; N 7.20 (Table. S1).
H2TPPCl4, UV-Vis in CH2Cl2, λmax/nm (logε): 426 (5.43), 523 (4.25), 602 (3.74), 668 (3.64) (Fig.
S15).1H NMR (400 MHz, CDCl3, TMS), δ/ppm: 8.50-8.88 (m, 4H, β-pyrrole), 8-8.28(m, 8H, o-
phenyl), 7.65–7.93 (m, 12H, m- and p-phenyl),-2.50,-3.0 (2H, NH). (Fig. S16). Anal. Calcd. For
C44H26Cl4N4: C, 70.23; H, 3.48; N 7.45%. Found: C, 63.86; H 3.32; N 6.88 (Table. S2).
H2TPPCl6, UV-Vis in CH2Cl2, λmax/nm (logε): 444(4.43), 545(3.56), 587(3.49), 698 (3.41) (Fig.
S17).1H NMR (400 MHz, CDCl3, TMS), δ/ppm: 8.35-8. 53(m, 2H, β-pyrrole), 8.09-8.25 (m, 8H,
o-phenyl), 7.70–7.85 (m, 12H, m- and p-phenyl), -1.95 (2H, NH) (Fig. S18). Anal. Calcd. For

C44H24Cl6N4.CH2Cl2 .H2O.C6H14: C, 60.62; H, 4.19; N 5.54%. Found: C, 58.70; H 3.95; N 5.23
(Table. S3).
H2TPPCl8, UV-Vis in CH2Cl2, λmax/nm (logε): 451 (4.83), 553 (3.56), 602 (3.55), 715 (3.31) (Fig.
S19). 1H NMR (400 MHz, CDCl3, TMS), δ/ppm: 8.18- 8.25 (m, 8H, o-phenyl), 7.75-7.85 (broad,
12H, m- and p-phenyl), -1.25 (broad, 2H, NH). (Fig. S20). Anal. Calcd. For
C44H22Cl8N40.5CH2Cl2.0.5H2O.2C6H14: C, 60.91; H, 4.70; N 5.03%. Found: C, 60.04; H 4.51; N
4.76 (Table. S4).
2.3. Porphyrin diprotonated species
The diprotonated species were synthesized by the addition of excess amounts of HCOOH,
CF3COOH and aqueous HCl solution to the solution of the respective porphyrins in
dichloromethane. In the case of HCl, the organic phase was separated and dried to evaporate the
excess acid and solvent. In the case of the former, the acid was directly added to the
dichloromethane solution of the chlorinated porphyrins and the excess acid and dichloromethane
was removed by slow evaporation of solvent at room temperature. See [25] for more details. The
UV-vis and NMR spectra of the dications are shown in Electronic supplementary material, Fig.
S18 to S52). Immobilization of porphyrins on in the situ prepared nanostructured polymer
(nanoAmb) was described elsewhere [2,26].
2.4. General oxidation procedure
The photooxidation reactions were performed in a double walled cylindrical glass vessel. Water
circulation through the outer jacket was used to conduct the reactions at room temperature. In a
typical reaction, 6.6 ×10-4 mmol photosensitizer and cyclooctene (0.66 to 6.6 mmol) were added to
10 ml acetonitrile to achieve the desired molar ratio. The reaction mixture was analyzed by GC.

Control reactions in CDCl3 were used to characterize the oxidation product; the 13C NMR spectrum
confirmed the formation of cyclooct-1-en-3-yl hydroperoxide as the sole product (see Electronic
supplementary material, S54 and S53, also Fig. S55 and S56).
2.5. Chemical determination of singlet oxygen quantum yields
1,3-diphenylisobenzofuran (DPBF) as a specific quencher of singlet oxygen was used to determine
the singlet oxygen quantum yields of H2TPPClX. H2TPP with a ΦΔ value of 0.70 in acetonitrile
was used as a reference photosensitizer [32]. More details were described elsewhere [24-26]. No
catalyst degradation was observed at very short reaction times (less than 800 s), used for
determination of ΦΔ.
3. Results and Discussion
3.1. UV-vis spectra of H2TPPClx
The excitation of a given photosensitizer with the absorption of the light emitted by the light source
requires an efficient overlap between its absorption bands and the emission bands of the source.
From an application point of view, photosensitizers with red shifted absorption bands are more
desirable [33]. The UV-vis spectra of the chlorinated porphyrins, shown in Fig. 2 demonstrate a
continuous increase in the wavelength of the Soret band from H2TPP to H2TPPCl8 in both the Soret
and Q band regions. It is obvious that the position of the absorption bands in the UV-vis spectra is
an important factor determining the efficiency of the LED lamps in the photooxidation reaction (vide
infra).

Fig. 2. UV-vis spectra of the chlorinated porphyrins in dichloromethane.
Table 1. UV-vis spectral data of H2TPPClx in dichloromethane.
Porphyrin
Soret
λ/nm
417
IV
λ/nm
514
(4.58)
58
III
II
I
λ/nm
548
λ/nm
589
λ/nm
647
H2TPP
(logε, M-1cm-1)
H2TPPCl2
(logε)
(5.79)
421
(4.38)
547
(4.30)
595
(4.29)
650
(5.48)
426
(4.19)
-
(3.69)
523
(3.68)
602
(3.58)
668
H2TPPCl4
(logε)
(5.43)
444
(4.25)
545
(3.74)
587
(3.64)
698
H2TPPCl6
(logε)
-
-
(4.78)
451
(3.56)
553
(3.49)
602
(3.41)
715
H2TPPCl8
(logε)
(4.83)
(3.56)
(3.55)
(3.31)

Comparison of the data summarized in Table 1 with those of H2TPPBrx [19,22], reveals red shifts
in the position of the Soret and Q bands of H2TPPBrx which is more dominant for those with higher
degrees of bromination.
3.2. Photooxidation reactions
Previous studies on the photocatalytic activity of porphyrins showed that the type of solvent, light
source and photosensitizer to substrate molar ratio play crucial roles in the efficiency of
photooxidation reactions and the oxidative stability of the photosensitizers. Accordingly, several
control reactions (Tables 2 and 3) were performed to find the optimum reaction conditions.
3.3. Solvent effects
The oxidation of cyclooctene in the presence of H2TPPCl4 was conducted in different chlorinated
and non-chlorinated solvents. As seen from Table 2, a nearly complete reaction was achieved in
acetonitrile:water:toluene at a 10:9:1 volume ratio. The comparison of entries 1 and 2 in Table 1
shows that the presence of water is crucial for achievement of the maximum yield of cyclooct-1-
en-3-yl hydroperoxide. It is observed that in spite of the longer lifetime of singlet oxygen in
dichloroethane and CH2Cl2 (77 μs), the acetonitrile/water/toluene solvent mixture has been the best
solvent system for conducting the oxidation reaction. The lifetime of singlet oxygen in acetonitrile
(63 μs), toluene (29 μs) and water (3.8 μs) [34-37] is shorter than that in the former. It is notable
that H2TPP and the chlorinated derivatives had a limited solubility in acetonitrile and therefore the
presence of toluene, dichloromethane or dichloroethane was needed for complete dissolving of the
photosensitizer. The higher efficiency of the photooxidation reactions in water or water/acetonitrile
mixture was recently observed in the aerobic oxidation of sulfides catalyzed by porphyrins [19];
this observation has been attributed to the stabilization of the reaction intermediate by water as a

strong hydrogen bond forming solvent. It is noteworthy that in all used solvents, high degrees of
catalyst degradation were observed for a reaction time of 72 h.
Table 1. Oxidation of cyclooctene in the presence of H2TPPCl4 in different solvents.
Solvent
Catalyst:
alkene
Photodegradationa Yieldb
Entry Solvent
ratio (v/v)
(%)
83
(%)
11
99
12
18
5
1
2
3
4
5
6
Toluene:acetonitrile
9:1
10:9:1
9:1
1:5000
1:5000
1:5000
1:5000
1:5000
1:5000
Acetonitrile:H2O:toluene
Dichloromethane:methanol
Acetonitrile:H2O:toluene
Toluene
91
77.5
95
20:9:1
-
86
1,2-Dichloroethane
-
90
8
aMeasured based on the change in the absorbance at the λmax of the photosensitizer. bThe
formation of cyclooct-1-en-3-yl hydroperoxide as the sole product was confirmed by 13C NMR
(see S53, Fig. S53 and S54, also supplementary Information of [19,25]).

Table 3. Aerobic oxidation of cyclooctene under different reaction conditions in
acetonitrile:water:toluene at a 10:9:1 ratio.
Time Yieldb
LED
lamp
Catalyst: Photodegradationa
Entry Photosensitizer
TONc TOFd (h-1)
alkene
(%)
(h)
24
72
48
24
72
72
24
24
48
72
24
24
48
72
(%)
99
1
H2TPPCl2
H2TPPCl2
H2TPPCl2
H2TPPCl2
H2TPPCl2
H2TPPCl2
H2TPPCl4
H2TPPCl4
H2TPPCl4
H2TPPCl4
H2TPPCl4
H2TPPCl4
H2TPPCl6
H2TPPCl6
White
Blue
Red
1:1000
1:1000
1:1000
1:5000
1:5000
1:5000
1:1000
1:1000
1:5000
1:5000
59
93
10
57
90
7
990
990
41
13
4
2
99
3
22
220
4
White
Blue
Red
90
4500
4600
1500
1000
700
187
64
21
41
29
89
47
308
52
14
1
5
92
6
30
7
White
Blue
White
Blue
82
86
80
85
85
89
91
79
99
8
70
9
86
4300
3400
740
10
11
12
13
14
68
White 1:10000
7.40
Blue
White
Blue
1:10000
1:1000
1:1000
12.50 1250
68
6
6800
60

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
H2TPPCl6
H2TPPCl6
H2TPPCl6
H2TPPCl6
H2TPPCl8
H2TPPCl8
H2TPPCl8
H2TPPCl8
H2TPPCl8
H2TPPCl8
H2TPPCl8
H2TPP
Red
White
Blue
Red
1:1000
1:5000
1:5000
1:5000
1:1000
1:1000
1:1000
1:5000
1:5000
91
86
82
60
91
82
85
70
71
40
67
75
90
17
71
90
20
72
72
72
72
72
48
72
72
72
72
72
72
72
72
72
72
72
6.50
40
8
65
1
27
5
2000
400
8
1500
1000
200
21
14
4
White
Blue
Red
99
20
4
40
1
White
Blue
23
43
3
1150
2150
300
16
30
4
White 1:10000
Blue
White
Blue
Red
1:10000
1:1000
1:1000
1:1000
1:5000
1:5000
1:5000
3
300
4
80
75
20
82
82
80
800
11
10
3
H2TPP
750
H2TPP
200
H2TPP
White
Blue
Red
4100
4100
4000
57
57
55
H2TPP
H2TPP
a,b See the footnotes of Table 2. c Moles of product per mole of catalyst. d TON per unit time.

3.4. Effect of the light source
The oxidation of cyclooctene in the presence of H2TPPCl2 using 10 W white, blue and red
LED lamps (Table 3, entries 1-6) shows that the white and blue LED lamps are much more
efficient than the red one. Also, the catalyst degradation was significantly lower in the
presence of the white lamp. It is noteworthy that although the catalyst was much more
stable in the presence of the red lamp, no acceptable yield of cyclooct-1-en-3-yl
hydroperoxide was observed in the case of the red lamp. The preference of blue LED lamp
over the red one shows that the absorption at the Soret band region of the chlorinated
porphyrins is mainly responsible for the photosensitizer ability of the catalyst. The emission
spectra of the LED lamps demonstrated in Fig. S57 show that the emission bands of the
red, blue and white LED lamps appear at 595-656 (the emission maxima at 622 nm), 447-
496 (the emission maxima at 467 nm) and 421-469 & 502-637 nm (the emission maxima
at 476 nm), respectively. On the other hand, the comparison of entries 13-18 shows that in
the case of H2TPPCl6, the white LED lamp was much more efficient than the blue and red
ones.
This observation seems to be due to the shift of the Soret band of H2TPPCl6 to higher
wavelength (444 nm, in comparison with 421 nm for H2TPPCl2) leading to a more efficient
overlap between the emission bands of the white LED lamp and the absorption bands of
H2TPPCl6. Due to the very low yield of the product achieved in the presence of the red
LED lamp, this lamp was no longer used in the photooxidation studies.
3.5. Photosensitizer:olefin molar ratio

The increase in the catalyst to cyclooctene molar ratio from 1:1000 to 1:5000 for the reaction
performed in the presence of H2TPPCl4 (entries 7, 9, 11 for the white lamp and 8, 10, 12 for the
blue lamp) led to no detectable change in the yield of cyclooct-1-en-3-yl hydroperoxide, but a ca.
4.5-fold increase in the turnover number (TON) and turnover frequency (TOF) values. Further
increase in the molar ratio to 1:10000 led to a significant decrease in the efficiency of the oxidation
reaction. Interestingly, very similar degrees of catalyst degradation were observed in these molar
ratios.
The effect of catalyst to olefin molar ratio was also examined for the reaction catalyzed by
H2TPPCl8 (entries 19, 22, 24 for the white lamp and 20, 23, 25 for the blue lamp). Herein, the
increase in the molar ratio from 1:1000 to 1:5000 was associated with a remarkable decrease in
the conversion value. However, the oxidative degradation of the photosensitizer decreased from
91 to 43 %.
3.6. Effect of degree of chlorination
As seen from Table 3, the highest photocatalytic activity was observed for H2TPPCl2 and
H2TPPCl4; the oxidation of cyclooctene was completed in 24 h in the presence of these porphyrins,
but a much longer reaction time (72 h) was needed for completion of the reaction in the case of
H2TPPCl8 (entries 1, 7, 13 and 19). Furthermore, due to the nearly complete degradation of
H2TPPCl6 after 72 h, a highest cyclooct-1-en-3-yl hydroperoxide yield of ca. 70 % was achieved
in the presence of this porphyrin. On the other hand, the comparison of H2TPPCl2, H2TPPCl4 and
H2TPP shows a two or three-fold decrease in the reaction time in the presence of the chlorinated
porphyrins (entries 1, 7 and 26, also 4, 9 and 29). Interestingly, the lowest degree of catalyst
degradation was observed for H2TPPCl2 and the other chlorinated porphyrins were of comparable
catalyst stability to the non-chlorinated compound. Furthermore, while a comparable oxidative

stability was observed for H2TPP and H2TPPCl6, the latter showed a higher catalytic activity than
that of the former in the 1:5000 molar ratio of catalyst to olefin.
The positions with highest electron densities i.e. the α and meso carbon atoms [38] are expected to
be involved in oxidative degradation of porphyrins and metalloporphyrins. Furthermore, Dolphin
et al. showed that β position are also involved in porphyrin degradation. The introduction of
electron-withdrawing substituent at porphyrin periphery is usually accompanied with an increase
in its oxidative stability [14]. However, due to the influence of different factors including the
distribution of electron densities over the aromatic macrocycle, the steric effects of the meso and
β substituents and the energy of porphyrin HOMO levels [39], the relative oxidative stability of
porphyrins and metalloporphyrins cannot be explained on the basis of the known electronic effects
of the β and meso substituents [40]. In a previous work, the Mn(III) complex of meso-tetra(4-
nitrophenyl)porphyrin was found to have a decreased oxidative stability than the corresponding
complex of meso-tetra(phenyl)porphyrin [41]. In the case of H2TPPClx with four to eight chlorine
atoms a decreased oxidative stability of the photocatalyst was observed in comparison with that of
H2TPPCl2 and the corresponding non-chlorinated porphyrin. It seems that the electron-
withdrawing effects of the β chlorine atoms caused an increase in the electron densities at positions
susceptible to oxidative destruction (α, β and meso carbons).
3.7. H2TPPClx vs. H2TPPBrx
In a previous study [19], we have found that the series of H2TPPBrx suffer from extensive
photooxidative degradation in the aerobic oxidation of olefins, leading to relatively low
photocatalytic activity of these compounds. The comparison of the results summarized in Table 3
with those obtained using H2TPPBrx [19] under the same conditions (mole ratio of photosensitizer

to olefin, light source and solvent), shows higher oxidative stability and significantly higher
catalytic activity of H2TPPClx. Also, much shorter reaction times were needed to achieve
quantitative conversions in the oxidation of cyclooctene, especially in the case of the moderate
degrees of β-chlorination. It is noteworthy that, under the optimized conditions, conversion values
in the range of ca. 0 to 30 % were achieved in the presence of the brominated porphyrins.
3.7. Reactive oxygen species
Previous studies concerning the mechanism of photooxidation reactions mediated by porphyrin
derivatives revealed that singlet oxygen and/or superoxide anion radical species are usually
involved in the aerobic oxidation of organic compounds [15-17, 25,26]. In a type I mechanism,
free radicals formed through electron or hydrogen atom transfer between the excited states of
photosensitizer and organic substrate react with oxygen to form species such as superoxide radical
anion. Type II mechanism involves an energy transfer process between the triplet excited state of
photosensitizer and triplet molecular oxygen, leading to the formation of singlet oxygen species
[23].

Fig.3. The change in the UV–vis spectrum of DPBF (9.5 × 10-5) (a) in the presence of H2TPPCl2
in acetonitrile using a 10 W red LED lamp as the light source. Fig. b shows the changes at the λmax
of DPBF vs. time (blue curve); the red curve demonstrates the change in absorption band of DPBF
in the absence of H2TPPCl2.
In this study, DPBF as a well-known singlet oxygen scavenger [44] was used to determine the
nature of reactive oxygen species responsible for the oxidation of cyclooctene (Fig. 3). A solution
of DPBF in acetonitrile (9.5 ×10-5 M) was irradiated with a 10 W red LED lamp in the presence
and absence of H2TPPCl2 as the best photosensitizer of the series; fast disappearance of the
absorption band due to DPBF at 415 nm is in accord with the presence of singlet oxygen as the
main ROS. Furthermore, in a control reaction, the oxidation of cyclooctene was conducted in the
presence of 1,4-benzoquinone as the quencher of superoxide radical anion [45]. The result showed
no detectable change in the conversion value showing the absence of superoxide radical anion in
the catalytic cycle. This observation may be due to the electron-withdrawing effects of the chlorine
atoms which prevents electron transfer from the porphyrin photosensitizer to organic substrates. It
should be noted that previous studies [16,24,25,42,43] showed little or no contribution of an
electron transfer path to the photooxidation reactions catalyzed by porphyrin derivatives.
3.8. Singlet oxygen quantum yields
Singlet oxygen quantum yields (ФΔ) of H2TPPClx, measured through the reaction of the ROS with
1,3-diphenylisobenzofuran (DPBF) as a quencher of singlet oxygen (Fig. 4) and methylene blue
(ΦΔ= 0.52 in water and acetonitrile) [44] as a reference photosensitizer, are summarized in Table
4.

Fig. 4. Change in the UV-vis spectrum of DPBF upon irradiation with a 10 W red LED lamp in
the presence of H2TPPClx.

Table 4. Singlet oxygen quantum yield of H2TPPClx and H2TPPBrx in dichloromethane.a
b
Photosensitizer
Ф∆
Photosensitizer
Ф∆
H2TPP
0.70
0.08
0.19
0.10
0.37
H2TPP
0.70
0.21
0.13
0.04
0.02
H2TPPCl2
H2TPPCl4
H2TPPCl6
H2TPPCl8
H2TPPBr2
H2TPPBr4
H2TPPBr6
H2TPPBr8
aDetermined chemically using DPBF as a quencher of singlet oxygen and H2TPP as a reference
photosensitizer. bSee [19] for more details.
Singlet oxygen quantum yields of H2TPP and H2TPPClx decreased in the order: H2TPP> H2TPPCl8
> H2TPPCl4 > H2TPPCl6 ≥ H2TPPCl2. It is noteworthy that control reactions showed that no
catalyst degradation occurs in the short reaction times (< 800 s) used for evaluation of the singlet
oxygen quantum yields. The order of catalytic activity of the chlorinated porphyrins and H2TPP is
different from that observed for the singlet oxygen quantum yields.
The comparison of ФΔ of H2TPPClx and H2TPPBrx [19], shows that with the exception of
H2TPPCl2, the former show higher ФΔ values. Also, the two series of β-halogenated porphyrins
showed decreased ФΔ values than H2TPP. Furthermore, no close correlation was observed between
the degree of halogenation of H2TPP and ФΔ of the photosensitizers.
3.8. Porphyrin diacids vs. the free base porphyrins
The catalytic and photocatalytic activities of most porphyrins are usually limited by their low
stability under oxidative and photooxidative conditions [14,40]. As was shown previously
[16,19,24,25], protonation of porphyrins can be used to increase the photooxidative stability of

meso-tetra(aryl)porphyrins. As seen from Table 3, most of the chlorinated porphyrins showed a
comparable oxidative degradation to that of H2TPP. The oxidation of cyclooctene was carried out
using the dications of the brominated porphyrins with various weak and strong acids (Fig. 5). Fig.
4 shows decreased photocatalytic activity of the porphyrin diacids compared to the free base
porphyrins, although a 5 to 14% increase in the yield of cyclooct-1-en-3-yl hydroperoxide was
observed in the case of the dications of H2TPPCl4. Also, the porphyrin diacids showed comparable
or lower oxidative stabilities than those of the corresponding free base porphyrins (Fig. 6).
Fig. 5. Catalytic activity of free base H2TPPClx and their diprotonated species with CF3COOH and
HCOOH in the oxidation of cyclooctene; porphyrin and cyclooctene were used in a 1:5000 mole
ratio. The reactions were carried out in the presence of a 10 W, white LED lamp.

Fig. 6. Photooxidative degradation of free base H2TPPClx and their dications.
Fig. 7. DRS spectra of H2TPPCl2@nanoAmb (red curve), H2TPPCl4@nanoAmb (blue curve),
H2TPPCl6@nanoAmb (green curve) and H2TPPCl8@nanoAmb (purple curve).

Furthermore, H2TPPClx compounds were immobilized on nano-structured amberlyst 15
(nanoAmb) to study the catalytic activity and oxidative stability of the chlorinated porphyrins
under heterogeneous conditions. The DRS spectra of H2TPPClx@nanoAmb, shown in Fig. 7.
confirm the immobilization of H2TPPClx on the polymer support [24,26]. Porosimetry analysis is
in accord with the mesoporous structure of H2TPPClx@nanoAmb (Fig. 8, Table 5).
Table 5. BET analysis of nanoAmb and H2TPPClx@nanoAmb.
Photosensitizer
Mean pore diameter (nm)
Vt (cm3g-1)a
SBET (m2g-1)a
nanoAmb
H2TPPCl2@nanoAmb
H2TPPCl4@nanoAmb
50.321
38.067
33.986
0.5183
0.3229
0.3458
85
33
40
H2TPPCl6@nanoAmb
H2TPPCl8@nanoAmb
33.009
30.071
0.3790
0.3795
45
50
aSBET and Vt are the specific surface area and total pore volume, respectively.
Fig. 8. N2 adsorption/desorption isotherms of nanoAmb and H2TPPCl2@nanoAmb.
DLS measurements showed particle size diameters of 226.9 and 195.1 nm for
H2TPPCl2@nanoAmb and H2TPPCl4@nanoAmb, respectively (Fig. S58, S59). Also, FESEM images

confirmed the nanostructure of the immobilized porphyrins with particle size less than ca. 50 nm for the
two compounds (Fig. 9 and Fig. S60). EDX spectrum and elemental mapping gave evidence for uniform
distribution of different elements in the case of H2TPPCl2@nanoAmb and H2TPPCl4@nanoAmb (Fig.10
and Fig. S61).
Fig. 9. FE-SEM images and EDX elemental mapping of H2TPPCl2@nanoAmb.

Table 6. Photooxidation of cyclooctene and cyclohexene in the presence of nanoAmb supported
H2TPPClx.a
entry
Photodegradation
TOF
(h-1)
52
55
36
45
86
41
12
5
Photosensitizer
Yield (%) Time (h) TON
(%)b
15b
1
H2TPPCl2@nanoAmb
H2TPPCl4@nanoAmb
H2TPPCl6@nanoAmb
H2TPPCl8@nanoAmb
H2TPPCl2@nanoAmb
H2TPPCl4@nanoAmb
H2TPPCl6@nanoAmb
H2TPPCl8@nanoAmb
H2TPPCl4
75
53
35
20
83
39
12
5
48
48
48
48
48
48
48
48
48
48
48
48
3750
2650
1750
2150
4150
1950
600
2
23b
3
42b
4
58b
5
2c
6
12c
7
60c
8
95c
250
9
85b,d
21b,d
80c,d
30c,d
75
50
86
47
3750
2500
4300
2350
78
52
89
49
10
11
12
H2TPPCl4@nanoAmb
H2TPPCl4
H2TPPCl4@nanoAmb
aA 1:5000 molar ratio of catalyst to cyclohexene (see S54, Fig. S55 and S56) was used. b,c Under
irradiation with blue and white LED (10 W) lamps, respectively. d For the oxidation of cyclooctene.
The immobilization of H2TPPClx on nanoAmb decreased their oxidative degradation from 71-90%
(Table 3, entries 4, 5, 9, 10) and 57-86% (Table 3, entries 16, 17, 22, 23) to 15-58% (Table 6,
entries 1-4) and 2-60% (Table 6, entries 5-8), in the case of the blue and white LED lamps,
respectively. The effect was more pronounced in the case of H2TPPCl2 and H2TPPCl4. It is
observed that with the exception of H2TPPCl8 (in the presence of the blue LED lamp), the oxidative

degradation of the whole series was significantly decreased through immobilization on nanoAmb
by an acid-base reaction. However, the yield of cyclooct-1-en-3-yl hydroperoxide decreased under
the heterogeneous conditions, probably due to the slower kinetics of heterogeneous reactions than
the homogeneous ones. The effect of catalyst immobilization was also studied in the oxidation of
cyclohexene under the same conditions (Table 6, entries 9-12). As seen from Table 6, the results
are in agreement with a remarkable decrease (2.7 to 4 fold) in the oxidative degradation of the
catalyst with concomitant decrease in the catalytic activity of the heterogenized catalyst.
4. Conclusion
The series H2TPPClx compounds with 2 to 8 chlorine atoms at the β position were used as
photosensitizers in the aerobic oxidation of cyclooctene and found that (i) in spite of the fact that
H2TPPCl2 andH2TPPCl4 were the most efficient catalysts of the series, further increase in the
number of chlorine atoms led to no enhancement of the photocatalytic activity of the chlorinated
porphyrins; it decreased in the order H2TPPCl2 > H2TPPCl4 > H2TPPCl8> H2TPPCl6; (ii) the φF
value of H2TPPClx decreased as H2TPPCl8> H2TPPCl4 > H2TPPCl6 ≥ H2TPPCl2 which was less
than that of the non-chlorinated porphyrin; (iii) the highest oxidative stability was observed in the
case of H2TPPCl2. The other chlorinated porphyrins were as stable as the non-chlorinated
porphyrin, H2TPP; (iv) a significant decrease in the reaction time was observed for the oxidation
reaction catalyzed by H2TPPCl2 and H2TPPCl4 compared to that conducted in the presence of
H2TPP; (vi) complete or almost complete oxidation of cyclooctene was achieved in the presence
of H2TPPClx with two, four and eight chlorine atoms at the pyrrole rings, using the optimized
porphyrin to olefin molar ratio. In the case of H2TPPCl6, a maximum yield of ca. 60% was achieved
under the optimized reaction conditions; (vii) control reactions conducted in the presence of DPBF
and 1,4-benzoquinone gave evidence for the involvement of singlet oxygen as the main ROS in

these reactions; (vii) the chlorinated porphyrins showed significantly higher photocatalytic activity
than the corresponding brominated porphyrins. Also, the former usually show higher φF values and
oxidative stabilities than the latter; (viii) while diprotonation of the chlorinated porphyrins with
weak and strong acids led to no increase in their catalytic activity and oxidative stability under
homogeneous conditions, porphyrin diacids formed by acid-base reaction between H2TPPClx and
Amberlyst 15 nanoparticles led to significant increase in their stability towards oxidative
degradation.
Acknowledgements
Financial support of this work by the Institute for Advanced Studies in Basic Sciences
(IASBS) is acknowledged.
References
[1] J. Kou, D. Dou, L. Yang, Oncotarget. 8 (2017) 81591-81603.
[2] A. Heydari-turkmani, S. Zakavi, N. Nikfarjam, New J. Chem. 41 (2017) 5012–5020.
[3] J. Zheng, X. Li,Y. Qin, S. Zhang, M. Sun, X. Duan, H. Sun, P. Li, S. Wang, J. Catal. 371
(2019) 214-223.
[4] G. Mele, R. Del Sole, G. Vasapollo, E. Garcı́a-López, L. Palmisano, M. Schiavello, J. Catal.
217 (2003) 334-342.
[5] M. Cheng, Y. Xi, C. Jing, W. Jing, Z. L-N Zhu, De-M. Kong, ACS Appl. Mater.
Interfaces 11 (2019) 13158-13167.
[6] X. Zhang, M. Cibian, A. Call, K. Yamauchi, K. Sakai, ACS Catal. 9 (2019) 11263-11273.

[7] Y. Yilmaz (Ed.), Phthalocyanines and Some Current Applications, 2015, Published by
InTech, Croatia.
[8] A. E. O’Connor, W. M. Gallagher, A. T. Byrne, Photochem. Photobiol. 85 (2009) 1053–
1074.
[9] J. G. Levy, M, Obochi, Photochem. Photobiol. 64 (1996) 737-739.
[10] D. García‐Fresnadillo, ChemPhotoChem. 2 (2018) 512-534.
[11] M. Riethmüller, N. Burger, G. Bauer, Redox Biol. 6 (2015) 157-168
[12] Y.‐J. Yuan Dr. Z.‐T. Y. H.‐L. Gao, G. Zou, Ch. Zheng, W. Huang, Chem. Eur. J. 19 (2013)
6340-6349.
[13] E. F. F. da Silva, F. M. Pimenta, B. W. Pedersen, F. H. Blaikie, G. N. Bosio, T.
Breitenbach, M. Westberg, M. Bregnhøj, M. Etzerodt, L. G. Arnaut, P. R. Ogilby, Integr. Biol.
8 (2016) 177-193.
[14] I. D. Cunningham, T. N. Danks, J. N. Hay, I. Hamerton, S. Gunathilagan, C. J. Janczak, J.
Mol, A Chem. 185 (2002) 25-31.
[15] R. Nasrollahi, S. Zakavi, New J. Chem. 42 (2018) 1806-1815.
[16] A. G. Mojarrad and S. Zakavi, Eur. J. Inorg. Chem. 2017 (2017) 2854-2862.
[17] A. G. Mojarrad, S. Zakavi, RSC Adv. 6 (2016) 100931-100938
[18] R. Bonnett, Chem. Soc. Rev. 24 (1995) 19-33.