Photooxidation of 3,5-diaryl-1-phenyl-2-pyrazolines: Experimental and computational studies

Article abstract of DOI:10.1016/j.jphotochem.2019.112285

Photooxidation of various 2-pyrazolines is studied experimentally and computationally. Experimental results show that the electron-donating/withdrawing substituents increases/decreases the rate of this photoreaction. The proposed light-induced electron-tr

Full text of DOI:10.1016/j.jphotochem.2019.112285

Journal Pre-proof
Photooxidation of 3,5-diaryl-1-phenyl-2-pyrazolines: Experimental and
computational studies s
Marzieh Soltani, Hamid Reza Memarian, Hassan Sabzyan
PII:
S1010-6030(19)31717-4
DOI:
https://doi.org/10.1016/j.jphotochem.2019.112285
JPC 112285
Reference:
To appear in:
Journal of Photochemistry & Photobiology, A: Chemistry
Received Date:
Revised Date:
Accepted Date:
7 October 2019
11 November 2019
2 December 2019
Please cite this article as: Soltani M, Memarian HR, Sabzyan H, Photooxidation of
,5-diaryl-1-phenyl-2-pyrazolines: Experimental and computational studies s, Journal of
3
Photochemistry and amp; Photobiology, A: Chemistry (2019),
doi: https://doi.org/10.1016/j.jphotochem.2019.112285
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.
©
2019 Published by Elsevier.

1
Photooxidation of 3,5-diaryl-1-phenyl-2-pyrazolines:
Experimental and computational studies s
Marzieh Soltani, Hamid Reza Memarian*, Hassan Sabzyan*
Department of Chemistry, University of Isfahan, 81746-73441 Isfahan, I. R. Iran.
Graphical abstract
Highlights
Steric and electronic effects of the substituents and nature of solvent affect the
irradiation time.
Computational (TD)B3LYP/6-311++G(d,p) results describe tunneling electron-
transfer process.
Triplet excited states of 2-pyrazolines may have little contribution to the electron
transfer step.
Cyclic voltammetric measurements support the photochemical results.
ABSTRACT

2
Photooxidation of various 2-pyrazolines is studied experimentally and computationally.
Experimental results show that the electron-donating/withdrawing substituents
increases/decreases the rate of this photoreaction. The proposed light-induced electron-transfer
mechanism explains the steric and electronic effects of the substituents, co-planarity of the aryl
rings substitutions, orientations of the C
solvent on the irradiation time for completion of reaction. Computational (TD)B3LYP/6-
11++G(d,p) results, including frontier orbital energies, UV-Vis transitions, electrostatic
3
-aryl ring towards the C
3
=N
2
double bond and the
3
potential, CHELPG charges, are used to find and describe tunneling electron-transfer process
and its associated structural relaxations which form the rate-determining step of the reaction,
and to justify the effect of substitutions on the rate of reaction. The intermediate complex
formed after the first electron transfer proceeds the proton transfer step only in its triplet state.
Triplet excited states of 2-pyrazolines may have little contribution to the electron transfer step.
Cyclic voltammetric measurements support the photochemical results.
Keywords: Photooxidation, 2-Pyrazolines, Reaction mechanism, Substituent effects, Tunneling
electron transfer.
1
. Introduction
Study of the pyrazoline derivatives is significant due to their various biological properties
such as anti-inflammatory, anti-cancer, antidiabetic and antibacterial activities [1-4]. In
addition, oxidative aromatization of 1,3,5-triaryl substituted pyrazolines to their corresponding
1
,3,5-triaryl substituted pyrazoles using thermal [5-9] or photochemical methods [10-16] is
interesting due to their important role in history of heterocyclic chemistry and changing their
pharmaceutical activities [17,18] upon oxidation or photooxidation. Therefore, due to the light

3
sensitivity of these compounds, which is important for the drug designers, considerable
attention has been focused on the photochemical behavior of 2-pyrazolines. A number of works
are devoted to the photooxidation of 2-pyrazolines either under direct UV light irradiation in
different solvents [10,11] or using dye sensitizers, such as methylene blue, Rose Bengal, Eosin
Y and Riboflavin, under visible light for producing the singlet oxygen as oxidant [12-14].
Direct irradiation of tri- or tetrasubstituted 2-pyrazolines in benzene under oxygen atmosphere,
after long irradiation times (48 to 88 hours), results in photooxidation to their corresponding
pyrazoles, and formation of 2-anilinopropanenitrile as by-product in some cases [11]. Owing
to the light absorption of some aryl- and heteroarylpyrazolines above 400 nm, their photo-
aromatization is observed by irradiation with the visible light (λ > 400 nm) in carbon
tetrachloride solvent [15]. Electron-transfer (ET) induced conversion of 1,3,5-
triarylpyrazolines to 1,3,5-triarylpyrazoles is reported using platinum(II) polypyridyl complex,
as photocatalyst, by irradiation with the visible light (λ = 450 nm), in which the former
compound in the ground state transfers an electron to electronically excited platinum complex
[16].
Our research interest concerns the investigation of the steric and electronic effects of
substitutions on different positions of various heterocyclic compounds on their thermal [19]
and sonothermal oxidation [20], photooxidation [21], cyclic voltammetric behavior [22],
spectral characteristic and computational results [23,24].
Recently, we have reported the synthesis of various 3- or 5-aryl substituted 1-phenyl-2-
pyrazolines or 4,5-dihydropyrazoles (DHPZ), containing electron-donor and electron-acceptor
substituents on the 3- or 5-aryl positions, in order to elucidate the steric and electronic
(inductive and resonance) effects of the C
3
- and C -aryl substitutions on the DFT computational
5
and spectroscopic results [25], and cyclic voltammetric measurements [26] (Scheme 1). In
continuation to these works, we were interested to elucidate these effects on the light sensitivity

4
of 2-pyrazolines by exposing them to the UV light and also to propose a clear reaction
mechanism by taking the results of experimental and various computational studies into
account.
2
. Results and discussion
Since the nature of solvent, especially its polarity and electronic behavior, affect the rate
of thermal or photochemical reactions, irradiation at λ ≥ 314 nm wavelengths of compound 1a
as representative was carried out in acetonitrile, chloroform and ethanol solvents. The results,
presented in the footnote of Table 1, indicate that the photooxidation reaction is faster in
chloroform than in the other two solvents. Therefore, to explain the steric and electronic effects
of the aryl substitutions on the C
pyrazolines (1a-s) on the reaction completion time of their photooxidation process, irradiation
λ ≥ 314 nm) of 1 mM solution of these compounds was carried out in chloroform until attaining
3
- or C -positions of the heterocyclic ring in 1-phenyl-2-
5
(
maximum (mostly complete) conversion (Scheme 1). The solvent was evaporated and the pure
products were obtained by recrystallization from ethanol. The photochemical results are
presented in Table 1.
Scheme 1. Photooxidation of aryl substituted 4,5-dihydro-1-phenyl-2-pyrazolines.

5
Table 1
Photooxidation of C
3
- or C -aryl substituted 1-phenyl-2-pyrazolines (1a-s) in
5
chloroform solvent to their corresponding pyrazoles (2a-s), introduced in
Scheme 1.
Compound
Y
H
X
H
Product Time (min) [a] Conversion (%)
1
1
1
a
a
a
2a
2a
2a
10
~90
100
100
[b]
H
H
H
H
180
[
c]
240
1
b
4-Cl
3-Cl
2-Cl
H
H
H
2b
2c
2d
2e
2f
15
15
30
5
~90
100
100
~95
100
100
100
100
100
100
100
100
100
~95
~70
100
100
100
1
c
1
d
H
1
e
4-Cl
3-Cl
2-Cl
H
1
f
H
12
5
1
g
H
2g
2h
2i
1
h
4-MeO
3-MeO
2-MeO
H
0.5
2
1
i
H
1
j
H
2j
1
1
k
4-MeO
3-MeO
2-MeO
H
2k
2l
1.5
2
1
l
H
1
m
H
2m
2n
2o
2p
2q
2r
2s
1
1
n
4-NO
3-NO
2-NO
H
2
2
2
90
15
30
70
30
13
1
o
p
q
H
1
H
1
4-NO
3-NO
2-NO
2
2
2
1
r
H
1
s
H
[
a]
[b]
The times of the maximum progression of reaction.
The reaction is carried out in
[
c]
acetonitrile solution. The reaction is carried out in ethanol solution.
Dehydrogenation of 2-pyrazoline derivatives and formation of the corresponding
1
photooxidation products, pyrazoles (2a-s), is approved by comparison of their IR, H NMR,
1
3
and C NMR and UV-Vis spectra as follows:
Infrared spectra indicated a decrease in the intensity of the aliphatic CH stretching due to
the formation of the aromatic pyrazole ring upon oxidation and elimination of two
hydrogens.
1
In H NMR spectra of 2a–s, one of the proton of the geminal 4-CH
2
hydrogens and its
vicinal 5-CH hydrogen of the AMX system in the NMR spectra of 1a-s disappeared and

6
simultaneously the absorption peak of the remaining 4-CH proton is shifted to lower field
(higher frequency) upon photooxidation and formation of the aromatic heterocyclic ring.
1
3
The C NMR spectra of pyrazole compounds show shifts of the C
4
and C resonances
5
towards the aromatic region (lower field) due to the occurrence of the aromatic pyrazole
1
13
ring. These observations in the H NMR and C NMR spectra of compound 1a and the
corresponding oxidation product 2a, as representatives, are presented in Fig. 1 and the
remaining spectra are provided in the Supporting Information.
Owing to the formation of the aromatic pyrazole ring, a decrease of the conjugation of the
C
3
-aryl ring with the C
3
=N double bond (the imine moiety) in the aromatic pyrazole ring
2
is expected. As a result, hypsochromic shift in the UV spectra of 2a-s is observed, which is
presented for the representative compounds 1a and 2a in Fig. 2, and the UV spectral data
of all 1a-s and 2a-s compounds are given in Table 2.
HX
HM HA
8
7
6
5
4
3
2
150
120
90
60
30

7
H
8
7
6
5
4
3
2
150
120
90
60
30
1
13
Fig. 1. Comparison of H NMR and C NMR spectra of compound 1a with those of the corresponding
oxidation product 2a showing the elimination of the benzylic (5-CH) and one of the aliphatic (4-CH )
2
hydrogen atoms under formation of the aromatic pyrazole ring upon photooxidation.
4
3
2
1
0
0
0
0
0
1
2
a
a
2
30
280
330
380
430
λ (nm)
Fig. 2. UV spectra of compound 1a and its corresponding oxidation
product 2a in chloroform showing the hypsochromic shift in the UV
spectra of 2a compared to that in 1a due to aromatization of the
heterocyclic ring.
Table 2
Comparison of the UV-Vis absorption data of 3- or 5-aryl substituted 1-phenyl-2-pyrazolines
1a-s) and their corresponding 1-phenyl-2-pyrazoles (2a-s) in CHCl
(
3
.
1
Y
X
λ
max (nm), ε
2
λmax ( nm), ε
a
H
H
355 nm (20893), 313 (sh, 9120),
a 293 nm (sh, 7943), 252 (30903)
241 (19953)
b
4-Cl
H
367 nm (22909), 315 (sh, 8318),
b 299 nm (sh, 8128), 262 (35481)
243 (20417)

8
c
3-Cl
2-Cl
H
H
367 nm (21878), 312 (sh, 6918),
46 (16596)
c
298 nm (sh, 4365), 255 (22387)
2
d
357 nm (15488), 314 (sh, 7762),
46 (13490)
d Tailed to 315 nm, 253 (29512)
2
e
f
H
H
4-Cl 355 nm (18197), 242 (16982)
3-Cl 356 nm (19055), 316 (sh, 9332),
e
f
290 nm (sh, 8710), 258 (30199)
294 nm (sh, 2818), 256 (11749)
240 (16218)
g
H
2-Cl 355 nm (17783), 316 (sh, 8318),
g 291 nm (sh, 10000), 257 (22909)
h 302 nm (sh, 5888), 260 (25704)
240 (15488)
h 4-CH
3
O
H
355 nm (30199), 308 (sh, 19498),
250 (33113)
i
j
3-CH
2-CH
H
3
3
O
O
H
H
360 nm (18621), 241 (13183)
357 nm (21878), 251 (19055)
i
297 nm (sh, 10233), 254 (28184)
297 (sh, 10965), 252 nm (23442)
j
k
4-CH
3
O 357 nm (19498), 315 (sh, 8912),
k Tailed to 320 nm, 257 (29512)
239 (18621)
l
H
H
3-CH
2-CH
3
3
O 356 nm (16596), 241 (14791)
O 360 nm (18621), 315 (sh, 8318),
l
293 nm (9772), 253 (27542)
m
m Tailed to 330 nm, 276 (23436),
243 (17378)
254 (24351)
n
4-NO
2
H
446 nm (22387), 267 (16218)
n 329 nm (21878), 263 (sh, 15488),
2
39 (25119)
o
3-NO
2-NO
2
2
H
H
367 nm (22387), 241 (20893)
350 nm (13490), 296 (sh, 9120),
o Tailed to 340 nm, 255 (41687)
p Tailed to 365 nm, 296 (34674)
p
246 (18621)
q
r
s
H
H
H
4-NO
3-NO
2-NO
2
2
2
350 nm (20417), 249 (20893)
351 nm (15849), 247 (18197)
353 nm (22909), 313 (sh, 12303),
q 308 nm (sh, 13183), 265 (21878)
r
s
292 nm (sh, 8128), 256 (39811)
291 nm (sh, 10000), 251 (26915)
243 (22909)
The results presented in Table 1 indicate that the tendency of electron-receiving of the
solvent, and also the electron-donating nature of the additional substituent especially on the C
5
-
aryl ring decrease the time of irradiation. Based on the proposed reaction mechanism for the
electron-transfer induced photooxidation process, shown in Scheme 2, and the failure of the
oxidation process in the absence of the UV light, the aryl substituted 1-phenyl-2-pyrazolines

9
or 4,5-dihydropyrazole (DHPZ) is promoted to its excited state (DHPZ*) by absorbing the light
photons. This process is followed by an effective electron transfer (ET) from the excited 2-
pyrazoline molecule (DHPZ*) as an electron-donating species to chloroform as a suitable
electron-acceptor molecule, which results in creation of the dihydropyrazoline radical cation
•
+
•−).
(
DHPZ ) with the positive charge on the N
Following this ET step and in competition with possible back electron transfer (BET) process,
the presence of the positive charge on the N atom is the driving force for elimination of one
proton from its neighborhood, namely the C
1
atom, and the chloroform radical anion (CHCl
3
1
•+
5
-position of the DHPZ species accompanied by
•−
the chloride ion eliminated from the CHCl
3
intermediate. This step leads to the formation of
•
the hydropyrazolyl radical (HPZ ), with the single electron centered on the benzylic C
5
atom,
and the dichloromethyl radical (CHCl
2
•) intermediates. Finally, the removal of the second
•
hydrogen atom from the hydropyrazolyl radical intermediate (HPZ ) completes the reaction
under formation of pyrazole (PZ) and dichloromethane.
Scheme 2. The mechanism proposed for the light-induced photooxidation of 3- or 5-aryl
substituted 1-phenyl-2-pyrazolines.

10
In order to elucidate the tendency of electron-receiving of the solvent involved in the
proposed photoreaction mechanism, irradiation (λ ≥ 314 nm) of 1a as a representative sample
was carried out in acetonitrile, chloroform and ethanol solvents. Whereas ~90% conversion of
1
a is achieved after 10 min irradiation in chloroform, total conversion of 1a to 2a is observed
after 180 and 240 min by carrying the irradiation in acetonitrile and ethanol, respectively. This
indicates that photooxidation reaction in chloroform solution is faster than that in the two other
solvents. The UV-Vis absorption spectra of the parent compound 1a in all three solvents
acetonitrile, chloroform and ethanol, presented in Fig. 3, indicate that the absorption intensity
of this compound especially above 314 nm (the experimental condition of photooxidation
reaction) in these solvents are not significantly different. Molar absorption coefficients (ε) at
the maximum absorption wavelength (λmax) for compound 1a at 354 nm in acetonitrile (ε =
-1
-1
-1
-1
1
7054 Lmol cm ), at 359 nm in chloroform (ε = 16264 Lmol cm ) and at 354 nm in ethanol
-1
-1
(
ε = 19301 Lmol cm ) are obtained. By comparison of these data, it is expected that in ethanol
the excitation of 1a should provide larger number of excited molecules (1a*) for donation of
an electron to the solvent molecules. Therefore, shorter irradiation time is expected when the
photooxidation reaction is carried out in ethanol solvent. Unfortunately, results showed that
photooxidation of 1a in ethanol was not effective. The reduction potentials of acetonitrile and
+
chloroform are reported to be ‒3.1 and ‒1.77 V (vs. FC /FC, Ferrocene), respectively [27,28],
which indicate that chloroform acts as a better electron-acceptor species. Consequently, it is
expected that the electron-transfer process from excited pyrazoline to chloroform molecule be
faster than that to acetonitrile and ethanol, which confirms the mechanism proposed in Scheme
2
, involving the solvent as an electron-acceptor molecule in this photooxidation reaction. In
addition to these findings, the results of computational studies, discussed later, explain
reasonably the electron transfer of photo-excited DHPZ to chloroform solvent.

11
20
15
10
5
0
CHCl3
CHCl
C H OH
3
2
5
EtOH
CH CN
3
3
CH CN
240
280
320
360
400
440
λ (nm)
Fig. 3. The UV-Vis absorption spectra of 1a obtained in chloroform,
acetonitrile and ethanol.
The electronic and steric effects of the substitutions of the C
3
- and C -aryl rings on the
5
required irradiation time of photooxidation reaction, concluded by comparison of the results
given in Table 1, indicate that the extent of these substitution effects (π-donor/π-acceptor
character of the C
depends on the location of the aryl substitution (C
In order to explain these effects in more details, photoreaction of a selected set of three 2-
pyrazolines (the parent compound 1a, and the compounds with para-NO substituted C - and
-aryl rings, respectively 1n and 1q) is followed by the UV-Vis spectroscopy. Comparative
slopes of the ln(A /A) at the corresponding appropriate wavelengths versus time, derived from
3
-aryl group and σ-donor/σ-acceptor character of the C
5
-aryl group) [25]
3
- or C -positions of the heterocyclic ring).
5
2
3
C
5
0
their UV-Vis reaction spectra (Fig. 4) represent the comparative substituent effect on the rate
of photooxidation reaction assuming first order kinetics. As mentioned in Table 1, the
irradiation time for complete conversion of 1h (30 Sec.) is very shorter than those of the other
compounds, and therefore, UV spectroscopic technique cannot follow the photooxidation of
this compound. The observed isosbestic points in the UV-Vis reaction spectra of the three
selected compounds shown in Fig. 4 indicate the occurrence of a clean reaction for conversion
of the reactant to the product, either without involvement of any intermediates or with
involvement of short-lived undetectable intermediates within UV-Vis measurement time.

12
1
0
.5
1
2
1
0
.5
2
1
a
y = 1.90×10-2 x - 0.0258
R² = 0.9877
.5
1
λ = 382
.5
0
.5
0
2
40
290
340
λ (nm)
390
440
0
0
0
10
20 30
Time (s)
40
50
2
1
0
.5
2
1
n
-4
y = 2.73×10 x + 0.0157
0
.16
R² = 0.9389
λ = 480
.5
1
0.12
0
0
.08
.04
0
.5
0
2
40
340
440
540
100
200
300
400
λ (nm)
Time (s)
0
.25
2
1
0
.5
1
q
y = 3.52 ×10-4 x - 0.0013
R² = 0.9935
0
.2
0.15
.1
2
.5
1
λ = 370
0
.5
0
0
.05
0
2
40
280
320
360
λ (nm)
400
440
200
400
600
Time (s)
Fig. 4. The UV-Vis reaction spectra of photooxidation of 1a (top), 1n (middle) and 1q
bottom) in chloroform and the corresponding ln(A /A)-time diagrams (right column) fitted
(
0
linearly to obtain the rate constants. All three spectra show hypsochromic shifts denoting
consumption of the starting material upon irradiation.
A comparative analysis of the ln(A
0
/A)-time diagrams (Fig. 4) indicates that the
consumption rate for compound 1a is higher than those of compounds 1n and 1q (compare the
−2
−4
−4
−1 −1
rate constants of 1.90×10 , 2.73×10 and 3.52×10 molL s obtained for 1a, 1n and 1q,
respectively). This trend is compatible with the observed experimental completion reaction
times of 10 for 1a vs. 90 and 70 min for 1n and 1q, respectively (Table 1) for the
maximum/complete photochemical conversions of these compounds. These results clearly

13
indicate that the electron-withdrawing nature of the para-nitro group on the C
3
- and C -aryl
5
substitutions (in 1n and 1q, respectively) increases the required irradiation time for completion
of their photochemical reaction due to their decreased tendency for electron transfer to solvent,
which will be more elaborated computationally later. The trend of these rate constants is
compatible with the trend observed and reported for the reaction completion (maximum
progression) times.
The comparative irradiation times of the photooxidation reaction observed for 1n and 1q
can be correlated with their UV-Vis absorption data given in Table 2. The electron-withdrawing
resonance effect of the para-nitro group on the C
charge on the Caryl-atom attached to C -atom, as demonstrated in the resonance structure 1n-RS
I in Scheme 3. While, in 1q with para-nitro group on the C -aryl ring, co-planarity of the C
aryl ring with the C =N double bond allows an effective delocalization of the N -lone pair
3
-aryl ring in compound 1n induces a positive
3
5
3
-
3
2
1
towards the positive charge or directly towards the para-nitro group, thus known as through
conjugation, as shown in the resonance structure 1n-RS II in Scheme 3. Owing to this effective
through conjugation, lower LUMO level and smaller HOMO‒LUMO energy difference are
expected. As a result, excitation energy for 1n is smaller and thus its UV-Vis spectrum shows
a bathochromic shift with respect to that of 1q. Considering higher intensity of the longer
wavelength radiations of the mercury lamp, a more effective excitation would thus be expected
for 1n which consequently would result in a faster photooxidation reaction. In contrast to this
expectation, experimental results (Table 1) indicate that 1n has the slowest photooxidation
reaction. Since all three resonance structures of 1n (1n, 1n-RS I & 1n-RS II) can absorb UV-
Vis light, the bathochromic shift in the UV-Vis spectrum of 1n (compared to that of 1q) can
be attributed more probably to the contribution of the zwitterionic resonance structure 1n-RS
II, which cannot donate an electron to the electron-willing solvent molecule, especially from
the positively charged N -atom. This is supported by observing no conversion of compound 1n
1

14
upon irradiation with visible light ( ≥ 400 nm) during the control experiment. This observation
along with the observed rates of photooxidation reactions, and the reported photooxidation of
some 2-pyrazolines by irradiation at λ > 400 nm [15], support the argument that more probable
zwitterionic resonance structure 1n-RS II with the positive charge on the N -atom, in spite of
1
having higher ability of light absorption, is not a suitable candidate to act as an electron-donor
species in the photoreaction. Therefore, irradiation time for compound 1n is actually longer.
Scheme 3. Through conjugation of the N
nitro group on the C -aryl ring (in compound 1n, Table 1) causes bathochromic shift in its
UV-Vis spectrum and decrease the electron-donating ability of the heterocyclic ring.
1
-lone pair towards the electron-withdrawing para-
3
25
20
15
10
5
0
1
a
1n
1q
240
290
340
390
440
490
540
590
λ (nm)

15
Fig. 5. Comparison of the experimental UV-Vis absorption spectra of 1n and 1q containing
respectively para-nitrophenyl group on the C
3
- and C -positions of the heterocyclic ring with
5
that of the parent compound 1a.
In contrast to the behavior of the electron-withdrawing character of the para-nitro group
in decreasing the electron-donating ability of the heterocyclic ring, which causes an increase
of the required irradiation time, the para-methoxy group on the C
3
-aryl ring (compound 1h)
decreases drastically the required irradiation time. Delocalization of the oxygen lone pair of
3
2
this group, by rehybridization of the oxygen atom from sp to sp , as illustrated in Scheme 4,
increases the electron density of the heterocyclic ring. Consequently, a shorter irradiation time
for compound 1h (30 sec) is observed.
Scheme 4. Electron-donating effect of the para-methoxy group on the C -aryl ring in
3
compound 1h (Table 1) increases the electron-donating ability of the heterocyclic ring.
Comparison of the irradiation times for chlorine containing compounds 1b, 1c and 1d
(having chlorine atoms on different positions of the C
3
-aryl ring) and 1e, 1f and 1g (having
chlorine atoms on different positions of the C -aryl ring) with those of the 2-pyrazolines having
5
methoxy or nitro groups on the same positions (Table 1) indicates that their photooxidations
are faster than those of the corresponding nitro-containing 2-pyrazolines and slower than those
of the corresponding methoxy derivatives. This trend is attributed to the lower electron-
donating ability of the chlorine atom compared to that of the methoxy group. Lower
electronegativity and weaker electron-donating resonance effect of the chlorine atom compared
to those of the oxygen atom of the methoxy group because of ineffective orbital overlap

16
between Cl 3p orbital and the Caryl 2p orbital compared to that between O 2p orbital and Caryl
2
p orbital, are the reasons of this observation.
These effects of the electron-withdrawing and electron-donating character of the
substituted aryl rings either on the C
3
- or on the C
5
-position of the heterocyclic ring (compared
to that of the unsubstituted parent compound 1a) on the electron-transfer induced oxidation of
2
-pyrazolines are also studied using cyclic voltammetric measurements [26]. The oxidation
+
peak potential of 1a in acetonitrile appears at 1.050 V vs. Fe/Fe . The π-acceptor or σ-acceptor
character of the para-nitrophenyl group on the C
the oxidation peak potential to 1.146 V or 1.097 V, respectively. While, the π- donating or σ-
donating character of the para-methoxyphenyl group on the C - or on the C -position, in 1h or
k, decreases the oxidation peak potential respectively to 0.747 and 0.905 V. All these
3
- or on the C -position, in 1n or 1q, increases
5
3
5
1
voltammetric data support the argument that electron-donating ability of the heterocyclic ring,
affected by the nature of the substituent, is the important characteristics determining the
photochemical reaction rate. In other words, electron transfer from the excited 2-pyrazoline
(
DHPZ*) to electron-acceptor solvent CHCl
3
, is affected by the extent of the electron-
- and C -aryl substitutions.
withdrawing and electron-donating characters of C
3
5

17
1
1
1
4
2
0
8
6
4
2
0
2
4
6
1a
1h
1k
1n
1q
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
-
-
-
Potential / V
Fig. 6. The voltammograms of selected 2-pyrazolines 1a, 1h, 1k, 1n & 1q (Table 1)
demonstrating the π- or σ-donating and π- or σ-accepting nature of their heterocyclic ring
affected by the methoxy and nitro substitutions on the C
3
- or C -aryl rings.
5
3
. Computational studies
Concerning the mechanism of photooxidation of 2-pyrazolines, some of works have not
mentioned any mechanism for this photochemical conversion [10,11]. The results of the
present study and another work [15], support the proposed mechanism illustrated in Scheme 2,
in which the electron transfer from photo-excited molecule to electron-accepting solvent is the
rate determining step forming the molecule radical cation intermediate. Recently, one work
reporting the photooxidation of 2-pyrazolines, suggests that the proton detachment from the 2-
•+
pyrazoline radical cation (DHPZ ), which results in the formation of hydropyrazolyl radical
•
(
HPZ ), is the rate determining step [29], because no correlation could be found between the

18
ionization potential of the investigated 2-pyrazolines with the obtained rate of reaction. In our
opinion, such correlation, which requires considering a direct electron transfer from the HOMO
of the molecule to the LUMO of the solvent molecule after ionization of molecule, is not
correct. Furthermore, results of the present experimental and theoretical photochemical studies
indicate that the electronic nature (electron-donating or electron-withdrawing) of the
substituted aryl rings on the C
3
- or C -positions affects the reaction completion (irradiation)
5
times and the oxidation peak potentials (Fig. 6). In order to address the above discussion on the
mechanism of the photooxidation reaction (Scheme 2) and to elucidate the steric and electronic
effects of the aryl substitutions on the C
required irradiation times for completion of the photooxidation reactions considered in the
present work, and also the key involvement of CHCl in the electron transfer process in the
3
- and C
5
-positions of the heterocyclic ring on the
3
present study, density functional theory (DFT) at the B3LYP/6-31++(d,p) level as implemented
in the Gaussian 09 package [30] is applied to optimize structures and to calculate the electronic,
bonding and spectroscopic characteristics of 2-pyrazolines (DHPZ), 2-pyrazoline radical
•
+
•
cations (DHPZ ), hydropyrazolyl radical intermediates (HPZ ), and the corresponding
pyrazoles (PZ). Some important structural parameters obtained in these computations are listed
in Table 3. A complete list of structural data are provided in Table S1 of the Supporting
Information.

1
9
Table 3
Characteristic bond lengths (Å ), dihedral angles (°) and bond angles of 4,5-dihydropyrazolines (DHPZ), 4,5-dihydropyrazoline radical cations
•+
•
(
DHPZ ), hydropyrazolyl radicals (HPZ ), and pyrazoles (PZ) obtained by B3LYP/6-311++G(d,p) computations.
DHPZ
DHPZ•+
HPZ•
PZ
Bond Lengths (Å)
-C -N
Bond angles (°)
Bond Lengths differences (Å)
N
1
-N
2
N
2
-C
3
C
3
-C
4
C
4
5
C
5
1
N
1
-C
6
C
3
-C12
C
5
-C18
C
5
-H
α
-177.1
-178.9
-179.0
-167.3
-177.8
-179.1
-178.8
-166.6
-177.9
-179.7
-179.1
-174.1
-178.0
-179.9
-179.6
-174.1
-178.0
-179.0
-179.6
-164.5
, Y = H. α = N
β
γ
θ
ΔBD
1
ΔBD
2
ΔBD
3
ΔBD
4
ΔBD
5
ΔBD
6
1
a
1.3666
1.3388
1.3821
1.3561
1.3728
1.3333
1.3875
1.3585
1.3656
1.3386
1.3808
1.3568
1.3494
1.3423
1.3656
1.3498
1.3690
1.3397
1.3906
1.3529
1.2905
1.3151
1.2983
1.3354
1.2900
1.3206
1.2975
1.3353
1.2912
1.3155
1.2990
1.3353
1.2964
1.3100
1.3139
1.3371
1.2896
1.3149
1.2954
1.3365
1.5155
1.5049
1.5106
1.4161
1.5155
1.5070
1.5110
1.4167
1.5153
1.5042
1.5115
1.4162
1.5149
1.5026
1.5107
1.4154
1.5162
1.5063
1.5074
1.4146
1.5536
1.5437
1.5041
1.3831
1.5531
1.5446
1.5040
1.3831
1.5535
1.5439
1.5042
1.3833
1.5538
1.5441
1.5045
1.3825
1.5542
1.5443
1.5056
1.3843
1.4822
1.5042
1.3878
1.3755
1.4819
1.5021
1.3882
1.3750
1.4824
1.5067
1.3903
1.3760
1.4860
1.5062
1.3672
1.3773
1.4783
1.5000
1.3690
1.3755
1.3933
1.3836
1.4199
1.4287
1.3924
1.3907
1.4175
1.4278
1.3917
1.3840
1.4183
1.4283
1.3974
1.3781
1.4320
1.4310
1.3959
1.3834
1.4288
1.4304
1.4640
1.4434
1.4613
1.4732
1.4625
1.4300
1.4596
1.4715
1.4637
1.4436
1.4609
1.4735
1.4540
1.4518
1.4413
1.4676
1.4642
1.4423
1.4622
1.4730
1.5213
1.5188
1.4283
1.4734
1.5211
1.5189
1.4279
1.4734
1.5204
1.5157
1.4287
1.4719
1.5210
1.5185
1.4363
1.4733
1.5202
1.5198
1.4148
1.4688
1.0938
1.0901
-7.4
-7.4
-47.2
-51.6
-9.5
-34.7
-34.7
-9.3
-44.6
-35.4
-42.3
-9.0
-44.1
-33.7
-45.1
-9.2
-45.2
-35.9
-43.7
-14.7
-45.4
-31.2
-38.0
-9.3
112.9
113.6
112.4
111.9
112.7
113.7
112.3
111.8
112.9
113.6
112.4
112.0
113.1
113.4
113.9
112.1
112.7
113.5
113.0
111.8
•+
1
a
- 0.0025
- 0.0022
- 0.0047
- 0.0025
- 0.0004
+ 0.0220
+ 0.0202
+ 0.0243
+ 0.0202
+ 0.0217
- 0.0099
- 0.0085
- 0.0096
- 0.0097
- 0.0099
•
1
a
-
-
- 0.0930
- 0.0932
- 0.0917
- 0.0847
- 0.1054
- 0.0944
- 0.0937
- 0.0921
- 0.1188
- 0.0495
- 0.0491
- 0.0493
- 0.0493
- 0.0486
2
a
1
h
1.0945
1.0901
•+
1
h
-4.5
•
1
h
-
-
-45.6
-50.9
-6.5
2
1
h
k
1.0936
1.0903
•+
1
k
-5.0
•
1
k
-
-
-46.0
-50.5
-4.5
2
1
k
n
1.0923
1.0901
•+
1
n
-4.2
•
1
n
-
-
-56.0
-52.6
-11.1
-3.9
-58.0
-54.4
2
1
n
q
1.0941
1.0897
•+
1
q
•
1
q
-
-
- 0.1093
2
q
-38.8
1
a: X = H, Y = H; 1h: X = H, Y = 4-MeO; 1k: X = 4-MeO, Y = H; 1n, X = H, Y = 4-NO2; 1q: X = 4-NO
2
2
-C
3
-C12-C13, β = N
2
-N
1
-C
6
-C11, γ = N
5
1
-C
5
-C18-C19, θ = N
2
-N
1
-C
5
.
•+
•
•+
•
•+
•
ΔBD
1
= (C
5
-C18) DHPZ ‒ DHPZ; ΔBD
2
= (C
5
-C18) HPZ ‒ DHPZ; ΔBD
3
= (C
5
-N
1
) DHPZ ‒ DHPZ; ΔBD
4
= (C
5
-N
1
) HPZ ‒ DHPZ; ΔBD
5
= (C
4
-C
) DHPZ ‒ DHPZ; ΔBD
6
= (C
4
-C
5
) HPZ ‒ DHPZ.

20
Analysis of these data (Table 3) indicates that the C
3
-aryl ring in DHPZ with the interring
=N
dihedral angle α (N -C -C12-C13) around 177-178° is co-planarly orientated toward the C
2
3
3
2
double bond of the heterocyclic ring, which allows an effective electron delocalization in this
π-conjugated system (as hetero-styrene type). In this optimized structure, the lone pair on the
N
1
-atom can be cross-conjugated with both N
the delocalization of the N lone pair towards its attached phenyl ring, the corresponding
interring dihedral angle β (N -N -C -C11) is around 4.5-11.1°. The steric hindrance between
the N -phenyl and C -aryl rings forces the C -aryl ring to deviate from co-planarity with the
heterocyclic ring, therefore, the corresponding interring dihedral angle γ (N -C
around 31.2-35.4°. The N -N -C bond angle labeled as θ in all investigated compounds is
between 111.8°-113.9°, which allows the N
delocalization of the N lone pair towards the N
Analysis of these data indicates also that the electron removal in the key step of reaction is
2
=C
3
double bond and N
1
-phenyl ring. Owing to
1
2
1
6
1
5
5
1
5
-C18-C19) is
2
1
5
3
/2
1
-atom having a hybridization of ~sp and possible
1
2
=C double bond and also N -phenyl ring.
3
1
occurred preferably from the lone pair in ~sp3/2 hybridized orbital on the N
-atom rather than
1
2
from the sp hybridized orbital on the N
2
-atom. This process leads to the formation of 2-
-atom. The radical center in DHPZ•+
intermediate (as representative 1 ) can be stabilized via displacement over the N ‒N ‒C
atoms, as a heteroallyl system (1 -RS I) and possibly further delocalization over the C -aryl
-phenyl ring (1 -RS III)
, C ‒C bond
bond lengths are observed in comparison to those
•+
pyrazoline radical cation (DHPZ ) intermediate on the N
1
•+
1
2
3
•
+
3
•
+
•+
ring, as a benzylic radical (1 -RS II) or delocalization over the N
illustrated in Scheme 5. Consequently, the shortening of the N ‒N
lengths, and increasing the N ‒C and C ‒N
in neutral DHPZ molecule. In addition to these changes, the shortening of C
bond lengths are observed. The magnitude of some of these changes of C
bond lengths are denoted as ΔBD , ΔBD and ΔBD
show also that the orientations of the N -phenyl, C - and C
1
1
2
3
‒C12 and N
1
6
2
3
5
1
5
‒C18 and C
4
‒C
5
5
‒C18, N
1
‒C
5
and C
4
‒
C
5
1
3
5
, and mentioned in Table 3. These data
-aryl rings towards the heterocyclic
1
3
5

21
•+
ring in DHPZ intermediate did not change effectively, which are confirmed by calculated
interring dihedral angles β, α and γ, respectively. The occurrence of the positive charge on the
•+
N -atom in the radical cation intermediate (DHPZ ) is the driving force for the ejection of a
1
proton from its neighborhood in the α-position, namely, C -H under formation of
5
•
hydropyrazolyl radical intermediate (HPZ ). The occurrence of the radical center on the C
5
-
atom in HPZ intermediate, requires rehybridization of the sp orbital of the C
-atom to sp2,
•
3
5
which allows possible and effective stabilization of the radical center by interaction with its
attached groups in the α-position and is rationalized in term of one electron molecular orbital
(
OEMO) theory [31]. The important factors for the stability of the radical center in HPZ•
intermediate can be:
i) delocalization of the single electron (SOMO) on the C
aryl ring as being a benzylic radical, in which the extent of the radical stabilization depends on
the nature and position (para/meta/ortho) of the additional substitution on the C -aryl ring and
also the parallel orientation of the aryl ring π orbitals with the SOMO of the radical center;
ii) interaction of the single electron with the non-bonding electron pair (NBMO) on the N
atom of the pyrazoline ring as a donor-substituted radical center; and finally,
iii) hyper-conjugation of the single electron with the C ‒H σ-bonds.
(
5
atom (benzylic carbon) over the C -
5
5
(
1
-
(
4

22
•+
Scheme 5. Single electron delocalization in the radical cation intermediate (1 ) towards the
•+
•+
•+
N
2
=C
3
double bond (1 -RS I), C
3
-aryl ring (1 -RS II) and N -phenyl ring (1 -RS III).
1
Concerning the interactions mentioned above for the stabilization of the radical center, the
changes of the bond lengths between the C
5
-atom and the N
1
-, C - and C18- atoms, from DHPZ
4
•
molecule to HPZ intermediate, elucidate the preference and extent of these interactions.
Comparison of the extent of the changes of C
5
‒C18, N
1
‒C
and ΔBD
-aryl ring and N
moiety in position 4 of the heterocyclic ring.
Stabilization of the carbon radicals by each of the above-mentioned factors, results in more
5
and C
given in Table 3 indicates that the
-atom than hyper-
4
‒C
5
bond lengths from DHPZ
•
to HPZ intermediate labeled as ΔBD
2
, ΔBD
4
6
single electron has stronger interaction with the C
conjugation with CH
5
1
2
•
+
favored proton detachment from C
5
-position of the radical cation intermediate (DHPZ )
•
forming hydropyrazolyl radical intermediate (HPZ ). Keeping this in mind, owing to the

23
•
various possible interactions of the radical center in the HPZ intermediate with its attached N
atom or groups in its α-position, the radical center can be considered as a donor-substituted
radical because of the interaction of its SOMO with the lone pair on the N -atom. Also, the
extent of the radical delocalization over the C -aryl ring (as benzylic radical), reflected in the
negative value of ΔBD (i.e. shortening of the C ‒C18 bond length, Table 3), depends on the
interaction of the SOMO with the LUMO or HOMO of the C -aryl group. Comparison of the
ΔBD values for compound 1k containing the p-methoxyphenyl group on the C -position
-0.0917 Å) and compound 1q containing the p-nitrophenyl group (-0.1054 Å) with those of
1
-
1
5
2
5
5
2
5
(
the parent compound 1a (-0.0930 Å), indicates that SOMO-LUMO interaction (in 1q) is more
favored than SOMO-HOMO interaction (in 1k).
Simultaneous attachment of the radical center with the electron-withdrawing substituent
(
capto) and electron-donating atom or group (dative) is assigned as a capto-dative stabilized
radical center [32]. The concept of capto-dative radical stabilization, in which the synergistic
effect of acceptor and donor substituents on the same radical center results in an enhanced
stabilization over the sum of the individual donor- and acceptor-substituted radical center, has
been developed and experimentally verified by Viehe [33]. The theoretical evaluation of the
synergetic capto-dative stabilization of free radicals is calculated and reported by Schleyer et
al [34]. An important point supporting the proposed reaction mechanism (Scheme 2) is the ease
•+
of the proton removal from DHPZ intermediate by the base present in the medium, as a fast
reaction step. Concerning the effective capto-dative stabilization of the radical center in 1q•,
having the para-nitrophenyl group on the C -position, it is expected that if the proton
5
detachment was the rate determining step, therefore, the photooxidation of 1q to its
photoproduct 2q would be the fastest photooxidation reaction among the 2-pyrazolines
considered in the present study. This is while the experimental rate of the photooxidation

24
reaction of 1q is slower (i.e. its completion time is longer) than all of these 2-pyrazolines except
for 1n. Therefore, the proton detachment is not the rate determining step in our study.
Electron-transfer induced reductive cleavage of the carbon-halogen bond in halogen-
containing compounds, such as aryl and alkyl halides including CCl , is studied
4
electrochemically and photochemically [35-38]. Based on the results of these studies, two types
of mechanisms are proposed for the carbon-halogen bond cleavage; (i) a stepwise mechanism
in which the electron transfer to halogen containing compound and bond breaking are
successive and (ii) the concerted electron transfer-bond breaking process. According to the
results of these studies [35-38] and what obtained in the present work, it is suggested that upon
electron transfer from the photo-excited DHPZ (MH
2
) to the CHCl
solvent, besides DHPZ•+
3
intermediate either the formation of the chloroform radical anion (CHCl
•−), as illustrated in
3
−
Scheme 2, the formation of the pair of the chloride ion (Cl ) and dichloromethyl radical
CHCl •
), or the formation of the complex [Cl ······ H‒CCl
−
•] can be occurred (Scheme 6).
(
2
2
•+
Further, proton detachment by either of these species from the DHPZ intermediate results in
the formation of HPZ•.
Scheme 6. Three possible ways in the electron transfer process from the photo-excited DHPZ
molecule (MH
2
) to the chloroform (CHCl ) solvent.
3

25
The (DFT) B3LYP/6-311++g(d,p) optimization of the structure of chloroform radical
anion (CHCl •
−) starting from the optimized structure of its neutral CHCl
molecule (Fig. 7) by
•] radical anion
3
3
−
adding an electron, resulted only in the formation of the [Cl ······ H‒CCl
2
complex intermediate. Therefore, the electron transfer process and (the immediate) reductive
cleavage of the carbon-chlorine bond in chloroform (as a halogen-containing compound) result
−
•] radical anion complex, as shown in Fig. 7, rather
in the formation of the [Cl ······ H‒CCl
2
•
−
−
•] radical anion
•)
intermediate forming MH•,
than the simple chloroform radical anion CHCl
3
. The [Cl ······ H‒CCl
2
−
complex is ready to lose its chloride ion (Cl ) and form the dichloromethyl radical (H‒CCl
2
•+
in the next step of the reaction or to abstract a proton from MH
2
•). The H‒CCl
•
radical intermediate is stabilized by
HCl and dichloromethyl radical (H‒CCl
2
2
interaction of its single electron with the lone pairs on the chlorine atoms. The shortening of
the remaining C‒Cl bond lengths and increasing of Cl‒C‒Cl bond angel are evidents for this
interaction.
−
Fig. 7. DFT optimized structures of CHCl
CCl •
breaking as a result of electron gaining.
3
(left) and its radical anion complex [Cl ······ H‒
2
] (right) in chloroform (CPCM model solvent) showing the carbon-chlorine bond
The trend observed for the relative rates of photooxidation reaction of 2-pyrazolines in
CHCl solvent can be explained by comparative analysis of the frontier orbital energies of these
molecules and the solvent. The B3LYP/6-311++G(d,p) computational results in chloroform
3

26
CPCM solvent are used to obtain the HOMO (S
0
) energies and the experimental UV-Vis
, S , …) energies are
excitations are used to obtain the energy gaps. Then, the LUMO (S
1
2
calculated for these 2-pyrazolines in chloroform (EHOMO + energy gap = ELUMO), and the
corresponding orbital energy diagrams are plotted in Fig. 8. Based on the proposed reaction
mechanism (Scheme 2), the first step in the photochemical oxidation is the excitation of 2-
pyrazoline molecule (MH
absorbing the mercury lamp radiation. In the second step, this excited electron is transferred to
the low laying vacant orbital(s) of the solvent, i.e. the LUMO of CHCl (S), which results in
•+) and chloroform radical
] or S (Fig. 7). The prerequisite for this electron transfer (ET)
is that the energy level of the electron-donating orbital(s) of the molecule E (LUMOs) should
be higher than that of the electron-accepting orbital(s) of the solvent E (LUMOs), i.e. ΔEMS
(LUMOs) – E (LUMOs) > 0. Note that chloroform (with λmax = 250 nm, 260 nm & 280 nm)
has no efficient absorption under experimental conditions (λ ≥ 314 nm). Furthermore, solvent
excitation (if any) would not lead to photochemical reaction via ET from S to MH , since
(LUMOs) is lower than E (LUMOs) for all 2-pyrazolines.
2
) by promotion of electron from its HOMO to its LUMO by
3
•+
the occurrence of the 2-pyrazoline radical cation (DHPZ , or MH
2
−
•
•−
anion complex [Cl ······H‒CCl
2
M
S
=
E
M
S
2
E
S
M

27
0
1
2
3
4
5
6
S3
S3
S3
S2
S2
-
-
-
-
-
-
T2
T
T
2
1
2
S2
S1
S
S
2
L
L
S1
L
T
S
L
2
S1
1
-1.7040
-
1.8450
-1.8264
-
1.9416
T2
S1
L
-
2.8279
T1
T1
T1
T1
T1
H
E =
HOMO
−8.7219
H
S0
H
S0
S0
H
-5.2068
H
S0
-
5.3350
-5.2998
S0
-5.4844
-
5.6083
1
a
1h
1k
1n
1q
CHCl
3
Fig. 8. Frontier molecular orbitals (FMO) engaging in the photochemical molecule →
solvent (MH * → S) electron transfer (ET) in the photooxidation reaction of representative
-pyrazolines (Table 1) in CHCl solvent. The B3LYP/6-311++G(d,p) HOMO energies, and
2
2
3
the corresponding LUMO energies calculated from the experimental UV-Vis transitions are
plotted in bold solid lines. The dashed vertical arrows indicate the mercury lamp maximum
energy (3.95 eV) available for transitions from HOMO to higher excited states. The dotted
lines show the triplet state(s) of the compounds obtained by TD-B3LYP/6-311++G(d,p)
computations.
Analysis of the orbital energies demonstrated in Fig. 8 shows that the ΔEMS values follow
the order 1h > 1k > 1a > 1q > 1n, which is consistent with the corresponding reaction
completion (irradiation) times of the photooxidation reaction reported in Table 1. Furthermore,
the reaction completion times observed for the photooxidation reaction is very long for the C
and C -nitro-substituted compounds 1n and 1q, respectively. These exceptional required longer
irradiation times for completion of the reaction can be attributed to their less effective
tunneling) electron transfer process due to smaller ΔEMS, because of the presence of electron-
withdrawing NO substituent. In addition, possible back electron transfer (BET) process, which
3
-
5
(
2

28
is more effective for these two compounds due to their larger E
S
(LUMO) – E (HOMO)
M
•+
•+
differences, and instability of their formed radical cation intermediates (1n and 1q ) can also
be considered to justify their smaller photooxidation reaction rates. Therefore, the BET process
is more possible in competition with the proton removal process (Scheme 2) for these two
compounds. These trends arise from their significantly different HOMO and LUMO orbital
energies of 1n and 1q compared to those of the other 2-pyrazolines. The exceptionally longer
required irradiation time for photooxidation reaction of 1n has already been explained above
in terms of possible involvement of the resonance structures in the ionized radical cationic state
for this molecule (Scheme 3). The ET and BET processes are demonstrated schematically in
Fig. 9 (See Fig. S1 for more details).
1
h
1h*
S
1h•+
S•−
1n•+
S
1n*
1n
Fig. 9. Schematic diagram showing transition probabilities at the corresponding wavelengths
and their comparative efficiencies of electron transfer (ET) and back electron transfer (BET)
processes of representative 2-pyrazolines 1h and 1n. The widths of the red and blue arrows
show the efficiencies of the ET and BET processes, respectively. The MH
2
*S tunneling
ET process may occur via singlet or triplet pathways as discussed in the text later.

29
A similar orbital energy diagram, as shown in Fig. 8, in which the orbital energies of all
DHPZs including LUMO energies are obtained computationally, is presented in Fig. S2 of the
Supporting Information. Based on this diagram, it is expected that the photooxidation reaction
would proceed only for compound 1h, since the prerequisite energy criterion ΔEMS > 0 is
satisfied only for this compound. This is not consistent with experimental photooxidation
reaction completion times reported in Table 1. This disagreement between computational and
experimental observations is quite natural due to the gas-to-liquid shifts in the orbital energies
due to interaction with explicit solvent molecules, which are not considered in our calculations
carried out with the CPCM model solvent. It can be concluded that this gas-to-liquid shift
would result in positive ΔEMS = E
M
(LUMO) – E
S
(LUMO) values for all of the 2-pyrazolines,
while maintaining order of their relative E
M
(LUMO) values such that the comparative
computational HOMO and LUMO energies (Fig. S2) are relatively compatible with
corresponding experimental energies (Fig. 8).
The comparative solvent effect on the photooxidation reaction completion times of the
representative 2-pyrazoline 1a discussed earlier (Fig. 3) can also be explained based on the
relative values of ΔEMS = E
M
(LUMO) – E (LUMO). This effect is demonstrated in Fig. 10 for
S
the non-substituted parent compound 1a in three different solvents, acetonitrile, chloroform
and ethanol (Their UV spectra are presented in Fig. 3). It can be clearly seen in Fig. 10 that the
ΔEMS values for these three solvents are significantly different and follows the order
ΔEMS(chloroform) > ΔEMS(acetonitrile) >> ΔEMS(ethanol), which is compatible with the
observed photooxidation reaction completion times for compound 1a reported in the footnote
of Table 1. Furthermore, it is expected that the nature of solvent weakly influences the HOMO
and LUMO energies of this 2-pyrazoline compounds, which can thus slightly affect the ET
process. Note that the ΔEMS(ethanol) value is negative and thus theoretically no photochemical
reaction is expected to occur in this solvent.