DFT calculations of 2,4,6-trinitrophenylbenzoate derivatives: Structure, ground state properties and spectral properties

Article abstract of DOI:10.1016/j.molstruc.2011.09.024

2,4,6-Trinitrophenylbenzoate derivatives 3(a-g) were synthesised and their optimum molecular geometries, ground state properties, IR, UV and NMR spectra were calculated using the B3LYP method with 6-31G Pople basis set. The predicted geometry showed that the aryl-oxygen bond is stronger than acyl-oxygen bond. Bond lengths are affected by the presence of electron donating or electron withdrawing groups in benzoyl moiety. The orientation of the two aryl rings with respect to each other is neither in the same plane nor perpendicular. Hammett plot showed good linearity between substituent constant and the HOMO as well as LUMO energies. Mulliken and NBO charges indicate that the carbonyl carbon (C7) is more positively charged than the ipso carbon (C10), while the inspection of the coefficient on the carbonyl carbon and the ipso carbon atom for LUMO showed that its value is greater for C10 compared to C7. The correlation between the experimental and the theoretical results indicates that density functional theory B3LYP method is able to provide satisfactory results for predicting IR, UV and NMR properties.

Full text of DOI:10.1016/j.molstruc.2011.09.024

Journal of Molecular Structure 1006 (2011) 303–311
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
DFT calculations of 2,4,6-trinitrophenylbenzoate derivatives: Structure, ground
state properties and spectral properties
⇑
Mahmoud F. Ibrahim, S.A. Senior, Mohamed A. El-atawy , Samir K. El-Sadany, Ezzat A. Hamed
Chemistry Department, Faculty of Science, Alexandria University, P.O. 426 Ibrahemia, Alexandria 21321, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 July 2011
Received in revised form 12 September 2011
Accepted 12 September 2011
Available online 22 September 2011
2,4,6-Trinitrophenylbenzoate derivatives 3(a–g) were synthesised and their optimum molecular geome-
tries, ground state properties, IR, UV and NMR spectra were calculated using the B3LYP method with 6-
31Gꢀꢀ Pople basis set. The predicted geometry showed that the aryl-oxygen bond is stronger than acyl-
oxygen bond. Bond lengths are affected by the presence of electron donating or electron withdrawing
groups in benzoyl moiety. The orientation of the two aryl rings with respect to each other is neither in
the same plane nor perpendicular. Hammett plot showed good linearity between substituent constant
and the HOMO as well as LUMO energies. Mulliken and NBO charges indicate that the carbonyl carbon
(C7) is more positively charged than the ipso carbon (C10), while the inspection of the coefficient on
the carbonyl carbon and the ipso carbon atom for LUMO showed that its value is greater for C10 com-
pared to C7. The correlation between the experimental and the theoretical results indicates that density
functional theory B3LYP method is able to provide satisfactory results for predicting IR, UV and NMR
properties.
Keywords:
2,4,6-Trinitrophenyl benzoate
DFT calculations
Regioselectivity
Picric acid
Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.
1. Introduction
Recently, quantum chemical calculations have been proven to
be an important tool for prediction of the relative electron den-
Esters have been studied extensively due to its important role in
biological, chemical, environmental industrial processes and many
natural molecules [1–4]. Ester group is also used as a molecular
segment in designing ferroelectric and antiferroelectric liquid crys-
tals (FLC and AFLC) [5]. As a result, esters formation and degrada-
tion processes have been the subject of extensive experimental and
theoretical studies [2–6]. It has long been known that either the
acyl-oxygen bond or aryl-oxygen bond of phenyl esters can be
cleaved depending on: (i) the structure of the ester [6,7] (ii) the
nature of reagent [8,9] (iii) the basicity of the leaving group anion
compared to that of the attacking nucleophile, (iv) the nature of
substituent in the aryl or the acyl group containing the ester
[8,9] and (v) the relative ‘‘hardness’’ and ‘‘softness’’ of the reaction
site and reagent. Competition between the two electrophilic cen-
ters in the ester substrate is well documented for reactions of
nitrophenyl sulfonates, sulfates and phosphates with nucleophiles
[10–12]. On the other hand, most carboxylic esters undergo
acyl-oxygen scission [10,13]. Simultaneous acyl and aryl-oxygen
scission were also observed for the methanolysis of picryl benzoate
under neutral conditions [7,14]. The modes of scission and the rate
constants for Ar (kAr) relative to CO (kco) were owing to the strong
activation of the 1-position (ipso) of the phenolic group [8].
sities on each reaction site, investigation of the relationship be-
tween structures and spectral properties of the organic
molecules and for the interpretation of experimental data arising
from industrial interest and applications. The density functional
theory (DFT) method, which involves electron correlation effects,
has proved suitable for calculating the energy-minimized struc-
ture and spectroscopic properties of phenyl benzoate derivatives
[15,16]. In addition, the time-dependent density functional the-
ory (TD-DFT) method has also been shown to be able to give
good results in calculations of the electronic absorption spectra
of these compounds. DFT calculations were performed in order
to determine structure of 2,4,6-trinitrophenylbenzoate deriva-
tives (3a–g), and to compare the calculated results with the
experimental data. Such calculations may be used to predict
the regioselectivity of these esters as precursors to study nucle-
ophilic substitution reactions involving acyl-oxygen versus aryl-
oxygen scissions.
2. Experimental
The UV spectra were recorded in methanol solution (C = 10ꢁ4 M)
on a 160-A UV–VIS Shimadzu spectrophotometer. The IR spectra
were recorded by FT-IR spectrometer, the 1H NMR spectra were car-
ried out at ambient temperature (ꢂ25 °C) on a (JEOL) 500 MHz
spectrophotometer using tetramethylsilane (TMS) as an internal
standard.
⇑
Corresponding author.
E-mail address: mohamed_elatawy@yahoo.com (M.A. El-atawy).
0022-2860/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.09.024

304
M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
0
0
2.1. Synthesis of 2,4,6-trinitrophenylbenzoate derivatives (3a–g)
H
3 ,5 -picryl), 8.09 (d, J = 8.4, 2H, H2,6-aroyl), 7.55 (d, J = 8.4, 2H,
H3,5-aroyl). C13H6ClN3O8: requires: C, 42.47; H, 1.64; N, 11.43%.
A solution of picric acid (1) (1 g, 43 mmole) and the respective
acid chloride (benzoyl chloride (2a), p-anisoyl chloride (2b), p-
toluoyl chloride (2c), m-toluoyl chloride (2d), p-chlorobenzoyl chlo-
ride (2e), m-chlorobenzoyl chloride (2f), p-nitrobenzoyl chloride
(2g)) (0.62 g, 44 mmole) in (10 ml) chloroform was refluxed for 3 h
[6]. Work up afforded a precipitate which was filtered and crystal-
lized from chloroform–ether mixture as a crystal. The purity was
checked by TLC (1:2 ethyl acetate:n-hexane).
Found: C, 42.79; H, 1.63; N, 11.64%.
2.7. 2,4,6-Trinitrophenyl m-chlorobenzoate (3f)
Yield: 74%; m.p: 138 °C; UV (MeOH): kmax = 230 nm (
e
= 20,470),
IR (KBr): 1762 cmꢁ1 (C@O), 1547, 1346 cmꢁ1 (NO2 asym and sym.
respectively). 1H NMR [CD3Cl]: d 9.20 (s, 2H, H3 ,5 -picryl), 8.14 (s,
1H, H2-Aroyl), 8.06 (dd, 1H, H6-aroyl), 7.72 (dd, 1H, H4-aroyl),
7.53 (t, 1H, H5-aroyl). C13H6ClN3O8 requires: C, 42.47; H, 1.64; N,
11.43% .Found: C, 42.35; H, 1.45; N, 11.25%.
0
0
2.2. 2,4,6-Trinitrophenyl benzoate (3a)
Yield: 75%; m.p: 159 °C; UV (MeOH): kmax = 234 nm
2.8. 2,4,6-Trinitrophenyl p-nitrobenzoate (3g)
(e
= 18,910), IR (KBr): 1758 cmꢁ1 (C@O), 1545, 1341 cmꢁ1 (NO2
asym and sym. respectively). 1H NMR [(CD3Cl)]: d 9.18 (s, 2H,
Yield: 76%; m.p: 184 °C; UV (MeOH): kmax = 250 nm (e = 19,250),
0
0
H
3 ,5 -picryl), 8.16 (d, J = 7.6, 2H, H2,6-benzoyl), 7.75 (t, J = 7.6, 1H,
IR (KBr): 1767 cmꢁ1 (C@O), 1548, 1349 cmꢁ1 (NO2 asym and sym.
respectively). 1H NMR [(CD3)2SO]: d 8.55 (s, 2H, H3 ,5 -picryl), 8.12
(d, J = 8.4, 2H, H2,6-aroyl), 8.28 (d, J = 8.4, 2H, H3,5-aroyl).
0
0
H4-benzoyl), 7.57 (t, J = 7.6, 2H, H3,5-benzoyl). C13H7N3O8: re-
quires: C, 46.86; H, 2.12; N, 12.61%. Found: C, 46.96; H, 2.07; N,
12.95%.
C13H6N4O10: requires: C, 41.28; H, 1.60; N, 14.81%.Found: C,
41.28; H, 1.55; N, 15.08%.
2.3. 2,4,6-Trinitrophenyl p-anisate (3b)
3. Calculations
Yield: 68%; m.p: 140 °C; UV (MeOH): kmax = 260 nm (e = 23,670),
IR (KBr): 1745 cmꢁ1 (C@O), 1544, 1344 cmꢁ1 (NO2 asym and sym.
All calculations were performed using the Gaussian 98 package
[17]. A conformational search for obtaining the most stable con-
former was done using the semi-empirical PM3 method, the most
stable conformer was subjected to full geometrical optimizations
using the DFT and Becke’s three-parameter hybrid exchange func-
tional in combination with the gradient-corrected correlation func-
tional of Lee, Yang and Parr B3LYP/6-31Gꢀꢀ method [18–20]
without any constraints to calculations. All calculated structures
were found to be true minima, i.e., no imaginary frequencies were
observed. After the ground state geometry optimization, the exci-
tation spectrum of each molecule has been computed with TD-
DFT [21] at the 6-31Gꢀꢀ level. 1H NMR chemical shifts of the previ-
ously optimized 2,4,6-trinitrophenyl benzoate derivatives 3(a–g)
have been calculated by using gauge-including atomic orbital
GIAO/B3LYP [22–24] density functional method with 6-31Gꢀꢀ basis
set. To evaluate the relative chemical shifts, the tetramethylsilane
(TMS) shielding constants calculated at B3LYP/6-31Gꢀꢀ method.
respectively). 1H NMR [(CD3Cl)]: d 9.15 (s, 2H, H3 ,5 -picryl),8.11
(d, J = 8.4, 2H, H2,6-anisoyl), 7.02 (d, J = 9.1, 2H, H3,5-anisoyl), 3.92
(s, 3H, OCH3), MS (M = 363). C14H9N3O9: requires: C, 46.29; H,
2.50; N, 11.57%. Found: C, 45.93; H, 2.24; N, 11.32%.
0
0
2.4. 2,4,6-Trinitrophenyl p-toluate (3c)
Yield: 67%; m.p: 166 °C; UV (MeOH): kmax = 240 nm
(e
= 22,040), IR (KBr): 1749 cmꢁ1 (C@O), 1544, 1341 cmꢁ1 (NO2
asym and sym. respectively). 1H NMR [(CD3Cl)]: d 9.16 (s, 2H,
0
0
H
3 ,5 -picryl), 8.04 (d, J = 8.4, 2H, H2,6-toluoyl), 7.36 (d, J = 7.6, 2H,
H3,5-toluoyl), 2.48 (s, 3H, CH3), MS (M = 347). C14H9N3O8: requires:
C, 48.43; H, 2.61; N, 12.10%. Found: C, 48.15; H, 2.61; N, 12.39%.
2.5. 2,4,6-Trinitrophenyl m-toluate (3d)
Yield: 67%; m.p: 148 °C; UV (MeOH): kmax = 235 nm
(e
= 23,950), IR (KBr): 1752 cmꢁ1 (C@O), 1544, 1341 cmꢁ1 (NO2
4. Results and discussion
asym and sym. respectively). 1H NMR [CD3Cl]: d 9.17 (s, 2H,
0
0
The reaction of equivalent amounts of picric acid and aroyl
chlorides in chloroform gave 2,4,6-trinitrophenyl 4- or 3-
substituted benzoates (3a–g) as indicated in Scheme 1. The purity
of the compounds was confirmed from their elemental analysis
and spectral measurements.
H
3 ,5 -picryl), 7.96 (m, 2H, H2,4-toluoyl), 7.55 (d, J = 7.6, 1H, H6-tol-
uoyl), 7.45 (t, J = 7.6, 1H, H5-toluoyl), 2.46 (s, 3H, CH3). C14H9N3O8:
requires: C, 48.43; H, 2.61; N, 12.10%.Found: C, 48.63; H 2.72; N
12.23%.
2.6. 2,4,6-Trinitrophenyl p-chlorobenzoate (3e)
5. Molecular structure and optimized geometry
Yield: 78%; m.p: 166 °C; UV (MeOH): kmax = 240 nm
(e
= 22,430), IR (KBr): 1752 cmꢁ1 (C@O), 1538, 1338 cmꢁ1 (NO2
The optimized geometrical parameters (bond length, angles and
dihedralangles) are listed in Table 1. The numbering of atoms for the
asym and sym. respectively). 1H NMR [CD3Cl]: d 9.18 (s, 2H,
O2N
HO
O2N
O2N
O
C Cl
O
-HCl
NO2
C O
NO2
X
X
O2N
3-or 4--Substituted benzoyl chloride
2(a-g)
Picryl 3- or 4-substituted benzoate
3(a-g)
Picric acid
(1)
a,X=H; b,X=4-OCH3;c,X=4-CH3;d,X=3-CH3; e,X=4-Cl;f,X=3-Cl; g, X=4-NO2
Scheme 1. Synthesis of 2,4,6-trinitrophenyl benzoate derivatives 3a–g.

M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
305
Fig. 1. The optimized structures and atom labeling of compounds 3a–g.
optimized geometries of 2,4,6-trinitrophenylbenzoates 3(a–g) are
shown in Fig. 1. To the best of our knowledge, only the unsubstituted
ester (3a) is previously prepared [7] and all the experimental data on
geometric structure are not available in the literature.
shows a small increase in bond distance. While electron withdraw-
ing groups such as 3-Cl, 4-NO2 showed a slight decrease in bond
distance. The 4-Cl substituent is expected to behave like the other
electron withdrawing groups; however, it has no effect on the
bond distance of carbonyl group. This is attributed to the opposite
polarization effect of halogen atoms (i.e. inductive opposes
resonance).
Table 1 indicates that the effect of the substituents is reversed
about C1AC7 bond length. The presence of electron donating
groups such as (4-OCH3, 4-CH3, 3-CH3) shorten the bond distance,
While the presence of electron withdrawing groups (3-Cl, 4-NO2)
lengthen the bond distance. As expected, the 4-chloro substituent
deviates from the behavior of the other electron withdrawing
groups and shorten the distance presumably due to its opposite
polarization effect.
Table. 1. Shows that the acyl-oxygen bond length (C7AO9) var-
ies between 1.398 and 1.417 Å while the aryl-oxygen bond length
(C10AO9) varies between 1.347 and 1.355 Å depending on the sub-
stituent at the acyl moiety. The shortening of the aryl-oxygen bond
relative to the acyl-oxygen bond can be attributed to the large con-
tribution of the lone pair of O9 with the trinitro groups of the picryl
ring in resonance. As a result, the aryl-oxygen scission is expected
to be more difficult and require higher activation energy compared
to the acyl-oxygen scssion when these compounds subjected to
nucleophilic reactions. The C7AO8 is the carbonyl group bond
length which was calculated to be (1.204 Å) for the unsubstituted
ester (3a). In general, the presence of moderate or weak electron
donating groups such as (4-CH3, 3-CH3) have no effect on C7AO8
bond length, whereas strong electron donating group (4-OCH3)
The obvious effect of the 4-OCH3 and 4-Cl substituents in bond
lengthening of C7AO8 and bond shortening of C1AC7 is due to the
cross conjugation of the lone pair of this group or atom with the

306
M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
Table 1
Some structure parameters of optimized geometry of 2,4,6-trinitrophenylbenzoate derivatives (3a–g) determined by B3LYP/6-31Gꢀꢀ method.
Structural parameters
H
4-OCH3
4-CH3
3-CH3
4-Cl
3-Cl
4-NO2
Ra
C7AO9
O9AC10
C7AO8
C1AC7
C11AN22
C13AN19
C15AN16
1.407
1.353
1.204
1.473
1.478
1.475
1.478
1.417
1.347
1.205
1.462
1.477
1.473
1.477
1.410
1.352
1.204
1.469
1.477
1.474
1.477
1.409
1.352
1.204
1.472
1.478
1.475
1.477
1.407
1.351
1.204
1.472
1.478
1.474
1.478
1.404
1.352
1.203
1.476
1.478
1.475
1.478
1.398
1.355
1.203
1.480
1.478
1.476
1.479
Aa
C1AC7AO9
C1AC7AO8
O8AC7AO9
C7AO9AC10
O9AC10AC11
O9AC10AC15
C11AC10AC15
111.46
127.54
121.00
118.54
118.80
124.11
117.01
110.79
128.18
121.01
119.82
118.83
124.44
116.73
111.42
127.72
120.86
118.57
118.81
124.15
116.95
111.52
127.62
120.86
118.51
118.83
124.10
116.98
110.79
127.54
121.65
119.84
118.79
124.27
116.94
110.77
127.43
121.78
119.78
118.76
124.27
116.96
110.80
126.98
122.20
119.83
118.78
124.11
117.11
Da
O9AC10AC11AN22
O9AC10AC15AN16
O8AC7AC1AC6
C7AO9AC10AC11
C7AO9AC10AC15
C5AC6AC1AC7
C3AC2AC1AC7
O9AC10AC11AC12
O9AC10AC15AC14
ꢁ0.40
ꢁ6.06
1.91
4.15
ꢁ53.02
126.22
180.00
179.84
177.60
ꢁ177.90
ꢁ0.21
ꢁ0.54
ꢁ5.04
1.37
4.87
ꢁ54.74
124.80
179.99
179.83
178.23
ꢁ178.44
ꢁ4.94
ꢁ0.40
5.74
5.73
5.86
1.37
ꢁ0.91
ꢁ2.51
ꢁ2.09
ꢁ1.49
5.20
ꢁ6.06
64.50
63.90
64.77
ꢁ55.18
124.40
ꢁ179.92
179.74
178.31
ꢁ178.48
56.16
ꢁ119.48
179.99
ꢁ179.92
177.39
ꢁ174.85
ꢁ119.64
179.95
ꢁ179.91
177.48
ꢁ174.89
ꢁ118.86
179.93
ꢁ180.00
177.28
ꢁ174.74
ꢁ123.63
180.00
ꢁ179.80
ꢁ178.72
178.87
a
R, bond length (Å); A bond angle (°); D, dihedral angle (°).
O2N
O
O2N
O
C
H3CO
C
O
NO2
H3CO
O
NO2
O2N
O2N
(I)
(II)
O2N
O2N
O+
O-
C
O
O-
N
O
H3CO
C O
(IV)
H3CO
N
O-
O
(III)
O2N
O2N
Fig. 2a. The resonance structures of (3b).
Also C1AC7 bond has some double bond character, so it becomes
shorter compared to the unsubstituted derivative. This effect is
more pronounced in case of 4-OCH3 substituent which has higher
ability to donate it’s lone pair than the electronegative chlorine
atom, Fig. 2a. These resonance structures are verified from molec-
ular electrostatic potential (MEP) [25] surface of compound 3a,
Fig. 2b.
Simply, MEP measures how attractive (blue) or repulsive (red)
for any region of the molecule is to a proton placed at any point
surrounding the molecule. MEP shows that the attractive region
(blue) accumulates over the oxygen of the nitro groups and over
the carbonyl oxygen, while the oxygen of the methoxy group and
of the phenolic part of the ester appear in the red region. The bond
length of the CAC bond is between 1.389–1.406 and 1.390–1.402
for aryl and trinitrophenyl rings respectively (not shown in the ta-
ble) which is much shorter than the typical CAC single bond
(1.54 Å) and longer than the C@C double bond (1.34 Å) [26].
Table 1 shows that the bond angles C7AO9AC10 for compounds
3a–g are in the range of 118.50–119.83°. The calculations indicated
that the O9 atom is out of plane of the picryl ring by 1.3–5.3°, and it
forms only 0.20–6.06° dihedral angle with the nitrogen atom of the
Fig. 2b. The molecular electrostatic calculation of compound 3b.
carbonyl group which leads to resonance structures (I–IV). These
resonating structures show that the carbonyl group has some sin-
gle bond character so it becomes elongated than the normal one.

M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
307
-3.3
-3.4
-3.5
-3.6
-3.7
-3.8
nitro group, these small values are enough to suggest a negligible
deviation from coplanarity with the picryl ring and explains the
ability of the lone pair of O9 to involve in resonance with the nitro
groups of the picryl ring. This is supported from the observed de-
crease in three CAN bond lengths than that normal Csp3AN single
bond due to the formation of resonanting structures III.IV, Table
1. Also the C13AN19 of the 4-nitro group is shorter than those of
ortho C11AN22, C15AN16 indicating a larger contribution of the
4-nitro group in resonance compared to the ortho ones. The orien-
tation of the two phenyl rings with respect to each other can be esti-
mated from the calculated dihedral angles between the plane of the
aromatic rings and the ester group (angles around the OAPic and
COAAr bonds). These angles are C7AO9AC10AC11 and
O8AC7AC1AC6 and are denoted by Ø1 and Ø2 respectively. The
Ø2 is very close to 0° because of the strong conjugation between
the carbonyl and the aryl group. While Ø1 found to be 53–64.5°
depending on the substituents which indicate that the two phenyl
rings are neither in the same plane nor perpendicular to each other.
R2 = 0.991
σ
Fig. 3. Plot of LUMO energy in (eV) of compounds (3a–g) against Hammett
constants (r).
-3
-4
-5
-6
-7
-8
-9
6. Ground state properties
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
The ground state properties, namely: energies of the HOMO
(highest occupied molecular orbital) and LUMO (lowest unoccu-
pied molecular orbital) levels and the dipole moments of com-
pounds 3a–g are calculated by B3LYP with 6-31Gꢀꢀ basis set,
Table 2.
Inspection of Table 2 shows that the presence of electron donat-
ing groups push up the frontier molecular orbitals, while the pow-
erful electron withdrawing nitro group pushes them down. The
chloro atom in 3- or 4- position has negligible effect on rising the
energy of the HOMO and lowering the LUMO energy. Attempts to
illustrate the effect of substituents on the HOMO and LUMO ener-
gies were made by plotting the values of their energies against
R2 = 0.80
σ
Fig. 4. Plot of HOMO energy in (eV) of compounds (3a–g) against Hammett
constants (r).
Hammett substituent constants (
r), Figs. 3 and 4. It has been seen
that the values correlates linearly with HOMO and LUMO ener-
r
gies with negative slope. In other words, the substituent effect
on HOMO and LUMO is straightforward: the electron donating
groups (e.g. 4-OCH3, 4-CH3 or 3-CH3) increase the orbital energy,
while the electron withdrawing groups (e.g. 4-Cl, 3-Cl or 4-NO2)
decrease the orbital energy. It is noticed that the Hammett correla-
tion for the LUMO energies is better than for HOMO energies, with
correlation coefficients 0.99 and 0.80 respectively. Also the calcula-
tions show that the titled esters have a dipole with a negative end
at the O9 and a positive end at the aroyl ring except for compound
(3g, x = 4-NO2) where the positive end is oriented toward the picryl
group.
Table 3
Mulliken, NBO atomic charge and atomic orbital coefficient of LUMO for the selected
centers (C7,C10).
cpd
x
Atomic orbital coefficient
LUMO
NBO charge
Mulliken
charge
C7
C10
C7
C10
C7
C10
3a
3b
3c
3d
3e
3f
H
0.0163
0.1818
0.1836
0.1804
0.1807
0.1866
0.1876
0.1741
0.831 0.370 0.623
0.818 0.377 0.604
0.829 0.371 0.621
0.831 0.370 0.624
0.824 0.372 0.612
0.827 0.371 0.616
0.826 0.367 0.620
0.406
0.401
0.407
0.406
0.396
0.394
0.391
4-OMe 0.0189
4-Me
3-Me
4-Cl
3-Cl
4-NO2
0.0159
0.0158
0.0224
0.0230
0.0333
Klopman equation is studied in the present work in order to
investigate the relative reactivity of the two electrophilic centers
namely (C7, C10) toward nucleophilic attack. The simplified form
of klopman equation [27] is given below
3g
2
ꢁQnucQelec
2ðcnuc elecbÞ
c
In this equation, Qnuc and Qelec are the total charges on interact-
ing atoms of nucleophile and electrophile, is the local dielectric
D
E ¼
þ
ð1Þ
eR
EHOMO ꢁ ELUMO
e
|fflfflfflfflfflfflffl{zfflfflfflfflfflfflffl}
|fflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflffl}
electrostatic term
constant, R is the distance between atoms of nucleophile and elec-
trophile, Cnuc is the coefficient of atomic orbital of nucleophile in
HOMO, Celec is the coefficient of atomic orbital of electrophile in
LUMO, b is the overlap integral, EHOMO is the energy of HOMO
nucleophile and ELUMO is the energy of LUMO electrophile. Thus
the charge on atom in a molecule (q) and its electron density (C)
are essential requirements for the solution of Klopman equation.
In the present study the effective atomic charges calculated with
usage of both Mulliken and natural bond orbital (NBO), and the
atomic orbital coefficient of LUMO calculated using B3LYP method
for the compounds 3a–g are shown in Table 3. The charges and the
coefficient of each compound are confined to the two centers for
the nucleophilic substitution reaction, namely (C7, C10). Both
orbital term
Table 2
The energies of frontier molecular orbitals and the dipole moments of the compounds
(3a–g) evaluated by B3LYP/6-31Gꢀꢀ method.
HOMO(ev)
LUMO(ev)
E(LUMO ꢁ HOMO)
l (D)
H
ꢁ7.5267
ꢁ6.7827
ꢁ7.3035
ꢁ7.2031
ꢁ7.5035
ꢁ7.398
ꢁ3.4504
ꢁ3.3489
ꢁ3.4019
ꢁ3.424
4.0763
3.4338
3.9016
3.7791
3.9571
3.8317
4.4668
4.1319
6.1581
4.9322
4.5715
2.432
4-OCH3
4-CH3
3-CH3
4-Cl
ꢁ3.5464
ꢁ3.5663
ꢁ3.7535
3-Cl
3.4534
1.1625
4-NO2
ꢁ8.2203

308
M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
Fig. 7. The experimental and calculated kmax values of studied 2,4,6-trinitrophenyl
benzoate derivatives 3a–g. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)
for C10 compared to C7 for all esters under investigation, Table 3.
These values are consistent with the electron density diagram of
the LUMO’s of the esters 3a–g, Fig. 5.
Thus C7 can be considered as the hard electrophilic center while
C10 is the soft one. As a result, one can suggest that the regioselec-
tivity of the nucleophilic reaction depends on the nature of the
attacking nucleophile: where hard nucleophile which has low lying
HOMO and usually a negatively charged, making the energy gap
between LUMO of the ester and the HOMO of the nucleophile is
large, so the orbital term in Klopman equation would be very small
and can be neglected, in such case the reaction is referred to as
charge controlled and occur selectively on the carbonyl carbon
(C7) and it will be kinetically fast due to a large electrostatic attrac-
tion. While for soft nucleophile which has high lying HOMO and
not necessarily have a negative charge, make the HOMO of the
nucleophile and the LUMO of the ester closer to each other which
led to small denominator of the orbital term, consequently the
orbital term has large value and so large influence of the orbital
coefficients, c, and the resonance integral, in this case the reaction
Fig. 5. Electron density diagram of LUMO’s of esters 3a–g.
The Mulliken and NBO charges show that the carbonyl carbon (C7)
is more positively charged than the ipso carbon (C10), and the
charge does not change significantly with variation of the substitu-
ent especially within the NBO scheme.
The general inspection of the coefficient on the carbonyl carbon
and the ipso carbon atoms in LUMO shows that its value is greater
Table 4
Experimental UV data along with calculated kmax (nm) and oscillator strength values for compounds 3a–g.
cpd
x
Experimental
kmax (log
TD-DFT
SCRF-TD-DFT
Assignment of electronic transition (coeff.%)
e
)
kmax
f
kmax
f
3a
3b
3c
3d
3e
3f
H
234 (4.277)
260 (4.374)
240 (4.343)
235 (4.379)
240 (4.351)
230 (4.311)
250 (4.284)
222
228
225
222
226
222
231
0.227
0.563
0.438
0.21
0.452
0.227
0.4
225
232
229
225
228
225
237
0.324
0.676
0.549
0.28
0.549
0.265
0.472
HOMO ꢁ 1 to LUMO + 3 (30.7%)
HOMO to LUMO + 3 (36.8%)
HOMO to LUMO + 3 (35%)
HOMO ꢁ 1 to LUMO + 3 (26.7%)
HOMO to LUMO + 3 (27.3%)
HOMO ꢁ 2 to LUMO (26.5%)
HOMO ꢁ 1 to LUMO + 2 (35.7%)
4-OCH3
4-CH3
3-CH3
4-Cl
3-Cl
4-NO2
3g
Fig. 6. Theoretically predicted electronic absorption spectra of compounds 3a–g.

M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
309
Fig. 8. Electron density diagram for the
p
and pꢀ molecular orbitals corresponds to the electronic transitions of compounds 3a–g.
is referred to as FMO controlled and would occur selectively on
C10. However, the interactions on each center can exhibit a mix-
ture of charge and FMO contributions for nucleophiles of interme-
diate lying HOMO.
absorptions for compounds 3a–g are collected in Fig. 6. The differ-
ences between the theoretical gas-phase calculations and the
experimental results are quite large. This is a reason to extend
the calculation with the polarizable continuum TD-DFT (PCM-TD-
DFT) [28,29] method.
In PCM the solute part lying inside a cavity, whereas the solvent
part (methanol) represented as a structureless material. The sol-
vent is also characterized by its dielectric constant and other mac-
roscopic parameters. We have selected the non-equilibrium PCM
solutions because UV/Vis spectra are studied in the present work.
It has been seen from Table 4 that solvent leads to a bathochromic
shift for all kmax calculated. The statistical analysis of the UV data in
7. Electronic absorption spectra
The electronic spectra of the titled compounds 3a–g measured
in methanol along with the theoretical electronic absorption spec-
tra using B3LYP/6-31Gꢀꢀ level optimized structure and calculated
by TD-DFT are listed in Table 4. The theoretical electronic

310
M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
1788
1783
1778
1773
1768
1763
1758
1753
Table 4 shows that the mean absolute error is improved from
16 nm to 12.5 nm when the solvent effect is involved in calcula-
tion. The comparison between the experimental and the theoreti-
cal results indicates that the density functional B3LYP methods
provide satisfactory results to predict UV properties, Fig. 7.
Molecular orbital analysis show that the frontier molecular
orbitals are mainly composed of p atomic orbitals, therefore, the
singlet–singlet absorptions is corresponding to
p–
pꢀ type elec-
tronic transitions. To gain further insight into the nature of the
electronic transition, the electron density plots of molecular orbi-
tals mainly involved in the electronic transitions are presented in
Fig. 8. It is a common feature for compounds 3a-e that their main
transitions involve the LUMO + 3 as the unoccupied orbital, so it
can be considered as the pꢀ orbital for these compounds, however,
the pꢀ orbital in case of compounds 3f, 3g is LUMO and LUMO + 2
respectively. As seen from Fig. 8. strong delocalization is present in
1740
1745
1750
1755
1760
1765
1770
pꢀ orbitals, whereas the
isosurfaces of the wave functions on the benzoyl moiety, except
for compound 3f in which the electron density in orbital is local-
p orbitals show more localization of the
experimental wave number
p
Fig. 9. Plot of the observed versus theoretical wave number of carbonyl group of
compounds 3a–g.
ized on the picryl moiety. Thus, these states correspond to an elec-
tron transfer from benzoyl moiety to the whole molecular skeleton.
Substituted 2,4,6-trinitrophenylbenzoates 3a–g show batho-
chromic effects when the hydrogen atom at the 3- or 4-position
is replaced by a substituent, especially by introducing the strongly
ouxochromic methoxy group. Weakly electron donating groups
such as CH3 and Cl have also small bathochromic effects. On the
other hand, a large bathochromic shift is observed on introducing
a nitro group. The effect of substituents on absorption spectra
can be explained on the basis that the electron donating substitu-
ents in para-positions (4-OCH3, 4-CH3), the lone pair of oxygen or
3.2565
3.2515
R2 = 0.8571
3.2465
3.2415
3.2365
CAH
r-bond of methyl group elongate the conjugated system
and results in a decrease of the excitation energy causing a batho-
chromic effect (Red shift to kmax). In addition, the electron donating
ability of substituents is ordered as OCH3 > CH3. The chloro substi-
tuent results in minor changes to kmax relative to that of unsubsti-
tuted compound, which is an indicative of the induced electron
attraction, as well as the joining of unshared electron pairs of chlo-
rine in the conjugation. The bathchromic shift of the nitro group is
probably due to the very polarizable character of this group.
-1
-0.8
-0.6
-0.4
-0.2
0
σ+
0.2
0.4
0.6
0.8
1
Fig. 10. Plot of log experimental wave number of the carbonyl group versus r+
.
Table 5
Calculated and experimental IR absorption (cmꢁ1) for the C@O and NO2 bands of compounds 3a–g.
ꢀ
ꢀ
ꢀ
cpds
3a
x
m
(C@O)
masymðNO2Þ
msym (NO2)
ꢀ
mcalc
1772 (251.9)a
ꢀ
ꢀ
mcalc
ꢀ
ꢀ
mcalc
ꢀ
mexp
mexp
mcalc
H
1759(s)
1745(s)
1750(s)
1752(s)
1752(s)
1762(s)
1768(s)
1618
1613
1591
1615
1611
1587
1617
1613
1590
1617
1613
1590
1616
1613
1589
1616
1613
1589
1616
1614
1608
1589
(311.9)a
(230.6)a
(119.9)a
(276)a
1545(s)
1544(s)
1544(s)
1544(s)
1538(s)
1547(s)
1548(s)
1354
1341
1339
1354
1341
1338
1355
1341
1339
1355
1341
1339
1353
1340
1338
1353
1340
1338
1352
1345
1339
133S
(.67)a
1341.6(s)
1344(s)
1341(s)
1341(s)
1338(s)
1346(s)
1349(s)
(352)a
(395.6)a
(3.64)a
(351.4)a
(405)a
3b
3c
4-O Me
4-Me
3-Me
4-CI
3-CI
4-NO2
1766 (342.7)a
1769 (342.7)a
1771 (269.3)a
1770 (284.6)a
1772 (236.8)a
1773 (229.5)a
(220.2)a
(8.5)a
(296.4)a
(230.7)a
(121.2)a
(323.3)a
(230.1)a
(116.6)a
(337.1)a
(220.6)a
(10.5)a
(.59)a
(350.9)a
(403.31)a
(1.17)a
(349.8)a
(396.9)a
(3.12)a
(359.6)a
(390)a
3d
3e
3f
(356.8)a
(221)a
(3.05)a
(357.4)a
(387.4)a
(3.08)a
(198.6)a
(403.6)a
(407.1)a
(10.5)a
(347.7)a
(227.6)a
(115.4)a
(11.3)a
(s) Strong.
a
Peak intensity.

M.F. Ibrahim et al. / Journal of Molecular Structure 1006 (2011) 303–311
311
Table 6
Experimental and calculated chemical shift of all protons in ppm of arylbenzoate derivatives (3a–g).
Proton no
X = H
4-OCH3
4-CH3
3-CH3
4-Cl
3-Cl
4-NO2
H-25
H-26
H-27
H-28
H-29
H-30
H-31
H-32
H-33
H-34
H-35
8.55 (8.16)
7.90 (7.57)
8.08 (7.75)
7.86 (7.57)
8.48 (8.16)
9.43 (9.18)
9.53 (9.18)
8.75 (8.11)
7.12 (7.02)
8.45 (8.04)
7.75 (7.36)
8.15 (7.96)
7.84 (7.45)
7.97 (7.55)
8.72 (8.09)
7.78 (7.55)
8.66 (8.06)
7.91 (7.53)
7.92 (7.72)
7.96 (8.12)
8.03 (8.28)
7.26 (7.02)
8.34 (8.11)
9.45 (9.15)
9.46 (9.15)
7.60 (7.36)
8.35 (8.04)
9.41 (9.16)
9.50 (9.16)
2.66 (2.48)
2.57 (2.48)
2.06 (2.48)
7.72 (7.55)
8.42 (8.09)
9.51 (9.18)
9.48 (9.18)
8.07 (8.28)
8.18 (8.12)
9.08 (8.55)
8.78 (8.55)
8.20 (7.96)
9.41 (9.17)
9.53 (9.17)
2.01 (2.46)
2.56 (2.46)
2.64 (2.46)
8.32 (8.14)
9.54 (9.20)
9.49 (9.20)
3.85 (3.92)
4.18 (3.92)
3.85 (3.92)
Value between brackets corresponds to the experimental chemical shift.
8. The vibrational frequency
Appendix A. Supplementary material
Molecular structures of 2,4,6-trinitrophenylbenzoate deriva-
tives 3a–g were first fully optimized at B3LYP/6-31Gꢀꢀ level meth-
od. The observed experimental FT-IR spectra for the compounds
3a–g and theoretically predicted IR spectra obtained by DFT/
B3LYP method with 6-31Gꢀꢀ basis set are given in Table 5. In exper-
imental IR measurements, the most remarkable concern for these
esters is directed to the C@O and NO2 bands. The calculated vibra-
tional frequencies are scaled by 0.961 scaling factor [30], but their
values still higher than the observed ones for the majority of the
normal modes. Two factors may be responsible for the discrepan-
cies between the experimental and computed spectra: The first is
the fact that the experimental value is unharmonic frequency
while the calculated value is a harmonic frequency. The second
reason caused by the environment [31] where the calculations
have been actually done on a single molecule while the experimen-
tal values recorded in the presence of intermolecular interactions.
A linearity between the experimental and calculated wave number
for the carbonyl group, can be estimated by plotting the calculated
versus experimental wave numbers Fig. 9. The plot shows that the
strong electron donating substituent 4-OCH3 and the 3-CH3 substi-
tuent deviate from the fitted line. The deletion of these substitu-
ents improves the correlation significantly from 0.82 to 0.99.
The effect of substituent on the carbonyl vibrational frequency
was found to be consistent with the previously discussed carbonyl
bond distance, where the electron donating group increase the car-
bonyl group bond distance causing a decrease of its bond strength
and consequently it vibrates at lower frequency or wave number.
This effect is more pronounced for the strong donating 4-OCH3
group. While the electron withdrawing groups lead to shortening
and strengthening of the carbonyl group which explain its vibra-
tion at higher frequency. This observation is due to the resonance
between the substituent and the carbonyl carbon which is sup-
ported from the good linear relationship between the frequency
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2011.09.024.
References
[1] R.A.Y. Jones, Physical and Mechanistic Organic Chemistry, Cambridge
University Press, Cambridge, 1979.
[2] G.K. McEvoy (Ed.), AHFS Drug Information, the American Society of Health-
System Pharmacists, Inc., 1999.
[3] X. Liu, B.C. Lynn, J. Zhang, L. Song, D. Bom, W. Du, D.P. Curran, T.G. Burke, J. Am.
Chem. Soc. 124 (2002) 7650.
[4] H. Matsumoto, Y. Sohma, D. Shuto, T. Hamawaki, T. Kimura, Y. Hayashi, Y. Kiso,
Pept. Sci. (2000) 237.
[5] M.F. Bone, A.H. Price, M.G. Clark, D.G. McDonnel, Liq. Cryst. Ordered Fluids 4
(1982) 799.
[6] J.F. Kirsch, W. Clewell, A. Simon, J. Org. Chem. 33 (1968) 127.
[7] A. Kirkien-Konasiewicz, A. Maccoll, J. Chem. Soc. (1964) 1267.
[8] G. Guanti, C. Dell’Erba, F. Pero, Chem. Commun. 55 (1975) 823.
[9] G. Guanti, C. Dell’Erba, F. Pero, G. Leandri, J. Chem. Soc. Perkin II (1977) 966.
[10] H. Feuer, S. Patai (Eds.), The Chemistry of the Nitro and Nitroso Groups, Wiley-
Interscience, New York, 1969.
[11] E. Buncel, C. Chuaqui, P. Forsythe, S. Mahoney, A. Raoult, J.F. Willtshir, J. Chem.
Soc. Perkin II (1975) 478.
[12] (a) A.J. Kirby, W.P. Jencks, J. Am. Chem. Soc. 87 (1965) 3209;
(b) S.J. Benkovic, P.A. Benkovic, J. Am. Chem. Soc. 88 (1966) 5504.
[13] (a) A.J. Kirby, in: C.H. Bamford, C.F.H. Tipper (Eds.), Comprehensive Chemical
Kinetics, vol. 10, Elsevier, Amsterdam, 1972;
(b) M.L. Bender, Mechanisms of Homogeneous Catalysis from Protons to
Proteins, Wiley-Interscience, New York, 1971.
[14] A. Kirkien-konasiewicz, G. Sammy, A. Maccoll, Nature 203 (1964) 637.
[15] R. Wrzalik, K. Merkel, A. Kocot, J. Mol. Model 9 (2003) 248.
[16] T. Imase, S. Kawauchiand, J. Watanabe, J. Mol. Struct. 560 (2001) 275.
[17] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
V.G. Zakrzewski, J.A. Montgomery Jr., R.E. Stratmann, J.C. Burant, S. Dapprich,
J.M. Millam, A.D. Daniels, K.N. Kudin, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G.A. Petersson, Q. Cui,
K. Morokuma, D.K. Malick, A.D. Rabuck, J.B. Foresman, J. Cioslowski, J.V. Ortiz,
A.G. Baboul, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. anayakkara, M.
Challacombe, P.M. W. Gill, B. Johnson, W. Chen, M.W. Wong, J.L. Andres, C.
Gonzalez, M. Head-Gordon, E.S. Replogle, J.A. Pople, Gaussian 98, Gaussian Inc,
Pittsburgh, PA, 1998.
of C@O and (r
+) of substituents Fig. 10. However, the effect of
[18] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[19] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1998) 785.
[20] P.C. Hariharan, J.A. Pople, Chem. Phys. Lett. 16 (1972) 217.
[21] D. Jacquemin, E.A. Perpète, I. Ciofini, C. Adamo, Acc. Chem. Res. 42 (2009) 326.
[22] M. Sßenyel, Ö. Alver, C. Parlak, Spectrochim. Acta Part A 71 (2008) 830.
[23] C. Parlak, Ö. Alver, M. Sßenyel, Spectrochim. Acta. Part A 69 (2008) 1252.
[24] A. Bagno, F. Rastrelli, G. Saielli, Chem. Eur. J. 12 (2006) 5514.
[25] P. Politzer, J.S. Murray, K.D. Sen (Eds.), Molecular Electrostatic Potentials,
Concepts and Applications, Elsevier, Amsterdam, 1996.
varying substituent on the vibrational frequency of the NO2 groups
is negligible, Table 5.
9. NMR spectra
[26] J.J. Nie, D.J. Xu, Chin. Struct. J. Chem. 21 (2002) 165.
[27] G. Klopman, J. Am. Chem. Soc. 90 (1968) 223.
[28] C. Amovilli, V. Barone, R. Cammi, E. Cancès, M. Cossi, B. Mennucci, C.S. Pomelli,
J. Tomasi, Adv. Quantum Chem. 32 (1998) 227.
[29] M. Cossi, V. Barone, J. Chem. Phys. 115 (2001) 4708.
[30] E.G. Lewars, Computational Chemistry, Introduction to the Theory and
Applications of Molecular and Quantum Mechanics, Kluwer, 2003.
[31] H.T. Varghese, C.Y. Panicker, D. Philip, J.R. Mannekutla, S.R. Inamdar,
Spectrochim. Acta Part A 66 (2007) 959.
Experimental and calculated chemical shift for all protons of es-
ters 3a–g are listed in Table 6. In comparing the experimental and
theoretical result for the hydrogen chemical shifts, the correlation
values are found 0.996, 0.995, 0.947, 0.995, 0.960, 0.956, 0.80 for
X = H, 4-OMe, 4-Me, 3-Me, 4-Cl, 3-Cl, 4-NO2, respectively. It can
be seen that B3LYP findings are in good agreement with all the
experimental results.