Proton Affinity and Molecular Basicity of m- and p-Substituted Benzamides in Gas Phase and in Solution: A Theoretical Study

Abstract

Proton affinities (PAs) and basicities (GBs) of substituted benzamides in gas phase have been calculated at the DET/B3LYP level with a 6-311++G(2df,2p)//6-311+G(d,p) basis set. The influence of environment on PAs has been studied by means of SCRF solvent effect computations using PCM solvation model for water solvent. Results reveal that benzamides behave as oxygen base. Theoretical results show a good agreement with the experimental data. A good linear correlations between these quantities and the molecular electrostatic potential and the valence natural atomic orbital energies have been obtained. Substitution effect on PAs, GBs and structural properties has been considered.

Full text

Chemical
Physics
Letters
610–611
(2014)
321–330
Contents
lists
available
at
ScienceDirect
Chemical
Physics
Letters
jou
rn
al
hom
epage:
www.elsevier.com/locate/cplett
Proton
affinity
and
molecular
basicity
of
m-
and
p-substituted
benzamides
in
gas
phase
and
in
solution:
A
theoretical
study
Zaki
S.
Safia,∗,
Salama
Omarb
aChemistry
Department,
Faculty
of
Science,
Al
Azhar
University–Gaza,
Gaza,
Palestine
bSeccion
de
Ingeniería
Química,
Departamento
de
Química
Física
Aplicada,
Universidad
Autónoma
deMadrid,
Ctra.
Colmenar
Km,
15,
28049
Madrid,
Spain
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
17
June
2014
In
final
form
21
July
2014
Available
online
26
July
2014
a
b
s
t
r
a
c
t
Proton
affinities
(PAs)
and
basicities
(GBs)
of
substituted
benzamides
in
gas
phase
have
been
calculated
at
the
DFT/B3LYP
level
with
a
6-311++G(2df,2p)//6-311+G(d,p)
basis
set.
The
influence
of
environment
on
PAs
has
been
studied
by
means
of
SCRF
solvent
effect
computations
using
PCM
solvation
model
for
water
solvent.
Results
reveal
that
benzamides
behave
as
oxygen
base.
Theoretical
results
show
a
good
agreement
with
the
experimental
data.
A
good
linear
correlations
between
these
quantities
and
the
molecular
electrostatic
potential
and
the
valence
natural
atomic
orbital
energies
have
been
obtained.
Substitution
effect
on
PAs,
GBs
and
structural
properties
has
been
considered.
©
2014
Elsevier
B.V.
All
rights
reserved.
1.
Introduction
Protonation
reactions
(Eq.
(1))
play
an
important
role
in
organic
chemistry
and
biochemistry
[1–3]
and
are
the
first
steps
in
many
fundamental
chemical
rearrangements
[4,5]:
B
+
H+→
BH+,
(1)
where
B
is
basic
center
in
the
molecule.
The
capability
of
an
atom
or
molecule
in
the
gas
phase
to
accept
a
proton
can
be
characterized
by
calculating,
from
the
above
reaction,
the
proton
affinity
(PA)
and
the
molecular
basicity
(GB)
in
the
gas
phase,
which
give
a
deep
under-
stand
of
the
correlations
between
molecular
structures,
molecular
stability,
and
reactivity
of
the
organic
molecules
[6].
Using
standard
conditions,
the
PA
is
defined
as
the
negative
of
the
enthalpy
change,
H,
for
the
gas
phase
reaction
(Eq.
(1))
and
the
molecular
basicity
in
gas
phase
(GB)
is
defined
as
the
negative
of
the
free
energy
change,
baseG,
associated
with
the
same
reaction
in
Eq.
(1).
The
PA
and
GB
quantities
are
site-specific
values
and
therefore
must
be
calcu-
lated
at
each
chemically
different
binding
site
in
the
molecule.
This
means
that
some
molecules
will
have
multiple
PA
and/or
multiple
GB
values.
It
was
pointed
out
that
direct
measurements
of
the
PA
are
not
easy
[7]
which
are
possible
for
only
a
few
molecules,
mainly
olefins
and
carbonyl.
However,
these
measurements
have
been
used
to
establish
a
scale
of
PA
values
that
permits
PAs
to
be
determined
∗Corresponding
author
at:
Chemistry
Department,
Faculty
of
Science,
Al
Azhar
University
–
Gaza,
Gaza,
P.O.
Box
1277,
Palestine.
E-mail
addresses:
z.safi@alazhar.edu.ps,
zaki.safi@gmail.com
(Z.S.
Safi).
for
many
other
molecules,
including
unstable
species
and
reaction
intermediates.
The
basis
for
this
scale
is
described
by
Hunter
and
Lias
[8].
On
the
other
hand,
based
on
the
phenomenal
growth
in
computer
power
in
recent
years,
much
attention
has
been
given
to
the
possibility
of
calculating
these
parameters
by
quantum
meth-
ods.
Ab
initio
approaches
are
very
successful
in
providing
reliable
values
of
PA
and
GB
for
small
molecules
even
at
lower
levels
of
the-
ory
[9].
In
last
two
decades,
the
progress
in
the
density
functional
theory
(DFT)
approaches
makes
this
method
another
candidate
for
reliable
calculation
of
proton
affinities.
It
has
been
shown
that
the
PA
values
computed
using
B3LYP
method
are
as
effective
as
high
level
ab
initio
results
and
the
results
were
compared
up
to
∼1–2
kcal/mol
in
comparison
with
the
measured
values
[10–17].
Recently,
several
descriptors
have
been
employed
to
predict
the
PA
and
GB
such
as
the
molecular
electrostatic
potential
(MEP)
on
the
nuclei
of
the
basic
centers
[18,19],
the
valence
natural
atomic
orbital
energies
(NAO)
of
the
basic
centers
[18],
the 1H
NMR
chem-
ical
shift
(ı1H)
of
the
incoming
proton
[20]
and
the
topological
properties
at
the
bond
critical
point
(BCP)
of
the
B
H+bond
(includ-
ing
electron
density,
BCP,
and
its
Laplacian,
∇2
)
calculated
by
means
of
Bader
theory
of
atoms
in
molecules
[21].
Strong
linear
correlations
were
found
between
these
descriptors
and
the
values
of
PA
or
GB
in
the
gas
phase.
Those
results
suggested
that
such
these
quantum
descriptors
(MEP,
NAO,
ı1H
and
BCP)
can
be
easily
used
to
approximately
estimate
the
PAs
or
GBs
in
the
gas
phase.
Last
few
years
have
witnessed
great
progress
in
studying
the
substituent
effects
on
acid–base
behavior
in
both
experimental
and
theoretical
studies
as
one
of
the
most
important
factors
to
determine
molecular
basicity
[22–31].
Different
models
to
probe
the
substituent
effects
have
been
gathered
in
reference
[24]
by
http://dx.doi.org/10.1016/j.cplett.2014.07.050
0009-2614/©
2014
Elsevier
B.V.
All
rights
reserved.

322
Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
Deakyne.
Extensive
works
have
been
made
to
study
the
presence
of
the
substituent
on
the
aromatic
ring,
not
necessarily
a
phenyl
ring
[22–24,27,32,33].
The
essential
of
studies
dealing
with
ortho-,
meta-
and
para-position
effects
on
aromatic
carbonyl
compounds
have
been
summarized
by
Kukol
et
al.
in
their
introduction
[24].
Substituent
effects
were
subdivided
into
polar
and
steric
effects.
Other
studies
were
concerned
with
the
additivity
of
substituent
effects
when
two
or
more
substituents
are
present
on
the
aro-
matic
cycle
[22,24,27,32].
In
general,
polar
effects
were
rather
discussed
in
term
of
strong
resonance
interaction
with
electron-
donating/withdrawing
character
of
the
substituent
[22,24].
For
this
study,
benzamide
and
some
of
its
meta-
and
para-
substituted
derivatives
(
NH2,
CH3,
NO2,
OH,
OCH3,
CF3,
Cl)
have
been
chosen.
It
was
pointed
out
that
the
benzamide
nucleus
has
been
found
to
play
a
crucial
role
in
conferring
antibacte-
rial
activity
to
these
molecules
leading
to
the
inhibition
of
bacterial
infections
[33,34].
From
the
chemistry
point
of
view,
the
amide
group
contains
two
basic
centers,
oxygen
and
nitrogen,
which
attracts
a
special
attention
to
predict
which
of
them,
is
the
more
basic
one.
Based
on
ab
initio
methods,
Ya˜
nez
and
Catalan
[35]
showed
that
benzamide
molecule
is
an
oxygen
base
and
the
proton
added
in
gaseous
phase
will
be
bind
with
the
oxygen
atom.
On
the
other
hand,
the
PA
and
GBs
of
benzamide
and
some
of
its
para-
and
meta-substituted
derivatives
were
measured
in
the
gas
phase
using
experimental
techniques
such
as
spectrometric
techniques
and
kinetic
methods
[36,37],
and
confirmed
by
semiempirical
methods
[37].
The
main
purpose
of
this
work
is
to
calculate
the
PA
and
GB
of
benzamide
and
some
of
its
meta-
and
para-derivatives
(
NH2,
CH3,
OH,
OCH3,
NO2,
CF3and
Cl)
in
the
gas
phase
and
in
solution
(water)
using
density
functional
theory
at
B3LYP
6-311++G(2df,2p)//6-311+G(d,p)
level
and
to
compare
these
calcu-
lated
values
with
the
available
experimental
and
theoretical
data.
We
will
examine
that
the
four
quantum
descriptors,
namely,
MEP
on
the
nuclei
of
the
base,
NAO
of
the
base, 1H
NMR
chemical
shifts
of
the
incoming
proton
and
the
electron
density
at
the
O
H+
bond
critical
point
can
be
used
to
predict
the
gas-phase
basicity
and
the
proton
affinity
of
the
species
under
probe.
Possible
lin-
ear
correlations
between
these
quantities
and
the
corresponding
experimental
data
on
the
one
hand
and
the
quantum
descriptors
on
the
other
hand
will
be
tested.
Finally,
the
influence
of
substituent
on
the
intrinsic
basicities
(PA
and
basicity),
structural
geometries
(C7
O,
C1
C7
and
the
O
Habond
lengths)
and
the
Mulliken
net
charges
(on
the
carbonyl
oxygen
atoms,
on
the
amide
carbon
atom
and
on
the
incoming
proton)
in
gas
phase
and
in
solution
will
be
considered
and
discussed.
2.
Computational
methods
The
standard
hybrid
density
functional
theory
(DFT)
in
the
framework
of
B3LYP
[38,39]
functional
at
the
6-311+G(d,p)
basis
function
was
used
for
geometry
optimization.
The
harmonic
vibra-
tional
frequencies
of
the
different
stationary
points
of
the
potential
energy
surface
(PES)
have
been
calculated
at
the
same
level
of
theory
used
to
check
that
all
the
structures
are
minima
with
no
imaginary
frequencies,
as
well
as
to
estimate
the
correspond-
ing
zero-point
energy
corrections
(ZPE)
that
were
scaled
by
the
empirical
factor
0.9806
proposed
by
Scott
and
Radom
[40].
In
order
to
obtain
more
reliable
energies
for
the
local
minima,
final
energies
were
evaluated
by
using
the
same
functional
combined
with
the
6-311++G(2df,2p)
basis
set.
It
has
been
shown
that
this
approach
is
well
suited
for
the
study
of
this
kind
of
systems,
yield-
ing
PAs
and
GBs
good
agreement
with
experimental
values.
A
single
point
calculation
was
ensued
to
compute
the
molecular
electro-
static
potential
(MEP)
on
each
of
the
nuclei
followed
by
a
full
Natural
Bonding
Orbital
(NBO)
calculations
[41]
to
obtain
the
natural
atomic
orbital
energies.
Moreover,
another
single
point
calculation
at
the
B3LYP/6-311+G(d,p)
level
with
the
gauge-
independent
atomic
orbital
(GIAO)
method
[42]
was
performed
to
compute
the 1H
NMR
absolute
shielding
constants
(ı1H
values)
of
the
associated
proton.
Furthermore,
the
calculated
chemical
shiel-
ding
were
converted
into
chemical
shift
(ı)
by
subtracting
32.5976,
the 1H
shielding
of
tetramethylsilane
computed
at
the
same
level
of
theory
and
basis
set.
Finally
analysis
of
the
electron
densities
at
bond
critical
points
(bcps)
of
the
optimized
structures
was
cal-
culated
by
generating
the
wave
functions
through
a
single
point
calculation
on
the
geometrized
structures
and
analyzed
these
wave
functions
by
means
of
the
atoms
in
molecules
(AIM)
theory
pro-
posed
by
Bader
et
al.
[21,43]
as
implemented
in
AIM2000
program
package
[44].
We
have
additionally
carried
out
calculations
in
solu-
tion
using
the
integral
equation
formalism
polarizable
continuum
model
(IEF-PCM)
in
which
the
solvent
is
represented
by
an
infinite
dielectric
medium
at
the
same
level
of
theory.
All
calculations
were
performed
with
the
Gaussian
09
[45]
with
tight
self-consistent
field
convergence
and
in
addition
ultrafine
integration
grids
were
used
for
MEP
calculations.
The
gas-phase
PA
of
a
molecule
A
is
determined
according
to
Eq.
(2):
PA
=
−Eel +
ZPVE
+
Evib −5
2RT(2)
where
Eel,
ZPVE
and
Evib correspond
to
the
differences
between
total
electronic
energy,
zero-point
vibrational
energy
and
temperature-dependent
portion
of
vibrational
energy
of
the
reac-
tants
and
products
at
298.15
K,
respectively.
Under
the
standard
state
conditions,
the
values
of
the
enthalpy
and
entropy
of
the
gas-phase
proton
are
0
gH(H+)
=
2.5RT
=
1.48
kcal/mol,
which
corresponds
to
the
translational
proton
energy
(a
loss
of
three
degrees
of
freedom
is
3/2(RT)
plus
the
PV
work-term
(=RT
for
ideal
gas)).
On
the
other
hand,
thermodynamically,
molecular
basicity
in
gas
phase
(GB)
is
related
to
PA
and
the
change
of
entropy
S
at
constant
temperature
(298.15
K)
by
Eq.
(3):
GB
=
PA
+
TS
(3)
For
proton,
only
the
translational
entropy
Strans(H+)
is
not
equal
to
zero
and
is
determined
to
be
26.04
cal
mol−1K−1[46]
and
0
gG(H+)
=
2.5RT
−
TS0=
1.48
−
7.76
=
−6.28
kcal/mol
have
been
used
[47–49].
To
calculate
the
proton
affinity
and
molecular
basicities
in
the
solvent
(water
in
our
case),
the
reaction
to
considered
is
identi-
cal
to
one
used
in
gas
phase
(Eq.
(1))
except
all
constituents
are
solvated.
PCM
thermodynamic
results,
obtained
from
harmonic
fre-
quency
analysis,
were
used
to
evaluate
the
PA
value
according
to
the
following
equation
(Eq.
(4)):
PA
=
−protH
=
−H(BH+
solv −
H(Bsolv)
−
H(H+
solv)(4)
In
this
case
the
proton
enthalpy
is
determined
from
summing
the
solvation
enthalpy
of
the
proton
solvH(H+)
(−275.12)
[50],
which
is
very
close
to
the
experimental
value
(−275.14
kcal/mol)
[51],
with
proton
gas-phase
enthalpy
H(H+
g)
(1.48
kcal/mol).
In
order
to
calculate
the
molecular
basicity
in
solution,
the
constant
pKb(B)
for
a
base
B
is
given
by
the
well
known
thermodynamic
relation
(Eq.
(5)):
pKb(B)
=protG(sol)
2.303RT (5)
where
R,
T
and
protG(sol)
(1
atm,
standard
state)
are
the
gas
phase
constant,
the
temperature
(298.15
K)
and
the
standard
free

Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
323
enthalpy
of
base-protonation
reaction
in
solution,
respectively.
protG(sol)
is
determined
from
the
following
equation
(Eq.
(6)):
protG(sol)
=
G(BH+(sol))
−
G(B(sol))
−
G(H+(sol))
(6)
In
water,
the
aqueous
free
enthalpy
G(H+(aq))
is
deter-
mined
by
summing
aqG(H+)
=
−266.13
kcal/mol
[51,52]
and
G(H+(g))
=
−6.3
kcal/mol
[53].
3.
Results
and
discussion
3.1.
Proton
affinities
and
molecular
basicities
in
gas
phase
The
numerical
data
of
the
total
electronic
energy
difference
E,
enthalpy
change
H,
entropy
contribution
S
for
the
proton
asso-
ciation
reaction
of
the
systems
studied
in
this
work
(Eq.
(1)),
from
which
one
obtains
the
PAs
and
GBs
are
summarized
in
Table
1
together
with
the
available
theoretical
[35,38]
and
experimental
values
[36,37,54]
for
all
the
species
under
probe.
Total
energies,
zero
point
energies,
thermal
corrections
to
energies,
enthalpies
and
Gibbs
free
energies
and
entropies
for
all
the
species
under
probe
are
listed
in
Tables
S1
of
the
supplementary
materials.
It
is
important
to
mention
that
the
presence
of
the
amino
nitrogen
atom
(
NH2)
of
the
amide
group
suggests
a
second
site
of
protonation
besides
the
carbonyl
oxygen
atom
(C7
O8*).
So
that
in
order
to
investigate
which
of
these
two
sites
is
the
preferred
one,
computed
PA
=
PA(O)
−
PA(N)
at
B3LYP/6-
311++G(2df,2p)//6-311+G(d,p)
level
of
theory
are
listed
in
Table
S1
of
the
supplementary
material
for
all
the
considered
benzamides.
These
results
indicate,
regardless
the
substituent
nature,
that
the
carbonyl
oxygen
atom
is
more
basic
than
that
of
the
amino
nitro-
gen
atom.
The
PA
difference,
PA
=
PA(O)
−
PA(N),
between
the
two
centers
is
at
least
∼12
kcal/mol
in
favor
of
the
carbonyl
oxygen
atom.
Our
results
obtained
at
the
level
of
theory
mentioned
above
is
∼10
kcal/mol
lesser
than
that
reported
by
Catalan
and
Ya˜
nez
[35]
by
performing
an
ab
initio
study,
using
STO-3G
minimal
basis
set.
This
finding
suggest
that
the
carbonyl
oxygen
atom
is
the
preferred
site
for
protonation,
which
is
line
with
those
theoretically
reported
in
the
literature
[35].
As
indicated
from
Table
1,
for
the
parent
benzamide,
the
calculated
PA
and
GB
values
are,
respectively,
equal
to
212.6
and
206.4
kcal/mol,
which
are
very
close
to
the
measured
experimental
data
[8,36,37,54]
and
better
than
those
theoretically
calculated
[35,37].
A
close
look
at
Table
S2
reveals
that
based
on
DFT
calculations
the
mean
absolute
deviations
are
less
than
1–2
and
1–3
kcal/mol,
respectively.
On
the
other
hand,
our
theoreti-
cal
results
show
that
substituents
have
a
significant
effect
on
the
order
of
the
values
PA
and
GB
properties.
It
can
be
seen
that
elec-
tron
donating
groups
such
as
NH2,
OH,
OCH3and
CH3cause
the
PA
and
GB
to
be
higher
in
comparison
with
the
parent
benza-
mide
species,
while
the
reverse
is
true
in
the
case
of
the
electron
withdrawing
groups
such
as
NO2,
CF3and
Cl.
Indeed,
the
maxi-
mum
PA
and
GB
values
are
223.6
and
216.7
kcal/mol
correspond
to
p-aminobenzamide,
whereas
the
minimum
values,
202.1
and
194.5
kcal/mol,
correspond
to
p-nitrobenzamide.
It
is
worth
men-
tioning,
regardless
of
the
nature
of
the
substituted
group,
that
substitution
at
para
position
of
the
benzene
ring
increases
the
PA
and
GB
values
more
than
that
at
meta
position,
exceptionally
in
the
case
of
m-NO2species,
which
is
found
to
be
higher
than
the
ana-
log
p-NO2one.
It
is
found
that
the
PA
of
the
p-NH2is
∼7
kcal/mol
higher
than
that
of
the
m-NH2derivative.
Importantly,
our
theoreti-
cal
results
indicate
that
when
going
from
the
electron-withdrawing
to
electron-donating
substituents,
the
PA
and
GB
raise
by
at
least
20
kcal/mol.
In
order
to
explain
this
trend,
Grutzmacher
and
Calta-
panides
[37]
showed
that
there
is
a
dominant
␲-electron
resonance
effect
on
the
PA
affinity
of
benzamide,
which
acts
only
in
the
pres-
ence
of
strongly
electron
donating
susbtituents
such
as
NH2group
(Scheme
1
of
the
supplementary
materials).
This
␲-electron
reso-
nance
increases
the
electron
density
on
the
carbonyl
oxygen
atom
of
the
amide
group
and
consequently
increases
the
interaction
of
the
oxygen
atom
with
the
incoming
proton
(Eq.
(1))
(Scheme
1).
A
closer
inspection
of
Table
1
shows
an
interesting
linear
rela-
tionships
between
the
calculated
PA
and
GB
of
benzamide
and
its
derivatives
and
the
available
experimental
ones
(Figure
1),
i.e.,
the
PA
(H)
in
Figure
1a
and
the
GB
(G)
in
Figure
1b,
with
determi-
nation
coefficients
0.977
for
PA
and
0.976
for
GB.
Therefore,
linear
correlations
are
derived
between
the
calculated
PA
and
GB
and
the
corresponding
experimental
data
as
(Eqs.
(7)
and
(8)):
PAcalc =
−22.3696
+
1.10808
PAexp gas (7)
GBcalc =
−38.05066
+
1.18574
GBexp gas (8)
According
to
the
illustrated
data
in
Table
1
and
based
on
the
B3LYP
calculations,
the
enthalpy
(PA)
and
Gibbs
free
energy
(GB)
Table
1
Calculated
protonation
energies,
E,
entropies,
S,
proton
affinities,
PA
and
basicities,
GB,
in
gas
phase
calculated
at
B3LYP/6-311++G(2df,2p)//6-311++G(d,p)
level
of
theory
together
with
the
available
theoretical
(ab
initio
and
PM3
semiempirical
level)
and
experimental
values.
All
values
are
in
kcal/mol,
MEP(O8)
and
MEP(N9),
NAO(O8),
NAO(N9)
(in
a.u.),
BCP and
its
∇2
(in
a.u.)
of
the
O8
Ha+bond
and
the 1H
NMR
chemical
shifts
of
the
incoming
proton
(ı1H
in
ppm)
(for
p-
and
m-benzamide)
and
its
derivatives.
X
DE
TDS
PAcalc PAexpaGBcalc GBexp aMEP(O)
MEP(N)
NPA(O)
NPA(N)
BCP ∇2
ı1H
p-OCH3−209
−7.6
219.1
215.2
211.4
207.8
−22.409
−18.373
−1.634
−1.313
0.358
−2.559
7.1
m-OCH3−205
−7.8
214.9
215.3
207.2
207.9
−22.406
−18.37
−1.645
−1.323
0.3565
−2.555
7.6
p-OH
−207
−7.7
216.7
−
209
−
−22.407
−18.371
−1.644
−1.322
0.3575
−2.559
7.1
m-OH
−204
−7.8
213.3
−
205.5
−
−22.404
−18.368
−1.653
−1.331
0.3563
−2.555
7.6
p-NH2−214
−6.9
223.6
214.2a
221.6
216.1
214.4
−22.414
−18.378
−1.617
−1.295
0.3577
−2.547
6.7
m-NH2−207
−7.6
216.5
215.3
208.9
207.9
−22.405
−18.371
−1.64
−1.32
0.3557
−2.541
7.5
p-CH3−208
−8.6
217
215.3
209.4
207.9
−22.405
−18.378
−1.643
−1.322
0.3559
−2.544
7.4
m-CH3−205
−9.5
214.5
215.3
206
207.9
−22.404
−18.37
−1.647
−1.326
0.3555
−2.542
7.5
H
−203
−7.7
208.4a
212.6
224.5b
213.2
204.9
205.8
−22.403
−18.367
−1.656
−1.334
0.3551
−2.542
7.6
p-NO2−193
−7.5
202.1
197.2b
201.2
194.6
196.7
−22.381
−18.347
−1.732
−1.411
0.3535
−2.539
8
m-NO2−194
−7.8
203.1
204.3
195.3
196.7
−22.384
−18.35
−1.723
−1.4
0.3537
−2.541
7.9
p-Cl
−201
−7.7
210.7
209.7
203
202.3
−22.394
−18.362
−1.681
−1.358
0.3552
−2.544
7.5
m-Cl
−199
−7.8
208.9
209.7
201.1
202.3
−22.393
−18.361
−1.685
−1.363
0.3556
−2.539
7.7
p-CF3−196
−7.8
205.9
206.2
198.1
198.8
−22.39
−18.354
−1.706
−1.385
0.354
−2.538
7.8
m-CF3−197
−7.5
206.4
206.7
198.9
199.8
−22.391
−18.355
−1.702
−1.38
0.3553
−2.553
7.8
aValues
taken
from
Refs.
[37,38,54].
bValues
taken
from
Ref.
[36].

324
Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
Scheme
1.
differences
are
close
to
each
other
in
value
(the
PA
values
are
∼8
kcal/mol
higher
than
the
GB
values)
because
the
entropy
contri-
bution
for
the
species
studied
here
is
relatively
small,
as
evidenced
by
the
TS
values
in
the
table.
Our
results
show
that
there
are
perfect
linear
relationships
(Figure
S1
of
the
supplementary
mate-
rials)
between
the
E
values
and
the
PAs
and
GBs
values
with
correlation
coefficients
are
very
close
to
unity
(0.9998
and
0.9988,
respectively).
In
agreement
with
recent
studies
[18,19]
such
these
strong
correlations
permit
us
to
conclude,
among
the
above
men-
tioned
parameters,
that
the
PA
and
GB
of
the
species
under
probe
can
be
represented
by
E,
if
one
only
cares
the
trend,
not
their
absolute
values.
3.2.
Analysis
of
MEP
and
NAO
results
It
was
pointed
out
that
the
molecular
electrostatic
potential
(MEP)
on
the
nuclei
of
a
molecule
originally
invoked
to
study
elec-
trophilic
reactivity
[18,19,55].
The
MEP
of
a
molecule
is
a
real
physical
property,
and
it
can
be
determined
experimentally
by
X-
ray
diffraction
techniques
[56].
V(
r)
(MEP)
that
is
created
at
a
point

rby
electrons
and
nuclei
of
a
molecule
is
given
as
(Eq.
(9)):
MEP
=
V(
r)
=
A
ZA

RA−
r

−(
r)


r−
r

d
r(9)
where
ZAis
the
charge
on
nucleus
A,
located
at
RA,
and
(
r)
is
the
molecule’s
electron
density
[57].
The
sign
of
MEP
(V(r))
at
any
point
depends
on
whether
the
effects
of
the
nuclei
or
the
electron
are
dominant
there.
The
most
negative
values
are
associated
with
the
lone
pairs
of
electronegative
atoms,
because
of
the
larger
value
of
the
electronic
term
in
Eq.
(5)
in
comparison
with
the
nuclear
term,
and
these
Vmin points
represent
the
centers
of
negative
charges
on
the
molecule
[58].
Huang
et
al.
[18]
proposed
that,
since
MEP
is
a
negative
quantity,
Eq.
(9)
implies
that
the
more
negative
the
MEP
value,
the
stronger
the
molecular
basicity,
the
larger
the
proton
affinity
and
gas-phase
basicity.
Also
they
proposed
that
for
systems
with
multiple
sites
of
the
same
basic
element
type,
the
most
basic
site
(largest
energy
decrease)
has
the
most
negative
MEP
value.
The
numerical
values
of
the
MEP
on
the
nuclei
of
oxygen
and
nitro-
gen
atom
(MEPOand
MEPN)
for
all
species
investigated
here
are
summarized
in
Table
1.
Our
results
indicate
that,
despite
of
the
nature
of
the
substituent
group
and
its
position
on
the
benzene
ring,
the
MEP(O)
values
are
more
negative
than
MEP(N).
Indeed,
we
have
found
that
the
values
of
former
are
at
least
4.02683
a.u.
more
negative
than
the
values
of
the
later.
Based
on
these
findings,
one
expects
that
the
oxygen
atom
should
be
the
site
to
preferably
bond
with
the
incoming
pro-
ton,
which
is
in
line
with
our
tabulated
results
in
Table
S2
of
the
supplementary
materials.
These
numerical
data
lead
us
to
suggest
that
the
analysis
of
MEP
on
the
nuclei
of
the
basic
center
can
be
used
as
a
very
good
tool
to
predict,
not
only
the
basicity
of
the
same
element
in
one
molecule
[18,20]
but
also
the
basicity
for
different
elements
in
one
molecule
such
as
oxygen
and
nitrogen.
As
can
be
seen,
the
MEP(O)
value
becomes
more
negative
with
the
electron
donating
substituents,
while
the
reverse
is
true
with
the
electron
withdrawing
substituents,
relative
to
the
parent

Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
325
Figure
1.
(a)
Calculated
proton
affinities
(in
kcal/mol)
versus
experimental
values
in
gas
phase
(298.15
K)
(open
circles);
(b)
calculated
basicities
(GB)
(in
kcal/mol)
versus
experimental
values
in
gas
phase
(298.15
K)
(open
squares);
(c)
molecular
electrostatic
potential
on
the
oxygen
nucleus
(MEP(O)
in
a.u.)
versus
calculated
proton
affinities
(in
kcal/mol)
(black
open
circles)
and
calculated
basicities
(GB)
(blue
open
squares)
in
gas
phase
(298.15
K);
(d)
valence
nature
atomic
orbital
energies
around
the
oxygen
atom
(NPA(O)
in
a.u.)
versus
calculated
proton
affinities
(in
kcal/mol)
(black
open
circles)
and
calculated
basicities
(GB)
(blue
open
squares)
in
gas
phase
(298.15
K);
(—):
fitted
line.
benzamide
molecule.
It
is
also
found
that
the
electrostatic
term
depends
on
the
electron
donation
or
electron-withdrawal
charac-
ter
of
the
substituent
and
its
position
on
the
benzene
ring
(see
Table
1).
That
is
to
say,
comparison
of
the
p-
and
m-position
substituted
cases
shows
that
the
MEPOvalues
in
the
former
are
more
negative
than
those
in
the
later
derivatives
exceptionally
of
p-chloro
derivatives.
Indeed,
the
most
negative
MEPOvalue
(−22.413697
a.u.)
is
observed
for
p-NH2species,
while
the
least
negative
one
(−22.381435)
is
reported
for
p-NO2one.
Figure
1c
plots
the
relationships
between
the
values
of
MEP
on
the
nucleus
of
the
oxygen
atom
(MEPO)
vs.
PAs
(H)
(open
black
circles),
and
GBs
(G)
(open
blue
squares)
quantities.
The
figure
reveals
strong
linear
correlation
with
the
correla-
tion
coefficients
0.976
and
0.973,
respectively.
Therefore,
one
finds
that
these
quantities
can
be
easily
estimated
from
the
MEPOvalues
by
equations:
PAcalc =
−13
741.52
−
622.96
MEP(O)
and
GBcalc =
−13
711.20
−
621.26
MEP(O).
This
is
in
line
with
the
earlier
studies
[18]
because
it
was
pointed
out
that
MEP
may
be
treated
as
a
good
measure
of
the
PA
for
the
nitrogen
atom.
To
better
understand
the
PA
and
GBs
and
to
confirm
the
above
prediction
of
MEPO,
Table
1
also
displays
the
values
of
the
valence
natural
atomic
orbital
energies
of
oxygen
and
nitrogen
atoms,
NAO(O)
and
NAO(N),
for
all
the
investigated
species.
Comparison
of
NAO(O)
and
NAO(N)
indicates,
regardless
of
the
substituent
nature,
that
the
NAO(O)
energies
are
more
negative
than
NAO(N)
by
at
least
0.3217
a.u.,
confirming
the
MEP
prediction.
These
results
indicate
once
again,
for
all
the
compounds
under
probe,
that
the
oxygen
is
more
basic
than
the
nitrogen
atom
and
the
carbonyl
oxygen
atom
is
the
preferred
protonation
site.
Plots
of
NAO(O)
vs.
PAcalc.gas (open
black
circles)
and
GBcalc.gas
(open
blue
squares)
exhibits
very
strong
linear
relationships.
Correlation
coefficients
are,
respectively,
0.980
and
0.975
(Figure
1d),
yielding
the
linear
equations
as:
PAcalc =
497.09
+
170.79
NPA(O)
and
GBcalc =
488.30
+
170.16
NPA(O).
These
results
lead
us
to
conclude
that
such
these
relationships
can
be
used
to
qualitatively
estimate
both
PAs
and
GBs
values
if
the
experimental
values
are
not
available.
3.3.
Analysis
of
AIM
results
Another
topics
that
should
be
considered
here
is
the
electron
densities
at
the
O
H+BCP,
BCP,
and
their
Laplacian,
∇2
.
These
topo-
logical
properties
were
evaluated
at
the
B3LYP/6-311+G(d,p)
level
of
theory
by
means
of
AIM2000
approach.
The
values
of
BCP and
∇2
are
included
in
Table
1.
As
can
be
seen,
the
BCP values
are
posi-
tive,
while
the
∇2
values
are
negative,
which
suggest
that
the
O
H+
interaction
have,
in
principle,
a
covalent
character.
Indeed,
the
BCP
and
∇2
lie
in
the
ranges
of
0.3535–0.3577
e/a.u.3and
−2.5467
to
−2.5392
e/a.u.5,
respectively.
It
is
evident
from
Table
1
that
moving
from
electron
donating
to
electron
withdrawing
substituent
cause
a
decrease
in
the
O-H+distance
and
consequently
an
increase
in
the
electron
density,
BCP,
at
O
H+BCP
with
respect
to
the
parent
benzamide.
For
example,
BCP =
0.3551,
0.3577
and
0.3535
for
the
parent
benzamide,
p-NH2and
p-NO2species,
respectively.
Again,
this
behavior
is
well
understood
in
the
light
of
the
␲-electron
amide
resonance.
3.4.
Analysis
of 1H
NMR
chemical
shifts
The
values
of
the 1H
NMR
chemical
shift
(ı1H)
of
the
associated
proton
with
the
carbonyl
oxygen
atom
of
the
benzamide
group
in

326
Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
Table
2
protE,
PA,
protG
and
pK(B)
of
para-
and
meta-substituted
benzamides
calculated
at
B3LYP/6-311++G(2df,2p)//6-311++G(d,p)
level
of
theory
in
solution
(water)
together
with
the
available
experimental
pK(B)
values.
All
values
are
in
kcal/mol.
Substituent
X Water
(calculated)
pK(B)
(experimental)
protE
PA
protG
pK(B) *
**
***
p-OCH3−250.2
−15.1
−13.7
−10.0
–
–
–
m-OCH3−247.7 −17.6 −16.2
−11.8
–
–
–
p-OH
−249.2
−15.7
−14.7
−10.7
–
–
–
m-OH
−247.4
−17.8
−16.4
−12
–
–
–
p-NH2−248.6
−12.4
−10.8
−7.8
–
–
–
m-NH2−248.6
−16.6
−15.4
−11.2
–
–
–
p-CH3−248.7
−16.4
−15.4
−11.2
1.46
1.44
1.67a
m-CH3−247.8 −17.3 −16.7 −12.2 1.33
1.37
1.76a
H
−248 −17.3 −15.9 −11.6
1.4
1.43
1.74a,
1.45b,
1.38c,
1.65d,
1.54e,
1.45e,
1.54f
p-NO2−243.5
−21.5
−20.5
−14.9
2.36
2.28
2.70a,
2.13d
m-NO2−243.9
−21.1
−20.1
−14.6
2.04
2.01
2.42a
p-Cl
−247
−18.3
−16.9
−12.3
1.6
1.66
1.97a
m-Cl
−245.8
−19.3
−18.3
−13.3
1.63
1.65
2.09a
p-CF3−245.1
−20
−18.9
−13.8
–
–
–
m-CF3−245.2
−19.8
−19.1
−13.9
–
–
–
*Values
taken
from
Ref.
[59].
** Values
taken
from
Ref.
[59].
*** Values
taken
from
Refs.
[a60, b61, c62, d63, e64, f65].
all
the
investigated
compounds
are
included
in
Table
1.
According
to
the
illustrated
data
in
Table
1,
the
effect
of
substituent
on
the 1H
NMR
chemical
shifts
of
the
associated
proton
with
the
basic
cen-
ter
of
the
benzamide
and
its
derivatives
has
a
regular
trend.
It
is
observed
that
on
going
from
an
electron
withdrawing
substituent
group
to
an
electron
donating
substituent
one,
the
values
of
ı1H
is
upfield
shifted
(from
6.7
to
8.0
ppm).
The
highest
upfield 1H
res-
onance
is
found
for
p-NH2species
(6.7
ppm),
whereas
the
lowest
downfield
shift
(8.0
ppm)
is
reported
for
p-NO2species.
In
order
to
explore
the
correlation
between
NMR
chemical
shifts
of
the
incom-
ing
proton
and
the
gas-phase
PAs
and
GBs
of
the
base, 1H
(acidic
proton)
chemical
shifts
of
the
B
H+complexes
were
plotted
vs.
gas-
phase
PAs
and
GBs
of
benzamides
and
its
derivatives
(not
shown).
Correlation
coefficients
are
nearly
equal
to
0.91.
According
to
these
results
and
those
reported
in
Table
1,
a
larger 1H
chemical
shifts
indicates
a
stronger
intrinsic
base
strength,
which
lead
us
to
con-
clude
ı1H
values
can
be
used
as
a
good
descriptor
to
estimate
the
PAs
and
GBs
values
of
benzamide
and
the
its
derivatives.
3.5.
Solvent
effect
on
proton
affinities
and
basicities
To
discuss
theoretically
the
influence
of
the
environment
on
protE,
PAs,
protG,
pK(B)
and
others
properties
of
p-
and
m-
substituted
benzamide,
calculations
have
been
carried
out
in
aqueous
solution
(water),
is
high
dielectric
solvent
with
ε
=
78.8.
Several
experimental
studies
have
been
carried
out
to
calculate
the
pK(B)
of
some
benzamide
derivatives
in
aqueous
sulfuric
acid
at
298
K
and
their
values
are
included
in
Table
2
[59–65].
In
order
to
discuss
the
solvent
effects
on
protonation,
we
have
chosen
the
self
consistent
reaction
field
(SCRF)
continuum
solvation
method
using
IEF-PCM
mode
at
B3LYP/6-311++G(2ff,2p)//6-
311+G(d,p)
level
of
theory.
It
was
pointed
out
that
this
method
has
been
reported
to
be
more
reliable
in
calculating
proton
affinities
[66]
on
the
one
hand,
and
is
the
less
problematic
in
parametrization
and
convergence
on
the
other
hand.
It
was
effectively
demon-
strated
that
this
method,
which
is
not
expensive,
leads
to
reliable
barriers
of
internal
rotation
in
both
unprotonated
and
protonated
para-substituted
acetophenones
[67,68].
Calculated
protonation
energies
(protE),
proton
affinities
(PAs),
basicities
(protG)
and
pK(B)
are
gathered
in
Table
2
together
with
the
available
experimental
pK(B)
values
[59–65]
of
compounds
under
probe,
calculated
according
as
mentioned
above.
Total
set
of
data
are
summarized
in
Table
S3
of
the
supplementary
materials.
A
closer
look
at
the
results
in
Table
2
indicates
that
the
large
basicity
range
between
the
benzamides
derivatives
is
evidently
the
result
of
cooperation
of
both
induced
solvent
effects
(solvation)
and
sub-
stituent
effects.
Inspection
of
these
results
reveals
that
solvation
is
enormous
in
influencing
the
intrinsic
basicities
in
solution
when
compared
with
those
in
gas
phase.
Whereas
the
influence
of
the
substituent
effect
on
the
intrinsic
basicities
in
both
gas
phase
and
in
solution
is
almost
the
same.
The
most
remarkable
result
is
certainly
the
drastic
change
of
PA
and
GB
when
going
from
the
gas
phase
to
solvent.
What-
ever
the
substituent,
protonation
of
the
considered
compounds
in
solution
is
much
more
difficult
than
in
gas
phase.
The
reason
of
this
drastic
diminution
must
be
searched
for
in
Eq.
(7)
used
to
calculate
the
proton
affinity
in
solution.
As
a
matter
of
fact,
in
this
equation
the
enthalpy
difference
H(BH+
solv)
−
H(Bsolv)
is
of
the
same
order
of
magnitude
as
its
homolog
in
gas
phase:
the
former
is
in
the
range
[−257
→
−252]
kcal/mol
and
the
latter
in
the
range
[−214
→
−193]
kcal/mol;
so
proton
affinities
are
about
210
kcal/mol.
On
the
contrary,
the
proton
enthalpy
in
solution
H(H+
solv)
is
equal
to
−273.66
kcal/mol
for
H2O
(as
mentioned
above)
whereas
its
homolog
in
gas
phase
(5/2(RT))
is
only
≈1.48
kcal/mol.
This
difference
in
proton
enthalpies
on
going
from
gas
to
solu-
tion
explains
the
dramatic
decrease
of
PA
accompanying
solvation.
These
results
are
in
good
agreement
with
previous
reported
studies,
which
showed
that
(PAgas 
PASO2Cl2
PAH2O)
[68]
and
PAgas 
PAC6H6
PAH2O[69].
The
influence
of
substituent
on
PA
shows
the
same
trend
as
in
gas
phase.
In
water
solvent,
electron
donating
substituent
effect
algebraically
raises
PAs
and
lowers
pK(B)s.
In
water,
the
consid-
ered
series
of
amide
compounds
experiences
a
basicity
increase
of
about
7
pK(B)
units
on
going
from
the
nitro
to
the
amino
derivative.
Although
these
values
are
probably
excessive,
they
show
that
the
magnitudes
of
substituent
effects
are
not
comparable
and
exhibit
some
correlation
with
electro-donor/electro-acceptor
properties
of
X
substituents.
For
a
given
substituent
whose
electronic
properties
are
known,
it
should
even
be
possible
to
predict
the
probable
pK(B)
value
of
corresponding
derivative.
Close
look
at
Tables
1
and
2
indicates
that
quantitative
agree-
ment
between
theory
and
experiment
is
poor
for
water.
These
pK(B)
values
have
been
calculated
by
using
different
techniques
such
as
standard
least
mean
squares
treatment
[60],
Yates
and
McCleland
and
excess
acidity
methods
[59],
in
which
the
pK(B)
values
deter-
mined
in
same
conditions
(298
K,
in
aqueous
sulfuric
acid
solution

Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
327
Figure
2.
pK(B)calc vs.
pK(B)exp for
experimentally
studied
benzamides
in
aqueous
H2SO4by
using
standard
least
mean
squares
treatment
[60].
by
UV
spectroscopy)
are
to
be
compared
with
computed
pK(B)calc,aq.
Plot
of
pK(B)calc,aq vs.
pK(B)exp exhibits
qualitative
agreement.
Correlation
coefficients
are
0.967,
0.913
and
0.899,
respectively.
Figure
2
presents
only
the
plotting
of
pK(B)calc,aq vs.
pK(B)exp,
which
determined
by
the
standard
least
mean
squares
treatment[60].
If
one
assumes
linear
behavior,
the
fitted
equation
should
be
given
by
the
following
expressions:
pK(B)calc =
−5.414
−
3.638
pK(B)exp (R
=
0.967)
Standard
least
mean
squares
It
is
worth
mentioning
that,
however,
these
values
of
the
cor-
relation
coefficients
R
are
not
representatives
of
a
true
linear
relations
(for
which
R
≥
0.99),
They
are
indicative
of
a
quasi
lin-
ear
correlations.
According
to
our
results,
although
the
quantitative
comparison
with
experimental
methods
is
poor,
one
can
conclude
that
the
computation
of
the
proton
free
enthalpy
in
solution
with
high
accuracy
is
fundamental
in
solvation
study
if
reliable
experi-
mental
data
are
not
available.
3.6.
Substituent
effect
on
the
molecular
structure
The
most
noticeable
geometrical
changes
during
protonation
(either
in
gas
phase
or
in
solution)
would
concern
C1
C7
and
C7−O8*
bonds.
The
corresponding
bond
lengths
for
both
neutral
and
O*-protonated
species
together
with
O8*−Haone
are
collected
in
Table
3.
Note
that
results
of
only
p.X-benzamides
were
reported
in
the
table
in
order
to
reduce
the
volume
of
the
text;
however,
the
whole
series
is
taken
into
account
in
the
related
plots
and
given
in
Table
S4
of
the
supplementary
materials.
It
indicates
that
these
parameters
display
large
variations
when
substituent
X
is
changed.
For
a
given
substituent,
protonation
substantially
short-
ens
phenyl-amide
bond
length
(C1−C7)
and
increases
carbonyl
one
(C7
O8*)
(see
Table
3).
This
fact
can
be
explained
with
the
aid
of
basic
principles
concerning
electronic
effects
in
the
conjugated
chemical
systems.
Here,
whatever
X
may
be,
electron
circulation
is
always
toward
the
amide
group.
So,
when
the
carbonyl
oxygen
atom
is
protonated,
it
acquires
a
more
electron-accepting
charac-
ter
than
in
the
neutral
form,
which
decreases
the
␲
character
of
the
C7
O8*
bond
and
increases
the
C1
C7
one,
leading
to
lengthening
of
the
former
and
contraction
of
the
latter.
These
variations
seem
somewhat
more
pronounced
when
X
is
a
strong
electron-donating
substituent
(X
=
NH2,
OCH3,
CH3and
OH).
It
is
also
observed
that
the
variation
in
the
C1
C7
bond
lengths
is
more
pronounced
than
that
in
the
C7
O8*,
which
can
be
well
understand
by
consider-
ing
the
dominating
influence
of
the
mesomeric
effect
on
electron
Table
3
Gas
phase
and
solution
geometrical
parameters
of
the
neutral
and
O*-protonated
forms
in
gas
phase
and
in
solution
(water)
(length
in
Å)
at
the
equilibrium
ground
state
for
some
p.X-benzamide
derivatives.a
Substituent
X
C1
C7
C7
O8*
O8*
Ha
Neutral
Protonated
Neutral
Protonated
Gas
p-OCH31.497
1.429
1.222
1.32
0.968
p-OH
1.498
1.432
1.221
1.318
0.966
p-NH21.494
1.42
1.222
1.324
0.967
H
1.503
1.446
1.22
1.312
0.969
p-NO21.508
1.455
1.218
1.308
0.970
p-Cl 1.503 1.442 1.22
1.314
0.969
p-CF31.506 1.452
1.219
1.31
0.970
Water
p-OCH31.495
1.446
1.234
1.314
0.970
p-OH
1.496
1.447
1.234
1.314
0.970
p-NH21.49
1.433
1.236
1.321
0.969
H
1.502
1.46
1.232
1.309
0.971
p-NO21.508
1.47
1.23
1.304
0.972
p-Cl
1.502
1.459
1.232
1.308
0.971
p-CF31.506 1.467 1.231 1.305
0.971
aTotal
set
of
data
are
given
in
Table
S4
of
supplementary
materials.
migration
toward
the
amide
group
(see
Scheme
1
in
the
supple-
mentary
materials).
For
solution,
interpretation,
previously
given
to
explain
the
gas
phase
results,
still
valid.
Comparison
of
the
gas
phase
and
the
solu-
tion
results
shows,
for
a
specific
X
substituent
(m-
and
p-NH2),
for
unprotonated
form,
in
opposition
to
the
protonated
forms,
that
changing
the
environment
from
gas
phase
to
water
shortens
the
C1
C7
and
lengthens
the
C7
O8*
and
O8*
Ha.
Figure
3
shows
that
whatever
the
substituent
X,
calculated
PA
in
both
environments
(gas
and
solution)
is
in
a
good
linear
relationships
with
the
C
O
bond
length
in
both
protonated
and
unprotonated
forms
on
the
one
hand
and
with
the
O
H
bond
length
in
the
protonated
form
on
the
other
hand.
Correlation
coefficients
are
ranged
from
∼0.88
to
0.97.
These
relations
can
be
useful
to
esti-
mate
the
theoretical
PA
if
the
experimental
ones
are
not
available.
3.7.
Substituent
effect
on
the
Mulliken
charges
The
Mulliken
net
charges
q(O*)
on
the
carbonyl
oxygen
atom
both
before
and
after
protonation,
on
acidic
hydrogen
q(Ha),
q(C7)
as
well
as
the
electron
migration
toward
the
hydroxyl
group
(q(O8*Ha)),
and
the
change
in
the
net
charges
on
C
atom
and
oxygen
atom
before
and
after
protonation,
q(C7)
and
q(O8),
respectively,
are
partially
summarized
in
Table
4.
Total
set
of
data
are
considered
in
the
plots
and
given
in
Table
S5
of
the
supple-
mentary
materials.
Close
inspection
of
Table
4
indicates
that
local
charge
densities
on
all
atoms,
excepted
Haand
in
lesser
degree
O*,
do
not
perfectly
correlate
with
the
electronic
characteristics
(electron-donating/electron-withdrawing)
of
the
substituents
X.
For
Haatom,
although
the
change
in
net
charge
is
small
enough,
from
−0.007
to
0.003,
with
respect
to
the
parent
benzamide
species,
an
identifiable
trend
could
be
pointed
out:
the
more
the
character
of
the
substituent
is
electro-donating,
the
more
important
the
migra-
tion
of
electrons
toward
carbonyl
oxygen
atom
is
and
the
weaker
the
positive
charge
on
Ha.
For
example,
NH2group
being
the
strongest
electron
donating
substituent,
the
net
charge
on
the
Hais
0.013
a.u.
smaller
than
that
on
the
Hain
the
case
of
the
NO2,
which
is
the
strongest
electron
withdrawing
group.
For
oxygen
atom,
the
change
in
electron
density
(q(O*))
accompanying
the
protonation
does
not
seem
very
affected
by
the
nature
of
the
substituent.
The
reverse
is
true
in
the
case
of
the
carbon
atom
(C7),
the
(q(C7))
is
widely
affected
and
the
resulting
effect
is
not
clearly
related
to
electronic
characteristics
of
the
substituent
X.
For
example,
for

328
Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
Figure
3.
Left:
calculated
proton
affinities
(in
kcal/mol)
versus
d(C7
O*)
(top)
and
d(O*
H)
bond
lengths
(in
Å)
(bottom)
in
gas
phase
protonated
forms;
right:
the
same
as
on
the
left
but
in
water
protonated
form.
(—):
fitted
line.
p-NH2and
m-NH2species,
q(O*)
=
0.056
and
0.015
a.u.,
respec-
tively,
whereas,
for
p-NO2and
m-NO2species,
q(O*)
=
0.057
and
0.036
a.u.,
respectively.
Also
importantly,
we
report
the
electron
migration
toward
the
hydroxyl
group
in
the
protonated
form,
which
is
given
by
the
fol-
lowing
equation:
qO∗H=[q(O7∗)
−
1
+
q(Ha)]
It
is
found
that
qO∗His
nearly
not
influenced
by
the
nature
of
the
substituent
and
is
more
than
twice
the
one
toward
oxygen
atom
in
neutral
form
(≈−0.82,
−0.88
a.u.)
against
(≈−0.32,
−0.34
a.u.)
(see
Table
4).
So,
CO*
protonation
substantially
increases
the
elec-
tron
deficit
of
the
rest
of
the
molecule.
The
picture
does
not
affect
when
the
protonation
process
takes
place
in
solution,
and
the
same
interpretation
still
works.
Figure
4
presents
the
relationships
between
computed
PAs
versus
net
charges
on
the
oxygen
atom
before
and
after
protona-
tion,
q(O*),
on
the
one
hand
and
on
the
incoming
proton,
q(H),
on
the
other
hand
in
both
gas
phase
and
in
solution.
Figure
4
exhibits
clear
linear
correlation
between
PAs
and
Mulliken
net
charges
in
the
protonated
and
unprotonated
forms.
Correlation
coefficients
Table
4
Computed
Mulliken
net
charges
(in
a.u.)
on
C7,
O8
and
Haatoms
for
neutral
and
O*-protonated
forms
in
gas
phase
and
in
solution
(water),
T
=
298.15
K.
q(O*H)
=
q(O8)
−
1
+
q(Ha)
measures
the
electron
migration
toward
the
hydroxyl
group
for
p.X-benzamide
derivatives.a
Substituent
Neutral
Protonated
q(O8*-H)
q(C7)bq(O8*)c
C7
O*
C7
O*
H
Gas
p-OCH3−0.120
−0.344
−0.069
−0.152
0.290
−0.861
0.051
0.192
p-OH
−0.122
−0.345
−0.111
−0.152
0.292
−0.860
0.012
0.193
p-NH2−0.127
−0.350
−0.107
−0.168
0.288
−0.880
0.021
0.181
H
−0.083
−0.339
−0.027
−0.137
0.294
−0.843
0.056
0.202
p-NO2−0.180
−0.339
−0.208
−0.139
0.297
−0.842
−0.029
0.200
p-Cl
0.002
−0.327
0.038
−0.124
0.301
−0.823
0.036
0.204
p-CF3−0.199
−0.330
−0.242
−0.129
0.298
−0.831
−0.043
0.200
Water
p-OCH3−0.052
−0.456
−0.027
−0.165
0.322
−0.843
0.025
0.290
p-OH
−0.054
−0.456
−0.069
−0.165
0.321
−0.844
−0.015
0.291
p-NH2−0.055
−0.467
−0.059
−0.180
0.314
−0.866
−0.003
0.287
H
−0.021
−0.452
0.005
−0.160
0.326
−0.834
0.025
0.292
p-NO2−0.113 −0.450
−0.184
−0.162
0.330
−0.832
−0.071
0.288
p-Cl
0.055
−0.428
0.039
−0.144
0.335
−0.810
−0.016
0.283
p-CF3−0.145
−0.437
−0.226
−0.151
0.332
−0.819
−0.081
0.286
aTotal
set
of
data
are
given
in
Table
S5
of
supplementary
materials.
bq(C7)
=
C7protonated
–
C7neutral.
cq(O8*)
=
C8protonated
–
C8neutral.

Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
329
Figure
4.
Left:
calculated
proton
affinities
PA
(in
kcal/mol)
vs.
Mulliken
net
charge
on
carbonyl
oxygen
(q(O8*))
before
and
after
protonation
(top)
vs.
charge
density
on
hydrogen
hydroxyl
atom
(q(H))
(bottom).
Right:
the
same
as
on
the
left
but
in
water
protonated
form.
(—):
fitted
line.
are
ranged
from
∼0.85
to
0.98.
These
relationships
can
be
quali-
tatively
used
to
evaluate
the
intrinsic
properties
of
the
benzamide
and
its
derivatives.
4.
Concluding
remarks
The
present
computational
study
used
DFT/B3LYP,
BLYP,
B3PLYP,
BP86
and
M06-2X
functionals
at
the
6-311++G(2df,2p)//6-
311+G(d,p)
level
to
calculate
the
proton
affinity,
the
molecular
basicities
and
the
structural
properties
of
a
series
of
meta-
and
para-
substituted
benzamides
(NH2,
CH3,
OH,
OCH3,
NO2,
Cl,
and
CF3)
in
the
gas-phase
and
in
solution
(water).
Our
results
permit
us
to
draw
the
following
conclusions:
-
In
all
the
compounds
under
probe,
regardless
of
the
substituent
and
its
location
(in
para
or
in
meta
position),
the
carbonyl
oxygen
atom
of
the
amide
group
is
more
basic
than
the
nitrogen
atom
and
it
is
the
preferential
protonation
site,
which
leads
us
to
conclude
that
benzamide
and
its
derivatives
are
oxygen
bases;
-
proton
affinities
and
gas-phase
basicities
obtained
by
B3LYP
func-
tional
compare
nicely
the
available
experimental
proton
affinities
and
gas-phase
basicities
for
all
species
under
probe.
Mean
abso-
lute
deviations
are
∼1–2
kcal/mol.
We
have
also
demonstrated
that
the
proton
affinity
and
gas-phase
basicity
are
strongly
related
to
the
total
energy
difference.
-
proton
affinities
and
gas-phase
basicities
of
benzamide
and
its
derivatives
have
been
investigated
by
means
of
four
quantum
descriptors,
namely,
molecular
electrostatic
potential
on
the
nuclei
of
the
basic
center
(MEP),
valence
natural
atomic
orbital
energies
on
the
basic
atom
(NAO), 1H
NMR
(of
the
incoming
elec-
tron)
chemical
shift
and
topological
properties
of
the
electron
density
at
the
O
H+bond
critical
point
(BCP and
).
Results
obtained
revealed
a
very
strong
linear
correlations
between
these
descriptors
and
the
proton
affinities
and
gas
phase
basicities,
with
the
correlation
coefficients
are
very
close
to
unity.
The
calculation
results
suggest
that
the
MEP,
NAO,
BCP and 1H
NMR
chemi-
cal
shifts
are
good
quantum
descriptors
to
predict
the
gas-phase
proton
affinities
and
basicities
of
benzamide
and
its
derivatives;
-
proton
affinities
are
directly
influenced
by
the
electron-
donating/withdrawing
character
of
the
substituent.
Protonation
induces
migration
of
electrons
from
all
over
the
molecular
sys-
tem
leading
to
pronounced
charge
redistributions
compared
to
the
deprotonated
one;
-
in
gas
phase
and
in
solution,
generally
an
adequate
linear
rela-
tionships
were
obtained
between
proton
affinity
and
structural
and
electronic
charge
properties
of
the
carbonyl
group
before
and
after
protonation
and
the
hydroxyl
group
in
the
protonated
forms;
-
in
solution,
the
most
remarkable
result
is
certainly
the
drastic
change
of
proton
affinity
on
going
from
gas
to
solution
phase.
So
that
according
to
the
proposed
solvation
model,
one
concludes
that
the
protonation
is
not
energetically
favorable
in
solution.
-
computed
pK(B)values
in
aqueous
solution
are
in
good
qualitative
agreement
with
experiment.
Nevertheless,
they
exhibit
discrep-
ancies
with
regard
to
measured
ones.
Acknowledgment
A
generous
allocation
of
computational
time
at
the
Scien-
tific
Computational
Center
(CCC)
of
the
Universidad
Autónoma
de
Madrid
(Spain)
is
acknowledged.
Appendix
A.
Supplementary
data
Supplementary
data
associated
with
this
article
can
be
found,
in
the
online
version,
at
doi:10.1016/j.cplett.2014.07.050.

330
Z.S.
Safi,
S.
Omar
/
Chemical
Physics
Letters
610–611
(2014)
321–330
References
[1]
R.
Stewart,
The
Proton:
Appellation
to
Organic
Chemistry,
Academic
Press,
New
York,
1985.
[2]
F.A.
Carrol,
Perspectives
on
structure
and
mechanism
in
organic
chemistry,
Brooks-Cole,
New
York,
1998.
[3]
J.
Zhao,
R.Y.
Zhang,
Atmos.
Environ.
38
(2004)
2177.
[4]
S.
Enami,
H.
Mishra,
M.R.
Hoffmann,
A.J.
Colussi,
J.
Phys.
Chem.
A
116
(2012)
6027.
[5]
R.A.
Kennedy,
C.A.
Mayhew,
R.
Thomas,
P.
Watts,
Int.
J.
Mass
Spectrom.
223
(2003)
627.
[6]
E.
Uggerud,
Mass
Spectrom.
Rev.
11
(1992)
389.
[7]
D.A.
Dixon,
S.G.
Lias,
in:
J.F.
Liebman,
A.
Greenberg
(Eds.),
Molecular
Structure
and
Energetics,
Physical
Measurements,
vol.
2,
VCH,
Deerfield
Beach,
FL,
1987.
[8]
E.P.
Hunter,
S.G.
Lias,
J.
Phys.
Chem.
Ref.
Data
27
(1998)
413.
[9]
W.J.
Hehre,
L.
Ramdom,
P.v.R.
Schleyer,
J.A.
Pople,
Ab
Initio
Molecular
Orbital
Theory,
John
Willey
&
Sons,
New
York,
1986.
[10]
A.K.
Chandra,
A.
Goursot,
J.
Phys.
Chem.
100
(1996)
11599.
[11]
M.
Remko,
K.R.
Liedl,
B.M.
Rode,
J.
Chem.
Soc.
Perkin
Trans.
2
(1996)
1743.
[12]
T.K.
Ghanty,
S.K.
Ghosh,
J.
Phys.
Chem.
A
101
(1997)
5022.
[13]
M.
Remko,
K.R.
Liedl,
B.M.
Rode,
J.
Mol.
Struct.
Theochem.
418
(1997)
179.
[14]
M.
Remko,
B.M.
Rode,
J.
Phys.
Chem.
A
103
(1999)
431.
[15]
J.S.
Rao,
G.N.
Sastry,
Int.
J.
Quantum
Chem.
106
(2006)
1217.
[16]
A.
Luna,
O.
Mo,
M.
Yanez,
J.-F.
Gal,
P.C.
Maria,
J.-C.
Guillemin,
Chem.
Eur.
J.
12
(2006)
9254.
[17]
Z.S.
Safi,
G.
Frenking,
Int.
J.
Quantum
Chem.
113
(2013)
908.
[18]
H.
Yuang,
L.
Lianghong,
L.
Shubin,
Chem.
Phys.
Lett.
527
(2012)
73.
[19]
A.
Ebrahimi,
S.M.
Habibi-Khorasani,
M.
Jahantab,
Comput.
Theor.
Chem.
966
(2011)
31.
[20]
Yi.
Delian,
Z.
Hailu,
D.
Zongwu,
J.
Mol.
Catal.
A:
Chem.
326
(2010)
88.
[21]
A.D.
Becke,
Phys.
Rev.
A
38
(1988)
3098.
[22]
A.
Ebrahimi,
S.M.
Habibi-Khorasani,
M.
Jahantab,
Comput.
Theor.
Chem.
66
(2011)
31.
[23]
E.
Oteypková,
T.
Nevˇ
eˇ
cná,
J.
Kulhánek,
O.
Exner,
J.
Phys.
Org.
Chem.
16
(2003)
721.
[24]
A.
Kukol,
F.
Strehle,
G.
Thielking,
H.-F.
Gr¨
vtzmacher,
Org.
Mass
Spectrom.
28
(1993)
1107.
[25]
S.
Böhm,
J.F.
Gal,
P.C.
Maria,
J.
Kulhánek,
O.
Exner,
Eur.
J.
Org.
Chem.
(2005)
2580.
[26]
C.A.
Deakyne,
Int.
J.
Mass
Spectrom.
227
(2003)
601.
[27]
D.
Kovaˇ
cek,
Z.B.
Maksi´
c,
I.I.
Novak,
J.
Phys.
Chem.
A
101
(1997)
1147.
[28]
R.
Hadjadj-Aoul,
A.
Bouyakoub,
A.
Krallafa,
F.
Volatron,
J.
Mol.
Struct.
(THEOCHEM)
849
(2008)
8.
[29]
U.
Senapati,
D.
De,
B.R.
De,
J.
Mol.
Struct.
(THEOCHEM)
808
(2007)
153.
[30]
M.
Swart,
F.M.
Bickelhaupt,
J.
Chem.
Theor.
Comput.
2
(2006)
281.
[31]
S.
Pandit,
D.
De,
B.R.
De,
J.
Mol.
Struct.
(THEOCHEM)
778
(2006)
1.
[32]
Z.B.
Maksi´
c,
B.
Kovaˇ
cevi´
c,
D.
Kovaˇ
cek,
J.
Phys.
Chem.
A
101
(1997)
7446.
[33]
U.
Senapati,
D.
De,
B.R.
De,
Indian
J.
Chem.
47A
(2008)
548.
[34]
D.
Rai,
R.K.
Singh,
Indian
J.
Chem.
B
50
(2011)
931.
[35]
J.
Catalan,
M.
Ya˜
nez,
Tetrahedron
36
(1980)
665.
[36]
H.F.
Grutzmacher,
A.
Caltapanides,
J.
Am.
Soc.
Mass
Spectrom.
5
(1994)
826.
[37]
H.F.
Grutzmacher,
A.
Caltapanides,
Eur.
J.
Mass
Spectrom.
4
(1998)
349.
[38]
A.D.
Becke,
J.
Chem.
Phys.
98
(1993)
5648.
[39]
C.
Lee,
W.
Yang,
R.G.
Parr,
Phys.
Rev.
B
37
(1988)
785.
[40]
A.P.
Scott,
L.
Radom,
J.
Phys.
Chem.
100
(1996)
16502.
[41]
A.E.
Reed,
R.B.
Weinstock,
F.
Weinhold,
J.
Phys.
Chem.
83
(1985)
735.
[42]
K.
Wolinski,
J.F.
Hinton,
P.
Pulay,
J.
Am.
Chem.
Soc.
112
(1990)
8251.
[43]
R.F.W.
Bader,
Atoms
in
Molecules
–
A
Quantum
Theory,
Oxford
University
Press,
Oxford,
UK,
1990.
[44]
F.
Biegler-König.,
A.I.M2000,
University
of
Applied
Sciences,
Bielefeld,
Germany,
2000.
[45]
M.J.
Frisch,
et
al.,
GAUSSIAN
09,
Revision
B.01,
Gaussian,
Inc.,
Wallingford,
CT,
2009.
[46]
E.D.
Raczy ´
nska,
M.
Makowski,
E.
Górnicka,
M.
Darowska,
Int.
J.
Mol.
Sci.
6
(2005)
143.
[47]
J.M.L.
Martin,
J.P.
Francois,
R.
Gijbels,
J.
Comput.
Chem.
10
(1989)
346.
[48]
S.G.
Hwang,
Y.H.
Jang,
D.S.
Chung,
Bull.
Korean
Chem.
Soc.
26
(2005)
585.
[49]
I.A.
Topol,
G.J.
Tawa,
S.K.
Burt,
A.A.
Rashin,
J.
Phys.
Chem.
A
101
(1997)
10075.
[50]
J.A.
Mejías,
S.
Lago,
J.
Chem.
Phys.
113
(2000)
7306.
[51]
M.D.
Tissandier,
et
al.,
J.
Phys.
Chem.
A
102
(1998)
7787.
[52]
D.M.
Camaioni,
C.A.
Schwerdtfeger,
J.
Phys.
Chem.
A
109
(2005)
10795.
[53]
M.D.
Liptak,
G.C.
Shields,
Int.
J.
Quantum
Chem.
85
(2001)
727.
[54]
http://webbook.nist.gov/chemistry/
[55]
R.
Bonaccor,
E.
Scrocco,
J.
Tomasi,
J.
Chem.
Phys.
52
(1970)
5270.
[56]
C.H.
Suresh,
Inorg.
Chem.
45
(2006)
4982.
[57]
J.V.
Correa,
P.
Jaque,
J.
Olah,
A.
Toro-Labbe,
P.
Geerlings,
Chem.
Phys.
Lett.
470
(2009)
180.
[58]
S.R.
Gadre,
R.N.
Shirsat,
Electrostatics
of
Atoms
and
Molecules,
Universities
Press,
Hyderabad,
India,
2000.
[59]
G.
Stojkovi,
E.
Popovski,
J.
Serb,
Chem.
Soc.
71
(10)
(2006)
1061.
[60]
K.
Yates,
J.B.
Stevens,
Can.
J.
Chem.
43
(1965)
529.
[61]
B.
Garcia,
R.M.
Casado,
J.
Castillo,
S.
Ibeas,
I.
Domingo,
J.M.
Leal,
J.
Phys.
Org.
Chem.
6
(1993)
101.
[62]
U.
Haldna,
Prog.
Phys.
Org.
Chem.
18
(1990)
65.
[63]
J.T.
Edward,
S.C.
Wong,
J.
Am.
Chem.
Soc.
99
(1977)
4229.
[64]
G.
Stojkovi,
B.
Andonovski,
Proceedings.
3rd
Aegan
Analytical
Chemistry
Days,
Polihnitos,
Lesvos,
Greece,
2002,
pp.
498.
[65]
R.A.
Cox,
L.M.
Druet,
A.E.
Klausner,
T.A.
Modro,
P.
Wan,
K.
Yates,
Can.
J.
Chem.
59
(1981)
1568.
[66]
R.
Hilal,
A.A.
Abdel
Khalek,
S.A.K.
Elroby,
Int.
J.
Quantum
Chem.
103
(2005)
332.
[67]
A.
Haloui,
Y.
Arfaoui,
J.
Mol.
Struct.
(THEOCHEM)
950
(2010)
13.
[68]
A.
Haloui,
E.
Haloui,
J.
Mol.
Model.
19
(2013)
631.
[69]
E.
Klein,
J.
Rimarˇ
c´␫k,
V.
Lukeˇ
s,
Acta
Chimi
Slovaca
2
(2009)
37.