DFT and experimental study of the structure and vibrational spectra of 2-(benzylthio)-N-{pyridinylmethylidene}anilines

Article abstract of DOI:10.1016/j.molstruc.2012.08.026

2-(Benzylthio)-N {pyridine-4-ylmethylidene}aniline was prepared by reaction of S-benzyl ortho-aminothiophenol with 4-pyridine carboxaldehyde and characterized by NMR, IR and Raman spectroscopy and mass spectrometry. The structure and the vibrational analysis of the series of 2-, 3-, and 4-pyridine derivatives was performed based on a comparative computational methodology study with the density functionals B3LYP and B3PW91 and the basis sets LanZ2DZ and 6-31++G(d,p). Comparison of computational results with single crystal X-ray diffraction results of 2-(benzylthio)-N {pyridine-3-ylmethylidene}aniline allowed the evaluation of structure predictions and confirmed B3PW91/6-31++G(d,p) as most accurate for structure determination of the four investigated levels of theories. B3LYP and B3PW91 with the LanL2DZ basis set consistently outperformed calculations for IR and Raman vibrational estimations when compared to level of theories using the 6-31++G(d,p) basis set. Application of scaling factors for IR and Raman frequency predictions showed excellent agreement with experimental values and supported the assignment of the major contributors of the vibration modes of the three pyridine pendant compounds. Overall, B3PW91/LanL2DZ level of theory showed best performance in accuracy and low computational cost for structural and vibrational analysis for the series of 2-(benzylthio)-N-{pyridinylmethylidene}anilines.

Full text of DOI:10.1016/j.molstruc.2012.08.026

Journal of Molecular Structure 1034 (2013) 29–37
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
DFT and experimental study of the structure and vibrational spectra
of 2-(benzylthio)-N-{pyridinylmethylidene}anilines
J. Alberto Acosta-Ramirez a, Mathew C. Larade a, Samantha M. Lloy a, Edward D. Cross a, Beth M. McLellan a,
Jaime M. Martell a, Robert McDonald b, Matthias Bierenstiel a,
⇑
a Department of Chemistry, Cape Breton University, 1250 Grand Lake Road, Sydney, Nova Scotia, Canada B1P 6L2
b X-ray Crystallography Laboratory, Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2
h i g h l i g h t s
"
"
"
"
Synthesis and characterization of 2-(benzylthio)-N-{pyridinyl-4-methylidene}aniline.
X-ray structure analysis of 2-(benzylthio)-N-{pyridinyl-3-methylidene}aniline.
IR and Raman vibration assignments of 2-(benzylthio)-N-{pyridinylmethylidene}anilines.
B3PW91/LanZ2DZ suitable level of theory for structure and vibration calculation of 2-(benzylthio)-N-{pyridinylmethylidene}anilines.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 June 2012
Received in revised form 15 August 2012
Accepted 15 August 2012
Available online 1 September 2012
2-(Benzylthio)-N {pyridine-4-ylmethylidene}aniline was prepared by reaction of S-benzyl ortho-amino-
thiophenol with 4-pyridine carboxaldehyde and characterized by NMR, IR and Raman spectroscopy
and mass spectrometry. The structure and the vibrational analysis of the series of 2-, 3-, and 4-pyridine
derivatives was performed based on a comparative computational methodology study with the density
functionals B3LYP and B3PW91 and the basis sets LanZ2DZ and 6-31++G(d,p). Comparison of computa-
tional results with single crystal X-ray diffraction results of 2-(benzylthio)-N {pyridine-3-ylmethylid-
ene}aniline allowed the evaluation of structure predictions and confirmed B3PW91/6-31++G(d,p) as
most accurate for structure determination of the four investigated levels of theories. B3LYP and
B3PW91 with the LanL2DZ basis set consistently outperformed calculations for IR and Raman vibrational
estimations when compared to level of theories using the 6-31++G(d,p) basis set. Application of scaling
factors for IR and Raman frequency predictions showed excellent agreement with experimental values
and supported the assignment of the major contributors of the vibration modes of the three pyridine pen-
dant compounds. Overall, B3PW91/LanL2DZ level of theory showed best performance in accuracy and
low computational cost for structural and vibrational analysis for the series of 2-(benzylthio)-N-
{pyridinylmethylidene}anilines.
Keywords:
Synthesis
Schiff base
Vibrational analysis
Crystal structure
DFT
FTIR
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
non-innocent ligand redox properties as found, e.g., in thiolate
complexes [11–14]. Review articles on electron transfers in Cu con-
Metalloproteins with predominantly nitrogenAmetal bonds
through histidine and sulfurAmetal bonds through methionine
and cysteine amino acids [1] side groups typically complex to Zn
and Cu centers, such as the class of zinc-finger proteins [2,3], cyti-
dine deaminase [4] and type I blue copper proteins [5–9]. Nitrogen
and sulfur donor atoms show a rich diversity in coordination
chemistry for their hard and soft binding ability, respectively
[10]. The large diffuse orbitals of S allow single electron transfers
taining proteins are available by Rorabacher [15], Farver and Pecht
[16] and Warren et al. [17]. Recently, we have become interested in
the non-blue/type II copper protein peptidylglycine
ing monooxygenase (PHM), which is one of the two non-coupled
copper ion domains of the bifunctional peptidylglycine -amidat-
a-hydroxylat-
a
ing monooxygenase protein (PAM, EC 1.14.17.3). In contrast to
type I copper complexes containing histidine and cysteine Cu coor-
dination, type II complexes have histidine and methionine coordi-
nation. PHM is involved in the post-translational modification in
the penultimate step of the biosynthetic pathway of amidated
peptide hormones that has been found in mammals [18], insects
[19] and cneridians [20]. Using kinetic isotope effects (KIEs),
forming localized, stable
p radical ions which can have additional
⇑
Corresponding author. Tel.: +1 902 563 1391.
E-mail address: Matthias_Bierenstiel@cbu.ca (M. Bierenstiel).
0022-2860/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2012.08.026

30
J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
PHM displayed values in excess of the semi-classical limitation for
CAH cleavage (11 Å) coupled with non-Arrhenius behavior [21,22].
The modified Marcus model for ground state hydrogen tunneling
was studied using KIEs with an N-acylglycine library supporting
a ‘pre-organization’ model of the substrate–enzyme complex
allowing both hydrogen and deuterium particle transfer through
a decrease in inter-nuclear sampling with tighter bound substrates
[23]. PHM shows structural and mechanistic similarities to dopa-
mine b-monooxygenase (DbM, EC 1.14.17.1) and both contain
non-coupled dicopper active sites that catalyze dioxygen-, copper-
and ascorbate-dependent stereospecific hydroxylation reactions
[23–25]. One Cu active site copper site contains two His and one
Met amino acid side groups and is involved in O2 activation and
substrate oxidation, while the other Cu center has three His bond-
ing environment and is involved with electron transfer reactions.
Palaniandavar presented a good literature overview on diversity
of synthetic models of blue copper proteins involving thioether,
imidazole, benzimidazole and pyridine functional groups [5]. We
disseminated earlier the synthesis of a series of N,S-containing li-
gand systems with the ortho-aminothiophenol motif which is re-
dox active and exhibits non-innocent ligand properties [26]. In
addition, these N,S-ligands can be used to mimic His and Met res-
idues and serve, complexed to a Cu center, as small transition me-
tal models for the analysis of active sites such as PHM. The
challenge of developing these small, biomimetic transition metal
complex systems lies in finding suitable complexes that uphold
the integrity of the active site in its structure and preserve its
chemical activity while minimizing ‘‘protein clutter’’, i.e. the large
peptide backbone that is not directly involved in the protein activ-
ity but necessary for formation of entatic states [27]. Therefore,
having good computational models predicting the structures and
properties of these N,S-ligands and their corresponding transition
metal complexes will assist the analysis and evaluation of active
sites which have been analyzed by structural X-ray based methods
and vibrational (IR and Raman) methods and also will allow the
efficient computational modeling for screening of Cu complexes
for biomimetic enzyme studies.
dimensions 25 m ꢁ 0.25 mm ꢁ 0.25
lm. FTIR spectra were re-
corded on a Thermo Nicolet 6700 FTIR Spectrometer with 4 cmꢂ1
resolution as KBr pellet (approximately 10 mg compound in
300 mg anhydrous KBr) in the 4000–400 cmꢂ1 range. Experimental
FTIR intensities were assigned for strong (s) and weak (w). FT-
Raman spectra were recorded on a Thermo-Electron NXR 9650
FT-Raman Spectrometer with low powered laser intensity and a
4 cmꢂ1 resolution with solid sample in the range of 4000–
400 cmꢂ1. Experimental Raman activities were assigned for strong
(s) and weak (w). MS data was obtained with an Agilent 6890
instrument equipped with mass spectrometer under EI mode at
70 eV. Melting point was recorded on a ThermoScientific Mel-
Temp 1101D instrument and was uncorrected. Elemental analysis
was performed by M-H-W Laboratories, Phoenix, Arizona, USA.
2.2. Synthesis of 2-(benzylthio)-N-{pyridin-4-ylmethylidene}aniline
(4)
In a 250 mL flask equipped with a Dean–Stark trap and con-
denser, S-benzyl ortho-aminothiophenol (1) (212 mg, 0.99 mmol)
was dissolved in 80 mL toluene. 4-Pyridine carboxaldehyde
(93 lL, 0.99 mmol) was added via syringe. The reaction mixture
was heated at reflux for 7 days until water droplets were observed
in the Dean–Stark trap. After cooling to ambient temperature, the
solvent was removed under reduced pressure and the residue
was dissolved in a minimal amount of hot toluene (approx.
10 mL). Then, 30 mL hexane was added to produce a yellow precip-
itate. Precipitation was further forced by cooling the solution to
ꢂ20 °C for 12 h. The obtained yellow solid was filtered and dried
in vacuo (140 mg, 46%). m.p. 175–177 °C. Anal. Calcd. for
C19H16N2S: C = 74.97%, H = 5.30%. Found: C = 74.59%, H = 5.39%.
EI-MS: m/z (%): 304(29) [M+], 271(54), 213(70), 109(27), 91(100).
For 1H and 13C signal assignments see Fig. 1; 1H NMR (400 MHz,
CDCl3): d = 4.14, (s, 2H, CH2APh), 6.99 (m, 1H CH(15)), 7.19 (m,
2H, CH(6) and CH(13)), 7.28 (m, 2H, CH(4) and CH(14)), 7.34 (m,
2H, CH(5) and CH(7)), 7.77 (d, (J = 5.6 Hz), 2H CH(10)), 8.32 (s,
1H, CH imine), 8.74 (d, J = 5.6 Hz, 2H, CH(11)). 13C{1H} NMR
(100.6 MHz, CDCl3): d = 36.9 (CH2APh), 117.6 (CH(15)), 122.3
(CH(10)), 126.4 (CH(6)), 127.1 (CH(13)), 127.2 (CH(14)), 127.8
(CH(7)), 128. 5 (CH(4)), 128.9 (CH(5)), 132.5 (C(12)), 136.9 (C(2)),
142.8 (C(9)), 149.4 (C(3)), 150.5 (CH imine), 157.7 (CH(8)). IR
Herein, we report the synthesis and characterization of 2-
(benzylthio)-N-{pyridinyl-4-methylidene}aniline (4) and the
evaluation of computational methodology of the series of
N-methyl-imine pendant 2-(benzylthio)-N-{pyridinylmethylid-
ene}anilines (2–4) for structural and vibrational analysis. The
determination of the frequencies and the intensities of the IR and
Raman vibrations is of interest as the three compounds can be used
as planar N,S-biomimetic ligands in transition metal complex mod-
els of active sites of enzymes, in particular Cu protein complexes.
(KBr)
m
(cmꢂ1): 3057 (w), 3027 (w), 2922 (w), 2890 (w), 2860
(w), 1623 (s, C@N), 1596 (s), 1572 (w), 1553 (w), 1493 (s), 1463
(w), 1453 (s), 1435 (w), 1408 (s), 1321 (w), 1285 (w), 1268 (s),
1237 (w), 1182 (w), 1158 (w), 1067 (w), 1040 (w), 1026 (w),
1000 (w), 962 (w), 808 (s), 781 (s), 759 (s), 733 (s), 715 (s), 699
(s), 664 (w), 641 (w), 556 (w), 545 (w), 500 (s), 481 (s). Raman (so-
lid)
m
(cmꢂ1): 3059 (w), 1623 (s, C@N), 1597 (s), 1573 (s), 1556 (s),
2. Experimental section
1463 (w), 1287 (s), 1270 (w), 1245 (w), 1228 (w), 1203 (w), 1183
(s), 1160 (w), 1040 (w), 1002 (w), 988 (w).
2.1. General
All solvents and chemicals (reagent grade; technical grade
(ꢀ90%) for ortho-aminothiophenol) were purchased from commer-
cial sources (Sigma–Aldrich and Fisher Scientific) and were used
without further purification. S-benzyl ortho-aminothiophenol (1)
is commercially available and was identical to the produced com-
pound using S-alkylation of ortho-aminothiophenol with benzyl
chloride in methanolic solution [26]. 2-(Benzylthio)-N-{pyridin-2-
ylmethylidene}aniline (2) and 2-(benzylthio)-N-{pyridin-3-ylme-
thylidene}aniline (3) were synthesized as previously reported [26].
NMR spectra were recorded on a 400 MHz Bruker Avance II
spectrometer operating at 400.17 MHz for 1H and 100.6 MHz for
13C nuclei and referenced to Me4Si as d = 0 ppm. J values are given
in Hz. GC analysis was performed on a Varian GC-450 equipped
with a ICB-5 produced by J&K Scientific service number 6026 with
2.3. X-ray structure determination
Crystals of 3 were grown via slow diffusion of n-pentane into a
dichloromethane solution of the compound. Data for compound 3
was collected using a Bruker APEX II CCD detector/D8 diffractom-
eter [28] using Mo K
a radiation, with the crystal cooled to
ꢂ100 °C. The data was corrected for absorption through use of a
multi-scan model. The structure was solved using the direct meth-
ods program SHELXD [29]. Refinements were completed using the
program SHELXL-97 [30]. Hydrogen atoms were assigned positions
based on the idealized sp2 or sp3 hybridization geometries of their
attached carbon or nitrogen atoms, and were given thermal param-
eters 20% greater than those of their parent atoms. ORTEP-III by
Burnett and Johnson was used for crystal structure visualization.

J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
31
2
Fig. 1. Assignment of the chemical shifts of 1H and 13C of compound 4.
The Listing of crystallographic experimental data is found in Tables
1 and 2 and Supplementary data.
compound structures and calculated frequencies as well as intensi-
ties and activities for IR and Raman are available in the Supplemen-
tary data document. Frequencies are scaled by factors 0.961, 0.957,
0.964 and 0.960 for B3LYP/LanL2DZ, B3PW91/LanL2DZ, B3LYP/6-
31++G(d,p) and B3PW91/6-31++G(d,p), respectively, as recom-
mended by the National Institute of Standards and Technology
[38]. The absence of imaginary frequencies was confirmed.
2.4. Computational methods
Density functional theory (DFT) calculations for geometry opti-
mizations and vibrational frequencies determination were per-
formed using the hybrid three-parameter exchange functional of
Becke [31,32], (B3) in combination with the gradient-corrected
correlation functionals of Lee, Yang and Parr [33], (LYP) and Perdew
et al. [34] (PW91). The 6-31++G(d,p) [35] and LanL2DZ [36] basis
sets were used in conjunction with both functionals, forming the
B3LYP/LanL2DZ, B3PW91/LanL2DZ, B3LYP/6-31++G(d,p) and
B3PW91/6-31++G(d,p) levels of theory, respectively, using the
Gaussian G03 program [37]. The Cartesian coordinates of the
3. Results and discussion
3.1. Synthesis and characterization of 2-(benzylthio)-N-{pyridinyl-4-
methylidene}aniline (4)
2-(Benzylthio)-N-{pyridinylmethylidene}anilines 2–4 are read-
ily obtained in a Schiff base condensation reaction by reaction of
the aniline NH2 group with the respective pyridine carboxaldehyde
reagent (Scheme 1) and yields of more than 95% of the crude prod-
ucts were observed in reaction monitoring by thin layer chroma-
tography and 1H NMR analysis of the crude reaction mixture
[24]. Compounds 2 and 3 were isolated and purified by recrystalli-
zation from hot MeOH. However, the same synthetic methodology
was unsuccessful in preparing the 4-pyridine derivative (4), which
was instead synthesized by the azeotropic removal of water in
refluxing toluene for 7 days. Attempts to recrystallize 4 in alcoholic
Table 1
Crystallographic experimental details for 3.
Data
Formula weight
304.40
Crystal dimensions (mm)
Crystal system
0.36 ꢁ 0.31 ꢁ 0.11
Monoclinic
Space group
C2/c (no. 15)
i
solutions (MeOH, EtOH and PrOH) resulted in significant loss of
Unit cell parametersa
product by cleavage of imine bond and formation of S-benzyl
ortho-aminothiophenol (1) likely caused by imine bond activation
in the para-substituted pyridine. Alternatively, flash chromato-
graphic techniques on silica were unsuccessful as the acidic SiO2
triggered cleavage of imine bond and product always contained
small amounts (about 5% as determined in 1H NMR analysis) of
1. Recrystallization from hot toluene and forced precipitation by
addition of non-polar hexanes was the most suitable method for
the isolation and purification of 4, isolated as a yellow colored solid
in 46% yield.
Based on the detailed NMR analysis of 2 and 3 [26], the charac-
terization of compound 4 was also obtained from chemical shifts
and coupling constants from a series of 1H, 13C{1H}, jmod, COSY,
HSQC and HMBC NMR experiments (Fig. 1). The characteristic sin-
glet 1H signal of the imine function (H8) is observed at 8.32 ppm.
The pyridine 1H chemical shifts of protons H10 and H11 are found
at 7.77 ppm and 8.74 ppm, respectively. The chemical shifts of the
1H and 13C signals are matching with the resonances observed in
compounds 2 and 3 [26]. Mass spectrometric analysis produced
the molecular ion signal at 304 m/z. The base signal is 91 m/z
and is characteristic of tropylium cation derived by CAS bond
cleavage of the benzyl group as well as the C@N bond cleavage
resulting in the [C6H5N]+ fragment. Corresponding to base signal
is fragment 213 m/z which is obtained by the other major frag-
ments of the cleaved products of CAS and C@N bonds. Fragment
271 m/z can be explained by loss of HSꢀ in a fragmentation rear-
rangement mechanism of the molecular ion.
a (Å)
b (Å)
c (Å)
° (deg)
V (Å3)
Z
11.7591 (9)
7.9649 (6)
33.186 (3)
97.0755 (9)
3084.5 (4)
8
1.311
0.207
qcalcd (g cmꢂ3
)
l
(mmꢂ1
)
Diffractometer
Radiation (k [Å])
Bruker PLATFORM/APEX II CCDb
Graphite-monochromated Mo K
(0.71073)
a
Temperature (°C)
Scan type
Data collection 2h limit (°)
Total data collected
–100
x
scans (0.3°) (15 s exposures)
52.86
11890 (ꢂ14 6 h 6 14, ꢂ9 6 k 6 9,
ꢂ41 6 l 6 41
Independent reflections
Number of observed reflections (NO)
3157 (Rint = 0.0230)
2866 ½F2o ꢃ 2
r
ðF2o Þꢄ
Structure solution method
Refinement method
Direct methods (SHELXDc) [23]
Full-matrix least-squares on F2
(SHELXL–97) [24]
Absorption correction method
Range of transmission factors
Data/restraints/parameters
Goodness-of-fit (S)c [all data]
Gaussian integration (face-indexed)
0.9775–0.9289
3157/0/199
1.309
a
Obtained from least-squares refinement of 9368 reflections with
4.94° < 2h < 52.74°.
b
Programs for diffractometer operation, data collection, data reduction and
absorption correction were those supplied by Bruker.
P
2
1=2
S ¼ ½ wðF2o ꢂ F2c Þ =ðn ꢂ pÞꢄ
(n = number of data; p = number of parameters
c
varied; w ¼ ½r2ðFo2Þ þ ð0:0382PÞ þ 8:4639Pꢄꢂ1 where P ¼ ½MaxðFo2; 0Þ þ 2F2c ꢄ=3Þ.
2

32
J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
Table 2
Comparison of selected geometrical parameters at four levels of theory with experimental values from the X-ray structure of compound 3 for the evaluation of structure
predictions.
Exp.
LanL2DZ
B3LYP
6-31++G(d,p)
B3LYP
X-ray
Errora (%)
Errorb (%)
B3PW91
Errorb (%)
Errorb (%)
B3PW91
Errorb (%)
Selected bond distances (Å)
C1AS
C10AS
C2AN1
C20AN1
C21AN2
C25AN2
1.817
1.767
1.413
1.272
1.337
1.343
0.34
0.33
0.42
0.47
0.45
0.45
1.855
1.936
1.422
1.296
1.353
1.361
2.09
9.56
0.64
1.89
1.19
1.34
1.845
1.920
1.416
1.294
1.350
1.358
1.54
8.66
0.21
1.73
0.97
1.12
1.798
1.865
1.408
1.279
1.335
1.343
ꢂ1.05
5.55
ꢂ0.35
0.55
ꢂ0.15
0
1.787
1.850
1.403
1.278
1.332
1.340
ꢂ1.65
4.70
ꢂ0.70
0.47
ꢂ0.37
ꢂ0.22
Selected bond angles (°)
C1ASAC10
SAC10AC11
C2AN1AC20
N1AC20AC22
C21AN2AC25
102.2
0.06
0.06
0.05
0.05
0.05
98.5
110.3
122.0
122.1
117.9
ꢂ3.62
2.99
1.58
0.98
1.51
98.3
110.0
121.7
122.0
117.8
ꢂ3.82
2.74
1.29
0.97
1.42
100.2
110.2
119.7
122.9
117.6
ꢂ1.96
2.87
99.9
109.9
119.4
122.4
117.5
ꢂ2.25
2.62
107.1
120.1
120.9
116.2
ꢂ0.37
ꢂ0.58
1.48
1.23
1.19
1.09
Selected torsion angles (°)
C1ASAC10AC11
175.0
0.03
0.03
0.02
176.2
ꢂ178.3
ꢂ174.7
0.69
ꢂ0.06
ꢂ41.1
178.6
ꢂ179.4
179.7
2.04
0.53
45.15
177.1
ꢂ178.3
177.5
1.22
1.83
ꢂ47.4
177.9
ꢂ178.5
ꢂ173.0
1.66
0.06
ꢂ39.7
C2AN1AC20AC22
C6AC1AC11AC12
ꢂ178.4
123.8
a
Error in X-ray diffraction converted to % for comparison with computational data.
Error calculated from (calculated value ꢂ experimental value)/(experimental value).
b
2
2
Scheme 1. Synthesis of 2-,3- and 4-substituted pyridinylmethylidene derivatives of S-benzylortho-aminothiophenol.
3.2. Structure analysis
others [40]. The B3LYP functional is most popular density func-
tional method and can provide reliable predictions on the struc-
tures and vibrational frequencies [41,42]. However, the less
widely used B3PW91 functional can yield slightly better results
than B3LYP [43,44], especially in cases where there are large elec-
tronegativity differences between bonded atoms, such as with
imine and pyridine nitrogen functional groups. The basis sets
LanZ2DZ and 6-31++G(d,p) were chosen as they are (i) flexible
Suitable X-ray crystallographic crystals of the 3-pyridinyl deriv-
ative (3) were obtained by slow diffusion of n-pentane into a
dichloromethane solution at room temperature and the ORTEP plot
is shown in Fig. 2. Unfortunately, suitable crystals for X-ray analy-
sis could not be obtained for compounds 2 and 4. The details of
X-ray structure analysis data are given in Table 1 and selected
geometrical parameters are listed in Table 2. The crystal structure
of 3 has an expected planar geometry of the conjugated pyridine–
imine–thioaniline moiety which is measured by the dihedral angle
C2AN1AC20AC22 of ꢂ178.4°. The C@N imine bond also possesses
the thermodynamically favoured trans-configuration that was pre-
viously predicted [26]. The thioether group (C1ASAC10) has an an-
gle of 102.2° and the phenyl group is placed most distant from the
thioaniline ring with almost linear dihedral angle (ꢂ175.0°) of
C1ASAC10AC11. The phenyl ring is tilted away, approximately
perpendicular, from the conjugated pyridine–imine–thioaniline
moiety, by 123.8° in the C6AC1AC11AC12 dihedral angle. The
bond distances of SACaromatic (C1AS) and SACaliphatic (C10AS) are
in the expected ranges with 1.817 Å and 1.767 Å, respectively.
The imine C@N bond length (C20AN1) is 1.272 Å. There are no
unexpected irregularities with bond distances and angles when
they were compared with close related compounds reported by
Ariza-Castolo and co-worker of pyridinyl aldimines [39].
For assignment of the IR and Raman vibrational modes of the
three titular compounds (2–4), computational methods were em-
ployed. Hartree–Fock, Møller–Plesset, and local density models
yield poor results for vibrational frequency and intensity calcula-
tions, and among the gradient-corrected and hybrid density func-
tionals, none stands out as particularly better or worse than the
Fig. 2. Perspective view of the 2-(benzylthio)-N-{pyridin-3-ylmethylidene}aniline
(3) showing the atom labeling scheme. Non-hydrogen atoms are represented by
Gaussian ellipsoids at the 20% probability level. Hydrogen atoms are shown with
arbitrarily small thermal parameters.

J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
33
2
3
4
Fig. 3. Comparative views (‘‘top view’’ and ‘‘side view’’ parallel and perpendicular of ortho-aminothiophenol ring) of predicted compound structures (2–4) using B3PW91/
LanL2DZ.
enough to account for a heterogeneous charge distribution; (ii)
small enough to not be prohibitively expensive when studying
large molecules; and (iii) accommodate metal atoms for future
work when studying transition metal complexes with 2–4 as li-
gands. The expanded LanL2DZ basis set, which uses effective core
potentials for all atoms larger than Ne allowing for treatment of
relativistic effects while reducing computational expense, was
compared to the more commonly reported 6-31++G(d,p) basis set
for organic molecules. One particular advantage of LanL2DZ is
the modeling of metals beyond the 3d elements, while 6-
31++G(d,p) is limited to elements up to Kr. LanL2DZ in conjunction
with hybrid DFT functionals is well suited to predict heats of for-
mation and ionization potentials [45] and bond dissociation ener-
gies [46]. 6-31++G(d,p) allows the modeling of compounds and
modifies calculations for diffuse orbitals, observed with the thioe-
ther functional group, and polarization functions with strongly po-
lar covalent bonds that occur with imine and pyridine nitrogens.
The availability of the X-ray crystal structure of 3 allowed the
comparison of the structural results of computational methods
with experimental data and the evaluation of accuracies. However,
we note that the comparison is between a solid-state experimental
structure, recorded at 173 K, and an ideal gas theoretical structure
at 0 K. Bond distances and angles involving CAH and C@C bonds
are uncharacteristic for structure evaluations but are available in
the Supplementary data. The results of bond distance calculations
for all four methods are in good agreement with the X-ray crystal
structure, with B3PW91/6-31++G(d,p) having a slightly better
accuracy among the four methods. On using B3LYP and B3PW91
with LanL2DZ, one consistent problem is that the bond lengths
are consistently overestimated by up to 2% for CAN and C@N
bonds and 4–6.5% for CAS bonds. In contrast, the bond distance
calculations using the 6-31++G(d,p) basis set have no clear trend
in comparison to experimental X-ray crystallographic data. The
typical error range for the 6-31++G(d,p) basis set is acceptable
and lies between 3% from experimental values. The bond angle
calculations using the LanL2DZ underestimate the C1ASAC10 bond
angle at the order of ꢂ3% while the SAC10AC11 angle is overesti-
mated by 2.6%. The angles of bonds involving nitrogen are slightly
overestimated by up to +1.6%. Calculation results using the 6-
31++G(d,p) basis set show no distinct preferential trend in bond
angle determination. 6-31++G(d,p) is accurate to the same order
of magnitude as LanL2DZ, but errors are slightly smaller. All three
aromatic rings are planar and consistent within all computational
methods. For dihedral angle analysis, we only looked at three
structurally relevant values of the conjugated pyridine–imine–thi-
oaniline moiety (C2AN1AC20AC22), the thioether functional
group (C1ASAC10AC11) and the tilting of the phenyl group from
the thioaniline ring (C6AC1AC11AC12). The calculated results of
C2AN1AC20AC22 and C1ASAC10AC11 are accurate and only dif-
fer by up to 2% and are of same order of magnitude as predicted
bond distances and bond angles. The phenyl group is rather tilted
to 123.8° and deviates significantly from calculated values by 40–
47%. Recalculations of all four methods with a starting phenyl tilt
of 123.8° converged again to same torsion angles as listed in Table
2. Thus the unusual conformation is likely based on packing effects
in the crystal lattice as calculations are conducted in gas phase
without intermolecular interactions.
To visually compare the predicted structures of compounds
2–4, a ‘‘top view’’ and ‘‘side view’’ structure of each compound is
shown in Fig. 3 for the B3PW91/LanL2DZ model, which is in good
agreement of structures calculated with three other models (not
shown). In all cases, the imine group is in trans configuration which
is more thermodynamically stable than the cis isomer. Most dom-
inant is the tilting of the pyridine pendant ring against the aniline–
imine moiety. In the 2-pyridine compound (2), the ortho position of
the pyridine nitrogen is in alignment with the imine nitrogen,
probably due to secondary electron interactions whereas in 3 and
4 with a more distant pyridine an increased tilt is observed. All
other angles and distances are in good agreement in particular
the bond angle of the thioether function which is observed in its
typical 98–100° range.
Overall, all four DFT computational methods are suitable for
predicting the structure of the compound 3 with experimental
X-ray crystallographic results. There are minor preferences for
6-31G++(d,p) basis set over LanL2DZ but the advantage in slightly
improved accuracy is lessened by increased computational ex-
pense and a future limiting issue of elements up to Kr, i.e. only
3d row elements when investigating transition metal coordination
complexes. B3LYP and B3PW91 are both acceptable levels of
theory, with the latter being slightly more accurate. We therefore

34
J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
Table 3
Comparison of
m
(C@N) at four levels of theory with experimental values of the IR and Raman spectra for 2, 3 and 4.
Comp.
Experimental
LanL2DZ
B3LYPa
6-31++G(d,p)
Scaled B3LYPb
Scaled B3LYPb
B3PW91a
Scaled B3PW91b
B3LYPa
B3PW91a
Scaled B3PW91b
IR (cmꢂ1
1621
1623
)
2
3
4
1699 (4.81%)
1683 (3.69%)
1683 (3.70%)
1633 (0.74%)
1618 (ꢂ0.31%)
1617 (ꢂ0.37%)
1712 (5.61%)
1700 (4.74%)
1698 (4.62%)
1638 (1.05%)
1627 (0.25%)
1625 (0.12%)
1727 (6.54%)
1713 (5.55%)
1713 (5.55%)
1666 (2.76%)
1651 (1.73%)
1651 (1.73%)
1739 (7.28%)
1727 (6.41%)
1726 (6.35%)
1669 (2.96%)
1658 (2.16%)
1657 (2.10%)
1623
Raman (cmꢂ1
1620
1621
)
2
3
4
1699 (4.87%)
1683 (3.82%)
1683 (3.63%)
1633 (0.80%)
1618 (ꢂ0.19%)
1617 (ꢂ0.43%)
1712 (5.68%)
1700 (4.87%)
1698 (4.62%)
1638 (1.11%)
1627 (0.37%)
1625 (0.06%)
1727 (6.60%)
1713 (5.67%)
1713 (5.54%)
1666 (2.77%)
1651 (1.85%)
1651 (1.66%)
1739 (7.35%)
1727 (6.54%)
1726 (6.34%)
1669 (3.02%)
1658 (2.28%)
1657 (2.03%)
1624
a
Frequency (cmꢂ1) and error calculated from (calculated value ꢂ experimental value)/(experimental value).
b
Scale values: 0.961 for B3LYP/LanL2DZ; 0.957 for B3PW91/LanL2DZ; 0.964 for B3LYP/6-31++G(d,p); 0.960 for B3PW91/6-31++G(d,p). Frequency (cmꢂ1) and error
calculated from (calculated scaled value ꢂ experimental value)/(experimental value).
choose B3PW91/6-31++G(d,p) as the best suited method for struc-
ture determination, with a recommendation for B3PW91/LanL2DZ
in cases where the 6-31++G(d,p) basis set is not possible.
1590–1690 cmꢂ1 range [47]. The calculated vibrational IR and Ra-
man spectra of all four different DFT methods showed a reliable
agreement with experimental values although the frequencies
were consistently overestimated by 3–8%. Such systematic fre-
quency overestimation [38] is known for IR and Raman calcula-
tions using DFT methods, so to obtain a better picture of the
accuracy of individual modes, the calculated frequencies were ad-
justed using the scaling factors recommended by the National
Institute of Science and Technology; 0.961 for B3LYP/LanL2DZ;
0.957 for B3PW91/LanL2DZ; 0.964 for B3LYP/6-31++G(d,p); 0.960
for B3PW91/6-31++G(d,p). Katsyubu and co-workers [48] de-
scribed ab initio (MP2) and DFT (B3LYP, M05, M05-2X) predictions
for IR intensities and Raman activities and found that of these,
B3LYP offers the most cost-effective choice for the prediction of
molecular vibrational properties. Accurate IR intensities can be ob-
tained with modest basis sets, whereas medium-sized sets were
needed for accurate calculations of Raman activities [48]. Our re-
sults show that the errors of the calculated values for both IR
and Raman frequencies were matching within the DFT methods
for each of the compounds (2–4); in particular, the scaled fre-
quency values are in the 5–30 cmꢂ1, or 0.25 to approximately 3%,
3.3. Vibrational analysis
The computational study was extended to vibrational spectros-
copy for frequency as well as intensity (IR) and activity (Raman)
analysis in order to support the assignment of experimental values
of the vibration bands for 2–4. The vibrational analysis was con-
ducted by frequency calculations of each the geometry optimized
structures (Tables S8–S19 in Supplementary data). No imaginary
frequencies were found thus eliminating saddle points in the po-
tential hyperenergy surface. As the vibrational modes of larger
molecules, such as 2–4, can be highly complex, we focused first
on major bond interactions, i.e. the dominant and characteristic
m(C@N) vibrational band which is active in both the IR and Raman
modes (Table 3). The most characteristic band for each of the
compounds is the C@N imine bond stretch found in the range of
1620–1624 cmꢂ1 which is in congruent with typical observed
range of imine vibrations for related aryl–pyrdinyl imines in the
Table 4
Assignment of vibration modes, scaleda frequencies (cmꢂ1), relative and absolute intensities (IR) or activities (Raman) for four levels of theory compared to experimental values
for 2 in the range of 1700–400 cmꢂ1
.
LanL2DZ
B3LYP
6-31++G(d,p)
B3LYP
B3PW91
B3PW91
Assignment of major contributors to IR vibration mode
(C@N)
Sym. stretch of CAC in thioaniline ring
CAH rock in thioaniline and pyridine rings
CAH rock in thioaniline and pyridine rings
m
1633 (100%, 121)
1572 (29%, 35)
1441 (40%, 48)
1432 (27%, 33)
1162 (17%, 21)
1059 (5%, 6)
1638 (100%, 112)
1577 (30%, 34)
1443(41%, 46)
1435(30%, 34)
1167(20%, 22)
1061 (7%, 8)
1666 (100%,141)
1573 (31%, 59)
1438 (42%, 59)
1433 (2%, 2)
1169 (20%, 28)
1058 (4%, 6)
1014 (4%, 5)
845 (4%, 5)
1669 (100%, 129)
1580 (32%, 41)
1437 (46%, 59)
1431 (6%, 5)
1172 (17%, 22)
1057 (5%, 6)
1014 (5%, 6)
849 (4%, 5)
m
m
m
(CaromaticANimine) and sym. stretch of CAC in thioaniline and pyrdine rings
(CaromaticAS) and sym. CAC stretch thioaniline
(CaromaticAS) and CAH twist pyridine ring
1006 (7%, 8)
844 (22%, 27)
797 (8%, 10)
1013 (6%, 7)
845 (22%, 25)
799 (7%, 8)
C@N twist and CAH twist in thioaniline ring
d(CaromaticAS)
796 (1%, 2)
795 (2%, 2)
CAH wag in pyridine ring
782 (37%, 45)
761 (73%, 88)
744 (21%, 26)
701 (44%, 53)
668 (11%, 13)
630 (17%, 21)
781 (38%, 43)
760 (76%, 85)
741(31%, 35)
697(51%, 57)
668 (12%, 13)
640 (13%, 14)
765 (24%, 34)
759 (7%, 10)
742 (38%, 54)
685 (31%, 44)
662 (5%, 7)
765 (16%, 20)
762 (12%, 18)
739 (41%, 53)
683 (41%, 53)
672 (1%, 1)
CAH wag in thioaniline ring
CAH twist out-of-plane in pyridine ring
CAH wag in phenyl ring
m
m
(CaromaticAS) and CAC wag of thioaniline ring
(CaliphaticAS)
647 (4%, 5)
644 (5%, 6)
Assignment of major contributors to Raman vibration mode
(C@N)
Sym. stretch of CAC in thioaniline ring
Sym. stretch of CAC in pyridine ring
m
1633 (100%, 1171)
1567 (14%, 166)
1556 (36%, 418)
1162 (16%, 183)
1638 (100%, 1638)
1577 (15%, 207)
1564 (38%, 532)
1167 (20%, 271)
1666 (100%, 1025)
1560 (14%, 70)
1570 (44%, 450)
1169 (33%, 340)
1669 (100%, 1143)
1580 (13%, 152)
1577 (42%, 477)
1172 (33%, 381)
m
(CaromaticANimine) and sym. stretch of CAC in thioaniline and pyridine rings
a
Scale values: 0.961 for B3LYP/LanL2DZ; 0.957 for B3PW91/LanL2DZ; 0.964 for B3LYP/6-31++G(d,p); 0.960 for B3PW91/6-31++G(d,p). Relative intensity compared to
(C@N). Calculated IR intensity (km/mol). Calculated Raman activity (A4/amu).
m

J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
35
Table 5
Assignment of vibration modes, scaleda frequencies (cmꢂ1), relative and absolute intensities (IR) or activities (Raman) for four levels of theory compared to experimental values
for 3 in the range of 1700–400 cmꢂ1
.
LanL2DZ
B3LYP
6-31++G(d,p)
B3LYP
B3PW91
B3PW91
Assignment of major contributors to IR vibration mode
(C@N)
Sym. stretch of CAC in thioaniline and pyridine rings
Sym. stretch of CAC in pyridine ring
CAH rock in pyridine ring
m
1618 (100%, 141)
1563 (28%, 39)
1534 (24%, 34)
1443 (36%, 51)
1424 (6%, 8)
1161 (21%, 30)
1026 (8%, 11)
1006 (9%, 12)
834 (28%, 40)
769 (16%, 23)
764 (58%, 82)
712 (29%, 41)
698 (37%, 52)
665 (6%, 9)
1627 (100%, 141)
1572 (29%, 41)
1544 (23%, 32)
1445 (38%, 54)
1424 (5%, 7)
1167 (28%, 39)
1030 (13%, 9)
1010 (9%, 12)
835 (30%, 42)
768 (14%, 20)
763 (59%, 83)
710 (31%, 44)
696 (41%, 58)
664 (6%, 8)
1651 (100%, 167)
1569 (28%, 47)
1555 (23%, 38)
1456 (17%, 28)
1436 (20%, 34)
1170 (37%, 61)
1036 (10%, 6)
1013 (3%, 5)
1658 (100%, 166)
1576 (29%, 48)
1562 (22%36)
1454 (19%, 31)
1435 (19%, 32)
1172 (33%, 55)
1035 (12%, 7)
1014 (4%, 6)
CAH wag in phenyl ring
CAC wag in thioaniline and pyridine rings
m
m
(CaromaticAS) and sym. CAC stretch phenyl ring
(CaromaticAS) and sym. CAC stretch thioaniline
C@N twist and CAH twist in thioaniline ring
d (CaromaticAS)
CAH wag in thioaniline ring
CAH wag in pyridine ring
CAH wag in phenyl ring
843 (13%, 22)
759 (5%, 8)
745 (34%, 56)
692 (18%, 30)
683 (27%, 45)
672 (5%, 9)
840 (12%, 20)
762 (4%, 7)
741 (34%, 56)
690 (18%, 30)
682 (32%, 53)
673 (4%, 6)
m
m
(CaromaticAS) and CAC wag of thioaniline ring
(CaliphaticAS)
631 (15%, 21)
638 (8%, 11)
644 (4%, 6)
640 (4%, 6)
Assignment of major contributors to Raman vibration mode
(C@N)
Sym. stretch of CAC in thioaniline ring
Sym. stretch of CAC in pyridine ring
m
1618 (100%, 1259)
1571 (2%, 19)
1563 (72%, 902)
1161 (20%, 250)
1627 (100%,1405)
1580 (1%, 12)
1572 (72%, 1009)
1167 (24%, 344)
1651 (100%, 998)
1581 (28%, 279)
1569 (50%, 500)
1170 (35%, 348)
1658 (100%, 1067)
1598 (28%, 296)
1576 (47%, 497)
1172 (24%,258)
m
(CaromaticANimine) and sym. stretch of CAC in thioaniline and pyridine rings
a
Scale values: 0.961 for B3LYP/LanL2DZ; 0.957 for B3PW91/LanL2DZ; 0.964 for B3LYP/6-31++G(d,p); 0.960 for B3PW91/6-31++G(d,p). Relative intensity compared to
(C@N). Calculated IR intensity (km/mol). Calculated Raman activity (A4/amu).
m
respectively, when compared to experimental values. The LanL2DZ
basis set consistently outperformed the 6-31++G(d,p) basis set
with raw data and scaled data by a margin on the order of one
third. Our results agree with Katsyubu’s finding showing B3LYP
as a good method for IR and Raman frequency predictions. B3LYP
and B3PW91 levels of theory are almost identical with predictions
of frequencies and on average B3LYP slightly out performs
B3PW91, although this trend is too small to be significant.
typical range of 3050–2850 cmꢂ1 are found in the respective re-
gions of 2–4 and are too uncharacteristic, hence not listed in Tables
4–6. The scaled frequencies were normalized for relative intensities
(IR) and activities (Raman) compared to the largest vibration band,
i.e. m(C@N). The IR intensities were listed to a minimum threshold
of 20% relative intensity; the Raman activities result threshold was
10% as there are fewer Raman signals compared to IR. This
relatively conservative categorization allowed the comparison of
strong to weak experimental values to the calculation results based
on intensity. The calculated vibrational modes are visualized using
the WebMO program interface and major contributors to the
With the good agreement of m(C@N) modes in IR and Raman, we
then proceeded to analyze the vibration spectra in other regions.
The CAH stretching bands for aliphatic and aromatic carbons in
Table 6
Assignment of vibration modes, scaleda frequencies (cmꢂ1), relative and absolute intensities (IR) or activities (Raman) for four levels of theory compared to experimental values
for 4 in the range of 1700–400 cmꢂ1
.
LanL2DZ
B3LYP
6-31++G(d,p)
B3LYP
B3PW91
B3PW91
Assignment of major contributors to IR vibration mode
(C@N)
Sym. stretch of CAC in thioaniline and pyridine rings
Sym. stretch of CAC in pyridine ring
CAH rock in pyridine ring
m
1617 (100%, 99)
1625 (100%, 101)
1578 (49%, 49)
1544 (27%, 27)
1433 (38%, 38)
1651 (100%, 128)
1582 (30%, 38)
1547 (23%, 29)
1657 (100%, 122)
1589 (34%, 41)
1557 (24%, 29)
1569 (36%, 36)
1534 (27%, 27)
1429 (40%, 40)
b
b
–
–
b
b
CAH rock in thioaniline ring
CAC wag in thioaniline and pyridine rings
–
–
1436 (29%, 37)
1169 (38%, 48)
1036 (9%, 11)
803 (25%, 32)
760 (7%, 9)
1436 (36%, 40)
1172 (33%, 40)
1036 (10%, 12)
832 (29%, 35)
762 (6%, 7)
1161 (30%, 30)
1026 (13%, 13)
826 (52%, 51)
770 (23%, 23)
765 (82%, 81)
698 (52%, 51)
660 (11%, 11)
635 (14%, 14)
480 (21%, 21)
1166 (27%, 27)
1029 (14%, 14)
823 (50%, 50)
770 (20%, 20)
763 (81%, 80)
696 (55%, 54)
660 (10%, 10)
643 (22%, 22)
479 (24%, 24)
m
(CaromaticAS) and sym. CAC stretch thioaniline
C@N twist and CAH twist in pyridine ring
CAH twist in phenyl ring
CAH wag in thioaniline ring
CAH wag in phenyl ring
745 (43%, 55)
684 (34%, 43)
669 (8%, 10)
742 (45%, 52)
683 (43%, 52)
669 (6%, 7)
m
m
(CaromaticAS) and CAC wag of thioaniline ring
(CaliphaticAS)
b
b
–
–
CAC wag of thioaniline and phenyl rings
478 (12%, 15)
476 (15%, 18)
Assignment of major contributors to Raman vibration mode
m
(C@N)
1617 (100%, 1335)
1564 (20%, 270)
1630 (100%, 1368)
1579 (57%, 778)
1651 (100%, 1025)
1582 (27%, 275)
1570 (35%, 357)
1169 (41%, 419)
1657 (100%, 1095)
1589 (25%, 269)
1577 (35%, 381)
1172 (40%, 437)
Sym. stretch of CAC in thioaniline ring
Sym. stretch of CAC in pyridine ring
m
b
b
–
–
(CaromaticANimine) and sym. stretch of CAC in thioaniline and pyridine rings
1161 (30%, 1161)
1166 (24%, 334)
a
Scale values: 0.961 for B3LYP/LanL2DZ; 0.957 for B3PW91/LanL2DZ; 0.964 for B3LYP/6-31++G(d,p); 0.960 for B3PW91/6-31++G(d,p). Relative intensity compared to
m
(C@N). Calculated IR intensity (km/mol). Calculated Raman activity (A4/amu).
b
Unable to assign.

36
J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
molecule vibration were assigned. The focus was on significant
vibrations, in particular involvement of S and N heteroatoms, as
such vibrations will be shifted once 2–4 are complexed to transi-
ligands and comparison of experimental vibrational data of the
active sites of proteins. Furthermore, B3PW91/LanL2DZ is a suit-
able level of theory for theoretical compound model development
of S-thioether, N-imine and N-pyridine based ligand designs for
transition metal complexes such as type II Cu proteins and even
transition metals beyond 3d elements.
tion metals. The
m(SACaliphatic) and m(SACaromatic) are typically ob-
served in the regions of 790–630 cmꢂ1 and 1100–1080 cmꢂ1
,
respectively. Due to accuracy and complexity of vibration modes,
we cannot assign with unambiguous confidence the exact mode
of the experimental vibrations although we have high confidence
in the assignments of the strong bands. The similarity of the three
derivatives (2–4) is fitting with similarly assigned major vibrations
of IR and Raman frequencies. The largest vibration bands in IR spec-
tra of 2–4, besides the dominant C@N stretch vibrations, are the
CAH rocking and CAH wagging vibrations of the three aromatic
subunits, i.e. for 2 of B3LYP/LanL2DZ 1441,1432, 844, 782, 761
and 744 cmꢂ1, which can be assigned to the experimental values
of 1473, 1433, 874, 777, 736 and 714 cmꢂ1, respectively. The
assignments of vibrations across the four levels of theory are very
consistent throughout all compounds and values within each set
of basis set more similar. Nevertheless, the scaled values of
B3LYP and B3PW91 with the 6-31++G(d,p) basis set tend to be
more overestimated. The stretching or deformation vibrations
[47] of the thioether function, found in the ranges of 1000–
1070 cmꢂ1 and 630–660 cmꢂ1 assigned are of interest as those will
be affected once a metal is bound though a sulfur lone pair; how-
ever, these vibrations are typically weak with about less than one
Acknowledgments
We thank NSERC, Canada Foundation for Innovation, NSRIT,
Enterprise Cape Breton Corporation and CBU for financial support
of this research. B.M. McLellan is grateful to NSERC for an USRA
award. Special thanks to Dr. C.D. Keefe for support of FTIR and
FT-Raman spectrometry.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molstruc.2012.08.
026.
References
[1] E. Farkas, I. Sovago, Amino Acids, Peptides and Proteins, vol. 36, The Royal
Society of Chemistry, 2007.
[2] S.V. Razin, V.V. Borunova, O.G. Maksimenko, O.L. Kantidze, Biochemistry
(Moscow) 77 (2012) 217–226.
fourth of the m(C@N).
The Raman vibrations of 2–4 are less abundant than the com-
plementary IR vibrations and only four very active bands have
been determined as most of the calculated values are far below
the 5% threshold. B3LYP/LanL2DZ calculations resulted for com-
pound 2 for major frequencies of 1633 (C@N), 1567, 1556 and
1142 cmꢂ1, which excellently correspond to the four major exper-
[3] J.L. Michalek, A.N. Besold, S.L. Michel, J. Dalton Trans. 40 (2011) 12619–12632.
[4] S. Vincenzetti, F.M. Carpi, A. Vita, Curr. Top. Pept. Protein Res. 10 (2009) 1–22.
[5] M. Palaniandavar, Proc. Indian Natl. Sci. Acad. Part A 70 (2004) 293–310.
[6] O. Siiman, N.M. Young, P.R. Carey, J. Am. Chem. Soc. 98 (1976) 744–748.
[7] E.I. Solomon, R.G. Hadt, Coord. Chem. Rev. 255 (2011) 774–789.
[8] M. Malarvizhi, P.R. Athappan, Mater. Sci. Forum 699 (2011) 231–244.
[9] T.D. Wilson, Y. Yu, Y. Lu, Coord. Chem. Rev., in press, doi: http://dx.doi.org/
10.1016/j.ccr.2012.06.015.
[10] R.G. Pearson, Coord. Chem. Rev. 100 (1990) 403–425.
[11] N. Roy, S. Sproules, T. Weyhermüller, K. Wieghardt, Inorg. Chem. 48 (2009)
3783–3791.
[12] K. Ray, T. Weyhermüller, F. Neese, K. Wieghardt, Inorg. Chem. 44 (2005) 5345–
5360.
imental vibrations of 2 at 1620 (C@N), 1585, 1562 and 1194 cmꢂ1
.
These are the CAC stretching modes of the aromatic rings that
shown large changes in polarizability and observed in all three
derivatives and hence small deviations. All other vibrations are
far below the 5% threshold margin and were not assigned.
[13] P. Ghosh, E. Bill, T. Weyhermüller, F. Neese, K. Wieghardt, J. Am. Chem. Soc.
125 (2003) 1293–1308.
[14] P. Ghosh, A. Begum, E. Bill, T. Weyhermüller, K. Wieghardt, Inorg. Chem. 42
(2003) 3208–3215.
[15] D.B. Rorabacher, Chem. Rev. 104 (2004) 651–697.
4. Conclusion
[16] O. Farver, I. Pecht, Coord. Chem. Rev. 255 (2011) 757–773.
[17] J.J. Warren, K.M. Lancaster, J.H. Richards, H.B. Gray, J. Inorg. Biochem., in press,
doi: http://dx.doi.org/10.1016/j.jinorgbio.2012.05.002.
[18] B.A. Eipper, R.E. Mains, Annu. Rev. Physiol. 50 (1988) 333–344.
[19] Vanden. Broeck, J. Peptides 22 (2001) 241–254.
[20] C.J. Grimmelikhuijzen, I. Leviev, K. Carstensen, Int. Rev. Cytol. 167 (1996) 37–
89.
[21] W.A. Francisco, M.J. Knapp, N.J. Blackburn, J.P. Klinman, J. Am. Chem. Soc. 124
(2002) 8194–8195.
[22] W.A. Francisco, G. Wille, A.J. Smith, D.J. Merkler, J.P. Klinman, J. Am. Chem. Soc.
126 (2004) 13168–13619.
[23] N.R. McIntyre, E.W. Lowe, J.L. Belof, M. Ivkovic, J. Shafer, B. Space, D.J. Merkler,
J. Am. Chem. Soc. 132 (2010) 16393–16402.
[24] J.P.J. Klinman, Biol. Chem. 281 (2006) 3013–3016.
[25] N.R. McIntyre, E.W. Lowe, D.J. Merkler, J. Am. Chem. Soc. 131 (2009) 10308–
10319.
[26] E.D. Cross, U.A. Shehzad, S.M. Lloy, A.R.C. Brown, T.D. Mercer, D.R. Foster, B.L.
McLellan, A.R. Murray, M.A. English, M. Bierenstiel, Synthesis (2011) 303–317.
[27] P. Comba, Coord. Chem. Rev. 200–202 (2000) 217–245.
[28] Programs for Diffractometer Operation, u.c.i., Data Collection, Data Reduction
and Absorption Correction were those Supplied by Bruker.
[29] T.R. Schneider, G.M. Sheldrick, Acta Crystallogr. D58 (2002) 1772–1779.
[30] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112–122.
[31] A.D.J. Becke, Chem. Phys. 98 (1993) 1372.
The density functionals, B3LYP and B3PW91, with the basis sets
LanZ2DZ and 6-31++G(d,p) were compared in a computational
methodology study with experimentally available X-ray crystal
structure, FTIR and FT-Raman data for a series of 2-(benzylthio)-
N-{pyridinylmethylidene}anilines. 2-(benzylthio)-N-{pyridinyl-4-
methylidene}aniline (4) was synthesized by condensation reaction
from S-benzyl ortho-aminothiophenol with 4-pyrdine carboxalde-
hyde and X-ray structure of 2-(benzylthio)-N-{pyridinyl-3-methyl-
idene}aniline (3) was obtained. B3PW91/6-31++G(d,p) slightly
outperformed the other three level of theories for prediction of
molecular structure when compared to available X-ray structure
of 3. In general, all four levels of theories had the least accuracies
for bond lengths and bond angles involving the sulfur atom. The
LanL2DZ basis set is comparable to the 6-31++G(d,p) basis set to
produce good estimates for structure analysis. However, for vibra-
tional mode estimates LanL2DZ outperforms 6-31++G(d,p) within
less than 3% error range when the appropriate scaling factors are
used. Judging the effectiveness of the four investigated methods,
B3PW91/LanL2DZ holds an edge for structural and vibrational
accuracy as well for least computational cost. In addition, 2-(ben-
zylthio)-N-{pyridinylmethylidene}anilines are potential N,S-ligand
models for biomimetic protein studies involving His and Met coor-
dination. The assignments of major contributors to the IR and
Raman vibration bands of the three titular compounds assist in
vibrational analysis of transition metal complexes with these
[32] A.D.J. Becke, Chem. Phys. 98 (1993) 5648.
[33] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[34] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C.
Fiolhas, Phys. Rev. B 46 (1992) 6671.
[35] R. Krishnan, J.S. Binkley, R. Seeger, J.A.J. Pople, Chem. Phys. 72 (1980) 650.
[36] P. Hay, W.R.J. Wadt, Chem. Phys. 82 (1985) 270.
[37] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,

J. Alberto Acosta-Ramirez et al. / Journal of Molecular Structure 1034 (2013) 29–37
37
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,
[39] R. Montalvo-Gonzalez, A.J. Ariza-Castolo, Mol. Struct. 655 (2003) 375–389.
[40] W.J. Hehre, Guide to Molecular Mechanics and Quantum Chemical
H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E.
Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y.
Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J.
Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03 Revision D.03; Gaussian, Inc., Wallingford, CT, 2004.
[38] R.D.I. Johnson (Ed.), 2011, vol. NIST Standard Reference Database Number 101,
NIST Computational Chemistry Comparison and Benchmark Database, Release
15b, August 2011, <http://cccbdb.nist.gov/ed>.
A
Calculations, Wavefunction Inc., Irvine, CA, 2003.
[41] S.F. Sousa, P.A. Fernandes, M.J.J. Ramos, Phys. Chem. A 111 (2007) 10439.
[42] C.J. Cramer, D.G. Truhlar, Phys. Chem. Chem. Phys. 11 (2009) 10757.
[43] J.M. Martell, L.A. Eriksson, J.D.J. Goddard, Phys. Chem. A 1997 (1927) 101.
[44] J.M. Martell, P.T. Beaton, B.E.J. Holmes, Phys. Chem. A 106 (2002) 8471.
[45] Y. Yang, M.N. Weaver, K.M. Merz, J. Phys. Chem. A. 113 (2009) 9843–9851.
[46] X. Xu, D.G. Truhlar, J. Chem. Theory Comput. 8 (2012) 80–90.
[47] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identification of
Organic Compounds, fourth ed., John Wiley and Sons, New York, NY, 1981.
[48] E.E. Zvereva, A.R. Shagidullin, S.A.J. Katsyuba, Phys. Chem. A 115 (2011) 63–69.