Vibrational spectroscopic studies of N1-ethyl-5′-bromo-7-azaindirubin-3′-oxime and N1-ethyl-indirubin-3′-monooxime

Article abstract of DOI:10.1016/j.molstruc.2015.01.036

We have prepared N1-ethyl-5′-bromo-7-azaindirubin-3′-oxime due to its potential for being a pharmaceutical. Infrared and Raman spectra have been recorded and vibrational assignments have been suggested based mainly on our previous vibrational investigation of N1-isopropyl-5′-chloro-7-azaindirubin-3′-oxime and on group characteristic frequencies. Temperature variation study has revealed the presence of conformers due to the internal rotation of ethyl group. IR spectra collected for N1-ethyl-7-azaindirubin-3′-oxime have shown rather similar spectral features with that of N1-ethyl-5′-bromo-7-azaindirubin-3′-oxime. IR spectra of these compounds have revealed the association through hydrogen bonding in the solid state. IR spectra recorded for these samples after annealing at high temperatures indicated the thermal conversion temperature to be lowered than 270 °C. Results from thermal analyses have determined the beginning decomposition temperatures to be 250°C and the decomposition enthalpies to be 94 kJ/mol for both N1-ethyl-5′-bromo-7-azaindirubin-3′-oxime and N1-ethyl-7-azaindirubin-3′-oxime.

Full text of DOI:10.1016/j.molstruc.2015.01.036

Journal of Molecular Structure 1087 (2015) 51–59
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Vibrational spectroscopic studies of
N1-ethyl-50-bromo-7-azaindirubin-30-oxime
and N1-ethyl-indirubin-30-monooxime
b
b
c
a
Ying-Sing Li a, , Qi-Zheng Yao , Zhao-Hui Wang , Jingcai Cheng , Tuyen Thi T. Truong
⇑
a Department of Chemistry, University of Memphis, Memphis, TN 38152, USA
b School of Pharmacy, China Pharmaceutical University, Nanjing 210009, China
c JC (Wuxi) COMPANY, Inc., Wuxi, JS 214036, China
h i g h l i g h t s
ꢀ
ꢀ
ꢀ
ꢀ
The title compounds were synthesized and identified.
These compounds were studied by vibrational spectroscopy.
Vibrational assignments were suggested based on the previous work and the group characteristic frequencies.
Thermal analyses were conducted by DSC and temperature variation of IR spectra.
a r t i c l e i n f o
a b s t r a c t
We have prepared N1-ethyl-50-bromo-7-azaindirubin-30-oxime due to its potential for being a pharmaceu-
tical. Infrared and Raman spectra have been recorded and vibrational assignments have been suggested
based mainly on our previous vibrational investigation of N1-isopropyl-50-chloro-7-azaindirubin-30-oxime
and on group characteristic frequencies. Temperature variation study has revealed the presence of
conformers due to the internal rotation of ethyl group. IR spectra collected for N1-ethyl-7-azaindirubin-
30-oxime have shown rather similar spectral features with that of N1-ethyl-50-bromo-7-azaindirubin-
30-oxime. IR spectra of these compounds have revealed the association through hydrogen bonding in the
solid state. IR spectra recorded for these samples after annealing at high temperatures indicated the thermal
conversion temperature to be lowered than 270 °C. Results from thermal analyses have determined the
beginning decomposition temperatures to be 250 °C and the decomposition enthalpies to be 94 kJ/mol
for both N1-ethyl-50-bromo-7-azaindirubin-30-oxime and N1-ethyl-7-azaindirubin-30-oxime.
Ó 2015 Elsevier B.V. All rights reserved.
Article history:
Received 12 November 2014
Received in revised form 15 January 2015
Accepted 20 January 2015
Available online 29 January 2015
Keywords:
N1-ethyl-50-bromo-7-azaindirubin-
30-oxime
N1-ethyl-indirubin-30-monooxime
Indirubin
Raman spectra
IR spectra
Thermal stability
Introduction
have potent inhibitory potential toward CDKs [4–6]. Histological
and transmission electron microscopic (TEM) investigations indi-
Indirubin is an active ingredient in a traditional Chinese herbal
medicine, used in the treatment of chronic myelogenous leukemia.
It was reported that the compound could be helpful for treating glio-
blastoma – a deadly brain tumors [1]. It has also been found to exhi-
bit antileukemic effectiveness in chronic myelocytic leukemia [2].
The crystal structure of cyclin-dependent kinases-2 (CDK-2) in com-
plex with indirubin derivatives shows that indirubin interacts with
the kinase’s ATP-binding site through van der Waals interactions
and three hydrogen bonds [3]. Due to the high affinity with the
enzymes ATP binding site, indirubin and some of its derivatives
cate that indirubin-3-monoxime has antitumor effect [7].
Perpete et al. have evaluated the UV/Vis spectra of a series of
indirubin, isoindigo and other indigo/thio related dyes by using
time-dependent Density Function Theory (DFT) in conjunction
with the polarizable continuum model [8]. Results from the elec-
trochemical and IR spectroscopic studies on the interaction of
indirubin with DNA show that indirubin interacts with phosphate
group of DNA by hydrogen bond or electrostatic interaction [9].
Karapanayiotis et al. have considered the influence of various hal-
ogen on the Raman spectrum in their Raman spectroscopic studies
of indigo and its four 6,60-halogeno analogues [10].
The purpose of the present work is to study the vibrational
spectra of indirubin derivatives and their thermal stabilities. It is
⇑
Corresponding author. Tel.: +1 901 678 2621; fax: +1 901 678 3447.
E-mail address: yingli@memphis.edu (Y.-S. Li).
http://dx.doi.org/10.1016/j.molstruc.2015.01.036
0022-2860/Ó 2015 Elsevier B.V. All rights reserved.

52
Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
Scheme 1. Structural formulas of (a) EIM, (b) indirubin and (c) EBAIO.
expected that the vibrational spectroscopy can provide useful
information about the characteristics and stabilities of the active
component or its derivatives of a drug. It has recently been
reviewed that Raman scattering is a useful spectroscopic method
in the study of pharmaceuticals [11]. Due to the potential in phar-
maceutical applications, we had recorded the surface enhanced
Raman scattering (SERS) of N1-ethyl indirubin-30-monooxime
(EIM) by using a silver coated microscopic slide pre-treated with
indium tin oxide nanopowder [12]. The use of the oxide nanopow-
er for the pretreatment was to improve the surface morphology of
the product substrate for better Raman signal enhancement. From
the observed SERS bands and their vibrational assignments, it is
unlikely that the molecules absorbed horizontally on silver surface
Step 1: Synthesis of EBAI
To
a solution of N1-ethyl-7-azaindole-2,3-dione (0.22 g,
1.23 mmol) and 1-acetyl-3-hydroxy-5-bromoindole (0.31 g,
1.23 mmol) in ethanol (20 mL), several drops of concentrated
hydrochloric acid were added until pH = 1–2 [14]. The mixture
was stirred at 60–65 °C for 1 h [15]. The reaction mixture was
cooled down and filtered, and the filtrate was distilled to remove
the ethanol. The obtained residue was dissolved with chloroform
and the solution was washed with water for several times, and con-
centrated to afford EBAI. The crude product was then purified by
column chromatography (200–300 mesh silica gel; chloroform/ligr-
oin, 3:1, v/v) followed by recrystallization in acetone to get 0.24 g
red needle-like crystal powder (yield: 52%); mp: 249–251 °C(des.).
IR (KBr, í, cmꢁ1): 3432, 3319, 3139, 1654, 1594, 1456, 1400,
1384, 1205, 1101, 970; 1H nuclear magnetic resonance (NMR)
(300 MHz, CDCl3, ppm) d: 1.37 (t, 3H, J = 7.14 Hz, CH3), 4.03 (q,
2H, J = 7.14 Hz, NACH2), 6.93 (d, 1H, J = 8.46 Hz, 70-H), 7.05 (dd,
1H, J = 5.21, 7.68 Hz, 5-H), 7.63 (dd, 1H, J = 2.03, 8.46 Hz, 60-H),
7.85 (d, 1H, J = 2.03 Hz, 40-H), 8.21 (dd, 1H, J = 1.57, 5.21 Hz, 6-H),
8.97 (dd, 1H, J = 1.57, 7.68 Hz, 4-H), 10.43 (s, 1H, N0AH); ESI-MS
(m/z) [M+H]+: 370.0 (Br = 79), 372.0 (Br = 81). Anal. Calcd. (%) for
through aromatic
p-electrons. Result from this study suggested
that EIM interacted with the substrate through the lone pair elec-
trons of the oxime functional group. Our recent vibrational spec-
troscopic investigation of N1-isopropyl-50-chloro-7-azaindirubin-
30-oxime (ICAIO) revealed that the molecule has a planar ring
structure with a bisecting isopropyl functional group that includes
an eclipsed secondary CAH (CI2ACI2H) bond with the N1AC bond.
Vibrational assignments were suggested based essentially on the
results from DFT calculations [13]. As a continuation of our inves-
tigation of indirubin derivations, we have carried out the vibra-
tional spectroscopic study on N1-ethyl-50-bromo-7-azaindirubin-
30-oxime (EBAIO). As there is a close similarity of molecular struc-
ture between ICAIO and EBAIO, it would be reasonable to suggest
the vibrational assignments, based essentially on the vibrational
characteristics of ICAIO. Additionally, we have recorded infrared
(IR) spectra of EIM and carried out thermal analyses for ICAIO,
EBAIO and EIM. Scheme 1 shows the molecular structural formulas
of EBAIO, EIM and indirubin. The analogue of molecular structure
between EBAIO and EIM might suggest some similarities in their
vibration spectral features.
C
17H12BrN3O2 (369.2): C, 55.16; H, 3.27; N, 11.35; Found (%): C,
55.24; H, 3.33; N, 11.27.
Step 2: Synthesis of EBAIO
Experimental
Unless noted, all solvents and reagents were freshly distilled or
purified according to standard procedures. Indirubin was pur-
chased from Nanjing Zelang Pharmaceutical Technology Co. Ltd.;
it was purified by crystallization in acetone.
Melting points were taken on XT-4 micro melting point appara-
tus and are uncorrected. Analytical TLC was carried out with plates
pre-coated with silica gel 60 F254 (0.25 mm thick) and visualized
under UV lamp (254 nM). Flash column chromatography was car-
ried out on silica-gel 200–300 mesh.
0.3 g of hydroxylamine hydrochloride and 0.6 g KOH were
added to
a solution of 0.24 g EBAI (0.64 mmol) in ethanol
(10 mL). The reaction mixture was stirred at room temperature
for 0.5 h, and then poured into water. The mixture was neutralized
with glacial acetic acid and the formed precipitate was filtrated
and washed with water to provide 0.24 g EBAIO (yield: 98%), an
orange crystal powder; mp: 288–289 °C(des.). 1H NMR
(300 MHz, CDCl3, ppm) d:1.25 (t, 3H, J = 6.87 Hz, CH3), 3.93 (q,
2H, J = 6.87 Hz, NACH2), 7.03 (dd, 1H, J = 5.26, 7.59 Hz, 5-H), 7.44
(d, 1H, J = 8.39 Hz, 70-H), 7.62 (d, 1H, J = 8.39 Hz, 60-H), 8.10 (d,
1H, J = 5.26 Hz, 6-H), 8.31 (s, 1H, 40-H), 8.79 (d, 1H, J = 7.59 Hz,
Preparation of EBAIO(c)
There are two major steps in the preparation of EBAIO. The first
step is involved with the synthesis of N1-ethyl-50-bromo-7-
azaindirubin (EBAI) from the reaction between N1-ethyl-7-azain-
dole-2,3-dione and 1-acetyl-3-hydroxy-5-bromoindole.The second
step is the oximation of EBAI. The experimental detail for the sam-
ple preparation is given in the following:

Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
53
4-H), 11.74 (s, 1H, N0-H), 13.94 (s, 1H, NOH); ESI-MS m/z [MꢁH]-:
Infrared (IR) spectra were performed on a Nicolet Avitar 370 DTGS
infrared spectrophotometer; NMR spectra were recorded on Bruke
AV-300 (300 MHz) NMR spectrometer using trimethylsilane (TMS)
as an internal reference. The chemical shifts were reported in ppm
(d), and the coupling constants (J) were given in Hertz (Hz). Mass
spectra (MS) were collected on Agilent 1100 LC/MS ESI (70–
100 eV) mass spectrometer. Elemental analyses were carried out
on an Elementar Vario EL instrument (Heraeus GmbH, Hanau,
Germany).
For spectroscopic investigation, IR spectra were recorded in
either adsorption or transmission mode by using a Matson Polaris
FTIR spectrophotometer equipped with room temperature DTGS
detector and WinFirst spectroscopy software. The IR sample was
prepared by pressing each sample with KBr powder (IR grade,
Sigma–Aldrich, St. Louis, MO, USA) to form pellets. The relative
amount of sample was 3% by weight. For the low temperature
investigation, a liquid nitrogen IR cell described in the previous
study was applied [13]. A resolution of 2 cmꢁ1 and a total of 96
scans were collected for each IR spectrum. Raman spectra were col-
lected with a spectrophotometric system that consisted of a Spex
Model 1403 double monochromator, a Melles Griot Omnischrome
43 argon ion laser, and a Hamamatsu Model R-928 photomultiplier
tube. The 488 nm laser line was used as the excitation light source
and the solid sample was filled in a capillary tube for Raman scat-
tering measurements. In the Raman data collection, the resolution
was set at 4 cmꢁ1 and the laser power at the sample was 5–10 mW.
DSC plots were collected with a TA Q20 Differential Scanning
Calorimeter (DSC). The system was purged with pure nitrogen at
50 mL/min and the heating rate was normally set at 5 °C/min.
383.1 (Br = 79), 385.1 (Br = 81). Anal. Calcd. (%) for C17H13BrN4O2.
1
H2O (389.72): C, 52.39; H, 3.49; N, 14.39. Found (%): C, 51.89;
4
H, 3.73; N, 13.86.
Preparation of EIM(a)
There are two major steps in the preparation of EIM. The first
step is involved with the synthesis of N1-ethylindirubin (EI) from
the reaction between indirubin and ethyl iodide. The second step
is the preparation of EIM via the oximation of EI.
Step 1: Synthesis of EI
2.62 g (10 mmol) of indirubin were treated with 0.6 g
(15 mmol) of NaH at 0 °C in 40 mL of anhydrous N,N-dimethyl-
formamide (DMF), and subsequently 1.15 mL(15 mmol) of ethylio-
dide (EtI) in 5 mL anhydrous DMF were slowly added drop wise.
The resulting solution was stirred at room temperature. After the
reaction was completed (TLC checking petroleum ether/ethyl ace-
tate = 3/1), the reaction mixture was poured into water (ꢂ300 mL).
The product was extracted with dichloromethane three times
(150 mL ꢃ 3) and the combined organic layers were dried over
Na2SO4. The concentrated crude product was purified by silica-
gel column chromatography with petroleum ether/ethylacetate
(4/1) as eluent to afford 1.5 g of EI (yield: 52%), an purple crystal
powder; mp: >200 °C(des.); 1H NMR(300 MHz, CDCl3, ppm) d:
1.33(t, J = 7.25 Hz, 3H, CH3), 3.89(q, J = 7.24 Hz, 2 H, CH2), 6.87–
6.90 (m, 1H, H-Ph), 6.94–6.98 (m, 1H, H-Ph), 7.00–7.02 (m, 1H,
H-Ph), 7.11–7.13 (m, 1H, H-Ph), 7.27–7.29 (m, 1H, H-Ph), 7.52 (d,
J = 7.60 Hz, 1H, H-Ph), 7.74(d, J = 7.60 Hz, 1H, H-Ph), 8.90(d,
J = 7.60 Hz, 1H, H-Ph), 10.56(bs, 1H, NH-1); ESI-MS (m/z):291.2
[M+H]+; Anal. Calcd (%) for C18H14N2O2(290.32): C 74.47, H 4.86,
N 9.65; Found (%): C 74.33, H 4.85, N 9.41.
Results and discussion
IR and Raman spectra of EBAIO
Our previous theoretical calculation has revealed that the most
stable stereo isomer of ICAIO is in ZE form [13], in which the ring
system linkage and the 30-oxime functional group have cis (Z)
and trans (E) forms, respectively. As EBAIO and ICAIO have identi-
cal ring system with different halogen substitutions at C50 and dif-
ferent alkyl groups at N1, it is likely that the most stable
stereoisomer of EBAIO is in ZE form. Similar to ICAIO, EBAIO is
expected to have a highly conjugate ring system with all the heavy
atoms in the rings and in the oxime and carbonyl groups sharing
the same plane. This could be consistent with the potential func-
tion governing the twisted motion of the indoline and the pyrolid-
inone ring with respect to the C3@C20. In case that the potential
function is double minima, the ground vibrational state must have
energy level higher than the barrier to the twist. Depending on the
internal rotation of the ethyl group, EBAIO is expected to have
either Cs or C1 point group. In either case, all vibrational modes
are both IR and Raman active.
Step 2: Synthesis of EIM
Fig. 1 shows the FTIR and Raman spectra of EBAIO. In Fig. 1-B, a
distinct Raman band at 1571 cmꢁ1 has a much higher intensity
than the rest of the Raman bands in the frequency region. This
band is attributed to the C3@C20 stretching vibration. The corre-
sponding assignment for ICAIO has a very strong Raman band at
1563 cmꢁ1 [13]. Due to the absence of the symmetric point of
inversion in EBAIO, the corresponding IR band is not expected to
be inactive. This can be evidenced from Fig. 1-A, in which the
observed IR band at 1572 cmꢁ1 contributed by the same vibration
mode is rather strong. This observation is similar to that of ICAIO,
which has a strong IR band at 1568 cmꢁ1 due to the absence of Ci
symmetry [13]. The very intense central C@C stretch Raman bands
have also been observed and assigned for indigoid dye [16], indigo
[17–20], Maya blue [21–24], isoindigo [25], thioindigo and
The preparation process of EIM is similar to that of BEAIO. Using
1.2 g of EI (2.9 mmol) as starting material, 0.65 g EIM were
obtained (yield 75%) as an orange crystal powder; mp:
>200 °C(des.); 1H NMR(300 MHz, CD3COCD3, ppm) d: 1.28 (t,
J = 7.25 Hz, 3H, CH3), 3.94 (q, J = 7.25 Hz, 2 H, CH2), 6.96–6.99 (m,
1H, H-Ph), 7.05–7.08 (m, 2H, H-Ph), 7.08–7.11 (m, 1H, H-Ph),
7.19–7.22 (m, 1H, H-Ph), 7.33–7.36 (m, 1H, H-Ph), 7.43–7.46(m,
1H, H-Ph), 8.38 (d, J = 7.82 Hz, 1H, H-Ph), 8.71 (d, J = 7.82 Hz, 1H,
H-Ph), 10.56 (bs, 1H, NH-1); ESI-MS (m/z): 304.1 [MꢁH]ꢁ; Anal.
Calcd (%) for C18H15N3O2 (305.34): C, 70.81; H, 4.95; N, 13.76.
Found (%): C, 70.56; H, 4.89; N, 13.87.
In the above sample preparations, instrumental methods for the
identification of reaction intermediates and products were applied.

54
Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
3500
2500
1500
500
Wavenumber (cm-1
)
Fig. 1. Vibrational spectra of EBAIO: (A) Infrared; (B) Raman.
indirubin [20] in the frequency region from 1557 to 1594 cmꢁ1
.
resonance [33]. For indigo, the intra-hydrogen bonding shifted the
This can be regarded as a characteristic C@C central double bond
stretch. The presence of the symmetric point of inversion in indigo
and isoindigo makes the absence of the corresponding IR bands
due to the activation of exclusion principle. C@O stretch is
observed to have a very intense IR band at 1664 cmꢁ1 for EBAIO.
The corresponding C@O vibration has been assigned to the strong
IR bands of ICAIO [13], indigo [17–19,26,27], 4,4,40,40-tetra-
methyl[2,20-bipyrrolidinylidene]-3,30-dinone [26] and isoindigo
[25] in the frequency range 1630–1730 cmꢁ1. It was pointed out
in a FTIR study of 2-pyrrolidinone and oxindole that the lactam car-
bonyl stretch has lower frequency (1690, 1669 cmꢁ1) in solid than
that in vapor phase (1759, 1772 cmꢁ1) and the frequency decreases
as the ring size increases [28]. All these should be helpful to con-
firm our assignment of the carbonyl stretch IR band of EBAIO at
1664 cmꢁ1, which has a corresponding weak Raman band at
1668 cmꢁ1. The relatively low frequency in solid state might be
caused by the association with the carbonyl group. The stretches
of other double bonds, including the C@C and C@N in the rings
and the C@N in the oxime, are expected to have strong or moderate
IR intensity. These can be easily identified from the FTIR spectrum
observed NAH stretch frequency to 3268 cmꢁ1 [17,18]. The obser-
vation of m
(NAH) for isoindigo at 3648 cmꢁ1 indicated the absence
of hydrogen bonding in the molecule [25]. From the molecular
structural point of view, the separation between NH hydrogen
and carbonyl O could be too far in isoindigo to have any intra-
hydrogen bonding. In case of EBAIO, the presence of a moderate
IR band at 3142 cmꢁ1 and
a very weak Raman band at
3148 cmꢁ1 is an indication of the occurrence of hydrogen bonding.
From the molecular structure of EBAIO, there is a chance of intra
hydrogen bonding of N10H with C2@O. However, the intra-hydro-
gen bonding is not expected to occur for the oxime hydroxyl group
because there is no highly electron negative N or O located in an
appropriate nearby location. The free OAH stretch in oxime group
is expected to have frequency in the frequency range 3500–
3650 cmꢁ1 [30]. In the present study, the absence of any IR band
in the region and the presence of the red-shift OAH stretch at
3241 cmꢁ1 may suggest that the presence of inter-hydrogen bond-
ing in solid EBAIO. Such inter-molecular hydrogen bonding may
also occur with the NAH group. Thus, EBAIO molecules associate
with each other through hydrogen bonding in solid state. Similar
to ICAIO, many observed EBAIO vibrational bands are contributed
by mixing different internal vibrations. For simplicity, only the
major contributors to the vibrational bands are given in the Table.
In addition to the OAH and NAH stretches, CAH stretch is
expected to have vibration in the high frequency region. For both
EBAIO and EIM, one my classify CAH into two different types of
bonding based on the relative amount of p-orbital in the carbon
atom. The first one or type one includes the carbon with ꢂ33% p-
orbital in the conjugate ring system; the second one or type two
comprises the carbon with ꢂ25% p-orbital in the alkyl chain.
Microwave spectroscopic studies have determined the significant
difference of bond length between these types of CAH bond. For
examples, CAH bonds in pyrrole belong to type one and have bond
lengths of 1.076 0.004 Å [34] while CAHs in propane belong to
type two and have bond lengths of 1.096 0.002 Å [35]. From the
experimental bond lengths, one may suggest that the CAH bond
order of type one is slightly higher than that of type two. This is
in agreement with the results of vibrational study, in which of
the CAH stretches of pyrrole in the gaseous state fall in the range
from 3118 to 3149 cmꢁ1 [32] and the corresponding vibrations of
propane have frequencies in the range 2882–2977 cmꢁ1 [36]. It
was also noted that the frequencies of pyrrole in the liquid state
are about 15 cmꢁ1 lower than those in solid [32]. One may thus
expect that the CAH stretch of type one is higher than that of type
two. Our observed FTIR bands at 3100, 3062 and 3020 cmꢁ1 are
assigned to the CAH stretches in the ring. On a similar basis, the
in the frequency region between 1572 cmꢁ1
1664 cmꢁ1
(C@O).
(m(C@C)) and
(m
The observed FTIR and Raman spectral frequencies are listed in
Table 1. The vibrational assignments were suggested based essen-
tially on those of ICAIO [13] except the ethyl group, of which the
assignment is referred to group frequency [29,30]. In the FTIR spec-
trum of EBAIO collected at 24 °C, there was a broad band around
3452 cmꢁ1. This band completely disappeared after baking the
sample at 96 °C overnights. This high frequency broad band must
therefore be contributed by water presented in the pre-baked sam-
ple. Due to the very weak intensity of water Raman band, no water
band could be observed in the recorded Raman spectrum. Exclud-
ing the water band, the high frequency bands observed in FTIR and
Raman spectra are at 3142 and 3148 cmꢁ1, respectively. The spec-
tral frequencies are likely attributed to the NAH stretch in the
presence of hydrogen bonding. The fundamental vibrational analy-
sis of pyrrole in the gaseous phase gave the NAH stretch at
3531 cmꢁ1 [31,32]. From the experimental condition, it is unlikely
that the hydrogen bonding would occur in pyrrole. For pyrroles
and indoles, the characteristic frequency of free
m(NAH) is
expected to be higher than 3400 cmꢁ1; hydrogen bonding will shift
the frequency down to ꢂ3000 cmꢁ1 [30]. In the IR spectroscopic
study of the association of 2-oxoindolines in solutions, the hydro-
gen bonded NAH stretch fell in the frequency range 2800–
3100 cmꢁ1; the extent of red shift from the free NAH stretch
(at 3450 cmꢁ1) depended on the strength of H-bonding and Fermi

Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
55
Table 1
Table 1 (continued)
Observed vibrational frequencies (cmꢁ1) and suggested assignments of N1-ethyl-50-
bromo-7-azaindirubin-30-oxime (EBAIO) and N1-ethyl-50-indirubin-30-monoxime
(EIM).
EBAIO
FTIR
EIM
Raman
Assignments
FTIR
EBAIO
FTIR
EIM
b(ring)ben
546 w
500 w
510 m
509 w
480 w
466 vvw
430 vw
/(C40H,C60H,C70H),
s(OH)
Raman
Assignments
FTIR
480 vw)
467 vw
433 vw
424 vw
418 vw
404 vw
b(ring)pyr/b(ring)ben
/(C4H)
d(CCN1)
/(C4H)
3241 m
3240 vw
3207 vw
3148 vw
3137 vw
3125 vw
3106 vw
3080 w
m
m
(OH)H-bonding
(NH)H-bonding
3260 w
456 vw
3142 m
3143 m
3114 m
419 vvw,sh
400 vvw
367 vw
339 vvw
296 vw
b(ring)ind, b(CNO)
s
(OH)
3100 m
3062 m
3020 w,b
m
m
m
(C4H)pyr
(CH)ring
(C60H,C70H)i
3100 m
3058 w
3044 w
3062 w
3027 vw
2979 vw
2971 vw
2934 w
2918 vw
2904 vw
210 vw
199 vw
164 vw
2975 m
2934 m
m
m
a(CH3)
a(CH2)
2987 m
2939 w
Abbreviations: s, strong; m, moderate; w, weak; v: very; b, broad; sh, shoulder;
ind, indoline; me, methyl; ox, oxime; ben, benzene ring; pyr, pyridine; pyrr, pyr-
rolidinone; ind, indole; et, ethyl group; Z, central ethylene;
m
m
s(CH3)
s(CH2)
2894 vw
2840 m
m
, stretch; d, deforma-
2870 m,sh
2778 m,b
1664 s
tion; b, in-plane bend;
u
, out-of-plane bend; i, in phase; o, out-of-phase; , wag;
x
skel, skeletal;.
1668 w
m
(C@O)
1645 m
1632 m
1610 m,sh
1602 s
1587 s
1572 s
1608 m
1608 m
1597 m
1571 vs
1550 w
1500 w
1472 m
m
m
m
m
(C@C),
(C@C)ind
(C@C), (C@N)ox, or
(C@C)Z
m
(C@N)ox, or
m
m
(C@N)pyr
(C@N)pyr
1614 s
1596 s
1586 s
1564 s
observed bands at 2870, 2934 and 2975 cmꢁ1 are attributed to the
stretches in the ethyl group. The identifications of representation
or symmetry agree with those of characteristic group frequencies
[29,30].
m
d(CH3) or d(CH2)
d(CH3)
1503 w
1472 m
1461 s
Displayed in Fig. 2 are the FTIR spectra of EBAIO after the treat-
ments at different temperatures. The upper figure covers the spec-
tra with the frequency range from 2500 to 3500 cmꢁꢁ1 and the
lower figure from 400 to 2000 cmꢁ1. In both the upper and lower
figures, spectra A, B and C were recorded for samples after the
treatments at 96, 165 and 225 °C, respectively, for three hours.
As there is no spectral variation among these three spectra, the
sample of BEAIO is stable up to 225 °C. Spectrum D was collected
for the sample after baking at 257 °C. Some variations of spectrum
D from spectrum A, B or C in the frequency region from 1550 to
1750 cmꢁ1 indicate that the sample begin to have a chemical con-
version around 257 °C; but the thermal conversion has not yet
completed. Not much change of the spectrum has been observed
when the sample was heated up to 274 °C (not shown in the fig-
ure). Further heating of the sample up to 300 °C or higher shows
a significant change of the spectrum from the one recorded at
low temperature. Consequently, there is a possibility of slow kinet-
ics in the thermal conversion. In the high frequency region, three IR
bands at 2975, 2934 and 2871 cmꢁ1 are clearly observed even after
baking the sample at 300 °C. As these three peaks are attributed to
the vibrations of CAH stretches in the ethyl group, the decomposed
product of EBAIO may still be composed of ethyl group.
To identify the existence of conformer in EBAIO, we have
applied a liquid nitrogen cell to record the FTIR spectra at low tem-
peratures. Depicted in Fig. 3 are the FTIR spectra collected at 24,
ꢁ110 and ꢁ196 °C along with the one recorded at room tempera-
ture after annealing at 140 °C for three hours (see Spectrum A).
Spectrum B was collected with liquid nitrogen cell without anneal-
ing the sample at high temperature. It is seen that these two spec-
tra are very similar in both frequency and relative intensity.
However, in comparison with those of Spectra C and D collected
at ꢁ110 and ꢁ196 °C, respectively, we have noticed a change of rel-
ative intensities for the bands at 598 and 587 cmꢁ1. The relative
intensity of the band at 587 cmꢁ1 increases with lowering the sam-
ple temperature. Consequently, this band is contributed by the
low-energy conformer. The spectral quality does not allow us to
da(CH3)
1480 m
1468 s
1453 s
1435 m
b(C70H,NH)o b(CH,NH)o,
m(CN)ind
1456 s
b(CH,NH)o,
b(C70H, NH)o
m(CN)ind
1440 m,sh
1430w,sh
1404 w
1380 vw
1359 m
1350 m
1317 s
1438 m
1400 m
1350 vw
1319 w
b(CI2H),
ds(CH3)
b(CI2H),b(C60H,NH)o, b(OH)
d(CH)
b(CH,OH)
m
(ring)pyr
1407 w
1388 vw
1365 m
1334 s
1301 w
1288 vw
1280 vw
1230 s
1192 w
1164 m
1150 w
1137 s
1293 w
1260 w
1213 w
1204 w
1150 w
b(C5H,C6H)o,
b(CI2H), (ring)pyr
b(CH)pyr/ (ring)ben
m(ring)pyr
1261 m
1221 s
1190 w
1163 m
m
m
m
(ring)ind,b(C40H,C70H)o, b(OH)
b(C60H,C70H)o, b(NH)
1141 s
b(C60H,C70H)o, b(NH),
b(C4H,C5H)o
b(C60H,C70H), d(CH)et
b(C60H,C70H), d(CH)et
b(C4H, C6H)
b(CH)ben
b(CH)ben
b(CH)ind/b(CH)pyr, b(OH)
m
m
1126 vw
1113 vw
1099 m
1081 vw
1067 w
1058 m
1026 m
982 s
1123 m
1120 w
1092 w
1082 vw
1066 vw
1030 w
998 m
1080 vw
1058 vw
1020 vw
977 w
(NO), b(CH)ben
(ring)pyrr, (NO)
976 w
945 w
940 vw
m
944 vw
937 vw
927 vw
910 vw
910 w
888 w
859 vw
807 w
794 m
780 vw
765 m
755 vw
707 vw
701vw,sh
688 vw
667 w
655 w
631 w
598 w
585 vw
569 vw
558 m
b(NH),
b(NH),
m
m
(ring)
(ring)
858 w
d(CC)et, b(CI2H)
(CC)et
854 vw
m
795 vw
b(C6H,C7H)ind
b(C6H,C7H)ind
/(C4H,C5H,C6H)pyr
u
u
780 w
757 w
746 m
750 w
706 m
ring)ben,
u(ring)pyr
(NH), (C7H)
u
730 w
m
(CBr)
b(ring)pyr
b(ring)ben, b(ring)pyr
u (NH)
698 vw
689 vw
670 vw
657 m
620 vw
594 vw
b(NO), b(ring)ind, b(ring)pyr
b(NH,C7H)out-of-phase
Conformer
629 w
608 w
588 vw
574 vw
556 w
determine
DH between the conformers. Similar conformer bands
were also identified in the temperature variation study of ICAIO
[13]. There are two asymmetric tops in EBAIO. The potential energy
570 vw
555 vw
b(ONC), b(ring)ben

56
Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
Fig. 2. FTIR spectra of EBAIO after annealing for three hours at different temperatures: (A) 96 °C; (B) 165 °C; (C) 225 °C; (D) 257 °C; (E) 300 °C; (F) 325 °C.
Fig. 3. Variation of EBAIO FTIR spectrum with temperature: (A) 140 °C; (B) 24 °C; (C) ꢁ110 °C; (D) ꢁ196 °C.
governing the internal rotation of the oxime hydroxyl group is
expected to be similar with that of ICAIO. Result from the calcula-
tion for ICAIO gave an energy difference of 6.9 kCal/mol between
the conformers with the orientations of 0° and 180° for the hydro-
xyl group. Such a large energy difference gives the population of
the high energy conformer to be 0.001% according to Boltzmann

Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
57
distribution. Thus, it is unlikely that one could identify the confor-
mation due to the internal rotation of hydroxyl group from IR spec-
troscopy. Consequently, the observation of conformer from the
FTIR spectral intensity variation with temperature must arise from
the conformer due to the internal rotation of ethyl group.
for the vibrations listed in Table 1. Around the broad band at
3114 cmꢁ1, two partially resolved bands at 3143 and 3100 cmꢁ1
could be identified; so is the very weak shoulder (3058 cmꢁ1
)
located in the high frequency side of the band at 3044 cmꢁ1. The
one with relatively higher frequency around 3143 cmꢁꢁ1 may arise
from the NAH stretch involving with H-bonding; the other bands
at 3100, 3058 and 3044 cmꢁ1 can be attributed to the CAH
stretches of the rings. From Fig. 4, it is noticeable that three bands
at 2987, 2939 and 2840 cmꢁ1 are forming a pattern similar to that
of EBAIO in Fig. 2. Assignments of these three bands may therefore
be made to asymmetric CH3, asymmetric CH2 and symmetric CH2
stretches, respectively. These assignments are consistent with the
characteristic frequencies expected for the CAH stretches in ethyl
group. Furthermore, a very weak shoulder appears on the high fre-
IR spectra of EIM
EIM has essentially the same ring system as EBAIO except that
the N7 in EBAIO is replaced by carbon and the substitution of C50
bromine in EBAIO by H (see Scheme 1). Thus, the basic ring struc-
ture of EIM is similar to that of EBAIO. Depicted in Fig. 4 are FTIR
spectra of EIM recorded at room temperature and after the thermal
treatments at 96, 170, 257 and 300 °C. As the treatment tempera-
ture has increased up to 257 °C, the recorded spectrum D appears
to be different from those annealed at lower temperature (see
Spectra A, B and C in Fig. 4). Thus one may conclude that EIM
has begun to decompose around 257 °C. We have also collected
the FTIR spectrum of residue sample obtained after heating up to
400 °C in the DSC experiment. The resulted spectrum is the same
as Spectrum E. From the collected IR spectra of samples after the
thermal treatment at different temperatures, the decomposition
temperature of EIM is roughly the same as that of EBAIO. As the
observed IR spectral frequencies of EIM at room temperature are
essentially equivalent with those of EBAIO, it is convenient to list
the frequencies in Table 1 showing the equivalence of vibration
assignments.
quency side of
expected to have a higher frequency than
weak shoulder band at 2894 cmꢁ1 is assigned to
we were unable to observe the corresponding vibration in the IR
spectrum of EBAIO, very weak EBAIO Raman band at
m
s(CH2). As symmetric CH3 stretch is generally
s(CH2) [29,30], this
s(CH3). Although
m
m
a
2904 cmꢁ1 was observed and assigned to the same vibration. From
Fig. 4, it is of interest to note that the CAH stretches of ethyl group
present not only in the spectra of samples after the treatments at
24, 96 and 170 °C, but also in the spectrum of thermal conversion
product at 257 or 300 °C (see spectra D and E). This result may
indicate the existence of ethyl group in the decomposition product
of EIM similar to the case of EBAIO.
In the frequency region between 1650 and 1550 cmꢁ1, a moder-
ate band at 1645 cmꢁ1 is attributed essentially to the carbonyl
stretch. This is equivalent to the assignment of the intense IR band
of EBAIO at 1664 cmꢁ1. Other strong bands in this region, including
those at 1614, 1596, 1586 and 1564 cmꢁ1, can find the equivalent
bands in the FTIR spectrum of EBAIO and are contributed by the
C@C and C@N stretches or the combination of these. CH3 deforma-
tion has been characterized to have moderate and strong intensity
in the frequency region 1440–1475 cmꢁ1 [29]. One may thus
assign this moderate band at 1480 cmꢁ1 to d(CH3). Many CAH in-
plane bending vibrations have been characterized to have frequen-
cies near the range from 976 to 1468 cmꢁ1 [29,30]. Consequently,
in the FTIR spectrum of EIM, the observed intense or moderate
bands at 1334, 1230, 1137 and 998 cmꢁ1 are good candidates of
these vibrations. The group frequencies of CAH out-of-plane bend-
ing vibrations are expected to have frequency in the lower fre-
quency region [30].
In the high frequency region, the highest frequency band at
3465 cmꢁ1 appears to be very weak with a broad background. This
frequency is relatively low in comparison with both the free NAH
stretch around 3648 cmꢁ1 [25] and the free OAH stretch of oxime
group in the range 3500–3650 cmꢁ1 [30]. As the carbonyl stretch
overtone has been characterized as a weak IR band in the region
3550–3200 cmꢁ1 [30], it is not unreasonable to consider this high
frequency weak band as the first overtone of m(C@O). The appear-
ance of this overtone depends on the anharmonicity of the vibra-
tion, and the observed frequency can be different from the twice
frequency of the fundamental vibration. However, repeat experi-
ments have shown that the weak band disappeared upon a long
time heating of EIM. It is, therefore, suggested that the band is
attributable to the water vibration, which is similar to the case,
observed in the FTIR spectrum of EBAIO. Other weak or very weak
bands observed at 3260, 3114, 3110, 3058, 3044, 2987, 2939, and
2840 cmꢁ1 appear to find the corresponding peaks in the FTIR
spectrum of EBAIO. Similar assignments are consequently made
Fig. 5 displays the FTIR spectra of (A) ICAIO, (B) EBAIO, (C) EIM
and (D) indirubin. The vertical or nearly vertical lines in the figure
Fig. 4. FTIR spectra of EIM after annealing for three hours at different temperatures: (A) 96 °C; (B) 170 °C; (C) 225 °C; (D) 250 °C; (E) 300 °C.

58
Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
Fig. 5. The correspondence of vibrational modes among the FTIR spectra of (A) ICAIO; (B) EBADO; (C) EIM; (D) indirubin.
are used to show the correlations among the vibrational modes of
these molecules. The equivalences of vibrational assignments are
reasonable for those bands of different intensities. For each of these
four compounds, the C@O stretching band appears to be intense in
the frequency range 1645–1664 cmꢁ1. These stretching frequencies
are lower than those of aliphatic ketones and five or six-member
ring lactone [29,30]. McDermott has carried out the vibrational
studies and normal coordinate analyses of
rolidinone and N-methyl-2-pyrrolidinone and assigned the C@O
stretches to the strong IR bands at 1770, 1650 and 1670 cmꢁ1
c-butyrolactone, 2-pyr-
,
respectively [37]. The relatively high frequency in the lactone com-
pared with those of pyrrolidinones could not be attributed to the
intra-hydrogen bonding because the large separation between the
0
amino hydrogen and the O (ꢂ2.54 ÅA) makes it unlikely that this
would occur. However, according to the reported potential energy
distribution (PED), m(C@O) is 90.3% for c-butyrolactone while only
37.3% and 51.4% for 2-pyrrolactone and N-methyl-2-pryyolidinone,
respective [37]. PED might therefore be a factor contributing to the
difference of C@O stretch frequencies. Additionally, there could be
other factors lowering the C@O stretching frequency: One is the
hydrogen bonding with N10AH, which can be evidenced from the
Fig. 6. DSC curves of (A) EIM; (B) EBAIO; (C) ICAIO.
indicate that the exothermic conversions are irreversible processes
under our experimental condition. The present DSC experiments
have provided more accurate determination for the temperature
of chemical conversion. The experimental conversion tempera-
tures determined for all three samples fall in a narrow temperature
near 276 6 °C, and the enthalpies of conversion are not signifi-
cantly different. From the shapes of exotherms, it appears that
the kinetic rates of conversion for both EBAIO and EIM are slower
than that of ICAIO. The evidence of such slow conversion rate
might also be observed from the IR spectra of EBAIO and EIM after
the treatments at different temperatures. The chemical conversion
temperature of EBAIO has been measured to be 5 °C higher than
that of EIM. Such a small difference could not be accurately con-
firmed from our present IR spectroscopic data.
red-shift of m
(N10AH); another is the increasing conjugation with
carbonyl group due to the fusion of one conjugation ring or the dou-
ble bond connection of the pyrrolidinone with another polyaro-
matic ring.
Thermal measurements
Thermal analyses have been carried out for samples EBAIO and
EIM. For comparison, we have also measured the DSC curve for
ICAIO. Displayed in Fig. 6 are the DSC curves for these three sam-
ples. The chemical conversions for all these samples are observed
to be exothermic. Based on the experimental plots, the exothermic
conversion temperatures are determined to be 282, 276 and
271 °C, and the enthalpy changes for the chemical conversions
are obtained to be ꢁ93, ꢁ94 and ꢁ94 kJ/mole for ICAIO, EBAIO
and EIM, respectively. In the thermal measurements, the samples
were heat-scanned from 0 to 400 °C after baking out moisture
and reaching thermal equilibrium. In separated experiences, the
reversed scan from 400 to 0 °C failed to show endothermic conver-
sion; neither exothermic nor endothermic process could be
observed in the second upward scan from 0 to 400 °C. All these
Summary
In the present study, we have synthesized EBAIO and identified
with IR, NMR, MS and elementary analysis. FTIR and Raman spectra
of EBAIO and EIM have been recorded and vibrational assignments
have been suggested based on the previous work on ICAIO and the
group characteristic frequencies. The temperature variation study
of IR spectra has identified the presence of two conformers in

Y.-S. Li et al. / Journal of Molecular Structure 1087 (2015) 51–59
59
[16] A.V. Barnov, Y.S. Bobovich, V.I. Petrov, Opt. Spectrosc. 61 (1986) 315–319.
[17] E. Tatsch, B. Schrader, J. Raman Spectroc. 26 (1995) 467–473.
[18] A. Amat, F. Rosi, C. Miliani, A. Sgamellotti, S. Fantacci, J. Mol. Struct. 993 (2011)
43–51.
[19] E.-S. Min, S.I. Nam, M.S. Lee, Bull. Chem. Soc. Jpn. 75 (2002) 677–680.
[20] I.V. Aleksandrov, Spectroscopy 45 (1978) 341–342.
[21] M.S. del Rio, M. Picquar, E. Haro-Poniatowski, E. van Elslande, V.H. Uc, J. Raman
Spectrosc. 37 (2006) 1046–1053.
EBAIO. A series of FTIR spectra of EBAIO and EIM after baking at dif-
ferent temperatures have been collected. It was found that both
EBAIO and EIM begin to decomposed around 257 °C. Thermal anal-
ysis has confirmed these results and determined the enthalpies of
chemical conversion to be ꢁ94 kJ/mol for both EBAIO and EIM.
[22] P. Vandenabeele, S. Bode, A. Alonso, L. Moens, Spectrochim. Acta, Part A 61
(2005) 2349–2356.
References
[23] H.G. Wiedemann, K.-W. Brzezinka, K. Witke, I. Lamprecht, Thermochim. Acta
456 (2007) 56–63.
[1] S.P. Williams, P. Shante, M.O. Nowicki, F. Liu, R. Press, J. Godlewski, M. Abdel-
Rasoul, B. Kaur, S.A. Fernandez, E.A. Chiocca, S.E. Lawler, Cancer Res. 71 (2011)
5374–5380.
[2] G. Eisenbrand, F. Hippe, S. Jakobs, S. Muehlbeyer, J. Cancer Res. 13 (2004) 627–
635.
[3] C. Hoessel, S. Leclerc, J.A. Endicott, M.E. Nobel, A. Lawrie, P. Tunnah, M. Leost, E.
Damiens, Nat. Cell Biol. 1 (1999) 60–67.
[4] D. Marko, S. Schatzl, A. Friedel, A. Genzlinger, H. Zankl, J. Cancer Res. Clin.
Oncol. 130 (2004) 627–635.
[5] D. Marko, S. Schatzl, A. Friedel, A. Genzlinger, H. Zankl, Br. J. Cancer 84 (2001)
283–289.
[6] F.G.E. Perabo, C. Frossler, G. Landwehrs, D.H. Schmidt, A. Von Rucker, A. Wirger,
S.C. Muller, Anticancer Res. 26 (2006) 2129–2135.
[7] K. Ravichandran, A. Pal, R. Ravichandran, Microsc. Res. Tech. 73 (2010) 1053–
1058.
[8] E.A. Perpete, J. Preat, J.-M. Andre, D. Jacquemin, J. Phys. Chem. A 110 (2006)
5629–5635.
[9] B.-X. Ye, L.-J. Yuan, C. Chen, J.-C. Tao, Electroanalysis 17 (2005) 1523–1528.
[10] T. Karapanayoitis, S.E. Jorge, Analyst 129 (2004) 613–618.
[11] Y.-S. Li, J. Church, J. Food Drug Anal. 22 (2014) 29–48.
[12] Y.-S. Li, J. Cheng, K.T. Chung, Spectrochim. Acta Part A 69 (2008) 524–527.
[13] T. Robbins, Y. Wang, Q.-Z. Yao, Z.-H. Wang, J. Mol. Struct. 1048 (2013) 51–58.
[14] J. Tatsugi, T. Zhiwei, Y. Izawa, Arkivoc 1 (2001) 63–73.
[15] D. Raileanu, O. Constantinescu-Simon, E. Mosanu, C.D. Nenitzescu, Rev. Roum.
Chim. 12 (1967) 105–108.
[24] A. Domenech, J. Raman Spectrosc. 42 (2011) 86–96.
[25] S. Lunak, Chem. Phys. Lett. 477 (2009) 116–121.
[26] E. Wille, W. Luttke, Liebigs Ann. Chem. 12 (1980) 2039–2054.
[27] D.N. Shigorin, N.S. Dokunikhin, E.A. Gribova, Z. Fiz, Z. Fiz. Khim. 29 (1955) 867–
876.
[28] W.M. Coleman III, B.M. Gordon, Appl. Spectrosc. 42 (1988) 108–113.
[29] D. Lin-Vien, N.B. Colthup, W.G. Fateley, J.G. Grasselli, The Hand Book of
Infrared and Raman Characteristic Frequencies of Organic Molecules, AP,
Boston, 1991.
[30] G. Socrates, Infrared and Raman Characteristic Group Frequencies, third ed.,
John Wiley and Sons, Chichester, 2001.
[31] D.W. Scott, J. Mol. Spectrosc. 37 (1971) 77–91.
[32] T.D. Klots, R.D. Chirico, W.V. Steele, Spectrochim. Acta, Part A: Mol. Biomol.
Spectrosc. 50A (1994) 765–795.
[33] A. Koll, M. Rospenk, L. Stefaniak, J. Wojcik, J. Phys. Org. Chem. 7 (1994) 174–
177.
[34] L. Nygaard, J.T. Nielsen, J. Kirchheiner, G. Maltesen, J. Rastrup-Andersen, G.O.
Sorensen, J. Mol. Struct. 3 (1969) 491–506.
[35] D.J. Lide, J. Chem. Phys. 33 (1960) 1514.
[36] T. Ogawa, J. Hara, K. Hirota, Nippon Kaga. Z. 89 (1968) 19–22.
[37] D.P. McDermott, J. Phys. Chem. 90 (1986) 2569–2574.