Background defining during the imine formation reaction in FT-IR liquid cell

Article abstract of DOI:10.1016/j.saa.2005.07.020

Imine formation is a very important chemical reaction because of its relevance to biological process. Therefore, it is crucial to follow whole reaction process in detail. The current work performed to monitor the whole imination reaction in real time in l

Full text of DOI:10.1016/j.saa.2005.07.020

Spectrochimica Acta Part A 64 (2006) 93–100
Background defining during the imine formation
reaction in FT-IR liquid cell
∗
Hilmi Namli , Onur Turhan
Department of Chemistry, University of Balikesir, Soma Cad., 10100 Balikesir, Turkey
Received 25 April 2005; received in revised form 1 July 2005; accepted 2 July 2005
Abstract
Imine formation is a very important chemical reaction because of its relevance to biological process. Therefore, it is crucial to follow whole
reaction process in detail. The current work performed to monitor the whole imination reaction in real time in liquid cell by FT-IR spectroscopy.
The complex spectral futures due to solvent, unreacted reagents, acid catalysis and other additives are eliminated by defining a background
at the beginning or at any time during the reaction. This procedure also makes it possible to monitor the changes in the concentration of
each component in the liquid cell. The consumption of the functional groups of the reagents results in absorbance due to the direct difference
spectra while the appearance of functional groups is monitored as percentage transmittance. The concentration changes in the cell arising
from the reaction gives the product spectra without having to isolate it from the mixture. It is also possible to see the intermediates appearing
and disappearing during the reaction. This report also illustrates a brief application of the technique by time dependence of the peak highs in
absorption (ABS) mode.
©
2005 Elsevier B.V. All rights reserved.
Keywords: Background defining; In situ time-dependent FT-IR; Imine formation
1
. Introduction
reagent and the solvent. Having eliminated the absorptions
due to the reagent and the solvent will give clearer spec-
trum containing FT-IR peaks only due to the products and
intermediates.
Defining a background in such a way can be easily appli-
cable even to the bimolecular reaction which has large molec-
ular structure and quite difficult to distinguish each peaks [5].
We choose imine formation reaction between benzalde-
hyde and aniline to investigate by background defining
method. This reaction was already examined by Raman [6]
and LI-Raman [7] spectroscopy. Furthermore, reactions of
salicylaldehyde and 2-pyridinecarboxaldehyde with aniline
were also investigated.
Recent developments gave the scientist an opportunity to
follow the reaction in situ by FT-IR [1] and Raman [2] spec-
troscopy. Because of the FT techniques it is quite practical
obtaining a spectrum, which allows the scientist to have the
spectra in the liquid or gas phase consecutively at different
temperatures [3]. Liquid flow measurements were also con-
ducted to observe all stages of the reaction [4]. The main
problem of the interpretation on the IR spectra comes from
method itself.
To eliminate the peaks due to the background material,
defining a background is an essential requirement for FT-
IR measurements. The background could be potassium bro-
mide (KBr) disc for solid samples or pure solvent for liquid
cells. In the case of mixed solvents, the solvent mixture
can be employed as background. In this report we tried to
define a new background, which is composed of both the
Intermediates [8] play a fundamental role in the final
product of the chemical reaction. Often, these intermediates
cannot be monitored in real time however sometimes they
can be trapped or isolated, but this time their characterization
is a very time consuming process.
FT-IR as a reaction monitor can recognize the formation of
intermediates [9] as well as determine reduction of reactants
and formation of products. In most cases, it is not feasible
∗
Corresponding author. Tel.: +90 2662493358; fax: +90 2662491012.
E-mail address: hnamli@balikesir.edu.tr (H. Namli).
1
386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2005.07.020

9
4
H. Namli, O. Turhan / Spectrochimica Acta Part A 64 (2006) 93–100
to remove aliquots of a reaction mixture for measurement.
Currently, more sophisticated and costly methods [10] like
Fiber Optic Probe system capillary glass tube and microma-
chined flow cell [11] is used to allow real-time monitoring
of the reaction. Defining the background at any time of the
reaction allow us to eliminate all the peaks arising from any
component or even product formed till that time and let us to
follow any changes after that time.
better spectrum as solvent while reagents and product are
−1
more soluble and no limitation over 1250 cm
.
The imine formation reaction between benzaldehyde
and aniline was investigated by defining the beginning
of the reaction as background. Fig. 1 shows the FT-IR
spectra of reagents, benzaldehyde and aniline, the solvent,
chloroform, and reaction mixture, which is defined as a
background.
From the background defined reaction mixture spectra
(Fig. 1b) it is easy to identify the aldehyde and amine func-
2
. Experimental
tional groups in the mixture. Characteristic benzaldehyde
peaks appeared at 1702 cm for carbonyl and 2741 and
−
1
−
1
2
.1. Application of the background defining
2822 cm for COH hydrogen (Fig. 1c) and characteristic
−1
aniline aromatic stretching appeared at 1619 cm as well as
−1
FT-IR spectra were collected using Perkin-Elmer Model
NH2 hydrogen peaks at 3376 and 3453 cm in the chloro-
form (Fig. 1d) were all observed in the background spectra
(Fig. 1b).
BX 1600 instrument using 0.015 mm path length liquid CaF2
cell.
First recorded spectrum is used as background and the
relative intensities of the peaks were examined. Two cases
are, starting from the first recorded spectra after the scan
used as background, the changes in relative peak intensities
are clearly observed, indicating that the reaction continuous
in the cell. Two cases are realized in the spectra, where the
relative percentage transmittance increases (over 100%) or
decreases (under 100%) depending on the concentration of
the materials in the solution. For instance, when we consider
the most intense peaks, the C O at 1702 which is exactly the
same as benzaldehyde is going up with time while C N at
1
. Equivalent concentration of aldehyde and aniline was pre-
pared in chloroform and FT-IR spectra of each reagent
were recorded for comparison.
. The reagents were mixed and scanned in the CaF2 liquid
cell after stirring.
. The scanned spectra of the mixture were accepted as a
background for the latter real time scans.
. FT-IR spectra were collected sequentially in an optimum
time intervals that are determined by reaction rate.
. The increasing and decreasing peaks were compared with
the reagents and product.
2
3
4
5
6
−
1
1630 cm going down because of the product imine forma-
. The extra peaks appeared in the spectra were attributed to
the possible intermediates or interactions.
tion (Fig. 2).
As expected, the increasing peaks exactly coincide with
the reagents in the solvent, indicating a concentration
decrease. On the other hand, the decreasing peaks, which are
at the same frequency with the product, indicate formation of
a new compound. Therefore, it is also possible to estimate the
Ethyl alcohol and chloroform were used as solvents. The
changes in the C O and C N bonds were observed perfectly
in both solvents. However, no change in the amine peak was
observed in the EtOH due to intense OH. Chloroform gave
Fig. 1. FT-IR spectra of chloroform (a); background defined time of reaction mixture (b); benzaldehyde (c); aniline (d) (spectra were arbitrarily displaced in
vertical for clarity).

H. Namli, O. Turhan / Spectrochimica Acta Part A 64 (2006) 93–100
95
Fig. 2. Time resolved FT-IR spectra of the benzaldehyde and aniline in 1250–1800 cm 1 region (benzaldehyde (a), aniline (b), in situ real time scans of reaction
−
mixture (c), isolated product (d)).
position of the increasing peaks from the pure benzaldehyde
and aniline spectra in chloroform (Fig. 3).
Considering negative peaks relative to 100% transmit-
tance, it is possible to follow the product formation in solution
starting from the beginning of the reaction without any inter-
ruptions of the other components. In order to validate the
background defining method to other type of organic reac-
tions, we applied the same analysis to the reactions of 2-
salicylaldehyde(Fig. 6)and2-pridinecarboxaldehyde(Fig. 7)
with aniline (Scheme 1), both reactions gave similar observa-
tion with the benzaldehyde and aniline reaction, the positive
peaks were appeared exactly at the same frequencies with
the reagent while the negative with the isolated products in
chloroform. However, the peak intensities and frequencies
differ from each other. For instance, C O peak appears at
In addition, by defining the background at any time of
the reaction allows us to subtract the unreacted reagents and
already formed products. This results in decreasing intensi-
ties of the peaks due to decreasing reaction rate. Because of
the density differences of the chloroform and water, the for-
mation of the water during the reaction do not disturb the
FT-IR beam. Water related peaks have not been observed in
the spectra (Fig. 4).
Consequentscanningofthereactionmixtureafterdefining
background gave more intense peaks by the time depended
concentration changes. Fig. 5 shows the comparison of the
FT-IR spectra of the reaction mixture with the reagents,
benzaldehyde, aniline, and product, N-benzylideneaniline, in
chloroform.
−
1
1666, 1702 and 1714 cm in 2-piridincarbaldehyde, ben-
zaldehyde, and salicylaldehyde in chloroform, respectively.
Similar differences on the products spectra for imine C N
Fig. 3. Background defined baseline corrected FT-IR spectra of amine to the imines conversion 2650–3100 cm 1 (C H) region (c) (the spectra of aniline (a)
−
and benzaldehyde (b) were arbitrarily displaced in vertical for clarity).

9
6
H. Namli, O. Turhan / Spectrochimica Acta Part A 64 (2006) 93–100
−
1
Fig. 4. Background defined baseline corrected FT-IR spectra of amine to the imines conversion 3200–3759 cm region (a) (the spectra of benzaldehyde (b)
and aniline (c) were arbitrarily displaced in vertical for clarity).
functionality at 1630, 1628, and 1620 cm 1 were also obser-
−
hyde and aniline (0.25 M) were mixed together, it was placed
in to CaF2 cell and scanned as background. After scanning
the entire component in the cell as background, the solu-
tion was scanned in each 3 min and recorded for peak highs
manipulation.
The peak highs for all the main peaks are obtained using
the PE spectrum 2 in absorbance mode. The calculated values
are shown in Table 1.
ved according to the aromatic ring (Figs. 5–7).
All reactions were slow enough to follow each step. Usu-
ally 5–20 min time intervals were used to observe signif-
icant changes. The rate of the reactions were considered
according to the relative intensity changes on the background
defined spectra and found that 2-pyridinecarboxaldehyde
was faster than benzaldehyde while salicylaldehyde was
slower.
For the comparison of the peaks to each other, the other
−1
peaks values were plotted against the 1702 cm . The good
regressions for each line show that the peak highs may be
used for latter calculation like concentrations.
2
.2. Comparison of the peak highs
For a short application of the method, the reaction of
benzaldehyde and aniline monitored for 90 min in a 3 min
sequence. As soon as the same concentration of benzalde-
The negative values arise from consumption of the reagent
and positive values arise from appearance of the product. All
the peak highs are related to the same concentration changes
Fig. 5. FT-IR spectra of benzaldehyde (a); aniline (b); background defined in situ aniline and benzaldehyde reaction (c); isolated product in chloroform (d)
spectra of reagent and products were arbitrarily displaced in vertical for clarity).
(

H. Namli, O. Turhan / Spectrochimica Acta Part A 64 (2006) 93–100
97
Fig. 6. FT-IR spectra of (a) salicylaldehyde (b) aniline (c) background defined reaction (d) isolated product (N-salicylideneaniline) in chloroform (spectra of
reagent and products were arbitrarily displaced in vertical for clarity).
in the cell. So, calculating the concentration [X] at any time
may give the directly product concentration and subtracting
this value from the initial reactant concentration it is also
possible to calculate the rest of the reactant concentration.
The time versus peak highs plots should give rough infor-
mation about the rate of the reaction. As it is seen in Fig. 8 the
increase or decrease on the line are not the same but related
to the peak intensities for a single concentration by means of
the most intense peak for C O is the most lower value.
tion by this way is inconvenient for continuous monitoring of
chemical reactions since each time a sample must be taken.
Following each addition, the changes are very difficult to
analyze and each time cleaning process of the cell by pass-
ing trough solvent and refilling for the next scan is necessary
in order not to cause concentration differences. Subtracting
the initial spectrums (t = 0) of reactants and solvent from lat-
ter may give some suggestion about the reaction but these
subtraction may need tremendous work and subtracted spec-
tra may not give valuable results for further studies [13].
This shortcoming can be overcome by continuous method
by adding all the reagents at once and defining it a back-
ground. This allows us not to disturb the system until the end
of the experiment (Fig. 9). In this work, an easy application
of FT-IR spectroscopy for imine formation in a liquid cell
was demonstrated. Analysis of the increasing and decreasing
3
. Results and discussion
Usually the observation of the reaction in the liquid cell
was performed by addition of one of the reagent to the other
and recording the changes [12]. However, liquid cell detec-
Fig. 7. FT-IR spectra of (a) 2-pyridinecarboxaldehyde, (b) aniline, (c) background defined real time scans and (d) isolated product in chloroform (spectra of
reagent and products were arbitrarily displaced in vertical for clarity).

Table 1
The peaks height values (ABS) of in situ reaction of benzaldehyde and aniline in 3 min intervals
Time (min)
ABS
702 (cm 1)
−
−1
−1
−1
−1
−1
−1
−1
−1
1451.5 (cm )
1
1629.9 (cm
)
1618.1 (cm
)
1601.3 (cm
)
1591.7 (cm
)
1579.5 (cm
)
1499.4 (cm
)
1485.5 (cm
)
3
6
9
−0.0579
−0.1243
−0.1858
−0.2333
−0.2703
−0.2991
−0.3214
−0.3399
−0.3555
−0.3687
−0.3802
−0.3904
−0.3992
−0.4075
−0.4146
−0.4215
−0.4278
−0.434
0.0099
0.0223
0.0338
0.0429
0.0499
0.0557
0.06
0.0636
0.0666
0.0692
0.0713
0.0732
0.075
0.0765
0.0779
0.0792
0.0804
0.0814
0.0825
0.0836
0.0845
0.0854
0.0862
0.0871
0.0876
0.088
−0.0136
−0.028
−0.0099
−0.0209
−0.0321
−0.0412
−0.0485
−0.0542
−0.0589
−0.0629
−0.0662
−0.0691
−0.0717
−0.0741
−0.076
0.0102
0.0225
0.0337
0.0424
0.0491
0.0546
0.0586
0.0621
0.065
0.0673
0.0693
0.0711
0.0728
0.0742
0.0755
0.0767
0.0779
0.0788
0.0799
0.081
0.0109
0.0242
0.0367
0.0463
0.0539
0.0601
0.0647
0.0686
0.0718
0.0746
0.0769
0.0789
0.0808
0.0824
0.084
−0.02
0.0095
0.0213
0.0325
0.0412
0.048
0.0537
0.0579
0.0614
0.0643
0.0668
0.069
0.0707
0.0724
0.074
0.0753
0.0766
0.0778
0.0788
0.0799
0.081
0.0056
0.0128
0.0196
0.0248
0.029
0.0324
0.0349
0.037
0.0388
0.0403
0.0415
0.0426
0.0436
0.0445
0.0454
0.0461
0.0469
0.0475
0.0481
0.0489
0.0494
0.0499
0.0504
0.051
−0.0421
−0.0626
−0.0784
−0.091
−0.0412
−0.0517
−0.0597
−0.0659
−0.0706
−0.0747
−0.0782
−0.0811
−0.0838
−0.0863
−0.0882
−0.0901
−0.0917
−0.0934
−0.0949
−0.0963
−0.0976
−0.0987
−0.0998
−0.1008
−0.1018
−0.1028
−0.1047
−0.1064
−0.1075
−0.1085
−0.1091
−0.1097
1
1
1
2
2
2
3
3
3
3
4
4
4
5
5
5
6
6
6
6
7
7
7
8
8
8
9
2
5
8
1
4
7
0
3
6
9
2
5
8
1
4
7
0
3
6
9
2
5
8
1
4
7
0
−0.1005
−0.108
−0.1141
−0.1193
−0.1238
−0.1277
−0.1312
−0.1341
−0.137
−0.078
−0.0795
−0.0811
−0.0826
−0.084
−0.1393
−0.1417
−0.1438
−0.146
0.084
0.0854
0.0867
0.0878
0.0891
0.0903
0.0913
0.0923
0.0932
0.0942
0.0948
0.0953
0.096
−0.4395
−0.4446
−0.4494
−0.454
−0.0853
−0.0864
−0.0876
−0.0887
−0.0897
−0.0908
−0.092
−0.1477
−0.1493
−0.1509
−0.1524
−0.154
0.0818
0.0827
0.0835
0.0843
0.0849
0.0855
0.0862
0.087
0.082
0.0829
0.0837
0.0846
0.0849
0.0853
0.0859
0.0868
0.0874
0.0881
−0.4585
−0.4629
−0.4687
−0.4738
−0.4783
−0.4825
−0.4852
−0.4883
−0.1554
−0.1577
−0.1596
−0.1612
−0.1626
−0.1635
−0.1644
0.0512
0.0513
0.0518
0.0523
0.0526
0.0531
−0.093
0.0887
0.0895
0.0901
0.0907
−0.0939
−0.095
−0.0957
−0.0965
0.0875
0.088
0.097
0.0976

H. Namli, O. Turhan / Spectrochimica Acta Part A 64 (2006) 93–100
99
Scheme 1. Reactions of benzaldehyde (a), salicylaldehyde (b), 2-pyridinecarboxaldehyde (c), with aniline.
−
1
Fig. 8. The peak highs vs. 1702 cm plots in 3 min intervals for overall 90 min.
Fig. 9. Time to peak highs (ABS) plots of the benzaldehyde and aniline reaction.

1
00
H. Namli, O. Turhan / Spectrochimica Acta Part A 64 (2006) 93–100
Scheme 2. The concentration changes of the reaction that FT-IR recognizes.
peaks, after background subtraction were required in order to
have an acceptable spectra from a small volume of chemical
species in a liquid cell. As a result, a series of in situ real
time direct difference FT-IR spectrum were accumulated for
the imine formation reactions. We demonstrated the results
of FT-IR detection of bond breaking and formation during
the model reactions of imine formation in CaF2 liquid cell.
For an application the changes were scanned in each three
min for overall 90 min. The peak highs in absorbance mode
were calculated using the PE spectrum V2 software. The data
were directly transferred to the MS excel for peak highs to
peak highs and time to peak highs manipulations.
The proposed method could also be applicable to the other
accessories of the FT-IR, which would give an opportunity to
follow any difference including the intermediates which are
stable enough to scan by FT-IR.
The changes in concentration in the liquid cell were rec-
ognized by spectrophotometer and given as negative and
positive peaks according to the consumption or formation.
We have shown here an easy, inexpensive and simultaneous
way to monitor the C N bond formation and C O and N H
bond breaking in real time in the CaF2 liquid cell.
This study was carried out in order to establish an easy
method of monitoring the organic reactions in situ by mea-
suring the concentration changes during the reaction.
The decrease of the concentration of the reagents by react-
ing are appeared an upward peaks while the product forma-
tion downward peaks in transmittance mode. The decrease of
the concentration by consumption cause to the less absorp-
tion bands conversely more transmittance (upward peaks)
appearing on the same frequencies of the reagent in the sol-
vent. The increase of the concentration of the product cause a
new absorption bands (downward peaks). These all changes
were attributed to the same concentration difference for the
investigated reaction (Scheme 2).
References
[1] R. Schindler, B. Lendl, Anal. Commun. 36 (1999) 123.
[2] D.E. Pivonka, J.R. Empfield, Appl. Spectrosc. 58 (2004) 7.
[
3] (a) J. Ahola, M. Huuhtanen, R.L. Keiski, Ind. Eng. Chem. Res. 42
(
(
2003) 2756;
b) T. Nobukawa, M. Yoshida, S. Kameoka, S. Ito, K. Tomishige,
K. Kunimori, J. Phys. Chem. B 108 (2004) 4071.
[4] G.M. Hamminga, G. Mul, J.A. Moulijn, Chem. Eng. Sci. 59 (2004)
5479.
[
5] (a) H.B. McMahon, M. Fabian, F. Tomson, T.P. Causgrove, J.A. Bai-
ley, F.N. Rein, R.B. Dyer, G. Palmer, R.B. Gennis, W.H. Woodruff,
Biochim. et Biophys. Acta 1655 (2004) 321;
(
b) K.J. Rothschild, H. Marreo, Biophysics 79 (1982) 4045.
[6] M. Lee, H. Kim, H. Rhee, J. Choo, Bull. Korean Chem. Soc. 24
2003) 205.
(
[
[
[
7] M. Lee, J.-P. Lee, H. Rhee, J. Choo, Y.G. Chai, E.K. Lee, J. Raman
Spectrosc. 34 (2003) 737.
8] S. Hayashi, E. Tajkhorshid, K. Schulten, Biophys. J. 83 (2002)
4
. Conclusion
1
281.
9] V.A. Yaylayan, S.H. Majors, A.A. Ismail, J. Agric. Food Chem. 47
1999) 2335.
[10] M. Gallignani, M.R. Brunetto, Talanta 64 (2004) 1127.
FT-IR is a viable way of monitoring an organic reaction
in real-time where there are recognizable differences in func-
tional groups between reactants and the products. Defining
the reactants and solvent mixture as a background at the
beginning or at any time of the reaction makes the interpre-
tation of changes more feasible. The product spectra may be
obtainedseparatelywithoutisolatingitfromthereactionmix-
tureduetothepositiveadsorptionwhilereagentsarenegative.
(
[
11] P. Hinsmann, M. Haberkorn, J. Frank, P. Svasek, M. Harasek, B.
Lendl, Appl. Spectrosc. 55 (2001) 3.
12] (a) T. Amari, Y. Ozaki, Macromolecules 34 (2001) 7459;
b) J.E. Lynch, S.M. Riseman, W.L. Laswell, D.M. Tschaen, R.P.
Volante, G.B. Smith, I. Shinkai, J. Org. Chem. 54 (1989) 3792.
[13] D.E. Pivonka, K. Russell, T. Gero, Appl. Spectrosc. 50 (1999) 12.
[
(