- Email: [email protected]
cholesteryl palmitate blends
Spectrochimica Acta Part A 79 (2011) 45–50
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Two-dimensional infrared correlation spectroscopic studies of polymer blends—Specific interactions in polyethylene adipate/cholesteryl palmitate blends Maria Cristina Popescu ∗ , Cornelia Vasile Romanian Academy, “Petru Poni” Institute of Macromolecular Chemistry, 41A Grigore Ghica Voda Alley, 700487 Iasi, Romania
a r t i c l e
i n f o
Article history: Received 8 July 2010 Received in revised form 28 December 2010 Accepted 26 January 2011 Keywords: Specific interactions Liquid crystal Blends 2D correlation spectroscopy
a b s t r a c t Generalized two-dimensional infrared (2D IR) correlation spectroscopy has been applied to the study of the conformational changes and specific interactions in blends of polyethylene adipate (PEA) and cholesteryl palmitate (CP). IR spectra of CP, PEA, and their blends of different compositions: 10/90, 16/84, 32/68, 64/36, and 80/20 have been recorded. In order to apply 2D IR correlation analysis, the samples are divided into two sets: set A with high PEA content 0/100, 10/90, 16/84, 32/68 CP/PEA and set B with high CP content 64/36, 80/20 and 100/0 CP/PEA. The 2D IR synchronous correlation analysis separates the bands of PEA from those of CP. The cross-peaks in 2D IR asynchronous correlation analysis are indicative of the specific interaction or the conformational change in the blends. The bands of C O, OH and C–O vibrations of PEA, and CH3 and C–O vibrations of CP that are very sensitive to the intermolecular hydrogen bonding effect and consequently the partial miscibility of components, have been assigned by 2D correlation analysis. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Blending is one of the most important areas in polymer technology. Polymer blends found a considerably widespread range of applications, from consumer plastics of common everyday use to special materials. The continuing fervent interest in these materials is twofold. Firstly, the potential to combine characteristic properties of different polymers in a single processable material permits the expansion of range of applications of each component of the system. Secondly, if by blending a synergic effect appears, specific properties can be achieved. These permit the design of materials capable to perform determined functions. Specific interactions between chemical groups of the blend components play an important role in polymer mixtures. The physical properties of polymer blends are strongly influenced by blending conditions and processes that, in turn, affect the level of mixing of the blends, so there is a growing interest in the study of miscibility and phase behaviour of polymer blends. The incorporation of low molecular weight liquid crystals in amorphous polymers has been suggested as an alternative method to bring about modest improvements in the mechanical properties of polymeric composites [1–4]. By covering the textile materials
∗ Corresponding author. Tel.: +40 232 217454. E-mail address: [email protected] (M.C. Popescu). 1386-1425/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2011.01.042
with polymeric compounds one can obtain products with special properties and large applicability. Various techniques have been employed to investigate miscibility of polymer blends, such as microscopy, thermal analysis, dynamic mechanical analysis, dielectric measurements, X rays diffraction, and spectroscopy [5–8]. Differential scanning calorimetry (DSC) is one of the most widely used techniques for evaluating ˚ in terms of cooperative motion miscibility on a scale of 100–300 A, of polymer segments around the glass transition temperature (Tg ) [9]. Infrared (IR) absorption is sensitive to the local environment of the oscillating dipoles, and has proven to be a powerful technique for investigating intermolecular interactions [10]. Two-dimensional (2D) correlation analysis/spectroscopy has been known for about 15 years [11–13]. It has been employed to solve various problems predominantly in vibrational spectroscopy mainly related with bands’ overlapping. Several review articles and books are available on this subject [14]. The method is almost exclusively based on comparing/correlating the intensities of the bands. In the case of polymer blends, the technique of two-dimensional 2D IR correlation spectroscopy offers new possibilities for the analysis and interpretation of vibrational spectra. 2D correlation analysis is performed from a set of spectra collected from a system under an external perturbation that induces alterations in the spectrum of the system. In the 2D approach, the bands are spread over the second dimension, thereby simplifying the visualization of spectra consisting of overlapping bands. Inter- and intra-molecular
46
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
interactions can be investigated by correlating absorption bands from different functional groups or parts of a molecule. PEA/CP blends with different composition were studied by thermogravimetry (TG) and differential scanning calorimetry (DSC). Thermogravimetric data were used to establish the degree of interaction in blends [15,16]. It has been established that PEA/CP blends are characterized by good thermal stability and partial miscibility. In the present study, generalized 2D IR correlation spectroscopy has been applied to explore much deeply the intermolecular interactions in blends of polyethylene adipate (PEA) and cholesteryl palmitate (CP). 2. Experimental 2.1. Materials Polyethylene adipate (PEA) with end-capped hydroxyl groups is a commercial product purchased from Fibrex SA Savinesti, Romania (Mn = 2000 g/mol) with the following chemical formula:
Cholesteryl palmitate (CP) was used as a low-molecular-weight additive. It was purchased from Nopris SRL, Cluj Napoca, Romania, and used as received. It has the following structure:
CP is a compound with liquid crystalline properties characterized by two types of mesophases: i.e., cholesteric (Ch) and smectic (S). Monotropic phases are obtained from an isotropic melt (I) in agreement with DSC (4 ◦ C/min) determinations. The transitions between these phases are: (C → Ch) at 74.5 ◦ C, (Ch → I) at 80 ◦ C, and (Ch → S) at 73.5 ◦ C. 2.2. Blends preparation Semi-crystalline PEA and liquid crystalline CP were separately dissolved in 1,2-dichloroethane (DCE) to form 0.8 g/dL solutions. The solutions were mixed to obtain the following CP/PEA ratios (wt/wt): (1) 10/90, (2) 16/84, (3) 32/68, (4) 64/36, and (5) 80/20. The mixtures of solutions were stirred for 5 h. After that, the solvent was slowly evaporated at room temperature. To remove the residual solvent and moisture, the samples were dried in a vacuum oven at 50 ◦ C for 2–6 days and total removal of the solvent was checked by IR spectroscopy. 2.3. Investigation methods 2.3.1. FT-IR spectroscopy FT-IR spectra were recorded on solid samples in KBr pellets by means of a FT-IR DIGILAB Scimitar Series spectrometer (USA) with a resolution of 4 cm−1 . The concentration of the samples in the tablets was constant at 5 mg/500 mg KBr. Five recordings were performed for each sample after successive millings. The evaluations were made on the average spectrum obtained from these five recordings. Processing of the spectra was done by means of Grams/32 program (Galactic Industry Corp.).
Table 1 Bands assignment to PEA in FTIR spectra [19,20]. (cm−1 )
Assignments
2961 2881 1736 1465 1438 1412 1389 1274 1257 1168 1145 1083 1056 985 961
CH2 asymmetric stretching CH2 symmetric stretching C O stretching CH2 asymmetrical bending CH2 symmetric bending CH2 symmetrical bending CH2 asymmetrical wagging O–H deformation C–O–C asymmetric stretching C–O–C asymmetric stretching C–O–C symmetric stretching C–O–C stretching C–C stretching mode C–C stretching mode CH2 rocking CH2 rocking
2.3.2. 2D correlation spectroscopy 2D FT-IR correlation intensities were calculated using MATLAB program derived from the generalized 2D correlation method developed by Noda [17]. The FT-IR spectra of PEA and its blends with CP were divided into two sets: set A and set B according with procedure described by Ren et al. [18]. Set A contains the FT-IR spectra of (1) PEA, (2) 84% PEA–16% CP and (3) 68% PEA–32% CP. Set B contains the FTIR spectra of (4) 36% PEA–64% CP, (5) 20% PEA–80% CP and (6) CP. The spectra in set A were arranged in the order 1–2–3. The spectra in set B were arranged in the order 4–5–6. Thus, in both sets, the intensities of the bands of PEA are decreasing while those of CP are increasing. The reference spectra used for set A and B were those of the sample 1 and 4, respectively. In 2D correlation analysis, two kinds of correlation maps synchronous and asynchronous are generated from a set of dynamic spectra obtained from the modulation experiment [11]. According to reference [18], by using the function ˚ (which designates synchronous 2D correlation), the intensities between two bands can be written by the following relations: ˚[(PEA), (PEA)] > 0 ˚[(PEA), (CP)] < 0
˚[(CP), (CP)] > 0 ˚[(CP), (PEA)] < 0
Based on these relations, it is possible to ascribe and differentiate of the bands in the spectra of the blends to PEA or CP. An asynchronous correlation peak appears only if the intensity change of two bands at 1 and 2 has basically dissimilar trends. According to Noda rules [18] ˚[1 , 2 ] < 0 and [1 , 2 ] > 0 imply that the intensity change at 1 occurs at higher PEA content compared to that at 2 . So do ˚[1 , 2 ] > 0 and [1 , 2 ] < 0. If ˚[1 , 2 ] < 0 and [1 , 2 ] < 0 imply that the intensity change at 1 takes place at lower PEA content than that at 2 . So do also ˚[1 , 2 ] > 0 and [1 , 2 ] > 0. When an asynchronous peak [1 , 2 ] appears between two bands at 1 and 2 , it can be either due to the conformational change of the component polymers in their blends or to the specific interactions between the component polymers [18]. If the sequential order between 1 and 2 in set A is reversed in set B, then the band at 1 and 2 can be regarded as indicative of the specific interactions between PEA and CP. 3. Results and discussion The FT-IR spectra of PEA, CP and blends at room temperature are shown in Fig. 1 and were used to evidence the region of interest for 2D correlation spectroscopy study. The assignment of FT-IR bands of PEA and CP are listed in Tables 1 and 2. Generalized 2D correlation spectroscopy, is powerful in analyzing the rather complicated IR region, and may be able to identify
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
47
Fig. 2. Synchronous 2D FT-IR correlation spectra in the range of 3100–2700 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
Fig. 1. FT-IR spectra of the pure components and blends recorded at room temperature in 2700–3200 cm−1 (a) and 900–200-cm−1 (b) regions: (1) PEA, (2) 8 CP/92 PEA, (3) 16 CP/84 PEA, (4) 32 CP/68 PEA, (5) 64 CP/36 PEA, (6) 80 CP/20 PEA, and (7) CP.
spectral signatures of the specific interaction in polymer blends. The 2D study of the PEA/CP blends has dual goals. One is to separate the bands of PEA from those of CP in the highly overlapped complex spectra of the blends. Another is to study the intermolecular interaction between PEA and CP in the blends. The first goal is achieved by the synchronous 2D correlation analysis of the FT-IR spectra while the second one by the asynchronous 2D correlation analysis. If infrared spectra of immiscible mixtures with systematically varying compositions are obtained, the synchronous 2D correlation maps can be calculated, but not their asynchronous counterparts. This is because, in principle, there should be little or no interaction between the blend components and the intensities of all the bands in the spectra should change linearly with concentration. Small frequency or band width shifts have a pronounced effect on Table 2 Bands assignments to CP in FT-IR spectra [21]. (cm−1 ) 2961 2947 2916 2890 2869 2849 2827 1739 1465 1436 1415 1382 1328 1308 1283 1266 1242 1220 1197 1177 1135 1100 1067 1030 1011 996 980
Assignments CH3 asymmetric stretching CH3 asymmetric stretching CH2 asymmetric stretching CH stretching CH3 symmetric stretching CH2 symmetric stretching CH stretching band of O–CH3 C O stretching CH2 asymmetric bending or C C stretching CH2 or CH3 asymmetric deformation C–C aromatic ring CH3 symmetric bending CH deformation C–O asymmetric stretching C–O and C–C stretching C–O asymmetric stretching C–O–C asymmetric stretching C–O asymmetric stretching OC–O symmetric stretching C–O symmetric stretching C–O–C stretching C–O–C asymmetric stretching C–O–C asymmetric stretching C–C deformation C–C deformation CH bending CH bending
the 2D correlation spectra. In our study both synchronous and asynchronous spectra were obtained. This indicates the existence of inter-molecular interaction between blend components. To keep the discussion simple, we show just three regions of the spectra: the C–H stretching region between 3100 and 2700 cm−1 , C O stretching region between 1980 and 1600 cm−1 , and C–O stretching and CH bending region 1600–800 cm−1 from both set A and set B. 3.1. CH stretching bands In the spectral region from 3100 to 2700 cm−1 of PEA and CP the bands are due to aliphatic CH stretching vibrations, as listed in Tables 1 and 2. Fig. 2 shows the synchronous correlation maps of sets A and B for experimental (Fig. 2a and b) data in this region. In these spectra two auto-peaks at 2866 and 2917 cm−1 and one positive cross-peak at 2866 cm−1 vs. 2917 cm−1 were evidenced. The assignment of the bands to CP and PEA is consistent with the sign of the cross-peaks. The cross-peak is positive for bands due to the same components of blend and negative for bands due to different components [18]. The band at 2917 cm−1 must be due to CP, as can be seen from Fig. 1 and Table 2. It has positive correlation with the band at 2866 cm−1 . This fact indicates that the band at 2866 cm−1 is also due to the CP and that the intensity change of these bands has the same direction. Fig. 3 depicts the asynchronous correlation map of the composition-dependent FT-IR spectra in this spectral region. All peaks from Fig. 3a and b imply out-of-phase variations between CP and PEA bands. In Fig. 3a of set A, two new bands are evidenced at 2880 cm−1 and 2958 cm−1 . These bands have out of phase variations with CP bands at 2866 cm−1 and 2917 cm−1 , and most probably they come from PEA. In case of asynchronous spectrum of set B (Fig. 3b), five new peaks are evidenced at 2835 cm−1 , 2850 cm−1 , 2871 cm−1 , 2904 cm−1 and 2958 cm−1 . Comparing
Fig. 3. Asynchronous 2D FT-IR correlation spectra in the range of 3100–2700 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
48
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
Table 3 Synchronous and asynchronous 2D correlation intensities and order of intensity variation between two bands of set A. No.
˚
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
˚ (2958, 2848) > 0 ˚ (2958, 2914) > 0 ˚ (2880, 2848) > 0 ˚ (2871, 2835) > 0 ˚ (1737, 1708) > 0 ˚ (1765, 1737) > 0 ˚ (1384, 1467) > 0 ˚ (1384, 1284) > 0 ˚ (1384, 1230) > 0 ˚ (1384, 1207) > 0 ˚ (1384, 1064) > 0 ˚ (1384, 1041) > 0 ˚ (1269, 1064) > 0 ˚ (1230, 1170) > 0 ˚ (1211, 1170) > 0 ˚ (1170, 1128) > 0 ˚ (1170, 1064) > 0
a
(2958, 2848) < 0 (2958, 2914) < 0 (2880, 2848) < 0 (2871, 2835) < 0 (1737, 1708) > 0 (1765, 1737) < 0 (1384, 1467) > 0 (1384, 1284) > 0 (1384, 1230) > 0 (1384, 1207) > 0 (1384, 1064) > 0 (1384, 1041) > 0 (1269, 1064) > 0 (1230, 1170) < 0 (1211, 1170) < 0 (1170, 1128) > 0 (1170, 1064) > 0
Assignment
Ordera
(PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP)
2958 after 2848 2958 after 2914 2880 after 2848 2871 after 2835 1737 before 1708 1760 after 1737 1384 before 1467 1384 before 1284 1384 before 1230 1384 before 1207 1384 before 1064 1384 before 1041 1269 before 1064 1230 after 1170 1211 after 1170 1170 before 1128 1170 before 1064
1 after (before) 2 means the intensity change of the band at 1 occurs at higher (lower) CP contents than that at 2 .
Table 4 Synchronous and asynchronous 2D correlation intensities and order of intensity variation between two bands of set B. No.
˚
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
˚ (2958, 2846) > 0 ˚ (2958, 2916) > 0 ˚ (2880, 2848) > 0 ˚ (2871, 2835) > 0 ˚ (1741, 1714) > 0 ˚ (1753, 1741) > 0 ˚ (1380, 1467) > 0 ˚ (1380, 1270) > 0 ˚ (1380, 1245) > 0 ˚ (1380, 1205) >0 ˚ (1380, 1072) > 0 ˚ (1380, 1051) > 0 ˚ (1264, 1072) > 0 ˚ (1230, 1172) > 0 ˚ (1207, 1172) > 0 ˚ (1172, 1122) > 0 ˚ (1172, 1074) > 0
a
(2958, 2846) > 0 (2958, 2916) > 0 (2880, 2848) > 0 (2871, 2835) > 0 (1741, 1714) < 0 (1753, 1741) > 0 (1380, 1467) < 0 (1380, 1270) < 0 (1380, 1245) < 0 (1380, 1205) < 0 (1380, 1072) < 0 (1380, 1051) < 0 (1264, 1072) < 0 (1230, 1172) > 0 (1207, 1172) > 0 (1172, 1122) < 0 (1172, 1074) < 0
Assignment
Ordera
(PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP)
2958 before 2846 2958 before 2916 2880 before 2848 2871 before 2835 1741 after 1714 1753 before 1741 1380 after 1467 1380 after 1270 1380 after 1245 1380 after 1205 1380 after 1072 1380 after 1051 1264 after 1072 1230 before 1172 1207 before 1172 1172 after 1122 1172 after 1074
1 after (before) 2 means the intensity change of the band at 1 occurs at higher (lower) CP contents than that at 2 .
Fig. 3a and b, can be observed that the asynchronous peak at 2917 cm−1 vs. 2880 cm−1 appear only in set A, while those at 2917 cm−1 vs. 2904 cm−1 , 2917 cm−1 vs. 2946 cm−1 , 2850 cm−1 vs. 2835 cm−1 appear only in set B. All of these peaks should reflect the particular conformational features of set A and set B. The asynchronous peaks at 2958 cm−1 vs. 2914 cm−1 , 2958 cm−1 vs. 2848 cm−1 , 2871 vs. 2835 cm−1 and 2880 cm−1 vs. 2848 cm−1 appear in both Fig. 3a and b. They have opposite sign from set A to set B (see Tables 3 and 4). Thus, these asynchronous peaks are symptomatic of the specific interaction in the blends. The bands at 2917 cm−1 , 2848 cm−1 and 2871 cm−1 are due to CH2 and CH3 stretching vibrations of CP (Table 2), while the bands at 2958 cm−1 and 2880 cm−1 are due to CH2 stretching vibration of PEA (Table 1). The assignment to asymmetric CH2 stretching vibrations of the infrared bands near 2950 cm−1 and to symmetric CH2 stretching modes of chains in the TGT (trans-gauche- trans) conformation of the O–C–C–O sequence of bonds characteristic of the ordered chain conformation in the crystalline state of the bands near 2890 cm−1 was evidenced by Matsuura and Mingfu [22,23]. It is concluded that the CH2 and CH3 groups of CP and the CH2 groups of PEA are involved in the mixing. Probably –CH– chain fragments from PEA are entangled with CH3 –(CH2 )14 – chain fragment from CP assuming a partial miscibility. The CH3 stretching mode can be also involve in inter- and intra-molecular hydrogen bonds. The participation of the hydrogen bonding to asymmetric CH3 to the hydrogen bonds has been demonstrated by quantum mechanics and exper-
imentally proved for many systems with similar structure such as: poly(3-hidroxybutyrare) [24], poly(3-hydroxybutyrate-co-3hydroxyhexanoate) [25], poly(l-lactide)/poly(d-lactide) stereocomplex [26], etc. 3.2. C O stretching band The synchronous correlation spectra for this region reveal an auto-peak at 1735 cm−1 (Fig. 4a) and two auto-peaks at 1749 and 1728 cm−1 and one positive cross-peak at 1749 cm−1 vs. 1728 cm−1
Fig. 4. Synchronous 2D FT-IR correlation spectra in the range of 1980–1600 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
49
Fig. 5. Asynchronous 2D FT-IR correlation spectra in the range of 1980–1600 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
Fig. 7. Asynchronous 2D FT-IR correlation spectra in the range of 1600–800 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
(Fig. 4b) from set A and set B, respectively. The band at 1749 cm−1 has positive correlation with band at 1728 cm−1 , therefore both bands arise from same component. Therefore, it seems that the band at 1735 cm−1 due to C O stretching mode contains two bands at 1749 and 1728 cm−1 , reflecting two different ester conformers. The corresponding asynchronous correlation maps in Fig. 5 resolves three bands at 1765, 1737 and 1708 cm−1 for set A and 1753, 1741 and 1714 cm−1 for set B. They should represent three conformations of the free ester group [27]. Since the ester carbonyl groups may act as proton acceptor, the formation of H-bond between carbonyl and hydroxyl or methyl groups is very likely [28]. The involvement of esters groups in H-bonding as proved by existence of the two bands at 1765 cm−1 due to free and 1737 cm−1 due to H-bonded ester C O groups, respectively. The band at 1708 cm−1 belongs to the stretching vibration of bonded C O groups. The order of intensity variation represented by the cross-peaks at 1737 vs. 1708 and 1765 vs. 1737 is listed in Table 3 for set A and Table 4 for set B. The 5th row of Tables 3 and 4 show that the intensity of the band at 1737 cm−1 tends to occur at lower CP contents, in comparison with the intensity variation of the band at 1708 cm−1 . The band at 1737 cm−1 due to the C O stretching vibration of CP intensity increase primarily proportional to the weight percentage of CP content, while the 1708 cm−1 band intensity decrease is accompanied by the hydrogen bonding between PEA and CP. This result is reasonable, since more carbonyls can be hydrogenbonded with methyl’s when CP content increases. Comparing the 6th rows of Tables 3 and 4, it is seen that the behaviour of the band at 1765 cm−1 is opposite to that at 1737 cm−1 . The band at 1765 cm−1 must represent the ester conformation that gives hindrance to hydrogen bonding.
1257, 1168 and 1066 cm−1 are all due to the same component. The C–H deformation and C–O stretching bands of PEA from 1600 to 900 cm−1 region are very intense, while bands due to CP contribute little to the blend spectra from this region. It is reasonable to attribute to PEA all bands from 1600 to 900 cm−1 picked by 2D analysis. In synchronous spectrum of set B a band is evidenced at 1467 cm−1 that forms negative cross-peaks with all bands. All positive cross-peaks in Fig. 6b are about bands of the same component, while the negative cross-peaks are about bands of different components. The asynchronous maps of set A and set B in this region give cross-peaks that have opposite signs, as can be seen by comparing Fig. 7a with Fig. 7b. The order of intensity variation between two bands of set A and set B is listed in Tables 3 and 4, respectively. Tables 3 and 4 show that the event at 1384 cm−1 takes place before and after those at 1467, 1284, 1230, 1207, 1064 and 1041 cm−1 in set A and set B, respectively. Thus, the asynchronous peaks at 1684 vs. 1467, 1684 vs. 1284, 1684 vs. 1230, 1684 vs. 1207, 1684 vs. 1064 and 1684 vs. 1041 cm−1 in Fig. 7a and b should arise from the specific interaction that may be responsible for the blend compatibility. The same conclusion is also reached for the asynchronous peaks at 1269 vs. 1064, 1170 vs. 1230, 1170 vs. 1211, 1170 vs. 1128 and 1170 cm−1 vs. 1064 cm−1 by comparison of Tables 3 and 4. Consequently, the OH bending and C–O stretching vibrations of PEA and the CH3 bending and C–O stretching vibration of CP are indicative of the specific interaction between PEA and CP. It can be observed that all bands are shifted in one set comparing with the other, such frequency shifts of OH and C–O vibration are characteristic of H-bond formation. The results obtained by IR and 2D correlation spectroscopy are supported by the results obtained after treating of some textile materials with these blends [29]. On the treated materials was observed a decrease of the air permeability with increasing CP content, indicating compatibility between PEA and CP. Also, there is a noticeable decrease of hydrophilicity with increasing of the CP content. PEA treated samples are very hydrophilic due to the polymer structure, which shows the final polar –OH groups with high affinity for water. With increasing CP content, the samples became hydrophobic, due to involvement of OH groups to form hydrogen bonds with PEA.
3.3. C–O stretching and C–H bending vibrations The synchronous map in the region of 1600–800 cm−1 of set A is shown in Fig. 6a, where all cross-peaks are positive. It is noticed that all cross-peaks are positive, implying that the bands at 1388,
4. Conclusion
Fig. 6. Synchronous 2D FT-IR correlation spectra in the range of 1600–800 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
The present 2D FT-IR correlation spectroscopy study of composition-dependent spectral variations of PEA–CP blends revealed in detail the composition-induced structural changes in the blends. 2D correlation analysis of the spectra of the seven samples divided into two sets enables us to elucidate the band assignment and the structural changes for each set. The 2D synchronous correlation analysis separates the bands of PEA from those of CP and detects bands that are not identified from
50
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
the one-dimensional spectra of PEA and CP. The 2D asynchronous correlation analysis reveals many out-of-phase band intensity variations that are indicative of the conformational change or the specific interaction in the blends. In the first region, from 3100 to 2700 cm−1 , the CH2 and CH3 groups of CP and the CH groups of PEA are involved in blend formation The CH3 stretching mode being involve in inter and intramolecular hydrogen bonds. The band at 1736 cm−1 were resolved in three bands at 1765, 1737 and 1708 cm−1 for set A and 1753, 1741 and 1714 cm−1 for set B. The bands at 1765 cm−1 were assigned to free and 1737 cm−1 to H-bonded ester C O groups, respectively. The band at 1708 cm−1 belongs to the stretching vibration of bonded C O groups. Comparing the asynchronous spectra from set A and set B in this region, it can be concluded that carbonyls groups from PEA can be hydrogenbonded with methyl when CP content increases. From the last discussed region, the OH bending and C–O stretching vibrations of PEA and the CH3 bending and C–O stretching vibration of CP are indicative of the specific interaction between PEA and CP. Acknowledgements M.C. Popescu acknowledges the financial support of European Social Fund – “Cristofor I. Simionescu” Postdoctoral Fellowship Programme (ID POSDRU/89/1.5/S/55216), Sectoral Operational Programme Human Resources Development 2007–2013 References [1] K.F. Wissbrum, J. Rheol. 25 (1986) 619. [2] W. Huh, R.A. Weiss, L. Nicolais, Polym. Eng. Sci. 23 (1983) 779.
[3] A.K. Kalkar, A.A. Bhalwankar, Polym. Sci. 2 (1994) 573. [4] A.K. Kalkar, V.V. Kunte, A.A. Deshpande, Macromol. New Front., Proc. IUPAC Int. Symp. Adv. Polym. Sci. Technol. (1998). [5] O. Olabisi, L.M. Robeson, M.T. Shaw, Polymer–polymer miscibility, Academic Press, New York, 1979. [6] M.T. Shaw, Polymer blends and mixtures, in: D.J. Walsh, J.S. Higgins, A. Maconnachie (Eds.), NATO, Advanced Study Institute Series E89, Martinus Nijhoff Publishers, Boston, MA, 1985. [7] D.J. Walsh, S. Rostami, Adv. Polym. Sci. 70 (1985) 119. [8] L.A. Ultracki, Polymer Alloy and Blends, Hanser Publishers, Munich, 1989. [9] D.S. Kaplan, J. Appl. Polym. Sci. 20 (1976) 2615. [10] M.M. Coleman, J.F. Graf, P.C. Painter, Specific Interaction and the Miscibility of Polymer Blends, Technomic Publishing Inc., P A Lancaster, 1991. [11] I. Noda, Appl. Spectrosc. 47 (1993) 1329. [12] S. Sasic, A. Muszynski, Y. Ozaki, Appl. Spectrosc. 55 (2001) 343. [13] I. Noda, Y. Ozaki, Two-Dimensional Correlation Spectroscopy, Wiley Interscience, 2004. [14] I. Noda, Vib. Spectrosc. 36 (2004) 143. Appl. Spectrosc. 54 (7) (2000) 994 (special issue). Vib. Spectrosc. 36 (2004) (special issue). [15] D. Filip, C.I. Simionescu, D. Macocinschi, I. Paraschiv, J. Therm. Anal. 65 (2001) 821. [16] D. Filip, C.I. Simionescu, D. Macocinschi, Thermochim. Acta 395 (2002) 217. [17] I. Noda, Bull. Am. Phys. Soc. 31 (1986) 520. [18] Y. Ren, T. Murakami, T. Nishioka, K. Nakashima, I. Noda, Y. Ozaki, J. Phys. Chem. B 104 (2000) 679. [19] D.W. Mayo, F.A. Miller, R.W. Hannah, Course Notes on the Interpretation of Infrared and Raman Spectra, John Wiley and Sons Inc., 2003. [20] J. Yang, P. Pan, T. Dong, Y. Inoue, Polymer 51 (2010) 807. [21] I. Dreissig, S. Machill, R. Salzer, C. Krafft, Spectrochim. Acta Part A 71 (2009) 2069. [22] H. Matsuura, T. Miyazawa, Bull. Chem. Soc. Jpn. 41 (1968) 1798. [23] M. Lu, J. Runt, P. Painter, Macromolecules 42 (2009) 6581. [24] A. Padermshoke, Y. Katsumoto, H. Sato, S. Ekgasit, I. Noda, Y. Ozaki, Spechim. Acta Part A 61 (2005) 541. [25] H. Sato, J. Dybal, R. Murakami, I. Noda, Y. Ozaki, J. Mol. Struct. 35 (2005) 744. [26] J. Zhang, H. Sato, H. Tsuji, I. Noda, Y. Ozaki, Macromolecules 38 (2005) 1822. [27] P. Pan, Y. Inoue, Prog. Polym. Sci. 34 (2009) 605. [28] J. Zhang, H. Sato, I. Noda, Y. Ozaki, Macromolecules 38 (2005) 4274. [29] A. Grigoriu, D. Macocinschi, D. Filip, Int. Sci. Session, ALTEXIM II (2000) Sibiu.
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Two-dimensional infrared correlation spectroscopic studies of polymer blends—Specific interactions in polyethylene adipate/cholesteryl palmitate blends Maria Cristina Popescu ∗ , Cornelia Vasile Romanian Academy, “Petru Poni” Institute of Macromolecular Chemistry, 41A Grigore Ghica Voda Alley, 700487 Iasi, Romania
a r t i c l e
i n f o
Article history: Received 8 July 2010 Received in revised form 28 December 2010 Accepted 26 January 2011 Keywords: Specific interactions Liquid crystal Blends 2D correlation spectroscopy
a b s t r a c t Generalized two-dimensional infrared (2D IR) correlation spectroscopy has been applied to the study of the conformational changes and specific interactions in blends of polyethylene adipate (PEA) and cholesteryl palmitate (CP). IR spectra of CP, PEA, and their blends of different compositions: 10/90, 16/84, 32/68, 64/36, and 80/20 have been recorded. In order to apply 2D IR correlation analysis, the samples are divided into two sets: set A with high PEA content 0/100, 10/90, 16/84, 32/68 CP/PEA and set B with high CP content 64/36, 80/20 and 100/0 CP/PEA. The 2D IR synchronous correlation analysis separates the bands of PEA from those of CP. The cross-peaks in 2D IR asynchronous correlation analysis are indicative of the specific interaction or the conformational change in the blends. The bands of C O, OH and C–O vibrations of PEA, and CH3 and C–O vibrations of CP that are very sensitive to the intermolecular hydrogen bonding effect and consequently the partial miscibility of components, have been assigned by 2D correlation analysis. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Blending is one of the most important areas in polymer technology. Polymer blends found a considerably widespread range of applications, from consumer plastics of common everyday use to special materials. The continuing fervent interest in these materials is twofold. Firstly, the potential to combine characteristic properties of different polymers in a single processable material permits the expansion of range of applications of each component of the system. Secondly, if by blending a synergic effect appears, specific properties can be achieved. These permit the design of materials capable to perform determined functions. Specific interactions between chemical groups of the blend components play an important role in polymer mixtures. The physical properties of polymer blends are strongly influenced by blending conditions and processes that, in turn, affect the level of mixing of the blends, so there is a growing interest in the study of miscibility and phase behaviour of polymer blends. The incorporation of low molecular weight liquid crystals in amorphous polymers has been suggested as an alternative method to bring about modest improvements in the mechanical properties of polymeric composites [1–4]. By covering the textile materials
∗ Corresponding author. Tel.: +40 232 217454. E-mail address: [email protected] (M.C. Popescu). 1386-1425/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2011.01.042
with polymeric compounds one can obtain products with special properties and large applicability. Various techniques have been employed to investigate miscibility of polymer blends, such as microscopy, thermal analysis, dynamic mechanical analysis, dielectric measurements, X rays diffraction, and spectroscopy [5–8]. Differential scanning calorimetry (DSC) is one of the most widely used techniques for evaluating ˚ in terms of cooperative motion miscibility on a scale of 100–300 A, of polymer segments around the glass transition temperature (Tg ) [9]. Infrared (IR) absorption is sensitive to the local environment of the oscillating dipoles, and has proven to be a powerful technique for investigating intermolecular interactions [10]. Two-dimensional (2D) correlation analysis/spectroscopy has been known for about 15 years [11–13]. It has been employed to solve various problems predominantly in vibrational spectroscopy mainly related with bands’ overlapping. Several review articles and books are available on this subject [14]. The method is almost exclusively based on comparing/correlating the intensities of the bands. In the case of polymer blends, the technique of two-dimensional 2D IR correlation spectroscopy offers new possibilities for the analysis and interpretation of vibrational spectra. 2D correlation analysis is performed from a set of spectra collected from a system under an external perturbation that induces alterations in the spectrum of the system. In the 2D approach, the bands are spread over the second dimension, thereby simplifying the visualization of spectra consisting of overlapping bands. Inter- and intra-molecular
46
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
interactions can be investigated by correlating absorption bands from different functional groups or parts of a molecule. PEA/CP blends with different composition were studied by thermogravimetry (TG) and differential scanning calorimetry (DSC). Thermogravimetric data were used to establish the degree of interaction in blends [15,16]. It has been established that PEA/CP blends are characterized by good thermal stability and partial miscibility. In the present study, generalized 2D IR correlation spectroscopy has been applied to explore much deeply the intermolecular interactions in blends of polyethylene adipate (PEA) and cholesteryl palmitate (CP). 2. Experimental 2.1. Materials Polyethylene adipate (PEA) with end-capped hydroxyl groups is a commercial product purchased from Fibrex SA Savinesti, Romania (Mn = 2000 g/mol) with the following chemical formula:
Cholesteryl palmitate (CP) was used as a low-molecular-weight additive. It was purchased from Nopris SRL, Cluj Napoca, Romania, and used as received. It has the following structure:
CP is a compound with liquid crystalline properties characterized by two types of mesophases: i.e., cholesteric (Ch) and smectic (S). Monotropic phases are obtained from an isotropic melt (I) in agreement with DSC (4 ◦ C/min) determinations. The transitions between these phases are: (C → Ch) at 74.5 ◦ C, (Ch → I) at 80 ◦ C, and (Ch → S) at 73.5 ◦ C. 2.2. Blends preparation Semi-crystalline PEA and liquid crystalline CP were separately dissolved in 1,2-dichloroethane (DCE) to form 0.8 g/dL solutions. The solutions were mixed to obtain the following CP/PEA ratios (wt/wt): (1) 10/90, (2) 16/84, (3) 32/68, (4) 64/36, and (5) 80/20. The mixtures of solutions were stirred for 5 h. After that, the solvent was slowly evaporated at room temperature. To remove the residual solvent and moisture, the samples were dried in a vacuum oven at 50 ◦ C for 2–6 days and total removal of the solvent was checked by IR spectroscopy. 2.3. Investigation methods 2.3.1. FT-IR spectroscopy FT-IR spectra were recorded on solid samples in KBr pellets by means of a FT-IR DIGILAB Scimitar Series spectrometer (USA) with a resolution of 4 cm−1 . The concentration of the samples in the tablets was constant at 5 mg/500 mg KBr. Five recordings were performed for each sample after successive millings. The evaluations were made on the average spectrum obtained from these five recordings. Processing of the spectra was done by means of Grams/32 program (Galactic Industry Corp.).
Table 1 Bands assignment to PEA in FTIR spectra [19,20]. (cm−1 )
Assignments
2961 2881 1736 1465 1438 1412 1389 1274 1257 1168 1145 1083 1056 985 961
CH2 asymmetric stretching CH2 symmetric stretching C O stretching CH2 asymmetrical bending CH2 symmetric bending CH2 symmetrical bending CH2 asymmetrical wagging O–H deformation C–O–C asymmetric stretching C–O–C asymmetric stretching C–O–C symmetric stretching C–O–C stretching C–C stretching mode C–C stretching mode CH2 rocking CH2 rocking
2.3.2. 2D correlation spectroscopy 2D FT-IR correlation intensities were calculated using MATLAB program derived from the generalized 2D correlation method developed by Noda [17]. The FT-IR spectra of PEA and its blends with CP were divided into two sets: set A and set B according with procedure described by Ren et al. [18]. Set A contains the FT-IR spectra of (1) PEA, (2) 84% PEA–16% CP and (3) 68% PEA–32% CP. Set B contains the FTIR spectra of (4) 36% PEA–64% CP, (5) 20% PEA–80% CP and (6) CP. The spectra in set A were arranged in the order 1–2–3. The spectra in set B were arranged in the order 4–5–6. Thus, in both sets, the intensities of the bands of PEA are decreasing while those of CP are increasing. The reference spectra used for set A and B were those of the sample 1 and 4, respectively. In 2D correlation analysis, two kinds of correlation maps synchronous and asynchronous are generated from a set of dynamic spectra obtained from the modulation experiment [11]. According to reference [18], by using the function ˚ (which designates synchronous 2D correlation), the intensities between two bands can be written by the following relations: ˚[(PEA), (PEA)] > 0 ˚[(PEA), (CP)] < 0
˚[(CP), (CP)] > 0 ˚[(CP), (PEA)] < 0
Based on these relations, it is possible to ascribe and differentiate of the bands in the spectra of the blends to PEA or CP. An asynchronous correlation peak appears only if the intensity change of two bands at 1 and 2 has basically dissimilar trends. According to Noda rules [18] ˚[1 , 2 ] < 0 and [1 , 2 ] > 0 imply that the intensity change at 1 occurs at higher PEA content compared to that at 2 . So do ˚[1 , 2 ] > 0 and [1 , 2 ] < 0. If ˚[1 , 2 ] < 0 and [1 , 2 ] < 0 imply that the intensity change at 1 takes place at lower PEA content than that at 2 . So do also ˚[1 , 2 ] > 0 and [1 , 2 ] > 0. When an asynchronous peak [1 , 2 ] appears between two bands at 1 and 2 , it can be either due to the conformational change of the component polymers in their blends or to the specific interactions between the component polymers [18]. If the sequential order between 1 and 2 in set A is reversed in set B, then the band at 1 and 2 can be regarded as indicative of the specific interactions between PEA and CP. 3. Results and discussion The FT-IR spectra of PEA, CP and blends at room temperature are shown in Fig. 1 and were used to evidence the region of interest for 2D correlation spectroscopy study. The assignment of FT-IR bands of PEA and CP are listed in Tables 1 and 2. Generalized 2D correlation spectroscopy, is powerful in analyzing the rather complicated IR region, and may be able to identify
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
47
Fig. 2. Synchronous 2D FT-IR correlation spectra in the range of 3100–2700 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
Fig. 1. FT-IR spectra of the pure components and blends recorded at room temperature in 2700–3200 cm−1 (a) and 900–200-cm−1 (b) regions: (1) PEA, (2) 8 CP/92 PEA, (3) 16 CP/84 PEA, (4) 32 CP/68 PEA, (5) 64 CP/36 PEA, (6) 80 CP/20 PEA, and (7) CP.
spectral signatures of the specific interaction in polymer blends. The 2D study of the PEA/CP blends has dual goals. One is to separate the bands of PEA from those of CP in the highly overlapped complex spectra of the blends. Another is to study the intermolecular interaction between PEA and CP in the blends. The first goal is achieved by the synchronous 2D correlation analysis of the FT-IR spectra while the second one by the asynchronous 2D correlation analysis. If infrared spectra of immiscible mixtures with systematically varying compositions are obtained, the synchronous 2D correlation maps can be calculated, but not their asynchronous counterparts. This is because, in principle, there should be little or no interaction between the blend components and the intensities of all the bands in the spectra should change linearly with concentration. Small frequency or band width shifts have a pronounced effect on Table 2 Bands assignments to CP in FT-IR spectra [21]. (cm−1 ) 2961 2947 2916 2890 2869 2849 2827 1739 1465 1436 1415 1382 1328 1308 1283 1266 1242 1220 1197 1177 1135 1100 1067 1030 1011 996 980
Assignments CH3 asymmetric stretching CH3 asymmetric stretching CH2 asymmetric stretching CH stretching CH3 symmetric stretching CH2 symmetric stretching CH stretching band of O–CH3 C O stretching CH2 asymmetric bending or C C stretching CH2 or CH3 asymmetric deformation C–C aromatic ring CH3 symmetric bending CH deformation C–O asymmetric stretching C–O and C–C stretching C–O asymmetric stretching C–O–C asymmetric stretching C–O asymmetric stretching OC–O symmetric stretching C–O symmetric stretching C–O–C stretching C–O–C asymmetric stretching C–O–C asymmetric stretching C–C deformation C–C deformation CH bending CH bending
the 2D correlation spectra. In our study both synchronous and asynchronous spectra were obtained. This indicates the existence of inter-molecular interaction between blend components. To keep the discussion simple, we show just three regions of the spectra: the C–H stretching region between 3100 and 2700 cm−1 , C O stretching region between 1980 and 1600 cm−1 , and C–O stretching and CH bending region 1600–800 cm−1 from both set A and set B. 3.1. CH stretching bands In the spectral region from 3100 to 2700 cm−1 of PEA and CP the bands are due to aliphatic CH stretching vibrations, as listed in Tables 1 and 2. Fig. 2 shows the synchronous correlation maps of sets A and B for experimental (Fig. 2a and b) data in this region. In these spectra two auto-peaks at 2866 and 2917 cm−1 and one positive cross-peak at 2866 cm−1 vs. 2917 cm−1 were evidenced. The assignment of the bands to CP and PEA is consistent with the sign of the cross-peaks. The cross-peak is positive for bands due to the same components of blend and negative for bands due to different components [18]. The band at 2917 cm−1 must be due to CP, as can be seen from Fig. 1 and Table 2. It has positive correlation with the band at 2866 cm−1 . This fact indicates that the band at 2866 cm−1 is also due to the CP and that the intensity change of these bands has the same direction. Fig. 3 depicts the asynchronous correlation map of the composition-dependent FT-IR spectra in this spectral region. All peaks from Fig. 3a and b imply out-of-phase variations between CP and PEA bands. In Fig. 3a of set A, two new bands are evidenced at 2880 cm−1 and 2958 cm−1 . These bands have out of phase variations with CP bands at 2866 cm−1 and 2917 cm−1 , and most probably they come from PEA. In case of asynchronous spectrum of set B (Fig. 3b), five new peaks are evidenced at 2835 cm−1 , 2850 cm−1 , 2871 cm−1 , 2904 cm−1 and 2958 cm−1 . Comparing
Fig. 3. Asynchronous 2D FT-IR correlation spectra in the range of 3100–2700 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
48
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
Table 3 Synchronous and asynchronous 2D correlation intensities and order of intensity variation between two bands of set A. No.
˚
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
˚ (2958, 2848) > 0 ˚ (2958, 2914) > 0 ˚ (2880, 2848) > 0 ˚ (2871, 2835) > 0 ˚ (1737, 1708) > 0 ˚ (1765, 1737) > 0 ˚ (1384, 1467) > 0 ˚ (1384, 1284) > 0 ˚ (1384, 1230) > 0 ˚ (1384, 1207) > 0 ˚ (1384, 1064) > 0 ˚ (1384, 1041) > 0 ˚ (1269, 1064) > 0 ˚ (1230, 1170) > 0 ˚ (1211, 1170) > 0 ˚ (1170, 1128) > 0 ˚ (1170, 1064) > 0
a
(2958, 2848) < 0 (2958, 2914) < 0 (2880, 2848) < 0 (2871, 2835) < 0 (1737, 1708) > 0 (1765, 1737) < 0 (1384, 1467) > 0 (1384, 1284) > 0 (1384, 1230) > 0 (1384, 1207) > 0 (1384, 1064) > 0 (1384, 1041) > 0 (1269, 1064) > 0 (1230, 1170) < 0 (1211, 1170) < 0 (1170, 1128) > 0 (1170, 1064) > 0
Assignment
Ordera
(PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP)
2958 after 2848 2958 after 2914 2880 after 2848 2871 after 2835 1737 before 1708 1760 after 1737 1384 before 1467 1384 before 1284 1384 before 1230 1384 before 1207 1384 before 1064 1384 before 1041 1269 before 1064 1230 after 1170 1211 after 1170 1170 before 1128 1170 before 1064
1 after (before) 2 means the intensity change of the band at 1 occurs at higher (lower) CP contents than that at 2 .
Table 4 Synchronous and asynchronous 2D correlation intensities and order of intensity variation between two bands of set B. No.
˚
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
˚ (2958, 2846) > 0 ˚ (2958, 2916) > 0 ˚ (2880, 2848) > 0 ˚ (2871, 2835) > 0 ˚ (1741, 1714) > 0 ˚ (1753, 1741) > 0 ˚ (1380, 1467) > 0 ˚ (1380, 1270) > 0 ˚ (1380, 1245) > 0 ˚ (1380, 1205) >0 ˚ (1380, 1072) > 0 ˚ (1380, 1051) > 0 ˚ (1264, 1072) > 0 ˚ (1230, 1172) > 0 ˚ (1207, 1172) > 0 ˚ (1172, 1122) > 0 ˚ (1172, 1074) > 0
a
(2958, 2846) > 0 (2958, 2916) > 0 (2880, 2848) > 0 (2871, 2835) > 0 (1741, 1714) < 0 (1753, 1741) > 0 (1380, 1467) < 0 (1380, 1270) < 0 (1380, 1245) < 0 (1380, 1205) < 0 (1380, 1072) < 0 (1380, 1051) < 0 (1264, 1072) < 0 (1230, 1172) > 0 (1207, 1172) > 0 (1172, 1122) < 0 (1172, 1074) < 0
Assignment
Ordera
(PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (PEA, CP) (CP, PEA) (CP, PEA) (PEA, CP) (PEA, CP)
2958 before 2846 2958 before 2916 2880 before 2848 2871 before 2835 1741 after 1714 1753 before 1741 1380 after 1467 1380 after 1270 1380 after 1245 1380 after 1205 1380 after 1072 1380 after 1051 1264 after 1072 1230 before 1172 1207 before 1172 1172 after 1122 1172 after 1074
1 after (before) 2 means the intensity change of the band at 1 occurs at higher (lower) CP contents than that at 2 .
Fig. 3a and b, can be observed that the asynchronous peak at 2917 cm−1 vs. 2880 cm−1 appear only in set A, while those at 2917 cm−1 vs. 2904 cm−1 , 2917 cm−1 vs. 2946 cm−1 , 2850 cm−1 vs. 2835 cm−1 appear only in set B. All of these peaks should reflect the particular conformational features of set A and set B. The asynchronous peaks at 2958 cm−1 vs. 2914 cm−1 , 2958 cm−1 vs. 2848 cm−1 , 2871 vs. 2835 cm−1 and 2880 cm−1 vs. 2848 cm−1 appear in both Fig. 3a and b. They have opposite sign from set A to set B (see Tables 3 and 4). Thus, these asynchronous peaks are symptomatic of the specific interaction in the blends. The bands at 2917 cm−1 , 2848 cm−1 and 2871 cm−1 are due to CH2 and CH3 stretching vibrations of CP (Table 2), while the bands at 2958 cm−1 and 2880 cm−1 are due to CH2 stretching vibration of PEA (Table 1). The assignment to asymmetric CH2 stretching vibrations of the infrared bands near 2950 cm−1 and to symmetric CH2 stretching modes of chains in the TGT (trans-gauche- trans) conformation of the O–C–C–O sequence of bonds characteristic of the ordered chain conformation in the crystalline state of the bands near 2890 cm−1 was evidenced by Matsuura and Mingfu [22,23]. It is concluded that the CH2 and CH3 groups of CP and the CH2 groups of PEA are involved in the mixing. Probably –CH– chain fragments from PEA are entangled with CH3 –(CH2 )14 – chain fragment from CP assuming a partial miscibility. The CH3 stretching mode can be also involve in inter- and intra-molecular hydrogen bonds. The participation of the hydrogen bonding to asymmetric CH3 to the hydrogen bonds has been demonstrated by quantum mechanics and exper-
imentally proved for many systems with similar structure such as: poly(3-hidroxybutyrare) [24], poly(3-hydroxybutyrate-co-3hydroxyhexanoate) [25], poly(l-lactide)/poly(d-lactide) stereocomplex [26], etc. 3.2. C O stretching band The synchronous correlation spectra for this region reveal an auto-peak at 1735 cm−1 (Fig. 4a) and two auto-peaks at 1749 and 1728 cm−1 and one positive cross-peak at 1749 cm−1 vs. 1728 cm−1
Fig. 4. Synchronous 2D FT-IR correlation spectra in the range of 1980–1600 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
49
Fig. 5. Asynchronous 2D FT-IR correlation spectra in the range of 1980–1600 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
Fig. 7. Asynchronous 2D FT-IR correlation spectra in the range of 1600–800 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
(Fig. 4b) from set A and set B, respectively. The band at 1749 cm−1 has positive correlation with band at 1728 cm−1 , therefore both bands arise from same component. Therefore, it seems that the band at 1735 cm−1 due to C O stretching mode contains two bands at 1749 and 1728 cm−1 , reflecting two different ester conformers. The corresponding asynchronous correlation maps in Fig. 5 resolves three bands at 1765, 1737 and 1708 cm−1 for set A and 1753, 1741 and 1714 cm−1 for set B. They should represent three conformations of the free ester group [27]. Since the ester carbonyl groups may act as proton acceptor, the formation of H-bond between carbonyl and hydroxyl or methyl groups is very likely [28]. The involvement of esters groups in H-bonding as proved by existence of the two bands at 1765 cm−1 due to free and 1737 cm−1 due to H-bonded ester C O groups, respectively. The band at 1708 cm−1 belongs to the stretching vibration of bonded C O groups. The order of intensity variation represented by the cross-peaks at 1737 vs. 1708 and 1765 vs. 1737 is listed in Table 3 for set A and Table 4 for set B. The 5th row of Tables 3 and 4 show that the intensity of the band at 1737 cm−1 tends to occur at lower CP contents, in comparison with the intensity variation of the band at 1708 cm−1 . The band at 1737 cm−1 due to the C O stretching vibration of CP intensity increase primarily proportional to the weight percentage of CP content, while the 1708 cm−1 band intensity decrease is accompanied by the hydrogen bonding between PEA and CP. This result is reasonable, since more carbonyls can be hydrogenbonded with methyl’s when CP content increases. Comparing the 6th rows of Tables 3 and 4, it is seen that the behaviour of the band at 1765 cm−1 is opposite to that at 1737 cm−1 . The band at 1765 cm−1 must represent the ester conformation that gives hindrance to hydrogen bonding.
1257, 1168 and 1066 cm−1 are all due to the same component. The C–H deformation and C–O stretching bands of PEA from 1600 to 900 cm−1 region are very intense, while bands due to CP contribute little to the blend spectra from this region. It is reasonable to attribute to PEA all bands from 1600 to 900 cm−1 picked by 2D analysis. In synchronous spectrum of set B a band is evidenced at 1467 cm−1 that forms negative cross-peaks with all bands. All positive cross-peaks in Fig. 6b are about bands of the same component, while the negative cross-peaks are about bands of different components. The asynchronous maps of set A and set B in this region give cross-peaks that have opposite signs, as can be seen by comparing Fig. 7a with Fig. 7b. The order of intensity variation between two bands of set A and set B is listed in Tables 3 and 4, respectively. Tables 3 and 4 show that the event at 1384 cm−1 takes place before and after those at 1467, 1284, 1230, 1207, 1064 and 1041 cm−1 in set A and set B, respectively. Thus, the asynchronous peaks at 1684 vs. 1467, 1684 vs. 1284, 1684 vs. 1230, 1684 vs. 1207, 1684 vs. 1064 and 1684 vs. 1041 cm−1 in Fig. 7a and b should arise from the specific interaction that may be responsible for the blend compatibility. The same conclusion is also reached for the asynchronous peaks at 1269 vs. 1064, 1170 vs. 1230, 1170 vs. 1211, 1170 vs. 1128 and 1170 cm−1 vs. 1064 cm−1 by comparison of Tables 3 and 4. Consequently, the OH bending and C–O stretching vibrations of PEA and the CH3 bending and C–O stretching vibration of CP are indicative of the specific interaction between PEA and CP. It can be observed that all bands are shifted in one set comparing with the other, such frequency shifts of OH and C–O vibration are characteristic of H-bond formation. The results obtained by IR and 2D correlation spectroscopy are supported by the results obtained after treating of some textile materials with these blends [29]. On the treated materials was observed a decrease of the air permeability with increasing CP content, indicating compatibility between PEA and CP. Also, there is a noticeable decrease of hydrophilicity with increasing of the CP content. PEA treated samples are very hydrophilic due to the polymer structure, which shows the final polar –OH groups with high affinity for water. With increasing CP content, the samples became hydrophobic, due to involvement of OH groups to form hydrogen bonds with PEA.
3.3. C–O stretching and C–H bending vibrations The synchronous map in the region of 1600–800 cm−1 of set A is shown in Fig. 6a, where all cross-peaks are positive. It is noticed that all cross-peaks are positive, implying that the bands at 1388,
4. Conclusion
Fig. 6. Synchronous 2D FT-IR correlation spectra in the range of 1600–800 cm−1 , drawn from FT-IR spectra of set A (a) and set B (b).
The present 2D FT-IR correlation spectroscopy study of composition-dependent spectral variations of PEA–CP blends revealed in detail the composition-induced structural changes in the blends. 2D correlation analysis of the spectra of the seven samples divided into two sets enables us to elucidate the band assignment and the structural changes for each set. The 2D synchronous correlation analysis separates the bands of PEA from those of CP and detects bands that are not identified from
50
M.C. Popescu, C. Vasile / Spectrochimica Acta Part A 79 (2011) 45–50
the one-dimensional spectra of PEA and CP. The 2D asynchronous correlation analysis reveals many out-of-phase band intensity variations that are indicative of the conformational change or the specific interaction in the blends. In the first region, from 3100 to 2700 cm−1 , the CH2 and CH3 groups of CP and the CH groups of PEA are involved in blend formation The CH3 stretching mode being involve in inter and intramolecular hydrogen bonds. The band at 1736 cm−1 were resolved in three bands at 1765, 1737 and 1708 cm−1 for set A and 1753, 1741 and 1714 cm−1 for set B. The bands at 1765 cm−1 were assigned to free and 1737 cm−1 to H-bonded ester C O groups, respectively. The band at 1708 cm−1 belongs to the stretching vibration of bonded C O groups. Comparing the asynchronous spectra from set A and set B in this region, it can be concluded that carbonyls groups from PEA can be hydrogenbonded with methyl when CP content increases. From the last discussed region, the OH bending and C–O stretching vibrations of PEA and the CH3 bending and C–O stretching vibration of CP are indicative of the specific interaction between PEA and CP. Acknowledgements M.C. Popescu acknowledges the financial support of European Social Fund – “Cristofor I. Simionescu” Postdoctoral Fellowship Programme (ID POSDRU/89/1.5/S/55216), Sectoral Operational Programme Human Resources Development 2007–2013 References [1] K.F. Wissbrum, J. Rheol. 25 (1986) 619. [2] W. Huh, R.A. Weiss, L. Nicolais, Polym. Eng. Sci. 23 (1983) 779.
[3] A.K. Kalkar, A.A. Bhalwankar, Polym. Sci. 2 (1994) 573. [4] A.K. Kalkar, V.V. Kunte, A.A. Deshpande, Macromol. New Front., Proc. IUPAC Int. Symp. Adv. Polym. Sci. Technol. (1998). [5] O. Olabisi, L.M. Robeson, M.T. Shaw, Polymer–polymer miscibility, Academic Press, New York, 1979. [6] M.T. Shaw, Polymer blends and mixtures, in: D.J. Walsh, J.S. Higgins, A. Maconnachie (Eds.), NATO, Advanced Study Institute Series E89, Martinus Nijhoff Publishers, Boston, MA, 1985. [7] D.J. Walsh, S. Rostami, Adv. Polym. Sci. 70 (1985) 119. [8] L.A. Ultracki, Polymer Alloy and Blends, Hanser Publishers, Munich, 1989. [9] D.S. Kaplan, J. Appl. Polym. Sci. 20 (1976) 2615. [10] M.M. Coleman, J.F. Graf, P.C. Painter, Specific Interaction and the Miscibility of Polymer Blends, Technomic Publishing Inc., P A Lancaster, 1991. [11] I. Noda, Appl. Spectrosc. 47 (1993) 1329. [12] S. Sasic, A. Muszynski, Y. Ozaki, Appl. Spectrosc. 55 (2001) 343. [13] I. Noda, Y. Ozaki, Two-Dimensional Correlation Spectroscopy, Wiley Interscience, 2004. [14] I. Noda, Vib. Spectrosc. 36 (2004) 143. Appl. Spectrosc. 54 (7) (2000) 994 (special issue). Vib. Spectrosc. 36 (2004) (special issue). [15] D. Filip, C.I. Simionescu, D. Macocinschi, I. Paraschiv, J. Therm. Anal. 65 (2001) 821. [16] D. Filip, C.I. Simionescu, D. Macocinschi, Thermochim. Acta 395 (2002) 217. [17] I. Noda, Bull. Am. Phys. Soc. 31 (1986) 520. [18] Y. Ren, T. Murakami, T. Nishioka, K. Nakashima, I. Noda, Y. Ozaki, J. Phys. Chem. B 104 (2000) 679. [19] D.W. Mayo, F.A. Miller, R.W. Hannah, Course Notes on the Interpretation of Infrared and Raman Spectra, John Wiley and Sons Inc., 2003. [20] J. Yang, P. Pan, T. Dong, Y. Inoue, Polymer 51 (2010) 807. [21] I. Dreissig, S. Machill, R. Salzer, C. Krafft, Spectrochim. Acta Part A 71 (2009) 2069. [22] H. Matsuura, T. Miyazawa, Bull. Chem. Soc. Jpn. 41 (1968) 1798. [23] M. Lu, J. Runt, P. Painter, Macromolecules 42 (2009) 6581. [24] A. Padermshoke, Y. Katsumoto, H. Sato, S. Ekgasit, I. Noda, Y. Ozaki, Spechim. Acta Part A 61 (2005) 541. [25] H. Sato, J. Dybal, R. Murakami, I. Noda, Y. Ozaki, J. Mol. Struct. 35 (2005) 744. [26] J. Zhang, H. Sato, H. Tsuji, I. Noda, Y. Ozaki, Macromolecules 38 (2005) 1822. [27] P. Pan, Y. Inoue, Prog. Polym. Sci. 34 (2009) 605. [28] J. Zhang, H. Sato, I. Noda, Y. Ozaki, Macromolecules 38 (2005) 4274. [29] A. Grigoriu, D. Macocinschi, D. Filip, Int. Sci. Session, ALTEXIM II (2000) Sibiu.