Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Raman spectroscopy as a tool in differentiating conjugated polyenes from synthetic and natural sources Rafaella F. Fernandes a, Lenize F. Maia a, Mara R.C. Couri a, Luiz Antonio S. Costa b, Luiz Fernando C. de Oliveira a,⇑ a b
NEEM – Núcleo de Espectroscopia e Estrutura Molecular, Departamento de Química, Universidade Federal de Juiz de Fora, 36036-330 Juiz de Fora, MG, Brazil NEQC – Núcleo de Estudos em Química Computacional, Departamento de Química, Universidade Federal de Juiz de Fora, 36036-330 Juiz de Fora, MG, Brazil
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Pigments derived from polyenals
were unequivocally identified by Raman spectroscopy. Evaluation of differences and similarities between synthetic and natural polyenals. Use of the in plane rocking methyl group modes as fingerprint Raman bands in polyenals. Lasers of different energies characterizing polyenals in mixtures. Tissues of corals are pigmented by more than one conjugated polyenal.
a r t i c l e
i n f o
Article history: Received 11 March 2014 Received in revised form 19 May 2014 Accepted 3 June 2014 Available online 14 June 2014 Keywords: Carotenoids Lycopene Polyenal Raman effect Corals DFT calculations
a b s t r a c t This work presents the Raman spectroscopic characterization of synthetic analogs of natural conjugated polyenals found in octocorals, focusing the unequivocal identification of the chemical species present in these systems. The synthetic material was produced by the autocondensation reaction of crotonaldehyde, generating a demethylated conjugated polyene containing 11 carbon–carbon double bonds, with just a methyl group on the end of the carbon chain. The resonance Raman spectra of such pigment has shown the existence of enhanced modes assigned to m1(C@C) and m2(CAC) modes of the main chain. For the resonance Raman spectra of natural pigments from octocorals collected in the Brazilian coast, besides the previously cited bands, it could be also observed the presence of the m4(CACH3), related to the vibrational mode who describes the vibration of the methyl group of the central carbon chain of carotenoids. Other interesting point is the observation of overtones and combination bands, which for carotenoids involves the presence of the m4 mode, whereas for the synthetic polyene this band, besides be seen at a slightly different wavenumber position, does not appear as an enhanced mode and also as a combination, such as for the natural carotenoids. Theoretical molecular orbital analysis of polyenal-11 and lycopene has shown the structural differences which are also responsible for the resonance Raman data, based on the appearance of the (ACH3) vibrational mode in the resonant transition only for lycopene. At last, the Raman band at ca. 1010 cm1, assigned to the (ACH3) vibrational mode, can be used for attributing the presence of each one of the conjugated polyenes: the resonance Raman spectrum containing the band at ca. 1010 cm1 refers to the carotenoid (in this case lycopene), and the absence of such band in resonance conditions refers to the polyenal (in this case the polyenal-11). Ó 2014 Elsevier B.V. All rights reserved.
⇑ Corresponding author. Tel./fax: +55 32 21023310. E-mail address: [email protected] (L.F.C. de Oliveira). http://dx.doi.org/10.1016/j.saa.2014.06.022 1386-1425/Ó 2014 Elsevier B.V. All rights reserved.
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Introduction Natural pigments belonging to the class of unsubstituted linear polyacetylenes have been mainly identified by Raman spectroscopy as animal biochromes from parrot’s feathers, octocorals, pearls and shells of mollusks [1–6]. In many birds, red, orange and yellow feathers are colored by carotenoid pigments, but parrots are an exception [4–6]; these organisms have used other conjugated polyenes to color their plumage in red, indicating an alternative and unique system of pigmentation which seems to be conserved evolutionarily in parrots [7]. Red feathers from the parrot Ara macao are composed of a mixture of polyunsaturated linear aldehydes (polyenals) containing 6–9 conjugated double bonds named as psittacofulvins [8,9]. Brown, purple, red, orange and yellow pigments from shells of several species of mollusks were identified as unsubstituted polyacetylenes molecules ranging from 9 to 12 conjugated double bonds [10,11]. Similar coloration pattern observed in colored cultured freshwater pearls was also identified as linear polyacetylenes ranging from 6 to 14 double bonds [1]. So far, purple, red, orange and yellow coenenchymes and sclerites investigated from octocorals have also been attributed to such polyacetylenes ranging from 8 to 12 C@C unsaturations [1,4–6,12,13]. The bioactivity of red pigments in parrots feathers have been reported in ecological assays as a signal of individual quality and parental investment [14]. Red psittacofulvins reduced microbial damage to parrot feathers [15] and synthetic analogs named as parrodienes presented antioxidant, anti-tumoral and anti-inflammatory activities [4,12,16]. They are also useful in preventing the diseases caused by free radicals and on treatment of cardiovascular and inflammatory damages [16,17]. The biological activity in octocorals is unknown, however purple pigments are produced locally as purple halo in damaged tissues [4,18] as a consequence of an inflammatory immune response [19,20]. Tissues are composed of sclerites which are calcium carbonate structures secreted by scleroblasts and the increasing of purple pigment has been suggested to play a protective function in corals [21] and because of that, some studies related that the presence of polyenes in corals may be related with specific infectious process [4]. An example of this phenomenon was stated for the Gorgonia spp., in which a purple halo resulting from an increase of purple sclerites surrounding the necrotic tissues [18]. Another case is the appearance of an unusual violet pigmentation surrounding necrotic tissue of Pyllogorgia dilatata, a Brazilian endemic gorgonian. Maia et al. [4] have reported that the purpling in a Brazilian octocoral is attributed to psittacofulvin derivatives, which until then have been found as exclusive in parrots. Determination of the chemical structure of linear polyacetylenes from marine origin is still under investigation [22]. Experimental data obtained from Raman spectroscopy and theoretical calculations have suggested structures similar to polyenals from parrots, which is the only source from terrestrial environment. As mentioned above Raman spectroscopy has been pointed to be one of most suitable tools to investigate conjugated systems, being extensively used in the characterization of carotenoids and polyenals or unsubstituted polyacetylenes [8,23–25]. Polyenic molecules are strong Raman scatters due to the strong electron/ vibration coupling, and their vibrational modes can be selectively enhanced by resonance excitation nearby or close to the p–p* electronic transition [13]. The two bands of highest intensity observed in these organic pigments are the stretching modes of the C@C double bond (m1) and the CAC single bond (m2) of the main polyenic chain. The first one occurs between 1450 and 1680 cm1 and the latter between
435
1070 and 1210 cm1 [10,26]. The number of conjugated double bonds in the main chain or other molecules like as CaCO3 bound to the polyene affects the Raman shifts. Another aspect to be analyzed is that an excitation with lasers of different energies can result in drastic Raman spectrum changes, mainly based on spectral intensity, when a single component of a polyene mixture can have their Raman bands enhanced or suppressed to resolve the mixture [27–29]. If suitable exciting is chosen to be in resonance with an electronic transition of the biochrome it follows that is observed an intensification of the band for the specific chromophoric group (m1). This observation is due to the resonance Raman scattering [3,8,11,28] and has been demonstrated to be an useful tool for the study of biologically relevant molecules in their natural state [30]. Therefore, observing the relevance of polyconjugated molecules in biochemistry, both in terms of coloration and preventing diseases, as well as its significance in the marine environment, in this work we make use of Raman technique to characterize synthetic analogs of the polyenes found in octocorals. In order to evaluate the biological activity of linear polyacetylenes identified from corals, we are describing a synthesis of a carbon chain with a number of double bonds between ten and twelve and the comparative analysis with natural polyenes found in coral species, in order to provide a better knowledge on the subject. Theoretical calculations also have been performed for the same compounds in order to reinforce the basis of the vibrational analysis, as well as to show the electronic properties for the conjugated polyenes. Methodology Experimental section The synthetic route to obtain a polyene with a number of double bonds between ten and twelve is an adaptation of the methodology used by Stradi et al. [16], based on the autocondensation of crotonaldehyde in acidic medium and using pyrrolidine as catalyst. Crotonaldehyde, acetic acid and pyrrolidine were supplied by Aldrich. Under reflux system, 20 mL of crotonaldehyde were added to a flask and kept under magnetic stirring for 15 min under nitrogen flow. After this time, 0.20 mL of pyrrolidine and 0.20 mL of acetic acid was slowly added; the mixture was left under magnetic stirring in nitrogen atmosphere at 50 °C for 30 min. Since the reaction was quite exothermic, the heating was done in glycerine bath. The reaction was cooled in an ice bath and 200 mL of ethyl ether were added under magnetic stirring. The brown product obtained (yield of 85%) was composed by a mixture of polyenals ranging from 7 to 13 conjugated double bonds with predominantly the polyenal containing conjugated 11 double bonds known as tetracosaundecaenal (Fig. 1). Fourier transform Raman spectra were carried out using a Bruker RFS 100 instrument and a Nd:YAG laser operating at 1064 nm, equipped with a Ge detector cooled with liquid nitrogen. It was used 4 cm1 of spectral resolution, and good signal-to-noise ratios were obtained with 516 scans, using a range of laser powers at the sample between 25 and 75 mW. SENTERRA dispersive Raman microscope instrument operating at 532 nm and 785 nm was also used equipped with cooled charge coupled detector with the incident laser beam focused on the sample with a 50 objective. Good spectra were obtained using a range of laser powers between 0.2 and 2.0 mW and 3 or 5 accumulations for 3 s, with a spectral resolution of 3 cm1. Infrared spectral data in mode of attenuated total reflection (ATR) were obtained with a Bruker Alpha FT-IR spectrometer and spectral resolution of 4 cm1 and good signal-to-noise ratios were obtained with 128 scans.
436
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 1. Scheme of the polyenal synthesis and a representative structure of the tetracosaundecaenal.
Calculations Full unconstrained geometry optimization and frequency calculations for all distinct polyenals have been carried out at the Hartree–Fock (HF) and Density Functional Theory (DFT) using the three-parameter fit of the Becke´s exchange correlation potential (B3LYP) [31,32] as implemented in the Gaussian 09 program suite [Gaussian]. The Pople’s triple-f-quality basis sets 6-311+G(d, p) [33] have been applied for all atoms in polyenes of 10, 11 and 12 C@C double bonds. Vibrational Raman calculated frequencies were scaled by a factor of 0.97 when compared to experimental data. In order to evaluate the polyenes frontier molecular orbitals to establish a comparison with a carotenoid molecule such as lycopene, time dependent calculations (TD-DFT) were also performed [34]. All calculations were performed at the NEQC Dell and SGI servers located at Universidade Federal de Juiz de Fora.
Results and discussion In this investigation, the autocondensation synthesis of crotonaldehyde [16] has provided a mixture of different conjugated polyenes. The synthetic route was able to provide as the main product a polyunsaturated chain containing eleven double bonds. Such information is confirmed by the characterization through Raman spectroscopy using near-infrared laser excitation (1064 nm) and consequently, for the observation of main vibrational bands at ca. 1509, 1122 and 1010 cm1, assigned to [m1(C@C)], [m2(CAC)] and [m4(CACH3)] (Fig. 2, Table 1). The broad bands observed for these vibrational modes are probably due to the fact that the synthesis provides a mixture of unsaturated compounds. Infrared spectrum of the synthetic product was used to confirm the same bands observed in the Raman spectrum: the C@O (1670 cm1) and CACH3 (1005 cm1) stretching modes (Fig. 2). An interesting fact to be noted is the conjugative effect over the carbonyl group due to the double bonds, which causes the decrease of the absorption frequency if compared to less conjugated carbon chains. In attempt to identified polyenals from the mixture it was performed a detailed analysis of bands by decomposition of the
Fig. 2. Vibrational spectra of the synthetic product: IR-ATR (top) and FT-Raman (bottom).
spectra to improve the understanding of the vibrational characterization of the sample. For major bands ranging from 1490 to 1550 cm1 using laser line at 1064 nm it was estimated up to 9 bands at 1428, 1484, 1502, 1518, 1540, 1556, 1590, 1607 and 1630 cm1 assigned to the m1 mode (Fig. 3, panel A). Analyzing the behavior of synthetic sample excited using the 532 nm line it can be observed both combination and overtone bands. In the deconvolution analysis at 532 nm excitation line it has been observed five distinct band between 1498 and 1550 cm1: 1498, 1512, 1523, 1536 and 1550 cm1 (Fig. 3, panel B). Thus, the overtone band at 3029 cm1 could be attributed to the vibration (m1) at 1512 cm1 (2m1 = 3024 cm1) and the overtone band at 2256 cm1 corresponding to the vibration (m2) at 1130 cm1 (calculated as 2260 cm1). The combination of the fundamental bands (m1 + m2) has been observed at 2640 cm1, being the calculated value of 2650 cm1. Another five bands was observed in the region between 1002 and 1024 cm1: 1002, 1008, 1013, 1020 and 1024 cm1. Combination bands such as m1 + m4, m2 + m4 were not observed in the spectrum. The comparison between the number of observed bands in the spectra obtained at 1064 and 532 nm (Fig. 3, panels A and B) could indicate that the spectrum excited using green light is more selective for the identification of products with higher polyenic chains formed during the reaction. On the other hand, the spectrum obtained at 1064 nm has shown bands from products and contaminants, being more suitable to evaluate the purity of the products. To better understand the composition of reddish pigments we also analyzed the sample in different laser lines to identify different chromophoric groups, as well as different wavenumber values for each one of the vibrational modes, associated to the different conjugated polyenes (Table 1). Fig. 4(A and B) displays the Raman spectra obtained with 532 and 785 nm excitation where it can be clearly seen the enhancement of the band for the [m1(C@C)] chromophore group. Thus, using 532 nm laser line excitation the bands at 1514 [m1(C@C)], 1127 [m2(CAC)] and 1007 cm1 [m(SO2 4 )] are clearly seen; it is evidenced a greater contribution of the chain containing eleven double bonds (tetracosaundecaenal) within the polyene mixture, according to the theoretical studies [26,35]. In this way, under resonance or pre-resonance conditions ðk0 ¼ 532 nmÞ, there is a preferred signal enhancement for the band at 1514 cm1. Another important feature is the observation of several low-intensity peaks above 2000 cm1, which are due to overtones and combination tones of both the strong ACACA and AC@CA fundamental peaks. On the other hand, in the out of resonance condition ðk0 ¼ 785 nmÞ, it is clearly evident the sulfate band at 1007 cm1 besides other bands at 1500 [m1(C@C)] and 1133 [m2(CAC)] cm1, where is evident the contribution of the chain containing twelve double bonds. The resonance effect in Raman spectroscopy arises from the wavelength used to obtain the Raman spectrum and its relation within the electronic absorption band, causing the vibrational modes involved in the electronic transition to be selectively enhanced [23,36,37], so resonance Raman spectroscopy allows the site-specific investigation of chromophores within the molecule, pivotal to the function of the respective molecules [24]. In this sense, different chromophoric groups will be enhanced in intensity by using different laser lines to excite the Raman spectra. The characterization of conjugated polyenes such as polyenals and carotenoids must involve analysis of vibrational bands
437
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441 Table 1 Wavenumber positions of m1, m2, m3 and m4 modes (cm1) of the corals samples and the synthetic products by Raman spectroscopy. Samples
Raman wavenumbers (in cm1) and laser lines (in nm)
m1(C@C)
m2(CAC)
m4(CACH3)
m(CO2 3 )
532
785
1064
532
785
1064
532
785
1064
532
785
1064
Synthetic polyenals
1521
1496
1501
1130
1130
1120
–
–
–
1010
1009
1012
C. braziliensis yellow polyps
1524
1537 1523
1527
1134
1134 1134
1133 1133
1089
1089 1089
1089 1089
1020
1020 1020
1017 1017
C. braziliensis pink stalk
1506
1501
1539 1501
1121
1119 1119
1118 1118
–
1090 1090
1089 1089
1018
1021 1021
1017 1017
L. setacea
1516
1538 1520
1539 1519
1128
1135 1135
1136 1136
–
1089
1092
1018
1020 1020
1020 1020
L. punicea M. flamma R. muelleri P. dilatata purple halo
1514 1514 1506 1511
1507 1501 1499 1506
1507 1505 1500 1502
1125 1117 1120 1127
1122 1125 1115 1121
1123 1121 1116 1118
– 1086 – –
1089 1089 1089 1088
1090 1089 1089 1089
1017 1014 1017 1017
1020 1015 1020 1017
1020 1020 1019 1014
Fig. 3. Raman spectra of the synthetic product deconvolution (m1 mode) using two different excitation wavelengths: 1064 nm (panel A) and 532 nm (panel B).
attributed to (m1) C@C, (m2) CAC and (m4) CACH3 in polyenes or polyenals and (q3) CACH3 in carotenoids. Spectra recorded using 532 and 785 nm laser lines (Fig. 4, panels C and D) did not show a significant enhancement of the intensity of the band at 1010 cm1 corresponding to methyl groups. Theoretical calculations from conjugated polyenal with 11 double bonds (from now on called polyenal-11) showed that the band at 1106 cm1 (theoretical calculation unscaled) may be assigned to CACH3 mode in the end of carbon chain. Differing from carotenoids [38], we neither observed combination bands involving [m1(C@C) + m4(CACH3 in polyenals)] nor an intensification of the band at 1010 cm1, which may suggest that the mixture of obtained products does
not contain a methyl group in the middle of the conjugated carbon chain. Several other works of resonance Raman excitation of carotenoids have also shown that the CACH3 mode belongs to the electronic delocalization system over the carbon chain, which also supports the idea depicted in this investigation [3,8,38,39]. Previous studies on the octocoral Corallium rubrum have presented a different understanding on the polyenic based pigment composition. The divergence is mainly based on the fact that the found pigment is a carotenoid or a conjugated polyene completely demethylated [40,41]. In this sense, Brambilla et al. [22], based on the calculated wavenumber shifts caused by the conjugation, have reported that m4 mode is due to the in plane rocking of the ACH3 groups; as the number of methyl groups linked to the conjugated carbon chain decreases, the wavenumber of the vibrational mode increases and the intensity decreases. This fact could be justified by the decrease in the resonance effect on the polyunsaturated carbon chain containing the methyl groups. According to their data the main component of the pigment from C. rubrum is composed of polyene chain partially methylated. A good approach for this idea can be also found in a classical work by de Oliveira et al. [38], where authors have studied the resonance Raman effect on bixin, a C9 carotenoid obtained from Bixa orellana; in such system, authors have pointed out the massive conjugation mainly on the central 5 carbon atoms in the conjugated chain, despite the several double bond conjugations present in the structure. It is important to notice that even today there is a great interest in the use of resonance Raman conditions to analyze the spectroscopic properties from different polyenes; in a very recent paper, Tommasini and coworkers [42] present an review about the theory of overtones and combination bands in resonance Raman spectroscopy, using b-carotene as the model system; however, despite the physical approach authors have used in this paper, they do not address the chemical nature of the polyene system, or discuss the chemical importance of the ACH3 group in the structure of the carotene, since this group appears as one the chromophoric moieties in the resonance excitation profile; this is very evident in the bixin work, and of course the importance that the central part of the polyenic chain which is responsible for the electronic properties of the system [38]. In this sense, some previous works of our group have demonstrated that the spectroscopic difference between carotenes (or carotenoids) and other conjugated polyenes (such as parrodienes) is the presence of the band at ca. 1010 cm1, due to the ACH3 group, which is present in the resonance spectra of carotenes, but is not present in the resonance spectra of parrodienes [5]. In other words, the band at ca. 1010 cm1 will only be resonance enhanced if the ACH3 groups participate over the delocalized electronic cloud, such as in the case of b-carotene and other
438
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 4. Raman spectra of the synthetic product using two different excitation wavelengths: 532 nm (panel A – with Na2SO4 as a standard; panel C – without Na2SO4 as a standard) and 785 nm (panel B – without Na2SO4 as a standard; panel D – without Na2SO4 as a standard).
carotenes and carotenoids, such as bixin; however, if the conjugated polyene does not present the same ACH3 groups as carotenes, but has the same groups in the chemical structure, such as parrodienes, such band appears at the Raman spectrum in ordinary conditions, but cannot be seen in resonance conditions, as we are showing in our investigation. In order to evaluate the differences and similarities between synthetic, natural polyenals and carotenoids, in this study we have studied the octocoral Leptogorgia punicea. Subtle differences were observed from the decomposition of the in situ spectra using 532 laser line. The octocoral has shown 3 different bands for the m4 mode: 1003, 1013 and 1021 cm1, whereas 3 other bands have been also calculated for m1 mode between 1490 and 1536 cm1: 1490, 1515 and 1536 cm1. The overtone band at 3024 cm1 could be attributed to the overtone vibration (m1) in 1515 cm1 (2m1 = 3030 cm1) and the overtone band at 2245 cm1 corresponding to the vibration (m2) at 1128 cm1 (calculated for 2256 cm1). The combination of the fundamental bands (m1 + m2) was observed at 2628 cm1 (calculated for 2643 cm1) as well as the combination band 1013 cm1 (m4) + 1515 cm1 (m1) was observed as a broad feature at 2528 cm1. Differing from the synthetic polyenal the presence of the combination band m1 + m4 from L. punicea may suggest the presence of an additional methyl group, as proposed for C. rubrum [22]. Similar approach is shown in Table 2 that describes main vibrational modes observed by the deconvolution analysis using FT-Raman spectrum of four corals species. The main information obtained from these marine species is that they do not produce a unique pigment, and through the deconvolution m4 mode of corals we have found that some corals showed resemblance with methylated pigments, for example, L. punicea showed the m4 mode at 1006 and 1021 cm1 as well as demethylated pigments. Deconvolution analysis of m1 mode of corals showed that the species of L. punicea, L. setacea and Chromonephthea braziliensis (yellow polyps and pink stalk) have more than one band, confirming the hypothesis that it is the mixture of polyenes and not a single substance responsible for the coloration of corals and other organisms such as mollusk shells and feathers of macaws.
Table 2 Deconvolution bands of m1 and m4 (cm1) of the corals samples from the FT-Raman spectra. Samples
m1 (cm1)
m4 (cm1)
L. punicea
1509 1528
1006 1021
L. setacea
1517 1539
1004 1020
R. muelleri
1500 1514
1005 1020
C. braziliensis yellow polyps
1523 1540
1015
C. braziliensis pink stalk
1502 1538
1015
It is important to note that the theoretical calculated spectrum for the polyenal-11 (tetracosaundecaenal) are in good agreement with that obtained experimentally with respect to the wavenumbers of the main bands. As can be seen in Fig. 5, in which the DFT calculated spectrum is not scaled, the general profile of the bands is similar to experimental one, even without any adjustment to the Raman intensities. Fig. 5 also shows the bands in the region of 1400–1900 cm1 enhanced for better visualization. Assuming that would be interesting to have a scaling factor the DFT gas phase vibrational frequencies would require the adjustment of only 9% when compared to experimental data, which is a good fit thinking on other comparisons displayed in the literature [43]. The resulting theoretical spectrum assignments also show bands in 1422 and 1427 cm1 related to an umbrella motion-like of methyl group, d CH3; in 1489 cm1 (rocking) and 1502 cm1 (stretching and angular deformation of the conjugated methyl mode). Such information are important in showing that in fact, only one methyl must be present in the synthesized compound (terminal), since these bands are also present in the experimental spectrum. In general, the calculated vibrational frequencies are in good accordance with experimental values reported here.
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 5. Comparison between theoretical (bottom) and experimental (top) Raman spectra. The detached region (1400–1900 cm1) has been enhanced for a better visualization.
Fig. 6. Theoretical UV–Vis profile comparison between lycopene and polyenal-11.
Another, but no less important aspect is the analysis of the frontier molecular orbitals (MOs) obtained from TD-DFT calculations. In this analysis we performed a comparison of polyenal-11 with lycopene, a carotene with also 11 doubles bonds. It can be perceived by the overlaying in the theoretical UV–Vis spectra (Fig. 6) that despite the difference in displacement between the bands themselves (621 nm for lycopene and 594 nm for polyenal-11), the main contributions to the HOMO and LUMO of lycopene are exactly concentrated in the region where the molecule is straight, showing that this is an indication that the size of the box (thinking of the particle in the box model) may not be as important as suggested before [44]. Fig. 7 shows the MOs of these two compounds and allow us to confirm the p-electron delocalization, showing HOMOs concentrated in C@C doubles bonds and LUMOs concentrated in CAC single bonds, as expected. It is important to say that Fig. 7 just represents an illustration, a static representation, since in an ab initio calculation since it is know that there is no differentiation between those representations from the quantum point of view. In addition to synthesize the polyenal we are also presenting the occurrence of pigments related to conjugated polyenals in
439
different species of marine corals. The particular Raman shifts of the pigments are summarized for some representative cases in Table 1. The literature already describes the effect of chain length and the role of aldehydic function on the antiproliferative and antioxidant effects [45]. The spectrum obtained for the synthetic sample resembles the one observed for in situ sample and demineralized sclerite of L. punicea; since this kind of compounds is inserted into sclerites composed of calcium carbonate crystallized as calcite present in the coenenchymes [4,6,12,25] some differences are observed in the Raman shift values. The in situ sample shows bands at 1507 [m1(C@C)], 1123 [m2(CAC)] and 1090 cm1 [m(CO2 3 )] and demineralized shows bands at 1500 [m1(C@C)], 1116 [m2(CAC)] and a mode in 1011 cm1 [m4(CACH3)], which has been assigned to the rocking motion of ACH3 backbone in carotenes [39] or to a CH@CH wagging mode in polyene molecules [46]. Recent studies related the inclusion processes of carotenes with bcyclodextrins and the analysis by Raman spectroscopy, demonstrating the matrix effect on the band position as a consequence of the inclusion complex formation [47]; in the present case this effect can be attributed to the chemical interactions existing when calcium carbonate species are interacting with the pigment present in natural samples [48]. The mixture present in synthetic polyenals showed a predominant reddish-brown coloration, featuring octocorals as C. braziliensis we observed distinct coloration from different parts of the colony. According to the literature [6], the in situ Raman spectroscopic analysis characterization of this kind of coral using 632.8 and 1064 nm laser lines, in two different parts of the coral (the pink stalk and the yellow polyps) it can be seen a single band in the 1502 cm1 region. Therefore, this result proposed that only one type of pigment was present in the pink stalk, however recent analysis of the spectra made with the 1064 nm laser excitation, indicated that is probable is a mixture of pigments. This statement is grounded by the fact that two bands are present in the region of [m1(C@C)], at 1502 and 1540 cm1 for pink stalk which is not noticed in the yellow polyps. It has just a band in 1526 cm1. One very specific characteristic of isolated polyenals from corals is the insolubility in almost all the tested solvents; in other words, despite the conventional extractions method using different solvent systems is well described for carotenoids [49,50], such methods are not applied for polyenals from corals. Because of this feature, the pigment analysis using HPLC or other approaches to separate the different components of the mixture from the synthesis has become a limiting task. Finally, it can be emphasized that this study has provided spectroscopic data about the polyenal synthesis mainly containing 11 double bonds. The Raman spectroscopy technique using lasers of different energies can result in drastic changes, mainly based on spectral intensities and Raman shifts; different chromophoric groups and different wavenumber values for each one of the vibrational modes can be associated to polyenes containing different numbers of double bonds. Based on the behavior of the synthetic sample excited using the 532 nm line it can also be observed both combination and overtone bands, but differing from carotenoids, neither combination bands involving [m1(C@C) + m4(CACH3)] nor an intensification of the band at 1010 cm1. These results may suggest that the obtained product do not contain a methyl group in the central region of conjugated carbon chain, as carotenoids do present. About the analyzed corals, the main conclusion obtained from the Raman spectra analysis is the fact that these marine species do not produce a single pigment, and through the deconvolution of the m4 mode it can be found that some corals show resemblance with methylated pigments. The occurrence of more than one pigment in the tissues of corals [51] and phytoplanktonic organisms [52] may be related to the capture of sunlight through the water mass. Variations in light intensity that reaches the marine
440
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 7. HOMO ? LUMO gap for (a) polyenal-11 and (b) lycopene.
organisms may vary with turbidity, tide and season [53]. The production of pigments with different sizes of polyenic chains may provide adaptability of capturing light of different wavelengths. Conclusions In this work we have used Raman spectroscopy in the characterization of synthetic polyenals as well polyenals from natural sources such as species of octocorals. It can be observed some important differences between these compounds mainly analyzing the intensity and position of the m4 mode whereas the decomposition of the spectral analysis of this band can indicate that some corals do not produce a unique type of pigments: it may have a mixture involving methylated and demethylated polyenes present in the tissues. TD-DFT calculations of the frontier molecular orbitals (MOs) together with theoretical calculated Raman spectra of synthetic polyenal, natural polyenals and lycopene have provided clues to demonstrate the power of Raman spectroscopy in distinguishing pigments composed of similar chromophores such as carotenoid and demethylated polyene. At last, the band at ca. 1010 cm1 will only be resonance enhanced if the ACH3 groups participate over the delocalized electronic system, such as in the case of b-carotene and other carotenes and carotenoids, as bixin; however, if the conjugated polyene does not present the same ACH3 groups as carotenes, but has the same groups in the chemical structure, such as parrodienes, such band appears at the Raman spectrum in ordinary conditions, but cannot be seen in resonance conditions, as it can be seen in this investigation. Acknowledgements Authors are indebted to FAPEMIG, CNPq and CAPES/Ciências do Mar 1137/2010 (Brazilian agencies), PRONEX for financial support and the Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis (IBAMA) for collecting license (20649-1). They also thank Hiran Carvalho (UERJ), Antônio S.F. Couto (diving operator, Deep Trip, Arraial do Cabo city, RJ, Brazil) for collecting the coral samples and Clovis B. Castro for donating octocorals (UFRJ-MNRJ). References [1] S. Karampelas, E. Fritsch, J.Y. Mevellec, S. Sklavounos, T. Soldatos, Eur. J. Mineral. 21 (2009) 85–97. [2] J. Urmos, S.K. Sharma, F.T. Mackenzie, Am. Mineral. 76 (1991) 641–646.
[3] J.C. Merlin, M.L. Delé-Dubois, Comp. Biochem. Physiol. Part B: Comp. Biochem. 84 (1986) 97–103. [4] L.F. Maia, V.E. de Oliveira, M.E.R. de Oliveira, B.G. Fleury, L.F.C. de Oliveira, J. Raman Spectrosc. 43 (2012) 161–164. [5] L.F. Maia, V.E. de Oliveira, M.E.R. Oliveira, F.D. Reis, B.G. Fleury, H.G.M. Edwards, L.F.C. de Oliveira, J. Raman Spectrosc. 44 (2013) 560–566. [6] L.F. Maia, B.G. Fleury, B.G. Lages, J.P. Barbosa, A.C. Pinto, H.V. Castro, V.E. de Oliveira, H.G.M. Edwards, L.F.C. de Oliveira, J. Raman Spectrosc. 42 (2011) 653– 658. [7] K.J. McGraw, M.C. Nogare, Biol. Lett. 1 (2005) 38–43. [8] M. Veronelli, G. Zerbi, R. Stradi, J. Raman Spectrosc. 26 (1995) 683–692. [9] R. Stradi, E. Pini, G. Celentano, Comp. Biochem. Physiol. Part B: Biochem. Mol. Biol. 130 (2001) 57–63. [10] W. Barnard, D. de Waal, J. Raman Spectrosc. 37 (2006) 342–352. [11] C. Hedegaard, J.-F. Bardeau, D. Chateigner, J. Mollus. Stud. 72 (2006) 157–162. [12] F. Luwei, Y. Mingxing, J. China Univ. Geosci. 19 (2008) 146–151. [13] L. Bergamonti, D. Bersani, D. Csermely, P.P. Lottici, Spectrosc. Lett. 44 (2011) 453–458. [14] J.F. Masello, M.L. Pagnossin, T. Lubjuhn, P. Quillfeldt, Ecol. Res. 19 (2004) 421– 432. [15] E.H. Burtt, M.R. Schroeder, L.A. Smith, J.E. Sroka, K.J. McGraw, Biol. Lett. 7 (2011) 214–216. [16] R. Stradi, A. Bertelli, E. Pini, Polyunsaturated linear aldehyde and their derivatives with anti-radical and anti-tumoral activity, in: U.S Patent 2005/ 0175555 A1, 2005. [17] S. Mukherjee, D. Ray, M. Falchi, A. Bertelli, E. Pini, D.K. Das, J. Nat. Prod. 72 (2009) 871–875. [18] A. Alker, K. Kim, D. Dube, C.D. Harvell, Coral Reefs 23 (2004) 397–405. [19] S.P. Ellner, L.E. Jones, L.D. Mydlarz, C.D. Harvell, Am. Nat. 170 (2007) E143– E161. [20] C.V. Palmer, L.D. Mydlarz, B.L. Willis, Proc. Roy. Soc. B: Biol. Sci. 275 (2008) 2687–2693. [21] K. Kiho, A. Alker, K. Shuster, C. Quirolo, C.D. Harvell, Dis. Aquat. Organ. 69 (2006) 95–99. [22] L. Brambilla, M. Tommasini, G. Zerbi, R. Stradi, J. Raman Spectrosc. 43 (2012) 1449–1458. [23] M. Schmitt, J. Popp, J. Raman Spectrosc. 37 (2006) 20–28. [24] R.S. Stenner, J. Raman Spectrosc. 36 (2005) 276–278. [25] E. Fritschi, B. Rondeau, T. Hainschwang, S. Karampelas, EMU Notes Mineral. 12 (2012) 453–488. [26] T. Kupka, H.M. Lin, L. Stobin´ski, C.H. Chen, W.-J. Liou, R. Wrzalik, Z. Flisak, J. Raman Spectrosc. 41 (2010) 651–658. [27] H.E. Schaffer, R.R. Chance, R.J. Silbey, K. Knoll, R.R. Schrock, J. Chem. Phys. 94 (1991) 4161–4170. [28] H. Kuzmany, Pure Appl. Chem. 57 (1985) 235–246. [29] C. Heller, G. Leising, C. Godon, S. Lefrant, W. Fischer, F. Stelzer, Phys. Rev. B 51 (1995) 8107–8114. [30] A. Andreoni, C.A. Sacchi, O. Svelto, in: C.B. Moore (Ed.), Chemical and Biochemical Applications of Lasers, Academic Press, 1979, pp. 1–30. [31] A.D. Becke, J. Chem. Phys. 98 (1993) 5648–5652. [32] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785–789. [33] J.A. Pople, D.P. Santry, G.A. Segal, J. Chem. Phys. 43 (1965) S129–S135. [34] R.E. Stratmann, G.E. Scuseria, M.J. Frisch, J. Chem. Phys. 109 (1998) 8218–8224. [35] S. Karampelas, E. Fritsch, J.Y. Mevellec, J.P. Gauthier, S. Sklavounos, T. Soldatos, J. Raman Spectrosc. 38 (2007) 217–230. [36] R.J.H. Clark, T.J. Dines, Angew. Chem. Int. Edit. English 25 (1986) 131–158. [37] D.A. Long, in: The Raman Effect, John Wiley & Sons, Ltd., 2002, pp. 221–270. [38] L.F.C. de Oliveira, S.O. Dantas, E.S. Velozo, P.S. Santos, M.C.C. Ribeiro, J. Mol. Struct. 435 (1997) 101–107.
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441 [39] H. Okamoto, H.-O. Hamaguchi, M. Tasumi, J. Raman Spectrosc. 20 (1989) 751– 756. [40] E. Fritsch, S. Karampelas, Mar. Biol. 154 (2008) 929–930. [41] J. Cvejic, S. Tambutté, S. Lotto, M. Mikov, I. Slacanin, D. Allemand, Mar. Biol. 152 (2007) 855–862. [42] M. Tommasini, G. Longhi, S. Abbate, G. Zerbi, J. Raman Spectrosc. 45 (2014) 89– 96. [43] L.V. de Freitas, C.C.P. da Silva, J. Ellena, L.A.S. Costa, N.A. Rey, Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 116 (2013) 41–48. [44] J. Autschbach, J. Chem. Educ. 84 (2007) 1840. [45] D. Tagliazucchi, E. Verzelloni, E. Pini, G. Ronca, A. Conte, Pharmacologyonline 3 (2006) 765–771. [46] Y. Koyama, I. Takatsuka, M. Nakata, M. Tasumi, J. Raman Spectrosc. 19 (1988) 37–49.
441
[47] V.E. de Oliveira, E.W.C. Almeida, H.V. Castro, H.G.M. Edwards, H.l.F. Dos Santos, L.F.C. de Oliveira, J. Phys. Chem. A 115 (2011) 8511–8519. [48] V.E. de Oliveira, H.V. Castro, H.G.M. Edwards, L.F.C. de Oliveira, J. Raman Spectrosc. 41 (2010) 642–650. [49] T. Lacker, S. Strohschein, K. Albert, J. Chromatogr. A 854 (1999) 37–44. [50] D.B.R. Amaya, A guide to carotenoid analysis in foods. ILSI Press, Washington, DC, 2001, pp. 1–64. [51] D.F. Gleason, Ecosystems, evolution and ultraviolet radiation, statesboro, GA, 2001. [52] M. Stomp, J. Huisman, F. De Jongh, A.J. Veraart, D. Gerla, M. Rijkeboer, B.W. Ibelings, U.I.A. Wollenzien, L.J. Stal, Nature 432 (2004) 104–107. [53] M.K. Wicksten, Bull. Mar. Sci. 45 (1989) 519–530.
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Raman spectroscopy as a tool in differentiating conjugated polyenes from synthetic and natural sources Rafaella F. Fernandes a, Lenize F. Maia a, Mara R.C. Couri a, Luiz Antonio S. Costa b, Luiz Fernando C. de Oliveira a,⇑ a b
NEEM – Núcleo de Espectroscopia e Estrutura Molecular, Departamento de Química, Universidade Federal de Juiz de Fora, 36036-330 Juiz de Fora, MG, Brazil NEQC – Núcleo de Estudos em Química Computacional, Departamento de Química, Universidade Federal de Juiz de Fora, 36036-330 Juiz de Fora, MG, Brazil
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Pigments derived from polyenals
were unequivocally identified by Raman spectroscopy. Evaluation of differences and similarities between synthetic and natural polyenals. Use of the in plane rocking methyl group modes as fingerprint Raman bands in polyenals. Lasers of different energies characterizing polyenals in mixtures. Tissues of corals are pigmented by more than one conjugated polyenal.
a r t i c l e
i n f o
Article history: Received 11 March 2014 Received in revised form 19 May 2014 Accepted 3 June 2014 Available online 14 June 2014 Keywords: Carotenoids Lycopene Polyenal Raman effect Corals DFT calculations
a b s t r a c t This work presents the Raman spectroscopic characterization of synthetic analogs of natural conjugated polyenals found in octocorals, focusing the unequivocal identification of the chemical species present in these systems. The synthetic material was produced by the autocondensation reaction of crotonaldehyde, generating a demethylated conjugated polyene containing 11 carbon–carbon double bonds, with just a methyl group on the end of the carbon chain. The resonance Raman spectra of such pigment has shown the existence of enhanced modes assigned to m1(C@C) and m2(CAC) modes of the main chain. For the resonance Raman spectra of natural pigments from octocorals collected in the Brazilian coast, besides the previously cited bands, it could be also observed the presence of the m4(CACH3), related to the vibrational mode who describes the vibration of the methyl group of the central carbon chain of carotenoids. Other interesting point is the observation of overtones and combination bands, which for carotenoids involves the presence of the m4 mode, whereas for the synthetic polyene this band, besides be seen at a slightly different wavenumber position, does not appear as an enhanced mode and also as a combination, such as for the natural carotenoids. Theoretical molecular orbital analysis of polyenal-11 and lycopene has shown the structural differences which are also responsible for the resonance Raman data, based on the appearance of the (ACH3) vibrational mode in the resonant transition only for lycopene. At last, the Raman band at ca. 1010 cm1, assigned to the (ACH3) vibrational mode, can be used for attributing the presence of each one of the conjugated polyenes: the resonance Raman spectrum containing the band at ca. 1010 cm1 refers to the carotenoid (in this case lycopene), and the absence of such band in resonance conditions refers to the polyenal (in this case the polyenal-11). Ó 2014 Elsevier B.V. All rights reserved.
⇑ Corresponding author. Tel./fax: +55 32 21023310. E-mail address: [email protected] (L.F.C. de Oliveira). http://dx.doi.org/10.1016/j.saa.2014.06.022 1386-1425/Ó 2014 Elsevier B.V. All rights reserved.
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Introduction Natural pigments belonging to the class of unsubstituted linear polyacetylenes have been mainly identified by Raman spectroscopy as animal biochromes from parrot’s feathers, octocorals, pearls and shells of mollusks [1–6]. In many birds, red, orange and yellow feathers are colored by carotenoid pigments, but parrots are an exception [4–6]; these organisms have used other conjugated polyenes to color their plumage in red, indicating an alternative and unique system of pigmentation which seems to be conserved evolutionarily in parrots [7]. Red feathers from the parrot Ara macao are composed of a mixture of polyunsaturated linear aldehydes (polyenals) containing 6–9 conjugated double bonds named as psittacofulvins [8,9]. Brown, purple, red, orange and yellow pigments from shells of several species of mollusks were identified as unsubstituted polyacetylenes molecules ranging from 9 to 12 conjugated double bonds [10,11]. Similar coloration pattern observed in colored cultured freshwater pearls was also identified as linear polyacetylenes ranging from 6 to 14 double bonds [1]. So far, purple, red, orange and yellow coenenchymes and sclerites investigated from octocorals have also been attributed to such polyacetylenes ranging from 8 to 12 C@C unsaturations [1,4–6,12,13]. The bioactivity of red pigments in parrots feathers have been reported in ecological assays as a signal of individual quality and parental investment [14]. Red psittacofulvins reduced microbial damage to parrot feathers [15] and synthetic analogs named as parrodienes presented antioxidant, anti-tumoral and anti-inflammatory activities [4,12,16]. They are also useful in preventing the diseases caused by free radicals and on treatment of cardiovascular and inflammatory damages [16,17]. The biological activity in octocorals is unknown, however purple pigments are produced locally as purple halo in damaged tissues [4,18] as a consequence of an inflammatory immune response [19,20]. Tissues are composed of sclerites which are calcium carbonate structures secreted by scleroblasts and the increasing of purple pigment has been suggested to play a protective function in corals [21] and because of that, some studies related that the presence of polyenes in corals may be related with specific infectious process [4]. An example of this phenomenon was stated for the Gorgonia spp., in which a purple halo resulting from an increase of purple sclerites surrounding the necrotic tissues [18]. Another case is the appearance of an unusual violet pigmentation surrounding necrotic tissue of Pyllogorgia dilatata, a Brazilian endemic gorgonian. Maia et al. [4] have reported that the purpling in a Brazilian octocoral is attributed to psittacofulvin derivatives, which until then have been found as exclusive in parrots. Determination of the chemical structure of linear polyacetylenes from marine origin is still under investigation [22]. Experimental data obtained from Raman spectroscopy and theoretical calculations have suggested structures similar to polyenals from parrots, which is the only source from terrestrial environment. As mentioned above Raman spectroscopy has been pointed to be one of most suitable tools to investigate conjugated systems, being extensively used in the characterization of carotenoids and polyenals or unsubstituted polyacetylenes [8,23–25]. Polyenic molecules are strong Raman scatters due to the strong electron/ vibration coupling, and their vibrational modes can be selectively enhanced by resonance excitation nearby or close to the p–p* electronic transition [13]. The two bands of highest intensity observed in these organic pigments are the stretching modes of the C@C double bond (m1) and the CAC single bond (m2) of the main polyenic chain. The first one occurs between 1450 and 1680 cm1 and the latter between
435
1070 and 1210 cm1 [10,26]. The number of conjugated double bonds in the main chain or other molecules like as CaCO3 bound to the polyene affects the Raman shifts. Another aspect to be analyzed is that an excitation with lasers of different energies can result in drastic Raman spectrum changes, mainly based on spectral intensity, when a single component of a polyene mixture can have their Raman bands enhanced or suppressed to resolve the mixture [27–29]. If suitable exciting is chosen to be in resonance with an electronic transition of the biochrome it follows that is observed an intensification of the band for the specific chromophoric group (m1). This observation is due to the resonance Raman scattering [3,8,11,28] and has been demonstrated to be an useful tool for the study of biologically relevant molecules in their natural state [30]. Therefore, observing the relevance of polyconjugated molecules in biochemistry, both in terms of coloration and preventing diseases, as well as its significance in the marine environment, in this work we make use of Raman technique to characterize synthetic analogs of the polyenes found in octocorals. In order to evaluate the biological activity of linear polyacetylenes identified from corals, we are describing a synthesis of a carbon chain with a number of double bonds between ten and twelve and the comparative analysis with natural polyenes found in coral species, in order to provide a better knowledge on the subject. Theoretical calculations also have been performed for the same compounds in order to reinforce the basis of the vibrational analysis, as well as to show the electronic properties for the conjugated polyenes. Methodology Experimental section The synthetic route to obtain a polyene with a number of double bonds between ten and twelve is an adaptation of the methodology used by Stradi et al. [16], based on the autocondensation of crotonaldehyde in acidic medium and using pyrrolidine as catalyst. Crotonaldehyde, acetic acid and pyrrolidine were supplied by Aldrich. Under reflux system, 20 mL of crotonaldehyde were added to a flask and kept under magnetic stirring for 15 min under nitrogen flow. After this time, 0.20 mL of pyrrolidine and 0.20 mL of acetic acid was slowly added; the mixture was left under magnetic stirring in nitrogen atmosphere at 50 °C for 30 min. Since the reaction was quite exothermic, the heating was done in glycerine bath. The reaction was cooled in an ice bath and 200 mL of ethyl ether were added under magnetic stirring. The brown product obtained (yield of 85%) was composed by a mixture of polyenals ranging from 7 to 13 conjugated double bonds with predominantly the polyenal containing conjugated 11 double bonds known as tetracosaundecaenal (Fig. 1). Fourier transform Raman spectra were carried out using a Bruker RFS 100 instrument and a Nd:YAG laser operating at 1064 nm, equipped with a Ge detector cooled with liquid nitrogen. It was used 4 cm1 of spectral resolution, and good signal-to-noise ratios were obtained with 516 scans, using a range of laser powers at the sample between 25 and 75 mW. SENTERRA dispersive Raman microscope instrument operating at 532 nm and 785 nm was also used equipped with cooled charge coupled detector with the incident laser beam focused on the sample with a 50 objective. Good spectra were obtained using a range of laser powers between 0.2 and 2.0 mW and 3 or 5 accumulations for 3 s, with a spectral resolution of 3 cm1. Infrared spectral data in mode of attenuated total reflection (ATR) were obtained with a Bruker Alpha FT-IR spectrometer and spectral resolution of 4 cm1 and good signal-to-noise ratios were obtained with 128 scans.
436
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 1. Scheme of the polyenal synthesis and a representative structure of the tetracosaundecaenal.
Calculations Full unconstrained geometry optimization and frequency calculations for all distinct polyenals have been carried out at the Hartree–Fock (HF) and Density Functional Theory (DFT) using the three-parameter fit of the Becke´s exchange correlation potential (B3LYP) [31,32] as implemented in the Gaussian 09 program suite [Gaussian]. The Pople’s triple-f-quality basis sets 6-311+G(d, p) [33] have been applied for all atoms in polyenes of 10, 11 and 12 C@C double bonds. Vibrational Raman calculated frequencies were scaled by a factor of 0.97 when compared to experimental data. In order to evaluate the polyenes frontier molecular orbitals to establish a comparison with a carotenoid molecule such as lycopene, time dependent calculations (TD-DFT) were also performed [34]. All calculations were performed at the NEQC Dell and SGI servers located at Universidade Federal de Juiz de Fora.
Results and discussion In this investigation, the autocondensation synthesis of crotonaldehyde [16] has provided a mixture of different conjugated polyenes. The synthetic route was able to provide as the main product a polyunsaturated chain containing eleven double bonds. Such information is confirmed by the characterization through Raman spectroscopy using near-infrared laser excitation (1064 nm) and consequently, for the observation of main vibrational bands at ca. 1509, 1122 and 1010 cm1, assigned to [m1(C@C)], [m2(CAC)] and [m4(CACH3)] (Fig. 2, Table 1). The broad bands observed for these vibrational modes are probably due to the fact that the synthesis provides a mixture of unsaturated compounds. Infrared spectrum of the synthetic product was used to confirm the same bands observed in the Raman spectrum: the C@O (1670 cm1) and CACH3 (1005 cm1) stretching modes (Fig. 2). An interesting fact to be noted is the conjugative effect over the carbonyl group due to the double bonds, which causes the decrease of the absorption frequency if compared to less conjugated carbon chains. In attempt to identified polyenals from the mixture it was performed a detailed analysis of bands by decomposition of the
Fig. 2. Vibrational spectra of the synthetic product: IR-ATR (top) and FT-Raman (bottom).
spectra to improve the understanding of the vibrational characterization of the sample. For major bands ranging from 1490 to 1550 cm1 using laser line at 1064 nm it was estimated up to 9 bands at 1428, 1484, 1502, 1518, 1540, 1556, 1590, 1607 and 1630 cm1 assigned to the m1 mode (Fig. 3, panel A). Analyzing the behavior of synthetic sample excited using the 532 nm line it can be observed both combination and overtone bands. In the deconvolution analysis at 532 nm excitation line it has been observed five distinct band between 1498 and 1550 cm1: 1498, 1512, 1523, 1536 and 1550 cm1 (Fig. 3, panel B). Thus, the overtone band at 3029 cm1 could be attributed to the vibration (m1) at 1512 cm1 (2m1 = 3024 cm1) and the overtone band at 2256 cm1 corresponding to the vibration (m2) at 1130 cm1 (calculated as 2260 cm1). The combination of the fundamental bands (m1 + m2) has been observed at 2640 cm1, being the calculated value of 2650 cm1. Another five bands was observed in the region between 1002 and 1024 cm1: 1002, 1008, 1013, 1020 and 1024 cm1. Combination bands such as m1 + m4, m2 + m4 were not observed in the spectrum. The comparison between the number of observed bands in the spectra obtained at 1064 and 532 nm (Fig. 3, panels A and B) could indicate that the spectrum excited using green light is more selective for the identification of products with higher polyenic chains formed during the reaction. On the other hand, the spectrum obtained at 1064 nm has shown bands from products and contaminants, being more suitable to evaluate the purity of the products. To better understand the composition of reddish pigments we also analyzed the sample in different laser lines to identify different chromophoric groups, as well as different wavenumber values for each one of the vibrational modes, associated to the different conjugated polyenes (Table 1). Fig. 4(A and B) displays the Raman spectra obtained with 532 and 785 nm excitation where it can be clearly seen the enhancement of the band for the [m1(C@C)] chromophore group. Thus, using 532 nm laser line excitation the bands at 1514 [m1(C@C)], 1127 [m2(CAC)] and 1007 cm1 [m(SO2 4 )] are clearly seen; it is evidenced a greater contribution of the chain containing eleven double bonds (tetracosaundecaenal) within the polyene mixture, according to the theoretical studies [26,35]. In this way, under resonance or pre-resonance conditions ðk0 ¼ 532 nmÞ, there is a preferred signal enhancement for the band at 1514 cm1. Another important feature is the observation of several low-intensity peaks above 2000 cm1, which are due to overtones and combination tones of both the strong ACACA and AC@CA fundamental peaks. On the other hand, in the out of resonance condition ðk0 ¼ 785 nmÞ, it is clearly evident the sulfate band at 1007 cm1 besides other bands at 1500 [m1(C@C)] and 1133 [m2(CAC)] cm1, where is evident the contribution of the chain containing twelve double bonds. The resonance effect in Raman spectroscopy arises from the wavelength used to obtain the Raman spectrum and its relation within the electronic absorption band, causing the vibrational modes involved in the electronic transition to be selectively enhanced [23,36,37], so resonance Raman spectroscopy allows the site-specific investigation of chromophores within the molecule, pivotal to the function of the respective molecules [24]. In this sense, different chromophoric groups will be enhanced in intensity by using different laser lines to excite the Raman spectra. The characterization of conjugated polyenes such as polyenals and carotenoids must involve analysis of vibrational bands
437
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441 Table 1 Wavenumber positions of m1, m2, m3 and m4 modes (cm1) of the corals samples and the synthetic products by Raman spectroscopy. Samples
Raman wavenumbers (in cm1) and laser lines (in nm)
m1(C@C)
m2(CAC)
m4(CACH3)
m(CO2 3 )
532
785
1064
532
785
1064
532
785
1064
532
785
1064
Synthetic polyenals
1521
1496
1501
1130
1130
1120
–
–
–
1010
1009
1012
C. braziliensis yellow polyps
1524
1537 1523
1527
1134
1134 1134
1133 1133
1089
1089 1089
1089 1089
1020
1020 1020
1017 1017
C. braziliensis pink stalk
1506
1501
1539 1501
1121
1119 1119
1118 1118
–
1090 1090
1089 1089
1018
1021 1021
1017 1017
L. setacea
1516
1538 1520
1539 1519
1128
1135 1135
1136 1136
–
1089
1092
1018
1020 1020
1020 1020
L. punicea M. flamma R. muelleri P. dilatata purple halo
1514 1514 1506 1511
1507 1501 1499 1506
1507 1505 1500 1502
1125 1117 1120 1127
1122 1125 1115 1121
1123 1121 1116 1118
– 1086 – –
1089 1089 1089 1088
1090 1089 1089 1089
1017 1014 1017 1017
1020 1015 1020 1017
1020 1020 1019 1014
Fig. 3. Raman spectra of the synthetic product deconvolution (m1 mode) using two different excitation wavelengths: 1064 nm (panel A) and 532 nm (panel B).
attributed to (m1) C@C, (m2) CAC and (m4) CACH3 in polyenes or polyenals and (q3) CACH3 in carotenoids. Spectra recorded using 532 and 785 nm laser lines (Fig. 4, panels C and D) did not show a significant enhancement of the intensity of the band at 1010 cm1 corresponding to methyl groups. Theoretical calculations from conjugated polyenal with 11 double bonds (from now on called polyenal-11) showed that the band at 1106 cm1 (theoretical calculation unscaled) may be assigned to CACH3 mode in the end of carbon chain. Differing from carotenoids [38], we neither observed combination bands involving [m1(C@C) + m4(CACH3 in polyenals)] nor an intensification of the band at 1010 cm1, which may suggest that the mixture of obtained products does
not contain a methyl group in the middle of the conjugated carbon chain. Several other works of resonance Raman excitation of carotenoids have also shown that the CACH3 mode belongs to the electronic delocalization system over the carbon chain, which also supports the idea depicted in this investigation [3,8,38,39]. Previous studies on the octocoral Corallium rubrum have presented a different understanding on the polyenic based pigment composition. The divergence is mainly based on the fact that the found pigment is a carotenoid or a conjugated polyene completely demethylated [40,41]. In this sense, Brambilla et al. [22], based on the calculated wavenumber shifts caused by the conjugation, have reported that m4 mode is due to the in plane rocking of the ACH3 groups; as the number of methyl groups linked to the conjugated carbon chain decreases, the wavenumber of the vibrational mode increases and the intensity decreases. This fact could be justified by the decrease in the resonance effect on the polyunsaturated carbon chain containing the methyl groups. According to their data the main component of the pigment from C. rubrum is composed of polyene chain partially methylated. A good approach for this idea can be also found in a classical work by de Oliveira et al. [38], where authors have studied the resonance Raman effect on bixin, a C9 carotenoid obtained from Bixa orellana; in such system, authors have pointed out the massive conjugation mainly on the central 5 carbon atoms in the conjugated chain, despite the several double bond conjugations present in the structure. It is important to notice that even today there is a great interest in the use of resonance Raman conditions to analyze the spectroscopic properties from different polyenes; in a very recent paper, Tommasini and coworkers [42] present an review about the theory of overtones and combination bands in resonance Raman spectroscopy, using b-carotene as the model system; however, despite the physical approach authors have used in this paper, they do not address the chemical nature of the polyene system, or discuss the chemical importance of the ACH3 group in the structure of the carotene, since this group appears as one the chromophoric moieties in the resonance excitation profile; this is very evident in the bixin work, and of course the importance that the central part of the polyenic chain which is responsible for the electronic properties of the system [38]. In this sense, some previous works of our group have demonstrated that the spectroscopic difference between carotenes (or carotenoids) and other conjugated polyenes (such as parrodienes) is the presence of the band at ca. 1010 cm1, due to the ACH3 group, which is present in the resonance spectra of carotenes, but is not present in the resonance spectra of parrodienes [5]. In other words, the band at ca. 1010 cm1 will only be resonance enhanced if the ACH3 groups participate over the delocalized electronic cloud, such as in the case of b-carotene and other
438
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 4. Raman spectra of the synthetic product using two different excitation wavelengths: 532 nm (panel A – with Na2SO4 as a standard; panel C – without Na2SO4 as a standard) and 785 nm (panel B – without Na2SO4 as a standard; panel D – without Na2SO4 as a standard).
carotenes and carotenoids, such as bixin; however, if the conjugated polyene does not present the same ACH3 groups as carotenes, but has the same groups in the chemical structure, such as parrodienes, such band appears at the Raman spectrum in ordinary conditions, but cannot be seen in resonance conditions, as we are showing in our investigation. In order to evaluate the differences and similarities between synthetic, natural polyenals and carotenoids, in this study we have studied the octocoral Leptogorgia punicea. Subtle differences were observed from the decomposition of the in situ spectra using 532 laser line. The octocoral has shown 3 different bands for the m4 mode: 1003, 1013 and 1021 cm1, whereas 3 other bands have been also calculated for m1 mode between 1490 and 1536 cm1: 1490, 1515 and 1536 cm1. The overtone band at 3024 cm1 could be attributed to the overtone vibration (m1) in 1515 cm1 (2m1 = 3030 cm1) and the overtone band at 2245 cm1 corresponding to the vibration (m2) at 1128 cm1 (calculated for 2256 cm1). The combination of the fundamental bands (m1 + m2) was observed at 2628 cm1 (calculated for 2643 cm1) as well as the combination band 1013 cm1 (m4) + 1515 cm1 (m1) was observed as a broad feature at 2528 cm1. Differing from the synthetic polyenal the presence of the combination band m1 + m4 from L. punicea may suggest the presence of an additional methyl group, as proposed for C. rubrum [22]. Similar approach is shown in Table 2 that describes main vibrational modes observed by the deconvolution analysis using FT-Raman spectrum of four corals species. The main information obtained from these marine species is that they do not produce a unique pigment, and through the deconvolution m4 mode of corals we have found that some corals showed resemblance with methylated pigments, for example, L. punicea showed the m4 mode at 1006 and 1021 cm1 as well as demethylated pigments. Deconvolution analysis of m1 mode of corals showed that the species of L. punicea, L. setacea and Chromonephthea braziliensis (yellow polyps and pink stalk) have more than one band, confirming the hypothesis that it is the mixture of polyenes and not a single substance responsible for the coloration of corals and other organisms such as mollusk shells and feathers of macaws.
Table 2 Deconvolution bands of m1 and m4 (cm1) of the corals samples from the FT-Raman spectra. Samples
m1 (cm1)
m4 (cm1)
L. punicea
1509 1528
1006 1021
L. setacea
1517 1539
1004 1020
R. muelleri
1500 1514
1005 1020
C. braziliensis yellow polyps
1523 1540
1015
C. braziliensis pink stalk
1502 1538
1015
It is important to note that the theoretical calculated spectrum for the polyenal-11 (tetracosaundecaenal) are in good agreement with that obtained experimentally with respect to the wavenumbers of the main bands. As can be seen in Fig. 5, in which the DFT calculated spectrum is not scaled, the general profile of the bands is similar to experimental one, even without any adjustment to the Raman intensities. Fig. 5 also shows the bands in the region of 1400–1900 cm1 enhanced for better visualization. Assuming that would be interesting to have a scaling factor the DFT gas phase vibrational frequencies would require the adjustment of only 9% when compared to experimental data, which is a good fit thinking on other comparisons displayed in the literature [43]. The resulting theoretical spectrum assignments also show bands in 1422 and 1427 cm1 related to an umbrella motion-like of methyl group, d CH3; in 1489 cm1 (rocking) and 1502 cm1 (stretching and angular deformation of the conjugated methyl mode). Such information are important in showing that in fact, only one methyl must be present in the synthesized compound (terminal), since these bands are also present in the experimental spectrum. In general, the calculated vibrational frequencies are in good accordance with experimental values reported here.
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 5. Comparison between theoretical (bottom) and experimental (top) Raman spectra. The detached region (1400–1900 cm1) has been enhanced for a better visualization.
Fig. 6. Theoretical UV–Vis profile comparison between lycopene and polyenal-11.
Another, but no less important aspect is the analysis of the frontier molecular orbitals (MOs) obtained from TD-DFT calculations. In this analysis we performed a comparison of polyenal-11 with lycopene, a carotene with also 11 doubles bonds. It can be perceived by the overlaying in the theoretical UV–Vis spectra (Fig. 6) that despite the difference in displacement between the bands themselves (621 nm for lycopene and 594 nm for polyenal-11), the main contributions to the HOMO and LUMO of lycopene are exactly concentrated in the region where the molecule is straight, showing that this is an indication that the size of the box (thinking of the particle in the box model) may not be as important as suggested before [44]. Fig. 7 shows the MOs of these two compounds and allow us to confirm the p-electron delocalization, showing HOMOs concentrated in C@C doubles bonds and LUMOs concentrated in CAC single bonds, as expected. It is important to say that Fig. 7 just represents an illustration, a static representation, since in an ab initio calculation since it is know that there is no differentiation between those representations from the quantum point of view. In addition to synthesize the polyenal we are also presenting the occurrence of pigments related to conjugated polyenals in
439
different species of marine corals. The particular Raman shifts of the pigments are summarized for some representative cases in Table 1. The literature already describes the effect of chain length and the role of aldehydic function on the antiproliferative and antioxidant effects [45]. The spectrum obtained for the synthetic sample resembles the one observed for in situ sample and demineralized sclerite of L. punicea; since this kind of compounds is inserted into sclerites composed of calcium carbonate crystallized as calcite present in the coenenchymes [4,6,12,25] some differences are observed in the Raman shift values. The in situ sample shows bands at 1507 [m1(C@C)], 1123 [m2(CAC)] and 1090 cm1 [m(CO2 3 )] and demineralized shows bands at 1500 [m1(C@C)], 1116 [m2(CAC)] and a mode in 1011 cm1 [m4(CACH3)], which has been assigned to the rocking motion of ACH3 backbone in carotenes [39] or to a CH@CH wagging mode in polyene molecules [46]. Recent studies related the inclusion processes of carotenes with bcyclodextrins and the analysis by Raman spectroscopy, demonstrating the matrix effect on the band position as a consequence of the inclusion complex formation [47]; in the present case this effect can be attributed to the chemical interactions existing when calcium carbonate species are interacting with the pigment present in natural samples [48]. The mixture present in synthetic polyenals showed a predominant reddish-brown coloration, featuring octocorals as C. braziliensis we observed distinct coloration from different parts of the colony. According to the literature [6], the in situ Raman spectroscopic analysis characterization of this kind of coral using 632.8 and 1064 nm laser lines, in two different parts of the coral (the pink stalk and the yellow polyps) it can be seen a single band in the 1502 cm1 region. Therefore, this result proposed that only one type of pigment was present in the pink stalk, however recent analysis of the spectra made with the 1064 nm laser excitation, indicated that is probable is a mixture of pigments. This statement is grounded by the fact that two bands are present in the region of [m1(C@C)], at 1502 and 1540 cm1 for pink stalk which is not noticed in the yellow polyps. It has just a band in 1526 cm1. One very specific characteristic of isolated polyenals from corals is the insolubility in almost all the tested solvents; in other words, despite the conventional extractions method using different solvent systems is well described for carotenoids [49,50], such methods are not applied for polyenals from corals. Because of this feature, the pigment analysis using HPLC or other approaches to separate the different components of the mixture from the synthesis has become a limiting task. Finally, it can be emphasized that this study has provided spectroscopic data about the polyenal synthesis mainly containing 11 double bonds. The Raman spectroscopy technique using lasers of different energies can result in drastic changes, mainly based on spectral intensities and Raman shifts; different chromophoric groups and different wavenumber values for each one of the vibrational modes can be associated to polyenes containing different numbers of double bonds. Based on the behavior of the synthetic sample excited using the 532 nm line it can also be observed both combination and overtone bands, but differing from carotenoids, neither combination bands involving [m1(C@C) + m4(CACH3)] nor an intensification of the band at 1010 cm1. These results may suggest that the obtained product do not contain a methyl group in the central region of conjugated carbon chain, as carotenoids do present. About the analyzed corals, the main conclusion obtained from the Raman spectra analysis is the fact that these marine species do not produce a single pigment, and through the deconvolution of the m4 mode it can be found that some corals show resemblance with methylated pigments. The occurrence of more than one pigment in the tissues of corals [51] and phytoplanktonic organisms [52] may be related to the capture of sunlight through the water mass. Variations in light intensity that reaches the marine
440
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441
Fig. 7. HOMO ? LUMO gap for (a) polyenal-11 and (b) lycopene.
organisms may vary with turbidity, tide and season [53]. The production of pigments with different sizes of polyenic chains may provide adaptability of capturing light of different wavelengths. Conclusions In this work we have used Raman spectroscopy in the characterization of synthetic polyenals as well polyenals from natural sources such as species of octocorals. It can be observed some important differences between these compounds mainly analyzing the intensity and position of the m4 mode whereas the decomposition of the spectral analysis of this band can indicate that some corals do not produce a unique type of pigments: it may have a mixture involving methylated and demethylated polyenes present in the tissues. TD-DFT calculations of the frontier molecular orbitals (MOs) together with theoretical calculated Raman spectra of synthetic polyenal, natural polyenals and lycopene have provided clues to demonstrate the power of Raman spectroscopy in distinguishing pigments composed of similar chromophores such as carotenoid and demethylated polyene. At last, the band at ca. 1010 cm1 will only be resonance enhanced if the ACH3 groups participate over the delocalized electronic system, such as in the case of b-carotene and other carotenes and carotenoids, as bixin; however, if the conjugated polyene does not present the same ACH3 groups as carotenes, but has the same groups in the chemical structure, such as parrodienes, such band appears at the Raman spectrum in ordinary conditions, but cannot be seen in resonance conditions, as it can be seen in this investigation. Acknowledgements Authors are indebted to FAPEMIG, CNPq and CAPES/Ciências do Mar 1137/2010 (Brazilian agencies), PRONEX for financial support and the Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis (IBAMA) for collecting license (20649-1). They also thank Hiran Carvalho (UERJ), Antônio S.F. Couto (diving operator, Deep Trip, Arraial do Cabo city, RJ, Brazil) for collecting the coral samples and Clovis B. Castro for donating octocorals (UFRJ-MNRJ). References [1] S. Karampelas, E. Fritsch, J.Y. Mevellec, S. Sklavounos, T. Soldatos, Eur. J. Mineral. 21 (2009) 85–97. [2] J. Urmos, S.K. Sharma, F.T. Mackenzie, Am. Mineral. 76 (1991) 641–646.
[3] J.C. Merlin, M.L. Delé-Dubois, Comp. Biochem. Physiol. Part B: Comp. Biochem. 84 (1986) 97–103. [4] L.F. Maia, V.E. de Oliveira, M.E.R. de Oliveira, B.G. Fleury, L.F.C. de Oliveira, J. Raman Spectrosc. 43 (2012) 161–164. [5] L.F. Maia, V.E. de Oliveira, M.E.R. Oliveira, F.D. Reis, B.G. Fleury, H.G.M. Edwards, L.F.C. de Oliveira, J. Raman Spectrosc. 44 (2013) 560–566. [6] L.F. Maia, B.G. Fleury, B.G. Lages, J.P. Barbosa, A.C. Pinto, H.V. Castro, V.E. de Oliveira, H.G.M. Edwards, L.F.C. de Oliveira, J. Raman Spectrosc. 42 (2011) 653– 658. [7] K.J. McGraw, M.C. Nogare, Biol. Lett. 1 (2005) 38–43. [8] M. Veronelli, G. Zerbi, R. Stradi, J. Raman Spectrosc. 26 (1995) 683–692. [9] R. Stradi, E. Pini, G. Celentano, Comp. Biochem. Physiol. Part B: Biochem. Mol. Biol. 130 (2001) 57–63. [10] W. Barnard, D. de Waal, J. Raman Spectrosc. 37 (2006) 342–352. [11] C. Hedegaard, J.-F. Bardeau, D. Chateigner, J. Mollus. Stud. 72 (2006) 157–162. [12] F. Luwei, Y. Mingxing, J. China Univ. Geosci. 19 (2008) 146–151. [13] L. Bergamonti, D. Bersani, D. Csermely, P.P. Lottici, Spectrosc. Lett. 44 (2011) 453–458. [14] J.F. Masello, M.L. Pagnossin, T. Lubjuhn, P. Quillfeldt, Ecol. Res. 19 (2004) 421– 432. [15] E.H. Burtt, M.R. Schroeder, L.A. Smith, J.E. Sroka, K.J. McGraw, Biol. Lett. 7 (2011) 214–216. [16] R. Stradi, A. Bertelli, E. Pini, Polyunsaturated linear aldehyde and their derivatives with anti-radical and anti-tumoral activity, in: U.S Patent 2005/ 0175555 A1, 2005. [17] S. Mukherjee, D. Ray, M. Falchi, A. Bertelli, E. Pini, D.K. Das, J. Nat. Prod. 72 (2009) 871–875. [18] A. Alker, K. Kim, D. Dube, C.D. Harvell, Coral Reefs 23 (2004) 397–405. [19] S.P. Ellner, L.E. Jones, L.D. Mydlarz, C.D. Harvell, Am. Nat. 170 (2007) E143– E161. [20] C.V. Palmer, L.D. Mydlarz, B.L. Willis, Proc. Roy. Soc. B: Biol. Sci. 275 (2008) 2687–2693. [21] K. Kiho, A. Alker, K. Shuster, C. Quirolo, C.D. Harvell, Dis. Aquat. Organ. 69 (2006) 95–99. [22] L. Brambilla, M. Tommasini, G. Zerbi, R. Stradi, J. Raman Spectrosc. 43 (2012) 1449–1458. [23] M. Schmitt, J. Popp, J. Raman Spectrosc. 37 (2006) 20–28. [24] R.S. Stenner, J. Raman Spectrosc. 36 (2005) 276–278. [25] E. Fritschi, B. Rondeau, T. Hainschwang, S. Karampelas, EMU Notes Mineral. 12 (2012) 453–488. [26] T. Kupka, H.M. Lin, L. Stobin´ski, C.H. Chen, W.-J. Liou, R. Wrzalik, Z. Flisak, J. Raman Spectrosc. 41 (2010) 651–658. [27] H.E. Schaffer, R.R. Chance, R.J. Silbey, K. Knoll, R.R. Schrock, J. Chem. Phys. 94 (1991) 4161–4170. [28] H. Kuzmany, Pure Appl. Chem. 57 (1985) 235–246. [29] C. Heller, G. Leising, C. Godon, S. Lefrant, W. Fischer, F. Stelzer, Phys. Rev. B 51 (1995) 8107–8114. [30] A. Andreoni, C.A. Sacchi, O. Svelto, in: C.B. Moore (Ed.), Chemical and Biochemical Applications of Lasers, Academic Press, 1979, pp. 1–30. [31] A.D. Becke, J. Chem. Phys. 98 (1993) 5648–5652. [32] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785–789. [33] J.A. Pople, D.P. Santry, G.A. Segal, J. Chem. Phys. 43 (1965) S129–S135. [34] R.E. Stratmann, G.E. Scuseria, M.J. Frisch, J. Chem. Phys. 109 (1998) 8218–8224. [35] S. Karampelas, E. Fritsch, J.Y. Mevellec, J.P. Gauthier, S. Sklavounos, T. Soldatos, J. Raman Spectrosc. 38 (2007) 217–230. [36] R.J.H. Clark, T.J. Dines, Angew. Chem. Int. Edit. English 25 (1986) 131–158. [37] D.A. Long, in: The Raman Effect, John Wiley & Sons, Ltd., 2002, pp. 221–270. [38] L.F.C. de Oliveira, S.O. Dantas, E.S. Velozo, P.S. Santos, M.C.C. Ribeiro, J. Mol. Struct. 435 (1997) 101–107.
R.F. Fernandes et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 134 (2015) 434–441 [39] H. Okamoto, H.-O. Hamaguchi, M. Tasumi, J. Raman Spectrosc. 20 (1989) 751– 756. [40] E. Fritsch, S. Karampelas, Mar. Biol. 154 (2008) 929–930. [41] J. Cvejic, S. Tambutté, S. Lotto, M. Mikov, I. Slacanin, D. Allemand, Mar. Biol. 152 (2007) 855–862. [42] M. Tommasini, G. Longhi, S. Abbate, G. Zerbi, J. Raman Spectrosc. 45 (2014) 89– 96. [43] L.V. de Freitas, C.C.P. da Silva, J. Ellena, L.A.S. Costa, N.A. Rey, Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 116 (2013) 41–48. [44] J. Autschbach, J. Chem. Educ. 84 (2007) 1840. [45] D. Tagliazucchi, E. Verzelloni, E. Pini, G. Ronca, A. Conte, Pharmacologyonline 3 (2006) 765–771. [46] Y. Koyama, I. Takatsuka, M. Nakata, M. Tasumi, J. Raman Spectrosc. 19 (1988) 37–49.
441
[47] V.E. de Oliveira, E.W.C. Almeida, H.V. Castro, H.G.M. Edwards, H.l.F. Dos Santos, L.F.C. de Oliveira, J. Phys. Chem. A 115 (2011) 8511–8519. [48] V.E. de Oliveira, H.V. Castro, H.G.M. Edwards, L.F.C. de Oliveira, J. Raman Spectrosc. 41 (2010) 642–650. [49] T. Lacker, S. Strohschein, K. Albert, J. Chromatogr. A 854 (1999) 37–44. [50] D.B.R. Amaya, A guide to carotenoid analysis in foods. ILSI Press, Washington, DC, 2001, pp. 1–64. [51] D.F. Gleason, Ecosystems, evolution and ultraviolet radiation, statesboro, GA, 2001. [52] M. Stomp, J. Huisman, F. De Jongh, A.J. Veraart, D. Gerla, M. Rijkeboer, B.W. Ibelings, U.I.A. Wollenzien, L.J. Stal, Nature 432 (2004) 104–107. [53] M.K. Wicksten, Bull. Mar. Sci. 45 (1989) 519–530.
