Abstract

This paper describes the ordering degree of anionic, cationic, and zwitterionic surfactants with the increase of their packing density on Ge substrate by using Fourier transform infrared-attenuated total reflection (FTIR-ATR) spectroscopy. This work shows new insights on the conformational order of sodium dodecyl sulfate (SDS), N-hexadecyl-N-N-dimethyl-3-ammonio-1-propane-sulfonate (HPS), hexadecyl-trimethylammonium bromide (CTAB), and dodecyl trimethylammonium bromide (DTAB). DFT and semiempirical calculations are also performed in order to evaluate the effect of headgroup hydration and counterion. The CH2 asymmetric and symmetric stretching bands for the SDS molecule show a shift of 1.7 and 0.9 cm−1 to higher frequencies as the packing density increases, while it is observed a shift of 2.6 and 2.7 cm−1 for the HPS molecule, respectively. The DTAB molecule shows a shift of 4.5 cm−1 to lower frequencies for both CH2 asymmetric and symmetric stretching bands as the packing density increases, indicating the decrease of gauche conformations and the increase of all-trans conformations over the aliphatic chain.

1. Introduction

Surfactants have been widely studied due to its significance in both applied and fundamental processes: detergency, catalysis, flotation, lubrication, colloid stabilization, foaming, emulsification, protein denaturation, tension moderation in membranes, membrane permeation, and drug delivery [1]. Surfactants are also used to the syntheses of nano- and mesomaterials using their capability to form self-organized aggregate structures [2]. Therefore, the determination of the packing, ordering, and its relation to the properties of the surfactant aggregates is of fundamental importance [3, 4].

Sperline [5] and Sperline et al. [6] have studied several crystalline phases of sodium dodecyl sulfate (SDS) using transmission infrared techniques. They have pointed out some relevant aspects: (1) difficulty in assigning a degree of order to the packing of alkyl chains based on the asymmetric CH2 stretching band; (2) the molar absorptivities for the alkyl vibrational features of adsorbed structures may not be compared with those for SDS micellar solutions; (3) the relative abundances of the CH2 symmetric and asymmetric stretching bands of SDS change considerably with the crystalline phase. In other words, the nature of the packing of surfactant molecules may determine their molar absorptivities.

Snyder et al. [7], Flach et al. [8], and Dicko et al. [9] show that the frequency of the CH2 asymmetric stretching feature decreases with the conformational ordering in the hydrocarbon chain of polymers and phospholipids. However, the quantitative correlation between the magnitude of the shifts and the ordering extent has remained obscure and elusive. Since then, the asymmetric CH2 stretching band has been used as a reference for ordering in any kind of surfactant containing a methylene hydrocarbon chain. In fact, Prosser and Franses [10] and Scheuing and Weers [11] indicate that this feature correlates with the packing density and the conformational order of hydrocarbon chain tail of surfactants.

The cross-section of the sulfate headgroup is about  cm2 molecule−1 as reported by Vold and Vold [12] and Sigal et al. [13]. However, the cross-section of a tightly packed SDS monolayer is about 100 times larger,  cm2 molecule−1 [14, 15]. Besides, a condensed SDS monolayer film electrochemically reorganized shows a value of  cm2 molecule−1 [16], while the theoretically value is predicted to be cm2 molecule−1 [17, 18]. Therefore, the amount of surfactant material deposited over a solid substrate is dependent on the deposition procedure, packing and ordering of the surfactant.

The purpose of the present study is to prepare crystallites of surfactant on Ge substrates and assign the infrared features of these crystallites by using Fourier transform infrared-attenuated total reflection (FTIR-ATR). This study describes the vibrational features of an anionic SDS, cationic surfactants (hexadecyl-trimethylammonium bromide (CTAB) and dodecyl trimethylammonium bromide (DTAB)), and a zwitterionic surfactant (N-hexadecyl-N-N-dimethyl-3-ammonio-1-propane-sulfonate (HPS). In this study, the surfactants are directly prepared over an ATR Ge element to evaluate the packing density and organization of the alkyl chain conformation for each surfactant.

2. Materials and Methods

SDS (purity > 99%), CTAB (purity > 99%), DTAB (purity > 99%), and HPS (purity > 99%) were obtained from Sigma-Aldrich Co. Methanol (HPLC grade) was purchased from J. T. Baker and used as received.

The casting technique was applied to estimate the surfactant density over the ATR Ge element. The casting technique consisted in spreading an aliquot over one of the ATR Ge element faces by using a Teflon bar. It was assumed that the surfactant covers 90% of the ATR element. All surfactant solutions were made with deionized water (Millipore, Milli-Q, resistivity 18 MΩ cm), and prepared with 40% (V/V) of methanol. It is important to mention that the solutions were prepared below the c.m.c for all surfactants employed in this study. The volume added over the Ge substrate was varied from 2 to 10 μL. Then, the Ge substrate with the surfactant was put in a desiccator coupled with vacuum pump operating at a pressure of ~  Torr. This procedure eliminates the remaining solvent excess from surfactant in about 10 minutes. The whole deposition equipment was kept in a clean environment at room temperature to avoid complications with the presence of dust particles. It is important to assign that the depositions were performed layer-by-layer in order to vary the density from 1014 to 1018 molecules cm−2.

The infrared spectra of the transferred surfactant were collected on a Varian/Digilab FTS7000 spectrometer equipped with a high sensitivity narrow band liquid-nitrogen-cooled mercury-cadmium-tellurium (MCT) detector. The sampling was performed using Horizon ATR accessory (Harrick Scientific Inc.). The ATR theory and accessory are fully described in the literature [19]. It consists of a set of two plane mirrors to direct the infrared beam into the ATR Ge element and then to the MCT detector. The ATR Ge element is a single-pass trapezoid with dimension of  mm and a bevel angle ( ) of 45°. The length ( ) and thickness ( ) of the ATR Ge element determine the number of reflections ( ) by the formula , which gives 25 internal reflections. For each spectrum, 128 single-beam scans were averaged with 1 cm−1 resolution for the reference and sample. Prior to deposition, the ATR Ge substrate was cleaned with a suitable procedure [19] until the CH2 signal was eliminated. The reference spectrum was obtained by transmitting the infrared beam along the ATR Ge substrate alone, after which the sample spectrum was taken immediately after transferring the surfactant onto the ATR Ge element.

Theoretical Calculations
The calculations were carried out with the Gaussian software [20] and MOPAC package [21]. Stationary points on the potential energy surface were fully optimized, followed by the evaluation of the harmonic vibration frequencies in order to characterize their nature as minima. The absence of imaginary frequencies indicated that all optimized structures were true minima. Quantum chemical calculations were performed with Becke’s three parameter hybrid functionals (B3LYP), which include a mixture of HF exchange with DFT exchange correlation. B3LYP functional [22] uses the nonlocal correlation provided by the LYP expression [23]. The 6–311++G(d,p) basis set was employed. AM1 [24], RM1 [25], and PM6 [26] were employed for semiempirical calculations.

3. Results and Discussions

3.1. Deposition Techniques

The addition of methanol as a cosolvent in the standard surfactant solutions is an important aspect that must be commented. It was tried to evaluate the preparation of the surfactant solution with the absence of methanol. The problem is that some surfactants show a broad band in the spectral region between 1300 and 900 cm−1, which may be due to the water crystallization. The addition of several percentages of methanol as a cosolvent (20, 40, and 60%) was tried to solve this matter, nevertheless 40% seemed to avoid perfectly this problem to each surfactant. Prosser and Franses [10] had pointed out that the contamination by residual ethanol may cause two problems in the preparation of the cast films. First, they observed a shift of 6 cm−1 of the CH2 antisymmetric stretch for higher frequencies, which was also followed by a non clear region between 1500 and 800 cm−1 indicating a lack of crystalline structure in the cast film with ethanol. An explanation to this problem is that ethanol may penetrate the micelle [27]. After eliminating the solvent, some ethanol may intercalate between the surfactant molecules, disordering the crystalline structure. In our case, methanol is completely eliminated as it can be seen due to the absence of the infrared features in 3328, 1418, and 1030 cm−1, which refers to ν(O–H), δ(O–H), and ν(C–O) for methanol, respectively.

de Souza et al. [28] observed that drying procedures affect the morphology of the crystalline structure. The samples dried under room conditions seem to be more homogeneous than those dried under vacuum or by nitrogen flow [28]. Silva et al. [29] show that the hydrocarbon chains are more ordered using spontaneous water evaporation than those dried by nitrogen flow. They suggested that the spontaneous water evaporation reduces the effect of dragging by the drying front. However, the fast water evaporation in the process of nitrogen-flow drying “freezes” the disordered conformation chains. In addition, Halthur et al. [30] commented that drying procedures do not result in any irreversible changes in the conformational structure. In this study, in order to evaluate the influence of drying procedure, some tests were performed to observe the effects of spontaneous solvent evaporation at room temperature and by the use of vacuum pump. It is important to mention that none difference was observed in the vibrational spectra of the surfactants for small amount of materials. Nevertheless, a broad band in the region between 1300 and 900 cm−1 is observed for depositions with large amount of materials, larger than 1016 molecules cm−2. This is observed only for anionic and zwitterionic surfactants. However, it was preferred to use the vacuum pump drying for eliminating the solvent for all surfactants, keeping this procedure as standard.

3.2. Anionic Surfactant
3.2.1. C–H Stretching Region

The vibrational frequencies in the C–H stretching region for the SDS molecule can be observed in Figure 1(a). The asymmetric (CH3) and symmetric (CH3) CH3 stretching vibrational frequencies are located at 2955 and 2873 cm−1, respectively. The asymmetric and symmetric CH2 stretching vibrational frequencies ( (CH2) and (CH2)) are observed at 2917 and 2849 cm−1, respectively. The CH3 stretching intensities are lower than those for the CH2 stretching features as expected. The CH2 asymmetric and symmetric stretching features can be used to ascribe the packing and conformation of SDS molecules on Ge substrate due to the fact that the C–H stretching of alkyl chain assemblies is sensitive to the conformation of methylene chains. For example, when the (CH2) band presents values lower than 2852 cm−1, it is a good indicative of a more ordered crystalline structure, while values higher than that are representative of micelles and liquid crystals [5]. Besides, a shoulder at about 2860 cm−1 indicates the coexistence of noncrystallized SDS and hydrated crystalline SDS [31]. This shoulder is not observed in our study.

The CH2 asymmetric stretching appears at 2917 cm−1, suggesting a high ordering hydrocarbon chain in all-trans CH2 configuration [79]. The band shift from higher 2917 cm−1 to lower frequencies means that the number of gauche conformers decreases and the number of highly ordered all-trans conformers of alkyl chain increases. This reflects a change from more disordered (liquid phase) to a more crystalline (solid phase) conformation. The (CH2) band of SDS micelles are observed between 2936–2928 cm−1 [31] and it is clearly dependent on its concentration. In addition, Sperline et al. [6] show that the asymmetric stretching (CH2) for the SDS molecules in the liquid crystal phase is around 2924 cm−1. The (CH2) feature in our study is different from those described above, however, it is similar to those determined in the literature [5, 10]. Nevertheless, it is important to mention that each study used different preparation methods.

Figure 2(a) presents the relationship between packing density and the asymmetric stretching (CH2) feature. As the packing density increases on Ge substrate, it is observed that the band shifts to higher frequency values. The shift is only 1 cm−1 as the packing density of SDS molecules changes by a factor of 1000. The CH2 symmetric feature has the same behavior observed in the CH2 asymmetric band as presented in Figure 2(b). The (CH2) shift is similar to that observed to the (CH2) band. However, due to this small shift in (CH2) and (CH2) modes, it just indicates the high ordered all-trans conformers.

3.2.2. CH2 Scissoring Region

Additional information about the alkyl chain conformation is also obtained with the location of CH2 scissoring mode, δ(CH2). This feature is very sensitive to side-by-side chain interactions as well as the packing organization of the methylene chain [3235]. Low δ(CH2) intensities and frequencies around 1466 cm−1, and band broadening indicate a reduction of side-by-side chain interactions and an increase in chain motion, which is normally associated with the liquid state [32, 33, 35]. Besides, a frequency higher than 1472 cm−1 is an indicative of an all-trans conformation. Nevertheless, the δ(CH2) band in this study is located at 1468 cm−1 as presented in Figure 1(b), which is a characteristic of partially ordered chains [33, 34]. The shoulder at 1457 cm−1 is also an indicative of partially ordering conformation, which is generally associated with a combination of gauche defects and the asymmetric deformation of the terminal CH3 group [32, 33].

3.2.3. CH2 Wagging Region

The spectral region located in 1300–1400 cm−1 is a characteristic of the CH2 wagging modes. This region exhibits peaks which are related to gauche conformations [36, 37]. The peak at 1341 cm−1 indicates an end-gauche (e-g) conformation with a terminal CH3 group oriented in a gauche conformation relative to the CH2 group. In addition, the peak at 1354 cm−1 is an indication of two adjacent gauche bonds (d-g), and the peak at 1368 cm−1 is due to a gauche-trans-gauche sequence (g-t-g). Nevertheless, a very weak peak at 1341 cm−1 is observed in our study, suggesting an e-g conformation. This low intensity is also characteristic of the crystalline phase in the case of micelle structures [3538]. Furthermore, it is also observed a very weak band at 1378 cm−1 which refers to the CH3 umbrella mode.

3.2.4. Headgroup Vibrational Region

The SO2 asymmetric vibrational feature ( (SO2)) is the most intense band in the SDS spectrum. It is a combination of several overlapping peaks, and it is generally observed as double band [10]. As observed in Figure 1(c), (SO2) is located at 1219 and 1249 cm−1. The separation between these two peaks is an indicative of the conformational structure. The band separation of 30 cm−1 presented in our study is similar to that observed by Prosser and Franses [10], nevertheless higher than that found by Sperline [5] for the SDS crystalline conformation which ranges from 27 to 29 cm−1. However, the value observed in our study is lower than those predicted to the bulk (39–48 cm−1) [6, 38, 39] and liquid crystals (32-33 cm−1) [3941]. If there is an SDS interaction with charge particles, it is observed a splitting between 33–38 cm−1 [42].

The dipole components of sulfate headgroup have been used to evaluate the horizontal and vertical effects over the molecules. While the band at 1219 cm−1 is assigned to the component, the band at 1249 cm−1 and the SO2 symmetric stretching band ( (SO2)) are associated to the component. It is important to mention that the directions of the dipole components are orthogonal to each other [43, 44] as it was confirmed by our theoretical calculations. Li and Tripp [43] suggested that the relation between the (SO2) band intensities made allusion to important insights about the lateral electrostatic headgroup interaction. They indicated that the observation of an intensity band in 1249 cm−1 higher than that one in 1219 cm−1 may be an indicative of strong head-to-head lateral interaction. Nevertheless, the I1219/1249 ratio is basically constant and independent of the SDS packing density. A shoulder at 1278 cm−1 became more visible with the increase in the SDS packing density. This shoulder is associated with component of dipole moment, which is related to lateral interactions of the SDS headgroup. Therefore, this may be due to the repulsion between the lateral headgroup.

An important aspect about the SO2 symmetric vibrational feature is its observation at 1084 cm−1. Scheuing and Weers [11] observed the (SO2) mode shifted to 1060 cm−1. The explanation for this shift is the loss of interaction between the headgroup and counterions, which is assessed by our quantum chemical calculations.

Two shoulders at 1097 and 1066 cm−1 are also observed. Some studies suggest that these two peaks may be due to the packing organization [39]. The intensity ratio between these two peaks (I1097/1066) decreases from 5.88 at the packing density  molecules cm−2 to 0.85 at  molecules cm−2. In addition, the sharp bands at 1017, 993, and 983 cm−1 may be due to a crystalline phase, as it was reported previously by Sperline and Song [39].

Some studies suggest that the (SO2) mode changes in the presence of cationic surfactants [11, 43] or charged particles [42, 44]. Using the B3LYP/6–311++G(d,p) method, it was possible to determine that the frequencies of the (SO2) mode in the presence of Na+ counterion are reduced in about 30 cm−1, as compared with the absence of Na+ counterion (see Figure 3). Using semiempirical methods, the frequencies of the (SO2) mode in the presence of Na+ counterion are reduced in 16 and 81 cm−1 using the AM1 and PM6 semiempirical methods, respectively. Some simulations using the AM1 semiempirical method were performed in order to describe the hydrated headgroup with 24 water molecules without Na+ counterion and with Na+ counterion. It is observed an increase of 120 cm−1 for the (SO2) mode in the presence of Na+ counterion. On the other hand, it is calculated a decrease of only 10 cm−1 in the presence of water.

3.3. Cationic Surfactants
3.3.1. C–H Stretching Region

The vibrational frequencies of the CTAB molecule in the C–H stretching region are shown in Figure 5(a). The (CH3) and (CH3) frequencies are observed in 2943 and 2870 cm−1, respectively. The (CH3) feature located in a frequency lower than 2956 cm−1 is an indicative of the crystalline structure. The CH3 asymmetric stretching mode presents a value 12 cm−1 smaller than that observed for the SDS molecule. On the other hand, the (CH3) mode is only 3 cm−1 lower than that found for the SDS molecule, in agreement with the results reported in the literature [45]. It is important to assign that these bands seem to be less significant.

The (CH3) and (CH3) bands of the DTAB molecule are difficult to be observed due to their lower intensities. It is only possible to detect them with packing densities larger than 1015 molecules cm−2. It is important to mention that it is also difficult to distinguish between (CH3) and (N–CH3) bands, observing only a large band with low intensity in the region between 2940 and 2975 cm−1.

The shift in the (CH2) vibrational mode for the DTAB molecule is also observed. Figure 6(a) presents the behavior of the (CH2) feature for the DTAB molecule as the packing density increases. The (CH2) vibrational band shifts from 2921 to 2916.5 cm−1 for the DTAB molecule. The CH2 symmetric stretching (CH2) shifts 2 cm−1 as presented in Figure 6(b). Therefore, there is an increase of all-trans conformers on the methylene chain and a decrease of gauche ones.

Figure 5(a) shows that the (CH2) feature for the CTAB molecule is lower than 2917 cm−1 and the (CH2) frequency is lower than 2849 cm−1, meaning that it is a crystalline phase, which distinguishes from a micelle conformation [46, 47]. Figures 6(c) and 6(d) present straight lines for the changes of (CH2) and (CH2) with the packing density of the CTAB molecule. Li et al. [48] also determined a similar frequency for the (CH2) feature. Nevertheless, several investigations have found values around 2918 cm−1 [45, 4952]. It is important to mention that Campbell et al. [53] observed a (CH2) frequency lower than that found in our study, 2915 cm−1, for a packing density of  molecules cm−2, which is 3 times smaller than our lowest packing density. Considering the frequencies presented here, it is possible to assume that CTAB may follow the same behavior found for the SDS molecule.

3.3.2. CH2 Scissoring Region

The δ(CH2) scissoring feature for the CTAB molecule does not present any change with the increase of packing density as shown in Figure 5(b). The 1462 and 1472 cm−1 features are related to the δ(CH2) mode. However, the 1480 and 1487 cm−1 bands are characteristics of the (N–CH3) mode. When the δ(CH2) band is observed at 1472 cm−1, it is an indicative of high organization of the CH2 chain conformation. Another important aspect is the split of the δ(CH2) band. This is mainly due to the lateral interchain interaction between CH2 groups of adjacent chains, which in general characterizes an orthorhombic arrangement of the CH2 chains. It is important to mention that this splitting is absent in the monoclinic and triclinic conformation, or even in alkane chain assemblies of low packing density when lateral interchain interactions are weak [8, 54]. The splitting of the δ(CH2) band has also been determined in other studies for the CTAB molecule in crystalline phases [45, 4951].

The δ(CH2) mode for the DTAB molecule is observed at 1468 cm−1 as presented in Figure 4(c). Nevertheless, with the increase of the DTAB packing density, it is realized a splitting of this band in 1464 and 1472 cm−1 as shown in Figure 4(d). As it was previously mentioned, this mode is sensitive to the packing density arrangement in the alkyl chain assembly. The increase of trans conformers in the CH2 chain with the increase in the DTAB packing density is in agreement with the same observations found for the (CH2) and (CH2) modes.

3.3.3. CH2 Wagging Region

For the DTAB and CTAB molecules, the wagging modes between 1300–1400 cm−1 is presented in Figures 4(e) and 5(c), respectively. The three bands for CTAB are 1340, 1360, and 1370 cm−1. While the band with higher intensity in 1360 cm−1 refers to the d-g conformations, the lower intensity band at 1370 cm−1 indicates a g-t-g configuration. In addition, the band at 1340 cm−1 is an indication of an e-g conformation. Venkataraman and Vasudevan [45] observed higher intensities for these modes, indicating an organization lower than that found in our study. For DTAB, it is detected only a broad band in 1360 cm−1, characteristic of a d-g conformation. It was not observed any signal for densities lower than  DTAB molecules cm−2. Some authors commented that the absence of wagging signals is a strong evidence for the occurrence of a hydrocarbon chain melting phase transition [5557]. Nevertheless, the main reason for the absence of peaks may be due to the low absorptivity of DTAB wagging region.

3.3.4. Headgroup Vibrational Region

For the CTAB molecules, none change is observed in the region of the headgroup vibrational features. The (N–CH3) feature appears as a shoulder of weak intensity in 2949 cm−1 and a very weak band in 2959 cm−1. As the CTAB packing density increases, there is a small enhancement of these two vibrational modes. While the degenerate (N–CH3) band splits into three peaks at 3009, 3016, and 3030 cm−1, the (N–CH3) and ν(C–N) are observed at 1396 and 912 cm−1, respectively, which is in good agreement with the data in the literature [45].

The (N–CH3) mode of the DTAB molecule presents a different behavior as compared with what is observed for the CTAB molecule. The DTAB molecule has a large band in packing densities lower than molecules cm−2. As the packing density increases, the (N–CH3) feature splits in 7 cm−1, indicating a predominance of a trans conformation in the methylene chain. The splitting of the (N–CH3) band has been associated with differences in the headgroup packing. In general, it has also been linked with phase transitions in micelle structures [33]. Didodecyldimethylammonium (DDAB) bilayers also present a splitting of 7 cm−1 [33] in good agreement with our findings. Nevertheless, DTAB [33] and hexadecyltrimethylammonium sulfate micelles [32] showed a splitting of 11 cm−1.

The appearance of two bands for the (N–CH3) mode has also been correlated with a fully hydrated surfactant headgroup, or even related with the interaction of hydrated counterions in this region [33]. In order to understand the behavior of the bands (N–CH3) and (N–CH3), some theoretical calculations were performed (see Figure 7). Our AM1 semiempirical calculations show that the presence of counterion in a hydrated headgroup yield a splitting not only for the (N–CH3) mode, but also for the (N–CH3) feature. From our AM1 calculations it is observed a splitting of 26 and 30 cm−1 for (N–CH3) and (N–CH3) features, respectively. However, it was not possible to consider the hydrated headgroup in the absence of counterion due to some difficulty to find the stationary point with the AM1 method. In addition, the DFT results for the (N–CH3) mode is more sensitive for the (N–CH3) band to the interaction of counterion. While the (N–CH3) mode shifts only 2 cm−1, the (N–CH3) mode shifts 20 cm−1.

As it can be observed in Figures 4(a), 4(e), and 4(f), the (N–CH3) feature for the DTAB molecule shows only one peak at 3016 cm−1, while the (N–CH3) and ν(C–N) bands are observed at 1394 and 911 cm−1, respectively. With the increasing of the packing density, and similar to CTAB molecule, it is detected only an enlargement of the intensities.

3.4. Zwitterionic Surfactant
3.4.1. C–H Stretching Region

The bands of the alkyl chain for the HPS molecule are presented in Figure 8(a). The (CH3) and (CH3) bands are observed in 2944 and 2873 cm−1, respectively. Nevertheless, the (CH3) mode is not detected for the HPS molecule when the packing density is larger than 1017 molecules cm−2. For densities larger than that, it is observed only a peak in 2954 cm−1 related to the (N–CH3) mode. Besides, the fact that the (CH3) mode presents values lower than 2956 cm−1 is a good indicative of a crystalline conformation.

Different from the other surfactants evaluated in this study, a larger shift was found for the HPS zwitterionic surfactant. The (CH2) feature presented a shift from 2920 to 2923 cm−1. Therefore, the gauche conformers in the methylene chain seem to be detected in our investigation, even with low densities. As usual, the amount of gauche conformers increases when the packing density increases. This is particularly observed for the (CH2) band, which shows a similar shift, almost 3 cm−1.

3.4.2. CH2 Scissoring Region

The δ(CH2) feature for the HPS molecule shows a different behavior from the other surfactants previously investigated here. The δ(CH2) band for the HPS molecule is the only one that shows a significant shift to higher frequencies. For example, for packing densities smaller than 1017 molecules cm−2, it is observed a signal at 1465 cm−1, which is shifted to 1468 cm−1 as presented in Figure 8(b). This shift also indicates a high proportion of gauche conformations over the methylene chain.

Figure 9 presents the CH2 asymmetric (Figure 9(a)) and symmetric stretching (Figure 9(b)) features, and the CH2 scissoring mode (Figure 9(c)) for different HPS packing densities. The CH2 asymmetric stretching and the CH2 scissoring modes are linearly dependent on the HPS packing density. Surprisingly, the CH2 symmetric feature is not linearly dependent on the HPS density packing. Unfortunately, any plausible explanation for this behavior is available.

3.4.3. CH2 wagging region

The 1300–1400 cm−1 region is characteristic of wagging deformations that can be observed in Figure 8(c). The large band in 1345 cm−1 refers to the e-g conformation, while the 1366 cm−1 peak is associated with d-g ones. Nevertheless, the d-g conformers are only observed for packing densities larger than 1017 molecules cm−2. Besides, the 1378 and 1402 cm−1 bands are associated with CH2 umbrella mode and (N–CH3), respectively.

3.4.4. Headgroup Vibrational Region

None change is observed for the (N–CH3) and (N–CH3) features for the HPS molecule, similar to CTAB and DTAB molecules. It was not possible to assign the (N–CH3) band of the HPS molecule, perhaps due to its lower absorptivity. The feature located at 912 cm−1 is a small intensity and broad band, which refers to the ν(C–N) stretching. Nevertheless, the ν(C–N) mode is observed only for packing densities larger than 1016 molecules cm−2 as presented in Figure 8(d). With the increasing of packing density for the HPS molecule, it is only observed a broad band in the 900–1000 cm−1 region.

The 1133–1280 cm−1 region is associated with the (SO2) band. For densities lower than 1016 molecules cm−2, it is detected a doublet in 1246 and 1185 cm−1. Nevertheless, for packing densities between 1016 and 1017 molecules cm−2, it is observed only one signal in 1185 cm−1, which shifts to 1186 cm−1 for densities larger than 1017 molecules cm−2. Furthermore, it appears two shoulders with small intensities in 1211 and 1255 cm−1. The absence of a doublet and appearance of a singlet are a clear indicative of a change from a crystalline (solid-like conformation), to a more disordered conformation (liquid-like environment) [58, 59]. Although it is observed a split of about 44 cm−1 in our study, methyl methanesulfonate and ethyl methanesulfonate present splits of 25 and 34 cm−1, respectively [58, 59]. Furthermore, the frequencies of these shoulders for these molecules range from 1324 to 1367 cm−1, which are considerable higher than those found for the HPS molecule. Even higher frequencies for the (SO2) mode are also assigned for sulfonyl or sulfonic functional groups, which range from 1407 to 1456 cm−1 [6062].

The two features in 1038 and 1058 cm−1 are assigned to (SO2) mode. For packing densities larger than 1017 molecules cm−2, it is observed only one strong band in 1038 cm−1. This reflects a change from a more ordered to a more disordered conformation. Despite the fact that there is a split of 20 cm−1, the doublet splits for methyl methanesulfonate and ethyl methanesulfonate are 7 and 11 cm−1, respectively [58, 59]. It is also important to observe that these sulfonate groups present higher frequency values ranging from 1158 to 1183 cm−1, similar to those found for the asymmetric stretching. Considering the interaction of the sulfonate group with the hydronium ion, , DFT calculations presented a split of 18 cm−1 for the (SO2) mode, which is in good agreement with a split of 10 cm−1 in experimental conditions [61].

4. Conclusion

The degree of ordering for anionic, cationic, and zwitterionic surfactants is investigated for different packing densities on Ge substrate by using FTIR-ATR spectroscopy. New spectroscopic insights on the conformational order of sodium dodecyl sulfate (SDS), N-hexadecyl-N-N-dimethyl-3-ammonio-1-propane-sulfonate (HPS), hexadecyl-trimethylammonium bromide (CTAB), and dodecyl trimethylammonium bromide (DTAB) are presented. The effect of headgroup hydration and counterion was studied using DFT and semiempirical calculations. The CH2 and features for the SDS molecule show a shift of 1.7 and 0.9 cm−1 to higher frequencies as the packing density increases, while it is observed a shift of 2.6 and 2.7 cm−1 for the HPS molecule, respectively. The DTAB molecule shows a shifted of 4.5 cm−1 to lower frequencies for both CH2 and modes when the packing density increases. It is important to mention that these results are just a qualitative view of the organization of the alkyl chain conformation for each surfactant.

Acknowledgments

The authors are grateful to the FAPESP (grant no. 04/08227-5) and CNPq (grant no. 480631/2008-5) Brazilian agencies for the financial support. R. B. Viana also acknowledges CNPq for the fellowship. A. S. Pimentel thanks FAPERJ for an Award in the Jovem Cientista do Nosso Estado program (E-26/101.452/2010). A. S. Pimentel also thanks CNPq for the research support (grant no. 304187/2009-7 and 481481/2010-9). The authors are indebted to Professor Marcel Tabak (IQSC/USP) for the experimental and material support.