ISSN 09655441, Petroleum Chemistry, 2015, Vol. 55, No. 4, pp. 292–300. © Pleiades Publishing, Ltd., 2015. Original Russian Text © A.V. Straško, A.B. Shipovskaya, T.I. Gubina, O.N. Malinkina, A.G. Melnikov, 2015, published in Membrany i Membrannye Tekhnologii, 2015, Vol. 5, No. 1, pp. 39–47.
Usage of Cellulose Acetate Membranes for the Sorption–Luminescence Determination of Pyrene in Aqueous Media A. V. Straškoa, A. B. Shipovskayaa, b, T. I. Gubinaa, O. N. Malinkinaa, b, and A. G. Melnikova a
Saratov State Technical University, ul. Politekhnicheskaya 77, Saratov, 410054 Russia Institute of Chemistry, Saratov State University, Astrakhanskaya ul. 83, Saratov, 410012 Russia email:
[email protected]
b
Received August 20, 2014
Abstract—The applicability of cellulose acetate membranes (CAMs) as a solid matrix for the luminescence determination of pyrene in aqueous micellar solutions is shown. The effect of the concentrations of various surfactants, namely, anionic sodium dodecyl sulfate (SDS), cationic cetyltrimethylammonium bromide (CTAB), and nonionic polyoxyethylene (10) mono4isooctyl phenyl ether (TX100), on the fluorescence of pyrene in aqueous micellar solutions before and after sorption preconcentration and in an adsorbed state on a CAM has been studied. It has been found that the fluorescence intensity of pyrene on the solidphase matrix increases as a result of pyrene solubilization in surfactant hemimiceles formed on the sorbent surface. The highest degree of pyrene extraction on CAMs has been achieved in the presence of cationic CTAB micelles. The CAM has a negative surface potential (–31.5 ± 2.5 mV), which affects the hydrocarbon recovery. The degree of extractlion and the polarity index of a microenvironment of pyrene molecules in solutions decrease in the order CTAB → SDS → TX100. Keywords: cellulose acetate membrane, solidphase matrix, sorption preconcentration, luminescence analy sis, pyrene, aqueous micellar solutions DOI: 10.1134/S096554411504009X
Cellulose acetates belong to a class of largescale artificial polymers obtained from a renewable source of raw materials. It is likely that they are inferior to only cellulose and starch in the set of valuable proper ties and largescale applications in different branches of industry [1, 2]. For a period of many decades, cellu lose acetates have been traditionally used for the pro duction of different types of membrane elements (hol low fibers, porous film membranes, anisotropic multi layer membranes, etc.) for the reverse osmosis and ultra, hyper, and nanofiltration of natural water, wastewater, and biological fluids [3–6]. The advan tages of cellulose acetate membranes (CAMs) are high permeability, adequate stability in many media, in par ticular, in water, and a sufficient service life from sev eral months to several years. An important special fea ture of CAMs is the presence of a layer of bound water on their surface, which is responsible for surface charge localization.
cles from natural water [11], the separation of starch from lowmolecularweight sugars [12], the fraction ation of protein fractions [13], and the controlled release of insulin [14] has been studied. New approaches to improve the selectivity of CAMs in the processes of reverse osmosis, nanofiltration, etc., have been developed [15–17]. Studies on the production of film and hollowfiber membranes from the mixtures of cellulose diacetate with other polymers, in particular, with chitosan for the sorption extraction copper ions [18], with polyvinylpyrrolidone for pervaporation techniques [19], with cellulose triacetate for the dem ineralization of water [20], and with carboxymethyl cellulose for the isolation of albumin [21], and the characterization of their separating properties are of interest. The development of adsorption membranes from nanofibers prepared by electrospinning has been carried out [22, 23].
The CAMs are of considerable current interest. In the past decade, scientific research has been directed toward studying the mechanism of the transmembrane transfer of substances through CAMs [7, 8]. The use of CAMs for the production of highpurity solutions [9], the recovery of noble metals from process water in the jewelry industry [10], the removal of terrigenous parti
Polycyclic aromatic hydrocarbons (PAHs) are well known toxic environmental pollutants with carcino genic and mutagenic activities. Their concentrations in drinking water, wastewater, and aquatic environ ments should be monitored because of the ability of different PAHs to accumulate in environmental mate rials [24]. Pyrene is a readily available compound from
292
USAGE OF CELLULOSE ACETATE MEMBRANE
this class, and it is less toxic than other PAHs. There fore, it is a convenient test material for various studies. Recent studies showed that solidphase lumines cence (SPL) is the most promising method for the determination of PAHs and, in particular, pyrene in aqueous media [25–33]. It provides an opportunity to combine the sorption preconcentration of a substance on a solidphase matrix with its subsequent lumines cence determination directly in the sorbent phase without a stage of PAH desorption by an organic sol vent. This leads to a notable increase in the sensitivity and selectivity of determination. In the SPL method, phosphors are immobilized onto different solidphase matrices: aluminum oxide, quartz glass [27], zeolites [28], fiberglass [29], polyurethane foams [30], nylon membranes [31], silver [32] or gold [33] nanoparticles, etc. Filter paper (a cellulose matrix) is most com monly used [25, 26]. However, the efficiency of the preconcentration of hydrophobic PAHs on filter paper is low because of the hydrophilicity of this sorbent. Furthermore, at a long residence time in an aqueous medium, the sorbent is destroyed with the formation of an aqueous suspension with the dispersed phase of fibrous stuff. It is also well known that cellulose filters can be used for the preconcentration of PAHs sorbed on air borne particles in atmospheric air [34]. With consideration for the high efficiency of a cel lulose matrix in the sampling and analytical SPL determination of PAHs, the use of cellulose deriva tives, in particular, cellulose diacetate, for these pur poses is promising and important. Data on the simul taneous use of cellulose diacetate membranes for the sorption and solidphase preconcentration of pyrene from aqueous media and as matrices for luminescence signals are absent from the literature. In this work, we attempted to use CAMs for com bining the following two functions: as sorbents for the preconcentration of pyrene and as matrices for its determination by the SPL method. For this purpose, we prepared CAMs, evaluated their surface and phys icochemical properties, and studied the applicability of the membranes to luminescence analysis. Because the solubilization of a phosphor in surfactant micelles is a factor responsible for an improvement in the ana lytical characteristics of the luminescence determina tion of substances, in particular, pyrene [25, 26, 35], we used the following organized media (the selfasso ciated ensembles of diphilic surfactant molecules [35]) in the experiments: the aqueous micellar surfactant + pyrene solutions, which contained anionic, cationic, and nonionic surfactants. EXPERIMENTAL A flocculent sample of commercial cellulose diac etate with a viscosityaverage molecular weight of 77 kDa, an acetylation degree of 55%, and a moisture content of 3%, which is used for the production of tex PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
293
tile acetate yarns, was chosen for the preparation of CAMs. The solutions of 2 wt % cellulose diacetate in a mixture of analytical grade acetone + water (95 + 5), which is traditionally used in the manufacture of ace tate fibers, films, and membranes, were utilized. The CAMs were formed under standard conditions by a dry method. The solution was applied to plate glass, which was preliminarily degreased with ethanol and acetone, using a glass round spinneret. The complete ness of solvent removal was controlled based on weight changes. The thicknesses of CAMs in dry and wet states were 45 ± 5 and 55 ± 5 µm, respectively. Red ribbon filter paper according to TU (Technical Specifications) 03110 (EKROS) was used for com parison. The filter paper had the following character istics: thickness in a dry or wet state, 120 ± 10 or 150 ± 15 µm; density, 0.4 g/cm3; and porosity, 0.8. Pyrene of Purum grade (Fluka) with a purity of 96% was used. Aqueous micellar solutions with the pyrene concentration CP = 0.5 mM were used in the experiments. Anionic sodium dodecyl sulfate (SDS) (LenReaktiv), cationic cetyltrimethylammonium bromide (CTAB) (Sigma), and nonionic Triton X100 (TX100) (Sigma) were used for the preparation of the aqueous micellar solutions. The purity of all of the sur factants was 98–99%. The initial aqueous solutions of the surfactants and pyrene were prepared by a gravimetric method. A weighed portion of a surfactant was quantitatively transferred into a volumetric flask and dissolved in twicedistilled water, and the solution was brought to the mark (solution no. 1). An aliquot portion of solu tion no. 1 was placed in a volumetric flask, in which a weighed portion of pyrene was quantitatively dis solved, and the solution was brought to the mark with twicedistilled water (solution no. 2). The complete ness of pyrene dissolution was controlled based on the absorption spectrum. The surfactant + pyrene working solutions with the varied concentration of a surfactant (Csurfactant) and CP = const were prepared using a volu metric method by taking the aliquot portions of stock solution nos. 1 and 2 followed by their mixing and dilution. The stock solutions (nos. 1 and 2) were stored in the dark place to avoid photochemical degradation. The working solutions were prepared directly before the experiment. The critical micelle concentrations (CMC1/CMC2) of the surfactants used in this work in water at 25°С are the following: 8.0/50 mM for SDS [36], 0.9/21 mM for CTAB [37, 38], and 0.2/1.4 mM for TX100 [39]. The surfactant concentration in the aqueous surfactant + pyrene solutions was varied in a range of 0.01 CMC1–5 CMC2. The surface morphology of CAMs was evaluated by scanning electron microscopy (SEM) on a MIRA\\LMU scanning electron microscope (Tescan, Czech Republic) at a voltage of 8 kV and a conducting current of 60 pA. A layer of gold with a thickness of 5 nm was sprayed onto the samples using a K450X Carbon Coater (Germany): the spraying current was
294
STRAŠKO et al.
Table 1. Physicomechanical and surface energy characteristics of the solidphase matrices Sample CAM Filter paper CAM Filter paper
State Dry Wet
Breaking stress σ, MPa
Ultimate elongation ε, %
Surface potential ξ, mV
9.80 ± 2.00 19.3 ± 2.00 3.70 ± 1.20 0.17 ± 0.02
9.10 ± 5.00 1.30 ± 0.30 24.5 ± 7.00 0.15 ± 0.05
–31.5 ± 2.5 –10.0 ± 2.0 – –
20 mA, and the spraying time was 1 min. The surface potentials (ξ, mV) were measured on a Surface Poten tial Sensor (Nima KSV, Finland). Physicomechanical properties were studied on a TiraTest 28005 tensile testing machine of uniaxial ten sion (Germany) with a load cell of 100 N and a loading speed of 5 mm/min. The rupture stress (σ, MPa) and ultimate elongation (ε, %) were determined taking into account the crosssectional area and the initial length of the test sample. Before the use, the CAMs were repeatedly washed with twicedistilled water and dried in a desiccator over calcium chloride at room temperature. The filter paper was used without additional purification. A CAM or filter paper sample was placed in a specially designed 10mL sorption column, which consisted of a disposable sterile syringe (Master UNI PharmLine Limited, the United Kingdom) with a sorbent holder and a receiving reservoir. Sorption was conducted in the dynamic mode. For this purpose, a 10mL sample of the test solution of a surfactant + pyrene was passed five times through the column with a CAM (filter paper). Thereafter, the sample was removed from the column and dried in the same manner as the initial CAM, and the fluorescence spectra of pyrene were measured directly in the solidphase matrix. For eval uating the efficiency of sorption and the degree of pyrene extraction, the fluorescence spectra of the solutions before and after the sorption column were measured. Fluorescence analysis was performed on an LS 55 PerkinElmer luminescence spectrometer (the United States). A xenon lamp operating in the pulsed mode at a frequency of 50 Hz served as a radiation source. Monk–Gillieson monochromators were used. The measurements were carried out in a spectral range of 350–450 nm with a wavelength setting accuracy of 1 nm. The excitation wavelength was 320 nm, and the rate of scanning was 100 nm/min. For pyrene fluores cence measurements in solution and on a matrix, stan dard 1cm quartz cells and solid sample holders, respectively, were used. The index of polarity (I1/I3)—a quantitative mea sure of the local polarity of a medium in the microen vironment of a luminescent probe—was evaluated from the ratio of the maximum values of the fluores cence intensities (Ifl) of the first (I1) and third (I3) vibronic bands of the spectra of pyrene in solution and in a sorbed state [26].
The relative fluorescence intensity of pyrene (Ifl rel) in solution was determined as I5/1000, where I5 is a maximum value of Ifl of the fifth vibronic band of the spectrum. The values of Ifl rel of pyrene in a sorbed state on a CAM and filter paper were determined as I5/I5 (CMC2) and I5/I5 (CMC1), respectively, where I5(CMC2) and I5(CMC1) are the values of I5 for pyrene on the solidphase matrix after the dynamic absorption of the aqueous micellar solutions of a sur factant + pyrene with Csurfactant = CMC2 and CMC1, respectively. These surfactant concentrations were chosen because the intensity of pyrene fluorescence on the sorbent reached a maximum under the given experimental conditions. The recovery (R, %) of pyrene was determined from the relationship
R=
I 5' − I 5'' × 100 %, I 5'
where I 5' and I 5'' are the values of I5 for the stock solu tion and after sorption preconcentration on a CAM film (filter paper), respectively. RESULTS AND DISCUSSION Before studying the sorption preconcentration and luminescence determination of pyrene on a CAM, we evaluated its morphology and physicomechanical and surfaceenergy characteristics. According to SEM data, the structure of the membrane is a continuous lacy polymer network with a pore size of ~100– 500 nm (Fig. 1a). The rupture stress of the CAM in a dry state is somewhat lower than that of filter paper in the same state (Table 1). However, the ultimate elon gation of dry CAMs significantly exceeds an analogous characteristic of dry filter paper. In a wet state, the σ and ε of CAMs are considerably higher than those of filter paper. This is important with the use of such membranes as solidphase matrices for conducting a mass transfer process in aqueous media. Both of the sorbents are characterized by the negative values of the surface potential ξ (Table 1). The absolute value of ξ for the CAM is higher than that for filter paper. For optimizing conditions for the luminescence determination of pyrene in the aqueous micellar solu tions of SDS, CTAB, and TX100, at the first stage, we studied the dependences of the fluorescence intensity and polarity index of pyrene on surfactant concentra PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
USAGE OF CELLULOSE ACETATE MEMBRANE
295
(а)
2 μm
(b)
2 μm
(c)
2 μm
(d)
2 μm
2 μm
(e)
Fig. 1. SEM images of CAM surfaces: (a) the initial surface and (b–e) surfaces after the sorption preconcentration of pyrene sol ubilized in the micelles of (b, c) CTAB, (d) SDS, and (e) TX100 at Csurfactant = (b, d, e) CMC2 and (c) 2 CMC2. PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
296
STRAŠKO et al. а
Ifl 1000
5
800
4
600
3 2
400 200
1 0 375
400
425 λ, nm
400
425 λ, nm
400
425 λ, nm
b Ifl 500 5
400
4
300
2 200
3
100
1
0 375 c Ifl 600 5 450 4 300 3
150
2 1 0 375
Fig. 2. Fluorescence spectra of pyrene (a) in the initial aqueous solutions of SDS + pyrene, (b) after the sorption column, and (c) in a sorbed state on CAMs with the use of SDS + pyrene solutions with CSDS = (1) 0.08, (2) 0.8, (3) 8, (4) 40, and (5) 50 mM.
tion in the initial surfactant + pyrene aqueous solu tions and on the CAM after sorption preconcentra tion. Figure 2a shows the typical fluorescence spectra of the initial surfactant + pyrene aqueous solutions for the SDS + pyrene system used as an example. In the spectrum I fl = f ( λ) in a wavelength range of 370– 400 nm, there are five main vibronic bands (I1 ~375 nm, I2 ~380 nm, I3 ~385 nm, I4 ~390 nm, and I5 ~395 nm), of which the following three are wellresolved: I1, I3, and I5. Such spectra are typical of the solutions of PAHs, in particular, pyrene [25, 26]. It was established that the vibronic structure of the fluorescence spec trum of pyrene is sensitive to changes in the polarity of analyzed solutions. The fluorescence intensity of pyrene in aqueous solution increases with SDS con centration to reach a maximum value at CSDS = CMC2. The most considerable increase in Ifl was observed in the solutions with CSDS in the CMC1– CMC2 concentration range. It is well known that spherical micelles, which are capable of solubilizing lowpolarity PAH molecules, are formed in the aqueous solutions of SDS at a con centration of surfactant molecules higher than CMC1 by no more than an order of magnitude. The efficiency of the transfer of PAH molecules from an aqueous macrophase into a micellar micropseudophase is characterized by a distribution coefficient (Kd), which is 1.7 × 106 for pyrene [40]. This high value of Kd sug gests that more than 99% of pyrene molecules are bound to SDS micelles. Note that the micellar microphase (surfactant micelles) is referred to as a pseudophase because it does not have a stable inter face: micelles occur in dynamic equilibrium and exchange surfactant molecules with the environment (the rate of exchange is 10–5–10–8 s). It was found that the polarity index of pyrene in the aqueous micellar solutions of SDS depends on the concentration of this surfactant (Fig. 3, curve 2). At CSDS < CMC1, the value of I1/I3 is 1.58. This is consis tent with published data [26]. At CSDS = CMC1, the polarity index of pyrene decreases to I1/I3 = 1.10. At higher SDS concentrations (CSDS = 5 CMC1–CMC2), the values of I1/I3 for pyrene reach constant values at 1.03 ± 0.01. The function I 1 I 3 = f ( CSDS ) suggests the solubilization of pyrene in the hydrophobic part of a micelle. In this case, the role of SDS micelles con sists not only in the concentration of a fluorophore and the approach of system components but also in an increase in the hardness of fluorescent centers (a decrease in the vibrational excitation energy loss), the removal of water molecules from the nearest environ ment of pyrene molecules, the screening of pyrene from foreign fluorescence quenchers, and a decrease in the probability of nonradiative transitions. Natu rally, all of the above factors facilitate an increase in the fluorescence intensity of pyrene in the aqueous micel PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
USAGE OF CELLULOSE ACETATE MEMBRANE
lar solutions of SDS as the concentration of this sur factant is increased (Fig. 2a). The polarity index of the microenvironment of pyrene is ~1.02, as determined from the fluorescence spectra of a CAM dried after sorption preconcentra tion, and this value does not depend on the concentra tion of the SDS solution from which this PAH was sorbed (Fig. 3, curve 2'). Note that I1/I3 = 1.15 when the process was performed under the same conditions but with the use of filter paper as a solidphase matrix. This fact suggests that the CAM is a more hydrophobic matrix than filter paper, and it facilitates an increase in the intensity of a fluorescence signal; therefore, the use of the CAM as a solidphase matrix is more rea sonable. The spectra Ifl = f(λ) for the CTAB + pyrene and TX100 + pyrene systems are analogous to the func tions I1/I3 = f(Csurfactant). In both cases, an increase in the intensity of fluorescence and a decrease in the polarity index of pyrene in solution were observed as the surfactant concentration was increased (Fig. 3, curves 1 and 3). As in the SDS + pyrene system, the values of I1/I3 for pyrene in aqueous solution intensely decreased in the CTAB + pyrene and TX100 + pyrene systems until a surfactant concentration equal to CMC1. The values of I1/I3 determined from the flu orescence spectra of CAMs after the sorption precon centration of the hydrocarbon from the micellar solu tions of CTAB and TX100 were ~1.1 and ~1.02, respectively (curves 1' and 3'). Thus, the experimental results indicate that surfac tant micelles are formed in the analyzed multicompo nent solutions, and these micelles solubilize hydro phobic pyrene molecules in their internal nonpolar hydrocarbon regions. Moreover, in the case of the sol ubilization of pyrene in CTAB micelles, the polarity of the microenvironment of its molecules is somewhat higher than that in the micelles of SDS and TX100. This may be due to the fact that CTAB micelles are more permeable to the molecules of water than SDS and TX100 micelles. It was found that the decrease in the values of I1/I3 observed in the fluorescence spectra of pyrene in the aqueous micellar surfactant solutions can be described by the following order: CTAB > SDS > TX100 (see Fig. 3, curves 1–3). The efficiency of flu orescence increases with surfactant concentration. A maximum fourfold increase in fluorescence was observed in the presence of the micelles of anionic SDS. At the following stage, we evaluated the influence of surfactant concentration on the luminescence determination of pyrene. Figures 2b and 2c show the fluorescence spectra of pyrene in the aqueous solu tions of surfactants after the sorption column and in a sorbed state on the CAM with the use of the SDS + pyrene system as an example. For the convenient PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
297
I1/I3(1, 2) I1/I3(3) 1.8 1.6 1.6 1.4
1.4 I1/I3(1'–3') 1.3 2 3 1.1 1' 2', 3'
1.2
1
1.2 1.0 1.0
0.8 0.6 10 0
20 2
30 4
0.9 60 CSDS и СCTAB, mМ 40
50
6 9 CTX–100, mМ
Fig. 3. Dependence of the polarity index of pyrene on the concentrations of (1, 1') CTAB, (2, 2') SDS, and (3, 3') TX100 in aqueous micellar solutions from which solid phase sorption was performed in the luminescence deter mination of the test substance (1–3) in solution and (1 '–3') in a sorbed state on CAMs.
comparison of Ifl in the test systems, Table 2 summa rizes the relative fluorescence intensities of pyrene (Ifl rel) in the aqueous solutions of the surfactants before and after the sorption column and on the solid phase matrices after sorption preconcentration. The value of Ifl rel for the surfactant + pyrene stock solutions depends on the surfactant concentration in the system to increase with Csurfactant (Table 2). This sit uation persisted in a surfactant concentration range to CMC2–2 CMC2. As an example, data for the CTAB + pyrene system at CCTAB = 2 CMC2 are given. At higher surfactant concentrations, the value of Ifl rel decreased. An example is given for a solution of TX100 + pyrene at CTX100 = 5 CMC2. After the sorption preconcentration from the aque ous micellar solutions of the surfactants, the relative fluorescence intensity of pyrene considerably decreased in all of the solutions (Fig. 2b, Table 2). This fact is indicative of a high sorption capacity of CAMs for surfactant micelles with solubilized pyrene. This may be caused by the solubilization of pyrene in sur factant hemimicelles, which are formed on the sorbent surface. Figures 1b–1e show the visualization of aggregates formed on the CAM surface after the sorp tion preconcentration of the aqueous micellar solu tions of pyrene. Next, we studied the fluorescence of pyrene on the CAM after the solidphase preconcentration of the PAH in a micellar microphase (Fig. 2c, Table 2). As in the surfactant + pyrene solutions, the relative fluores
298
STRAŠKO et al.
Table 2. Effect of surfactant concentration on the fluorescence intensity and the recovery of pyrene
Surfac tant
CTAB
Surfactant concentration in the stock solution Csurfactant, mM
TX100
Critical micelle concentration
Pyrene recovery R, %
solution CAM
CAM
filter paper
0.05
0.02
37
–
0.18
0.09
0.18
50
27
CMC2
0.25
0.07
1.00
72
21
2 CMC2
0.41
0.02
0.90
95
20
0.08
0.01 CMC1
0.26
0.19
0.06
27
7
0.8
0.1 CMC1
0.56
0.29
0.15
48
30
8
CMC1
0.62
0.26
0.30
58
50
40
5 CMC1
0.95
0.38
0.72
60
4
50
CMC2
0.99
0.42
1.00
57
–
0.1 CMC1
0.21
0.18
0.23
14
–
0.2
CMC1
0.26
0.19
0.32
27
–
1.4
CMC2
0.39
0.20
1.00
50
–
7.5
5 CMC2
0.31
0.21
0.5
32
–
initial
after the column
0.1 CMC1
0.08
CMC1
21 42
0.09 0.9
SDS
Relative fluorescence intensity of pyrene at I5 Ifl rel, arb. units
0.02
cence intensity of pyrene in the sorbent phase increased as Csurfactant was increased in the solution from which the sorption preconcentration of the hydrocarbon was performed. A maximum intensity of the signal of pyrene on the CAM was observed at sur factant concentrations close to CMC2. For example, the maximum values of Ifl rel for the CTAB + pyrene, SDS + pyrene, and TX100 + pyrene systems occurred at CCTAB = CMC2–2 CMC2, CSDS = 5 CMC1–CMC2, and CTX100 = CMC2, respectively. As Csurfactant was further increased, the value of Ifl rel for pyrene in the CAM phase decreased. With the use of filter paper, the greatest relative intensity of pyrene flu orescence was observed with the use of micellar solu tions with Csurfactant = CMC1. The concentration and the nature of a surfactant in the test systems significantly affect the degree of pyrene extraction from the solution (Table 2). Under the specified experimental conditions, the greatest value of R for pyrene upon its sorption preconcentra tion from aqueous micellar solutions on CAMs or fil ter paper was reached at Csurfactant = CMC1–2 CMC2 or
only CMC1, respectively. In this case, the PAH recov ery in a filter paper phase was always lower than that in a CAM phase. The greatest recovery of pyrene was observed upon its sorption preconcentration on CAMs from the micellar solutions of CTAB. This is due to the fact that CTAB micelles with the solubilized molecules of pyrene have a localized positive charge at the micellar micropseudophase–aqueous phase interface, that is, at the micelle–solution interface. The positively charged CTAB micelles are better sorbed on a nega tively charged membrane surface, as compared with the negatively charged micelles of SDS and the neutral micelles of TX100. Comparatively high values of R (>50%) upon the sorption preconcentration on CAMs from the aqueous micellar solutions of negatively charged SDS with CSDS = CMC1–CMC2 are probably caused by a charge exchange of the polymeric matrix surface because of the adsorption of the surfaceactive anion. Amelina et al. [41] described an analogous change in the sign of a surface potential from negative to positive on passing PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
USAGE OF CELLULOSE ACETATE MEMBRANE
through an isoelectric state for cellulose fibers (initial and covered with a cellulose diacetate film) in the solutions of low and highmolecularweight surfac tants. With the use of nonionic TX100, a 50% recovery of pyrene was observed only at CTX100 = CMC2. In special experiments, we found that an increase in R to 75–80% can be reached with the use of TX100 + pyrene solutions with CP = 0.2 mM. CONCLUSIONS The experimental results show that CAMs exhibit high sorption capacity for the micelles of SDS, CTAB, and TX100 with solubilized pyrene. We found that the fluorescence of pyrene increased by a factor of 2–4 in aqueous micellar solutions in the concentration range Csurfactant = 0.01 CMC1–2 CMC2. The use of the orga nized media decreased the index of polarity of pyrene in the working solutions; this was explained by a decrease in the probability of nonradiative transitions because of a decrease in the polarity of the nearest microenvironment of the molecules of this PAH. The sorption preconcentration of pyrene from the aqueous micellar solutions on CAMs was accompanied by the formation of surfactant hemimicelles with solubilized pyrene on the sorbent surface. This led to an increase in the hardness of fluorescent centers and a notable increase in PAH fluorescence intensity on the solid phase matrix. We found that a maximum signal of pyrene fluorescence on CAMs was observed at CMC2 surfactant concentrations in solutions. We demon strated that the solidphase sorption of pyrene on the membrane can be used for the quantitative determina tion of the PAH by the SPL method. The greatest recovery of pyrene was achieved with the use of the micellar solutions of cationic CTAB. Under the spec ified experimental conditions, the degree of pyrene extraction from the solutions decreased in the order CTAB → SDS → TX100. Thus, the dynamic sorption of pyrene molecules solubilized in surfactant micelles on the surface of a solidphase (CAM) matrix can be used for the rapid and direct determination of the trace amount of this and other PAHs in aqueous media. In this case, the use of CAMs is very promising because the sorbent is readily available and inexpensive, and it almost does not swell in water, whereas the sorbed pyrene intensely fluoresces on a film. This makes it possible to use CAMs for the combination of two functions: the extraction of pyrene from aqueous media and its determinations by sorption–luminescence analysis. These matrices can be used in the development of sen sor systems for the ecological monitoring of aquatic environments and in pharmacological and toxicologi cal studies. PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015
299
ACKNOWLEDGMENTS This work was performed within the framework of state contract no. 4.1299.2014/K of the Ministry of Education and Science of the Russian Federation. REFERENCES 1. K. Kamide, Cellulose and Cellulose Derivatives (Elsevier, Amsterdam, 2005). 2. Scientific Principles of Carbohydrate Chemical Technol ogy, Ed. by A. G. Zakharov (LKI, Moscow, 2008), [in Russian]. 3. A. Schaefer, A. Fane, and T. Waite, Nanofiltration: Prin ciples and Applications (Elsevier, Amsterdam, 2004). 4. R. Zakaria, Hydrogen Separation Using Asymmetric Cel lulose Acetate Hollow Fiber Membranes (University of Waterloo, Canada, 2006). 5. V. V. Parashchuk and A. V. Volkov, Membrany, No. 1 (37), 25 (2008). 6. R. W. Baker, Membrane Technology and Applications (Wiley, Chichester, 2012), 3rd Ed. 7. M. Ulbricht, Polymer 47, 2217 (2006). 8. I. V. Vorotyntsev, I. I. Grinval’d, I. Yu. Kalagaev, et al., Membr. Membr. Tekhnol. 3, 227 (2013). 9. A. I. Bon, G. L. Bon, L. V. Melekhina, et al., RUF Patent No. 2303481; Byull. Izobret., No. 7 (2007). 10. S. L. Zakharov, A. V. Efremov, and Yu. A. Pavlov, Nauch. Vestn. Mosk. Gos. Gorn. Univ., No. 2 (23), 39 (2012). 11. A. G. Pervov, Yu. V. Kozlova, A. P. Andrianov, and N. B. Motovilova, Membrany, No. 1 (29), 20 (2006). 12. H. Rusli, S. Gandasasmita, and M. B. Amran, Iran. Polym. J. 22, 335 (2013). 13. M. G. Pozdeeva, T. O. Ryabukhova, and N. A. Oki sheva, Sorb. Khromatograf. Protsessy 13, 462 (2013). 14. L. I. Valuev, T. A. Valueva, I. L. Valuev, and N. A. Plate, Usp. Biol. Khim. 43, 307 (2003). 15. J. Su, Q. Yang, J. F. Teo, and T.S. Chung, J. Membr. Sci. 355, 36 (2010). 16. S. Zhang, K. Yu. Wang, T.S. Chung, et al., J. Membr. Sci. 360, 522 (2010). 17. Y. He, G.M. Li, H. Wang, et al., J. Taiwan Inst. Chem. En. 40, 289 (2009). 18. C. Liu and R. Bai, J. Membr. Sci. 284, 313 (2006). 19. H. Wu, X. Fang, X. Zhang, et al., Sep. Purif. Technol. 64, 183 (2008). 20. T. P. N. Nguyen, E.T. Yun, I.C. Kim, and Y.N. Kwon, J. Membr. Sci. 433, 49 (2013). 21. B. Han, D. Zhang, Z. Shao, et al., Desalination 311, 80 (2013). 22. L. Zhang, T. J. Menkhaus, and H. Fong, J. Membr. Sci. 319, 176 (2008). 23. R. Konwarh, N. Karak, and M. Misra, Biotechnol. Adv. 31, 421 (2013). 24. P. PlazaBolanos, A. G. Frenicha, and J. L. M. Vidal, J. Chromatogr., A 1217, 6303 (2010). 25. G. V. Mel’nikov, T. I. Gubina, and O. A. Dyachuk, Russ. J. Phys. Chem. 80, 1160 (2006).
300
STRAŠKO et al.
26. O. A. Dyachuk, T. I. Gubina, and G. V. Melnikov, J. Anal. Chem. 64, 3 (2009). 27. T. Saitoh, H. Itoh, and M. Hiraide, Talanta 79, 177 (2009). 28. W. B. Wilson, A. A. Costa, H. Wang, et al., Microchem. J. 110, 246 (2013). 29. G. Bernier and M. Lamotte, Rapid Chemical and Bio logical Techniques for Water Monitoring, Ed. by P. Quevauviller and R. Greenwood (Wiley, 2009), p. 275. 30. S. G. Dmitrienko, E. Y. Gurariy, R. E. Nosov, and Y. A. Zolotov, Anal. Lett. 34, 425 (2001). 31. V. Vasquez, M. E. Baez, M. Bravo, and E. Fuentes, Anal. Bioanal. Chem. 405 (23), 7497 (2013). 32. G. I. Romanovskaya, A. Yu. Olenin, and S. Yu. Vasil’eva, Russ. J. Phys. Chem. 85, 274 (2011). 33. H. Wang and A. D. Campiglia, Talanta 83, 233 (2010).
34. M. Ochsenkuhn–Petropoulou, K. Staikos, and G. Matuschek, J. Anal. Appl. Pyrol. 70, 73 (2003). 35. I. I. Parashchenko, T. D. Smirnova, S. N. Shtykov, et al., J. Anal. Chem. 68, 112 (2013). 36. A. P. Romani, A. E. H. Machado, N. Hioka, et al., J. Fluoresc. 19, 327 (2009). 37. R. Guo, X. J. Zhu, and X. Guo, Colloid Polym. Sci. 281, 876 (2003). 38. T. Liu and J. Wu, Colloid J. 70, 311 (2008). 39. D. Yu, F. Huang, and H. Xu, Anal. Methods 4, 47 (2012). 40. L. V. Levshin, S. N. Shtykov, I. Yu. Goryacheva, and G. V. Mel’nikov, Zh. Prikl. Spectrosk. 66, 201 (1999). 41. E. A. Amelina, I. V. Videnskii, N. I. Ivanova, et al., Vestn. Mosk. Univ., Ser. 2: Khim. 42, 49 (2001).
Translated by V. Makhlyarchuk
PETROLEUM CHEMISTRY
Vol. 55
No. 4
2015