CC BY
14
UDC 541.64
S. G. Karpova, A. A. Ol'khov, A. L. Iordanskii,
S. M. Lomakin, N. S. Shilkina, A. A. Popov, G. E. Zaikov
STRUCTURAL-DYNAMIC ANALYSIS OF FIBROUS MATRIXES BASED ON POLY(3-HYDROXYBUTYRATE)-CHITOSAN COMPOSITES
Keywords: structural-dynamic analysis, fibrous matrixes, poly(3-hydroxybutyrate, chitosan, blend composites.
With the use of scanning electron microscopy, differential scanning calorimetry, and electron paramagnetic resonance, the structural-dynamic analysis of ultrathin fibrous matrixes based on poly(3-hydroxybutyrate) and blend composites of this polymer with chitosan is performed. It is shown that the addition of a small amount of chitosan causes change in the morphologies of the matrixes and leads to a marked increase in their melting enthalpies. It is found that the studied fibers contain amorphous regions with various morphologies. The dynamics of the spin probe TEMPO in these regions is investigated, and its change under the influence of increased temperature, an aqueous medium, and ozone is examined. The mechanism controlling the effects of chitosan, temperature, and an oxidative aggressive medium on the structuring of fibers is advanced.
Ключевые слова: структурно-динамический анализ, волокнистые матрицы, поли(З-гидроксибутират), хитозан, смесевые
композиции.
Методами сканирующей электронной микроскопии, дифференциальной сканирующей калориметрии и электронного парамагнитного резонанса проведен структурно-динамический анализ ультратонких волокнистых матриц на основе поли (3-гидроксибутирата) и смесевых композитов данного полимера с хитозаном. Показано, что добавление небольшого количества хитозана вызывает изменение морфологии матриц и приводит к заметному увеличению их энтальпий плавления. Установлено, что исследованные волокна содержат аморфные области с различными морфологиями. Изучена динамика спинового зонда TEMPO в этих областях и ее изменение под воздействием повышенной температуры, водной среды и озона. Предложен механизм влияния хитозана, температуры и окислительной агрессивной среды на структурирование волокон.
Introduction
Biomedical and general engineering materials based on ultrathin and nanosized fibers derived from biopolymers and biocomposites are currently the subject of many studies [1-6].
Compared to traditional methods of preparing polymer fibers, the electrospinning method makes it possible to produce fibers with thicknesses less than a micrometer because solution or melt droplets formed at the end of a capillary are spun in the field of mechanical and electrostatic forces applied to the polymer solution or melt [7-9]. Variation in characteristics of the spinning solution (conductivity, viscosity), polymer properties (molecular mass and molecular-mass distribution), and manufacturing parameters of electrospinning (voltage on the electrode, distance between electrodes, and productivity) provides a way to affect the morphology, surface behavior, porosity, and geometry of fibers [10].
The physicochemical, dynamic, and transport characteristics of film and fibrous biopolymer materials based on the bacterial polyester poly(3-hydroxybutyrate) (PHB) that were designed for prolonged and targeted delivery of drugs to the human body were studied in [11-14]. Owing to high biocom-patibility, accelerated biodegradation, and satisfactory physicomechanical characteristics of films and fibers of PHB, this polymer may be regarded as one of the most promising medical polymers. In addition, PHB has found wide use in the creation of bone implants, junctions of nerve fibers, matrixes during the growth of cells and tissues, filters, and ultrafiltration membranes as well as in the design of artificial heart elements and the replacement of vessels [11, 15].
In order to program kinetics of the diffusion transportof drugs, combined (composite or blend) materials based on PHB and other biopolymers are currently used. Owing to their heterogeneity, these systems provide multipattern dynamics for the transport of drugs because of a change in the polymer structure as a result of crystallization, plasticization, and swelling as well as because of degradation of one of the components [16, 17].
Of particular importance are blends based on PHB and chitosan, which are amphiphilic and fully bioresorbable composite materials with improved sorption capacities and abilities to implement various kinetic profiles for the prolonged release of drugs in a wide time range (from several weeks to months). These blends were investigated with the use of PHB film matrixes containing 10-0% chitosan [18, 19].
This study concerns the effect of a small amount of chitosan on the structural organization of nonwoven blend composites based on PHB and has the aim to reveal specific features of diffusion and degradation processes occurring in heterogeneous systems. Microprobe EPR spectroscopy studies showed that water and an oxidizer (ozone) affect the structure of the fibrous material PHB-chitosan, an outcome that is important for forecasting the kinetics of drug release complicated by degradation of one of the components.
Research Objects and Analytical Methods
In this study, fibers were prepared from the natural biodegradable polymer polyhydroxybutyrate of the 16F series, which was synthesized via a microbiological method (BIOMERR, Germany). The initial polymer was a white powder with particle sizes of 5-20 ^m. The viscosity-average molecular mass of PHB was Mn =
2.06 x 105, density was d = 1.248 g/cm3, the melting temperature was Tm = 177°C, and the degree of crystallinity was ~65%. PHB ultrathin fibers were obtained via electrospinning [20] from spinning solutions: PHB and PHB-chitosan in chloroform. The concentration of PHB in a solution was 7 wt %. Russian-sourced chitosan (Bioprogress, Russia) was used as a finely dispersed powder. The molecular mass of chitosan was Mw = 4.4 x 105, and its degree of deacetylation was 82.3%.
The concentrations of chitosan were 0.05, 0.1, 0.2, 0.3, and 0.5 wt %. Chitosan-containing solutions were prepared with the use of an ultrasound bath. The sizes of chitosan particles were 50-150 nm. Particles were prepared via the mechanical mixing of an acidic chitosan solution in the mechanical-shift field (a stirrer with a rotor revolution of 1000-4000 rpm) followed by addition of an alkaline agent under continued stirring. The aqueous solution was then dried. The resulting chitosan particles were added to the spinning solution of PHB in chloroform and dispersed for a long time with the help of the ultrasound bath and microwave radiation. The sizes of chitosan particles were measured with the aid of an LA-960 analyzer (HORIBA Scientific). The formation of ultrathin fibers based on PHB was described in more detail in [21].
X-Band EPR spectra were measured on an EPR-V automated EPR spectrometer (Semenov Institute of Chemical Physics, Russian Academy of Sciences). To avoid saturation effects, the microwave power in the resonator was adjusted no greater than 7 mW. During spectral recording, the modulation amplitude was always much smaller than the width of the resonance line and did not exceed 0.5 G.
The stable nitroxide radical 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) was used as a probe. The radical was introduced into fibers from vapor at a temperature of 50°C. The concentration of the radical in the polymer did not exceed 10-3 mol/L.
The experimental spin-probe spectra in the low motion range (t > 10-10 s) were analyzed within the Brown-ian isotropic rotation model with the help of the program described in [22]. The spectra were simulated with the use of the main values of the g tensor and the tensor of hyperfine interaction of the radical: gxx = 2.0096, gyy = 2.0066, gzz = 2.0025, Axx = 7.0 H, Ayy = 5.0 H, and Azz = 35.0 G. Note that the value of Azz was determined experimentally from the EPR spectra of the nitroxide radical in the polymer at 77 K; it was almost the same as that reported in [23].
Correlation times of probe rotation, t, in the fastmotion region (t < 10-10 s) were determined from the EPR spectra via the following formula [24]:
t = AH+[(I+/I-)0'5 - 1] 6.65 x 10-10, (1)
where AH+ is the width of the spectrum component located in the weak field and I+/I- is the ratio of component intensities in the weak field and strong field, respectively. The error in measuring t was ±5.
Samples were annealed at 140°C and treated with an aqueous medium at 70°C at various heating temperatures.
The ozone oxidation of the samples was performed in an ozone-oxygen mixture with a partial concentration
of ozone of 5 x 10-5 mol/L. The degree of oxidation was estimated via IR spectroscopy on a Bruker IFS 48 spectrometer (United States) from the change in the absorption band at 1650-1700 cm-1, due to the carbonyl group.
The DSC studies of the samples were performed on a DSC 204 F1 instrument (Netzsch) in an inert atmosphere of argon at a heating rate of 10 K/min. The mean-statistic error during measurements of heat effects was approximately ±3%. The enthalpy of melting was calculated with the aid of the program NETZSCH Proteus-thermal analysis 4.8.4 via the standard technique (ASTM E793) [25]. The peaks were resolved with the use of the program NETZSCH Peak Separation 2006.01. Calculations were conducted via the Gauss-Newton combined method, where the Marquardt method was combined with optimization of the iteration-step length [26].
The structures of fibrous materials were studied via electron microscopy on a Hitachi TM-1000 scanning electron microscope (Japan) at an accelerating voltage of 20 kV. The surfaces of the nonwoven fibrous material were sprayed with the gold layer with a thickness of 100-200 A.
Results and Discussion
Effect of Small Amounts of Chitosan on the Geometrical Characteristics of PHB Fibers
During electrospinning, the parameters of the polymer solution influence both the morphology of the ul-trathin fiber and its diameter [27]. Variation in the average diameter of fibers makes it possible to design nonwoven materials with various packing densities of fibers, optimum mechanical characteristics, and a controlled state of the surface. In the case of medical application of fibrillar materials, specifically in the creation of matrixes for cell engineering, a decrease in the fiber diameter with a simultaneous increase in the packing density causes a positive dynamics of cell growth [2830].
As was shown in [29], the diameter of the PHB ul-trathin fiber may be changed through variation in the physicochemical characteristics of the solution [29]. In this study, a 7% PHB spinning solution in chloroform was selected as an optimum variant for electrospinning. This choice is primarily related to a good reproducibility of preparing fibers and their optimum diameters.
Figure 1 presents the micrographs of fibrous materials.
As is seen in Fig. 1a, the average diameter of PHB fibers is in the range 2-5 ^m. During mixing of PHB and chitosan in the spinning solution, the geometry of the fibrous material changes.
Along the whole length of individual fibers, specific thickenings and numerous junctions and splittings appear (Figs. 1b, 1c). The average diameter of an individual fiber obtained from the PHB-chitosan blend, as that in the case of the fibers based on the initial PHB, is in the range 2-5 ^m, and the small axes of ellipsoid thickenings are 20-50 ^m. As is well seen in Fig. 1, the number of ellipsoid structures increases with an increase in the concentration of chitosan.
The geometry changes in the fibers are unambiguously defined by the dispersed chitosan particles with sizes of 50-150 nm, which, because of incompatibility
with PHB molecules, play the role of fillers influencing the viscosity and conductivity of the spinning solution and, hence, the process of electrospinning. Two fundamentally important phenomena govern the structure of the cured fiber. The first is desorption of the solvent (chloroform) from the fiber volume and its subsequent surface evaporation. The second is the crystallization of the polymer in the fiber being cured. The issue of diffusion as the main mechanism of solvent desorption will be the subject of our further studies. Here, an attempt is made to answer the questions of how and to what extent the introduction of chitosan into the system changes the pattern of crystallization of the main component of the fiber: PHB.
Fig. 1 - Micrographs of fibrous materials: (a) PHB, (b) PHB-0.1% chitosan, and (c) PHB-0.2% chitosan
Thermophysical Characteristics of the Crystalline Phase of PHB-Chitosan Fibers
The contribution of the crystalline phase to the process of structuring in nonwoven blend composites was investigated via DSC. Thermograms of the PHB fibers were measured in the region of PHB melting at low concentrations of the minor component, chitosan: 0.05 to 0.5 wt %. A number of experiments with samples of various compositions showed that, in the first temperature scan, all these samples feature thermograms of similar shapes and have a single maximum that corresponds to the melting temperature of PHB. However, the positions of this maximum on the temperature scale are shifted relative to the melting temperature of the initial PHB (177°C). For example, after the addition of chi-tosan, even at the lowest concentration (0.05%), the melting temperature of PHB.
However, the positions of this maximum on the temperature scale are shifted relative to the melting temperature of the initial PHB (177°C). For example, after the addition of chitosan, even at the lowest concentration (0.05%), the melting temperature of PHB decreases to ~171°C, and after a further increase in the amount of this component, it remains practically constant: 172°C at 0.1% chitosan, 173°C at 0.3% chitosan, and 172°C at 0.5% chitosan. The exception is provided by the blend composite PHB-chitosan containing 0.2%
chitosan, for which an unexpectedly high melting temperature was observed: 179°C, which is close to the melting temperature of the initial PHB. The finding that the melting temperatures of the composites at low concentrations of chitosan are constant indicates the absence of compatibility of polymer components; that is, intermolecular interactions in the fiber are extremely weak. At the same time, with allowance for the overall reduction in the melting temperature of PHB by approximately 5°C, it may be supposed that even low concentrations of chitosan cause a change in the crystalline organization of the key fiber-forming polymer, PHB.
The general pattern of thermograms obtained for the same samples subjected to repeated heating in the cell of the DSC spectrometer is substantially different from the first-heating thermograms. After the repeated heating, individual maxima due to melting transform into bimodal maxima, as shown in Fig. 2, which presents thermograms for the melting of PHB fiber containing 0.3% chitosan. An exception is provided by the initial PHB sample and the above-mentioned "abnormal" fiber containing 0.2% chitosan. In fibers containing 0.2% chitosan, both first-heating and second-heating thermograms are practically the same in shape. Both thermograms exhibit welldefined single peaks with maxima at 179°C, which correspond to the melting temperature of PHB, but the peaks are somewhat different in terms of the specific enthalpies of melting, equal to 84.3 and 80.7 J/g for first and second temperature scans, respectively.
173°
161°
80 100 120 140 160 180 200
T, °c
Fig. 2 - Thermograms of PHB-0.3% chitosan: (1) first melting and (2) second melting
In addition, high-temperature melting peaks of the crystalline polymer shift by 2.4-12°C to lower temperatures during the second scanning. This result indicates that the initial ordered structure of PHB crystals in the fiber is considerably distorted as a result of their melting and subsequent cooling of the sample to room temperature in accordance with the procedure of DSC thermogram measurements. A change in thermophysical characteristics during the repeated temperature scan is most probably related to the fact that the samples underwent melting during the first heating to ~185°C. It appears that, during subsequent cooling, the structure of the fiber does not recover and does not attain its initial orientation and anisotropy. This phenomenon is accompanied by a certain reduction in crystallinity (a decrease in the specific enthalpy of melting) and worsening of the organization of the crystalline phase of the fiber (a total reduction in the melting temperature by 9-12°C).
The specific enthalpy of melting of the crystalline phase, AH, which was calculated via integration of the first-scan thermograms in the region of the phase transition of PHB, likewise changes after addition of chitosan to PHB. For example, if, for the fiber prepared from PHB, AH = 67.3 J/g, then, after addition of 0.05% chitosan, this value increases by 7% (to 72.0 J/g). As is clear in Fig. 3, the introduction of chitosan into the fiber in the low-concentration range entails a noticeable gain in the specific enthalpy of melting of PHB, with its maximum being observed at a chitosan content of 0.2%. After a further increase in the concentration of chitosan in the system, AH decelerates its growth and even decreases somewhat.
100
80
to
ЕЦ < 60.
- cuu
x 5
и
ж
100
80
i60
0 0.1 0.2 0.3 0.4 Chitosan concentration, wt %
0.5
Fig. 3 - Comparative characteristics of PHB-chitosan fibers at various concentrations of chitosan: (1) total concentration of the radical in the fiber, (2) the specific melting heat of PHB, and (3) the amount of dense regions in the intercrystallite space of PHB
The repeated temperature scanning for all samples leads to a marked overall decrease in AH in the vicinity of the high-temperature peak; however, the overall enthalpy, as calculated from the areas under both peaks, insignificantly differs from that of the initial fiber. The bimodal distribution of the crystalline phase of PHB in the fiber was previously observed for PHB-chitosan blend films. This circumstance was explained by appearance of two types of PHB crystallites: with a closer to perfect, well-organized packing of polymer chains and less organized, more defective crystals.
During the interpretation of the DSC data, it may be assumed that the gain in the specific enthalpy of melting of nonwoven blend composites as a result of the addition of a small amount of chitosan is due to the fact that this component facilitates enhancement of the nuclea-tion process and causes an increase in the amount of nuclei of heterogeneous crystallization of PHB. An increase in the concentration of these nuclei entails formation of a large amount of small crystallites that, because of smaller steric hindrances during the process of their growth, give rise to the crystalline phase with the minimum amount of defects relative to that of the crystalline phase of the PHB fiber lacking chitosan. As follows from the DSC studies, the difference in the degrees of crystallinity in the most favorable situation, namely,
at a concentration of chitosan nanoparticles in the fiber of 0.2%, may be as high as 25%.
Hence, an analysis of change in the melting temperature and in the value of AH makes it possible to state that the degree of crystallinity of PHB in the blend composites is higher than that of the initial PHB throughout the studied range of chitosan concentrations, from 0.05 to 0.5%. At the same time, the dependence of AH on the fiber composition shows a maximum at a chitosan concentration of 0.2%, an outcome that is apparently associated with the formation of a closer to perfect structure. A further decrease in the enthalpy of melting, reflecting the decline in the degree of crystallinity of PHB, becomes less pronounced after a further gain in the amount of chitosan in the blend, and, as was shown in [20], the crystallinity of PHB may be as high as 30% at a chitosan concentration of 80%. The gain in the degree of crystallinity that is observed in the initial range of chitosan concentrations is accompanied by the formation of crystallites with a high fraction of defects, as reflected by a reduction in the melting temperature (to 171-173°C) relative to that of the fiber based on the individual PHB (177°C).
Dynamics and Structure of the Amorphous Phase of PHB and PHB-Chitosan Fibers
It is well known that PHB belongs to the class of high-crystallinity polymers. Therefore, it is advisable to assume that the structural organization of the intercrystallite space in the studied nonwoven blend composite is immediately associated with organization and type of crystalline structure because of the presence of both tie chains and conformation hindrances arising during displacement of molecular coils in a narrow intercrystallite space. As a result, after the addition of small amounts of chitosan to PHB, an increase in the degree of crystallinity and change in the sizes of crystals, that is, a reduction in the intercrystallite space in the unit fibril, should lead to change in the structural-dynamic states of the noncrystallizable regions.
It is most convenient to investigate the structure and segmental dynamics of these regions via the microprobe EPR method with the use of stable radicals.
The EPR spectra of the TEMPO radical in PHB have a complicated pattern and represent the superposition of two spectra corresponding to two populations of radicals with different correlation times, ii and t2, where t1 characterizes the molecular mobility in less dense regions and t2 characterizes the molecular mobility in denser intercrystallite regions (Fig. 4).
The ratio of concentrations for regions with different densities was estimated via the mathematical treatment of the spectra through the special program NLSL, as was described in [25]. Calculations showed that the fraction of dense regions in the intercrystallite space of PHB, a2, is much higher than the fraction of less dense regions. As is clear from Fig 3, this value increases with the concentration of chitosan and attains its maximum in PHB fibers containing 0.2% chitosan, that is, changes similarly to the specific melting enthalpy of PHB, which reflects its degree of crystallinity.
A change in the concentration of dense intercrystallite regions of PHB causes a change in the
mobility of the radical. This result is direct evidence that the mobility of PHB changes. With the use of the abovementioned program, correlation times were calculated for fast and slow components of radical rotation. For the slow component, the correlation time changes considerably; it initially decreases and then increases with the amount of chitosan in the fiber (Fig. 5). An increase in t2 suggests deceleration of the molecular mobility of the radical and, accordingly, of the segmental mobility of PHB chains at high concentrations of chitosan.
3280 3320 3360 ' 3400
Magnetic induction, G
3280 ' 3320 ' 3360 ' 3400 Magnetic induction, G
Fig. 4 - (a) (1) Experimental and (2) theoretical EPR spectra of the nitroxide radical TEMPO in PHB-0.2% chitosan and (b) spectrum resolution into (3) slow and (4) fast components
т X 1010, s 160
120 -
80 -
40 -■-1-1-1-1-1-
0 0.1 0.2 0.3 0.4 0.^
Chitosan concentration, wt %
Fig. 5 - Correlation time t vs. amount of chitosan in fibers of the blend composite based on PHB
Hence (Figs. 3, 5), in the range of chitosan concentrations from 0 to ~0.2-0.3% there is a correspondence between the amount of the crystalline phase of PHB (the structural characteristic), the state of intercrystalline regions (the value of a2), and correlation time t2 (the dynamic characteristic): namely, t2 changes in the antibate manner to AH and a2. This phenomenon may be explained with the use of two arguments. On the other hand, the increase in the degree of crystallinity is accompanied by a gain in the fraction of dense regions in the boundary crystal-intercrystallite region. For this
sample series, the lowest crystallinity was observed for PHB samples. Exactly in the case of this polymer, the highest molecular mobility and the lowest fraction of the dense amorphous phase, a2, were registered. (The exception is the sample containing 0.5% chitosan.)
In addition to dynamic measurements, the treatment and quantitative interpretation of EPR data with the aid of the Win-EPR and SimFonia Bruker software packages were performed. With the use of these programs, the equilibrium concentrations of the radical sorbed in fiber samples based on the PHB-chitosan composite were calculated for the samples having equal masses. (The calculation data are presented in Fig. 3.) The most pronounced gain in the radical concentration is observed at high amounts of chitosan. At a chitosan content of 0.5%, the concentration of the radical (11.5 x 1018 spin/g) is a factor of nearly 2 greater than its concentration in the initial PHB (6.1 x 1018 spin/g). An increase in the amount of the radical in the fibrillar system is in direct agreement with the onset of the decrease in the specific enthalpy of melting of PHB, that is, with the decline in its degree of crystallinity (Fig. 3) and with an increase in the volume fraction of intercrystallite regions. In what follows, the data on t calculated from the EPR spectra are presented. It is important that, in this case, 5 x 10-11 < t < 10-9, whereas t1 (the fast component) calculated via the program NLSL is in the range 5 x 10-11 < t < 10-10 and the slow component t2 is in the range 10-10 < t < 10-7.
Additional information about the dynamic behavior of the PHB-chitosan fibrillar systems of various compositions was obtained in the study of the temperature dependence of the rate of radical rotation and in the determination of corresponding activation energy Et. The characteristic feature of the dependence of Et on the content of chitosan in PHB is that the value of Et sharply increases on passage from the individual PHB to the blend composites (Fig. 6). Such a sharp increase in the activation energy of probe rotation after incorporation of chitosan is probably associated not only with an increase in the crystallinity of the system (the specific enthalpy of melting) but also with a change in the state of the intercrystallite polymer phase, in which the content of denser regions increases.
E, J/mol
60 r
20
° 0 0.1 0.2 0.3 0.4 0.5
Chitosan concentration, wt %
Fig. 6 - Activation energy of probe rotation, Et, in PHB fibers containing various amounts of chitosan
The finding that Et remains almost the same with variation in the blend composition suggests that a slight change in the concentration of the crystalline phase in the fiber has practically no effect on the mechanism of the rotational mobility of the radical. However, during comparison of probe mobilities in the individual PHB and in its blend with chitosan, it is advisable to mention once again that the introduction of chitosan into the fibrillar matrix of PHB affects its crystallization. In this case, the nucleation regime passes from homogeneous to heterogeneous and the appearance of a large amount of fine crystals in the composite PHB-chitosan has a strong influence on intercrystallite regions of the nonwoven blend composite, thereby creating additional steric hindrances to the rotation of radicals distributed in these regions.
Influence of Annealing on the Dynamic Behavior oof PHB-Chitosan Fibers
It is well known that the heat treatment of polymer products at moderate or increased temperatures in an interval that is much lower than the melting temperature (for PHB, ~177°C) releases a number of internal stresses and favors additional crystallization.
Therefore, the isothermal heating (annealing) of PHB-chitosan fibers of various compositions was performed at a temperature of 140°C over various times from 0 to 120 min. For each heat-treated sample, the concentration dependence of relaxation time was obtained (Fig. 7).
10
>o
о/
0,2 0.3 0.4
Chitosan concentration, wt %
0.5
Fig. 7 - Concentration dependence of effective correlation time t at heating times of (1) 0, (2) 0.5, (3)
1, and (4) 2 h. Broken curves refer to the polynomial approximation of the data
It can be seen that, as the concentration of chitosan is increased, correlation times t for all composites except the unannealed one attain a maximum in the concentration range of chitosan from 0.25 to 0.3% and then tend to decrease. The most distinct maximum makes itself evident for the samples heated for a short time (curve 2). At relatively long times of heating, the height of the maximum decreases and the difference between the values of t for the samples heated for more than an hour (curves 3, 4) becomes less distinct. The latter result is related to approach of the structure of PHB to the equilibrium state. A weak concentration dependence of tc also provides evidence that the effect of chitosan on the intercrystalline regions of PHB weakens. Here, the
role of intermolecular hydrogen bonds is practically negligible because of low concentrations of amine and hydroxyl groups belonging to chitosan. Moreover, an increase in the mobility of radicals in the system with an increase in the time of heating (curves 1-3) indicates that, in this case, disordering of intercrystalline regions should be expected either owing to a reduction in the total crystallinity of PHB or as a result of redistribution of regions with denser organization and less dense organization of macromolecules in the intercrystallite space. The DSC thermograms of the annealed samples (140°C, 1 h) show marked increases in the specific heat of melting of PHB from 72 to 96 J/g with an increase in the content of chitosan to 0.05%, from 84 to 105 J/g with an increase in the content of chitosan to 0.2%, and from 75 to 85 J/g at a chitosan content of 0.5%. These data make it possible to assume that the additional crystallization in the course of heating involves macromole-cules situated in denser regions of the noncrystalline phase of PHB. In this case, in the intercrystallite space, a relative increase in the concentration of the less ordered amorphous phase and the transition (displacement) of radicals from denser regions to less dense regions may be expected; as a result, the mobility of probe rotation increases and, accordingly, the value of t decreases (Fig. 7).
Influence of an Aqueous Medium on the Structuring and Dynamics of Fibers
In order to estimate the effect of an aqueous medium on ultrafine fibers of the PHB-chitosan blend composite, the EPR spectra were measured for the samples exposed in distilled water at an increased temperature of 70°C. Calculated correlation times t are presented in Fig. 8 as functions of the times of exposure at various component ratios in the blend.
Fig. 8 - Plots of (a) the correlation time of the molecular probe TEMPO and (b) the ratio of intensities of the EPR spectrum, I+/I-, as functions of the time of exposure of fibrillar samples in the aqueous medium.
The amounts of chitosan are (1) 0, (2) 0.1, (3) 0.2, (4) 0.3, and (5, 6) 0.5%
It is of interest to compare correlation times obtained for the samples subjected to the combined effect of water and temperature and for the samples subjected to only heating (see above and Fig. 7). Figure 8 shows that the effect of water on the dynamic behavior of the radical in the various blend composites leads to the opposite tendency: Namely, a very strong deceleration of its rotation is observed with an increase in the concentration of chitosan in the system.
As was shown for film samples [18], the mixing of chitosan with PHB leads to hydrophilization of the system, its sorption capacity abruptly increases because of a high affinity of water for molecules of this polysac-charide, and simultaneously the plasticizing effect of water on PHB becomes more distinct. In this case, the diffusion coefficients of water and lowmolecular-mass drugs increase exponentially. It appears that similar processes occur in the fibers based on the PHB-chitosan blend. The plasticizing effect of water, which is the most pronounced in the intercrystallite space of PHB, in combination with the elevated temperature, creates favorable conditions for the additional crystallization of the polymer and, hence, leads to a more ordered state of intercrystallite regions. After the removal of water, the high-crystallinity state of PHB is "frozen" and the rotation of radicals in denser regions decelerates; as a consequence, correlation times increase abruptly. The higher the content of chitosan in the system, the more noticeable the reduction in the rate of radical rotation, in good agreement with our dynamic and diffusion data for the film composites PHB-polyurethane [12] and PHB-chitosan [18].
Another feature of the dynamic behavior of radicals in nonwoven blend composites appears as maxima on the kinetic curves of correlation times that are observed to higher or lower extents for all samples. All maxima occur in the same time interval, approximately in the range 40-45 min, which is independent of the content of chitosan in the system. This effect may be rationalized by the fact that, at the initial stage of the water-temperature effect, the fraction of densely packed PHB molecules in intercrystallite regions increases, and subsequently, a portion of them may pass into crystals. The majority of radicals here, as in the case of annealing, are displaced from denser regions and move to less dense regions; as a result, the transition from the slow regime to the faster regime of radical rotation occurs. At a fixed temperature and moisture values of the sample, the rate of translational diffusion of the radical from one type of region to the other type is almost invariable. This result explains the constant position of the maximum on the time scale during the process of water-temperature exposure of fibers.
Thus, the temperature and the time of storage in an aqueous medium on the nonwoven blend composites, in accordance with the dynamic behavior of the radical, exert the strongest effect on the structure of intercrystallite regions of PHB. In this case, a change in t indicates an increase in the rigidity of macromolecules with an increase in the concentration of chitosan due to
a gain in the fraction of ordered chains in the intercrystalline space. As in [3], conformational criterion h/L, where h is the mean distance between chain ends and L is the contour chain length [31], was used as a criterion of polymer chain rigidity, while the ratio of intensities of EPR spectra, I+/I-, reflecting the ratio between denser and less dense regions in the intercrystalline space of PHB, was used as the dynamic criterion in the form presented in Fig. 8b. The kinetic curve demonstrates the constant rate of redistribution between dense and less dense regions of the crystallizable PHB, a circumstance that provides an additional argument to explain the constant position of the maxima on the kinetic curves in Fig. 8a.
Influence Ozone on the Dynamics of Radical Rotation in PHB-Chitosan Fibers
In general, during the application of biomedical materials, their structures and segmental mobilities are affected by ozone along with the mechanical and temperature effects. Note that there are two sources of ozone in the atmosphere. First, ozone is formed during the work of powerful electric devices maintaining the vital activity of patients both during surgical operations and during therapy or monitoring in the hospital. Second, in several particular cases, ozone is still used to sterilize medical equipment. Despite the everyday contact of this aggressive compound with polymers, its effect on their morphological and dynamic characteristics remains a poorly studied field of polymer materials science.
Figure 9 plots the dependences of the correlation time of probe rotation on the concentration of chitosan in the nonwoven blend composite at various times of storage in the ozone-oxygen mixture. In the studied time intervals, the effect of ozone is the strongest in the case of the blend composites based on PHB fibers containing 0.05 and 0.5% chitosan, while the effect of ozone on the initial PHB is the weakest. As the time of the effect of ozone on the polymer is increased, the correlation time increases appreciably for all the studied polymers. Let us set forth in short a feasible hypothesis explaining the above results.
тх Ю10, s
О
O O.I 0.2 0.3 0.4 0.5
Chitosan concentration, wt %
Fig. 9 - Influence of the time of ozonolysis on the dynamics of radical rotation in the PHB fibrillar samples containing various amounts of chitosan. The times of ozonolysis are (1) 0, (2) 0.3, (3) 0.7, (4) 1.5, (5) 4.0, and (6) 7.0 h
In accordance with the published data [32], during the action of this ozone, folds occurring on the end surfaces of crystallites degrade above all because these are the most strained fragments of polymer chains. During ozonation, the physicochemical structuring of macro-molecules and the formation of the system of hydrogen bonds occur in parallel with the chemical rupture of bonds in a chain. In our opinion, change in the values of t during ozonolysis (Fig. 9) is associated with the fact that the processes of physical crosslinking of macro-molecules prevail when the accumulation of oxygen-containing groups in side chains enhances intermolecular interaction between segments and, as a consequence, decelerates their molecular mobility, a phenomenon that is manifested as a gain in the correlation time of the probe. The presence of maxima on the concentration dependences of t and the increase in their heights with the time of ozonolysis may be explained once more by the gain in the concentration of highly ordered tie chains in the intercrystallite regions of PHB and the redistribution of radicals from denser regions to less dense regions. A further increase in the correlation time is associated with an increase in the concentration of ozonoly-sis products (the concentration of oxygen-containing groups) and, as a consequence, with an increase in the rigidity of polymer molecules as a result of the formation of a network of hydrogen bonds that hinders the free motion of PHB segments.
Conclusions
The mutual effect of crystalline and intercrystalline (conditionally amorphous) regions in biodegradable high-crystallinity polymers and their composites remains a complex and poorly studied issue of modern polymer materials science. In this paper, the structural-dynamic studies of the influence of crystallization on the segmental mobility of PHB molecules in the intercrystalline regions of a nonwoven blend composite have been performed with the use of EPR and DSC methods. After the introduction of small amounts of chitosan that are much smaller than 1 wt %, the crystalline organization of PHB changes considerably, and the dynamics of probe rotation in the intercrystalline regions changes as a response to a change in the crystalline phase. Investigations of the state of the polymer matrix and of the role of the minor component, chitosan, has made it possible to interpret for the first time at the molecular and supramolecular levels the effects of a number of aggressive factors (such as increased temperature, an aqueous medium, and ozone) on the structural-dynamic characteristics of the nonwoven blend composite PHB-chitosan.
Acknowledgment
This work was supported by the Russian foundation for Basic Research, project nos. 14-03-00405-a, 13-0212407 ofi_m2, and 5-59-32401 RT-omi.
References
1. C. Mota, D. Puppi, D. Dinucci, M. Gazzarri, F. Chiellini, J. Bioact. Compat. Polym. 28, 320 (2013).
2. P. Vashisth, K. Nikhil, S.C. Pemmaraju, P.A. Pruthi, V. Mallick, H. Singh, A. Patel, N.C. Mishra, R.P. Singh, V. Pruthi, J. Bioact. Compat. Polym. 28, 652 (2013).
3. S.G. Karpova, A.L. Iordanskii, M.V. Motyakin, A.A. Ol'khov, O.V. Staroverova, S.M. Lomakin, N.G. Shilkina, S.Z. Rogovina, A.A. Berlin, Polym. Sci., Ser. A 57 (2), 131 (2015).
4. Z. Wang, B. Sun, M. Zhang, L. Ou, Y. Che, J. Zhang, D. Kong, J. Bioact. Compat. Polym. 28, 154 (2013).
5. C.-H. Hung and W.W.-F. Leung, Sep. Purif. Technol. 79 (1), 34 (2011).
6. F.-G. Banica, Chemical Sensors. Biosensors, Fundamentals and Applications (Wiley, New York, 2012).
7. A. Kulkarni, V.A. Bambole, and P.A. Mahanwar, Polym. -Plast. Technol. Eng. 49 (5), 427 (2010).
8. T. Dvir, B.P. Timko, D.S. Kohane, and R. Langer, Nat. Nanotechnol. 6 (1), 13 (2011).
9. P. Raghavan, D.-H. Lim, J.-H. Ahn, Ch. Nah, D.C. Sherrington, H.-S. Ryu, H.-J. Ahn, React. Funct. Polym. 72, 915 (2012).
10. A. Baji, Y.-W. Mai, S.-C. Wong, M. Abtahi, P. Chen, Compos. Sci. Technol. 70, 703 (2010).
11. A.A. Ol'khov, V.S. Markin, R.Yu. Kosenko, M.A. Gol'dshtrakh, A.L. Iordanskii, Russ. J. Appl. Chem. 88 (2), 308 (2015).
12. S.G. Karpova, A.L. Iordanskii, S.N. Chvalun, M.A. Shcherbina, S.M. Lomakin, N.G. Shilkina, S.Z. Rogovina, V.S. Markin, A.A. Popov, A.A. Berlin, The Reports of Phys. Chem. 446 (2), 176 (2012).
13. A.P. Bonartsev, V.A. Livshits, T.A. Makhina, V.L. Myshkina, G.A. Bonartseva, A.L. Iordanskii, eXPRESS Polym. Lett. 1 (12), 797 (2007).
14. E.L. Ivantsova, R.Yu. Kosenko, A.L. Iordanskii, S.Z. Rogovina, E.V. Prut, A.G. Filatova, K.Z. Gumargalieva, A.A. Berlin, S.P. Novikova, Polym. Sci., Ser. A 54 (2), 87 (2012).
15. H. Ueda and Y. Tabata, Adv. Drug Delivery Rev. 55, 501 (2003).
16. A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, and A.A. Ol'khov, in Chemical and Biological Kinetics. New Horizons (Chemistry, Moscow, 2005), Vol. 1, p. 640 [in Rus.].
17. Yu.N. Pankova, A.N. Shchegolikhin, A.L. Iordanskii, A.L. Zhulkina, A.A. Ol'khov, G.E. Zaikov, J. Mol. Liq. 156 (1), 65 (2010).
18. E.L. Ivantsova, A.L. Iordanskii, R.Yu. Kosenko, S.Z. Rogovina, A.V. Grachev, E.V. Prut, The Reports of Russian Academy of Science (in Rus.), 431, 500 (2010).
19. E.L. Ivantsova, R.Yu. Kosenko, A.L. Iordanskii, S.Z. Rogovina, E.V. Prut, A.G. Filatova, K.Z. Gumargalieva, S.P. Novikova, Polym. Sci., Ser. A 54 (2), 87 (2012).
20. Yu.N. Filatov, Electrospinning of Fibrous Materials (ESF-processes), Ed. by V.N. Kirichenko (Oil and Gas Inst., Moscow, 1997) [in Rus.].
21. A.A. Olkhov, O.V. Staroverova, A.P. Bonartsev, I.I. Zharkova, E.D. Sklyanchuk, A.L. Iordanskii, S.Z. Rogovina, A.A. Berlin, A.A. Ishchenko, Polym. Sci., Ser. A 8 (2), 100 (2015).
22. D.E. Budil, S. Lee, S. Saxena, and J.H. Freed, J. Magn. Reson., Ser. A 120, 155 (1996).
23. V.P. Timofeev, A.Yu. Misharin, and Ya.V. Tkachev, Biophysics 56 (3), 407 (2011).
24. A.L. Buchachenko and A.M. Vasserman, Stable Radicals (Chemistry, Moscow, 1973) [in Rus.].
25. Z.D. Stephen, Handbook of Thermal Analysis and Calorimetry (Elsevier, Amsterdam; Boston; London, 2002).
26. J. Opfermann, Rechentech. Datenverarbeitung 23 (3), 26 (1985).
27. A.A. Olkhov, O.V. Staroverova, A.L. Iordanskii, and G.E. Zaikov, AIP Conf. Proc. 1599, 558 (2014).
28. O.V. Staroverova, A.M. Shushkevich, G.M. Kuz'micheva, A.A. Ol'khov, V.V. Voinova, A.A. Ol'khov, I.I. Zharkova, K.V. Shaitan, E.D. Sklyanchuk, V.V. Gur'ev, Tekhnol. Zhivykh Sist. 10 (8), 74 (2013).
29. G.C. Rutledge and S.V. Fridrikh, Adv. Drug Delivery Rev. 59 (14), 1384 (2007).
30. I.I. Zharkova, O.V. Staroverova, V.V. Voinova, N.V. Andreeva, A.M. Shushkevich, E.D. Sklyanchuk, G.M. Kuz'micheva, A.E. Bespalova, E.A. Akulina, K.V. Shaitan, A.A. Ol'khov, Biomed. Chemistry. 60 (5), 553 (2014).
31. A.N. Ozerin, Candidate's Dissertation in Chemistry (Research Physical Chemistry Institute. Karpov, Moscow, 1977).
32. A. Keller and M. Willmouth, Makromol. Chem. 141, 91 (1994).
© S. G. Karpova - Ph.D., Researcher of Physical Chemistry of Synthetic and Natural Polymers Compositions Laboratory, Emanuel Institute of Biochemical Physics, A.A. Ol'khov - Ph.D, Associate Professor, Senior Researcher of Perspective Composites and Technologies Laboratory, Plekhanov Russian University of Economics, Moscow, e-mail: aolkhov72@yandex.ru, A. L. Iordanskii - Doctor of Chemistry, Full Professor, Head of Diffusion Phenomena in Polymer Systems Laboratory, Semenov Institute of Chemical Physics, S. M. Lomakin - Ph.D., Head of Polymers Chemical Resistance Laboratory, Emanuel Institute of Biochemical Physics, N. S. Shilkina - Researcher of Emanuel Institute of Biochemical Physics, A. A. Popov - Doctor of Chemistry, Full Professor, Head of Physical Chemistry of Synthetic and Natural Polymers Compositions Laboratory, Emanuel Institute of Biochemical Physics, G. E. Zaikov -Doctor of Chemistry, Full Professor of Plastics Technologies Department, KNRTU.
© С. Г. Карпова - к.х.н., сотр. лаб. физико-химии композиций синтетических и природных полимеров Института биохимической физики им. Н.М. Эмануэля РАН; А. А. Ольхов - к.т.н, доц., ст. науч. сотр. лаборатории перспективных композиционные материалов и технологий Российского экономического университета им. Г.В. Плеханова, e-mail: aolkhov72@yandex.ru; А. Л. Иорданский - д.х.н., проф., зав. лаб. диффузионных явлений в полимерных системах Института химической физики им. Н.Н. Семенова РАН; С. М. Ломакин - к.х.н., зав. лаб. химической стойкости полимеров Института биохимической физики им. Н.М. Эмануэля РАН; Н. Г. Шилкина - сотр. Института биохимической физики им. Н.М. Эмануэля РАН; А. А. Попов - д.х.н., проф., зав. лаб. физико-химии композиций синтетических и природных полимеров Института биохимической физики им. Н.М. Эмануэля РАН; Г. Е. Заиков - д.х.н., проф. каф. технологии пластических масс КНИТУ.
