The Role of Surface Charge of Nucleation Agents on the

Article pubs.acs.org/JPCB

The Role of Surface Charge of Nucleation Agents on the Crystallization Behavior of Poly(vinylidene fluoride) Ying Wu and Shaw Ling Hsu* Polymer Science and Engineering and NSF Materials Research Science & Engineering Center, University of Massachusetts, Amherst, Massachusetts 01003, United States

Christian Honeker, David J. Bravet, and Darryl S. Williams† Saint Gobain High Performance Research Center, 9 Goddard Road, Northboro, Massachusetts 01532-1545, United States ABSTRACT: The effect of the surface charge of nucleation agents on the crystallization behavior of poly(vinylidene fluoride) (PVDF) has been investigated. Ion− dipole interaction between the positive surface of nucleation agents and the partially negative CF2 dipoles of PVDF is established as a main factor for further lowering the free energy barrier for nucleation, and thus increasing significantly the crystallization kinetics. This is in contrast to the behavior observed for nucleation agents possessing either negative surface or neutral charges. Positive nucleation agents led to a remarkable increase in the crystallization temperature of PVDF (lower supercooling) as compared with that of neat PVDF. The dispersion of each type of nucleation agent is also important. The melting temperatures of nucleation agents need to be higher than the melting temperature of PVDF. The melting point and degree of crystallinity of PVDF can also be raised by using specific nucleation agents. The detailed crystallization kinetics and conformational changes of the PVDF chain have been investigated. With the addition of positive nucleation agents, the γ and β chain conformations, instead of the α phase, dominate.

■

INTRODUCTION Nucleation agents are widely used to control and enhance the mechanical properties, degree of crystallinity, specific morphological features, dimensional stability, and optical transparency (reduction of crystallite size), and to increase the processing speed of the semicrystalline polymers.1−5 Generally speaking, the effectiveness of nucleation agents is dependent on increasing the surface area and lowering the nucleation barrier to enhance the heterogeneous nucleation process.6−9 Examples of success have been proven in virtually all semicrystalline polymers including poly(vinylidene fluoride) (PVDF), the subject of our special interest.10,11 Nucleation agents for PVDF include organic compounds,12−15 inorganic metal salts,16,17 nanoparticles,11,18 nanoclay,5,19−21 and nanotubes.22,23 In the case of PVDF, crystallization kinetics in some circumstances has indeed been increased; thus, sample stability has been enhanced. In addition, the melting temperature has been raised. It has also been suggested the change in melting temperature may be caused by a change in crystalline phase from the usual α phase to the higher melting β or γ phases.5,10,19 During the heterogeneous nucleation process, lattice matching between nucleation agents and polymer may play an important role,24−26 but not necessarily an essential one.27−29 In this study, we have focused on the possibility of enhancing the nucleation process by incorporating a specific affinity between PVDF and the charged surfaces of the nucleation agents used. We were drawn to this subject based © 2012 American Chemical Society

on previous studies on a number of polymers that possess strong, large, permanent dipoles. First, the crystallization process of the γ phase of PVDF can be significantly affected even in the presence of a minute electric field.30,31 Second, although not understood, various organic compounds have proven to be extremely effective nucleation agents in dipolerich poly(lactic acid) (PLA).32,33 Third, the role of dipole− dipole interaction has proven to be significant in dictating the crystallization process and crystalline stability in PLA.34,35 In all of these cases, the role of partial charges or interchain interactions involving dipoles is significant. Nucleation agents enhance secondary nucleation by providing a surface on which the melt crystallizes. The foreign particle effectively increases the nuclei beyond the critical size in order for the crystalline unit to grow spontaneously. The overall free energy associated with nucleation combines the favorable term arising from chain aggregation and the unfavorable term associated with the interface formation.6,36,37 In this study, we propose and verify the addition of another favorable term based on the specific affinity between the polymer and nucleation agents. We have selected a number of nucleation agents to verify our hypothesis. Effective nucleation agents must disperse homogeneously in the PVDF melt in order to provide a high surface to Received: May 5, 2012 Revised: May 28, 2012 Published: May 30, 2012 7379

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

temperature for about 2 h. They were then cast on the slides and dried in a vacuum oven at room temperature. The dried nucleation agent films were dipped into saturated solutions containing the fluorescent molecules for 10 min. The films were then washed thoroughly with 1,4-dioxane, a nonsolvent for the nucleation agent, but a solvent for the fluorescent molecules. Finally, the films with adsorbed fluorescent molecules were characterized by fluorescence optical microscopy (Olympus DP71). Differential Scanning Calorimetry (DSC). The differential scanning thermograms were measured by using a TA Instrument Model Q100 equipped with an RCS cooling system and a nitrogen gas purge with a flow rate of 50 mL/min. The instrument was calibrated with an indium standard (Tm = 156.6 °C). The experiments were conducted in the temperature range of 0 to 220 °C. For the nonisothermal crystallization study, the samples were heated to 220 °C at a rate of 10 deg/min under nitrogen atmosphere and held at 220 °C for 5 min in order to eliminate the previous thermal history. The samples were cooled to 0 °C at a rate of 10 deg/min in order to evaluate the crystallization temperature. Typically the samples obtained were then reheated to 220 °C at a heating rate of 10 deg/min in order to evaluate the crystallinity obtained. In this study, a value of 104.6 J/g was used as the heat of fusion of the perfectly crystalline α phase PVDF. For isothermal crystallization behaviors, the samples with the previous thermal history removed were then crystallized with an undercooling of 20 deg below the crystallization temperature (ΔT = 20 °C) to perform isothermal crystallization for 300 min. We adopted the procedure developed earlier for polypropylene47,48 in order to determine the efficiency of the nucleation agents studied. Optical Microscopy. The morphologies of PVDF samples were observed by using an Olympus Vanox optical microscope equipped with a Canon SD400 digital camera. The films on the slides were rapidly heated to 220 °C and kept for 5 min at this temperature to ensure the melting of all the crystals. The sample was then cooled and maintained at room temperature until the spherulitic growth ceased. Polarized optical micrographs were taken in order to estimate crystal size. Small-Angle X-ray Scattering (SAXS). Small-angle X-ray scattering measurements were performed on a Rigaku Geigerflex RAD-3A with Cu Kα radiation (1.54 Å). The detector was placed at approximately 1.47 m from the sample. The samples were placed in a vacuum chamber, perpendicular to the incident beam. Samples were sandwiched between Kapton films within a metal holder and were mounted on the Linkam heating stage. The samples were heated to 220 °C and kept for 10 min at this temperature to ensure complete melting of all the crystals. Then samples were maintained at this temperature for 1 h. Subsequently, the samples were cooled to 10 °C lower than the melting temperature of the nucleation agents (still higher than the melting point of PVDF) and kept at this temperature for 1 h. SAXS images were corrected for the intensity of the primary beam, time, and sample thickness. The scattering of the Kapton films was subtracted from the normalized intensity. The scattering angle of the SAXS pattern was calibrated with the silver behenate standard. Attenuated Total Reflectance Infrared (ATR-IR). PVDF films were characterized by using ATR-IR spectroscopy to determine the crystalline phases of PVDF and the effects on PVDF from the addition of nucleation agents. The ATR-IR measurements were obtained by using a PerkinElmer 100

volume ratio and, therefore, a large number of nucleation sites per unit of volume.23,38−40 Therefore, we have selected a number of nucleation agents with melting temperature near or above that of PVDF. If the nucleation agents dissolve into the PVDF melt, then on cooling they will precipitate first. The resulting nanoscale nuclei serve as nucleating sites for PVDF, due to its large surface area, leading to enhanced nucleation of small polymer crystals. The degree of dispersion of nucleation agents in the melt can be probed by using optical microscopy and small-angle X-ray scattering.41−44 In addition, we have selected a number of nucleation agents with different surface partial charges: positive, negative, or neutral. The surface charge of nucleation agents was characterized by using the fluorescent labeling method.45,46 The specific nucleation agents selected for this study include positive phosphonium, pyridinium, pyrrolidinium, ammonium, and sulfonium salts; negative sulfate and phosphate salts; and neutral flavanthone. We have characterized the effects of each on the PVDF crystallization and morphology. In addition, the amount of β and γ crystalline phases produced has been shown to be dependent on the nucleation agents used. It should be emphasized that all of our results reported in this study were obtained under identical experimental conditions with the same polymer. The results that we report truly represent the different effect of various nucleation agents on the crystallization behavior of PVDF under the same circumstances.

■

EXPERIMENTAL SECTION Materials. The PVDF used in this study, Kynar 740, was supplied by the Saint Gobain Co., Ltd. Three types of nucleation agents have been studied (Scheme 1). First, positive nucleation agents (NAps), tetrabutylphosphonium hexafluorophosphate (NAp-1), ethyltriphenylphosphonium bromide (NAp-2), n-heptyltriphenylphosphonium bromide (NAp-3), N-acetonylpyridinium bromide (NAp-4), 1-butyl-1-methylpyrrolidinium bromide (NAp-5), tetrabutylammonium hydrogen sulfate (NAp-6), and triphenylsulfonium tetrafluoroborate (NAp-7), were purchased from Sigma and Alfa, respectively, and were used without further purification. Second, negative nucleation agents (NAns), sodium lauryl sulfate (NAn-1), sodium n-tridecyl sulfate (NAn-2), 1-naphthyl phosphate monosodium salt monohydrate (NAn-3), as well as neutral nucleation agent flavanthone (neutral) were all purchased from Alfa and used as received. Fluorescent molecules Naphthol Yellow S and Basic Blue 3 were purchased from Sigma and were used without further purification. Preparation of Films. The PVDF films with different concentrations of nucleation agents were obtained from the mixture of their separate acetone solutions. PVDF and nucleation agent solutions were separately prepared by dissolving them in acetone and stirring the mixture at 60 °C for 1 day. The mixtures were stirred for several hours at room temperature then dried in a vacuum oven for at least 1 day at room temperature to remove the residual acetone. To evaluate the dispersion of nucleation agents, a drop of solution was placed on a glass coverslip. These films were then heated above melting and studied by using different thermal profiles. It should be emphasized that all of our samples were prepared and characterized under identical experimental conditions with the same polymer. Characterization of Nucleation Agent Surface Charge (Fluorescent Labeling Method). Nucleation agents were first dissolved in acetone by stirring the mixture at room 7380

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

Scheme 1. The Chemical Structures and Melting Temperatures of the Nucleation Agents Used

have a positive surface charge and (2) they are all phosphonium salts with different melting temperatures above that of the α phase of PVDF. We chose NAp-4, 5, -6, and -7 to change the chemical structure of the nucleation agent while maintaining the positive surface charge. Their melting temperatures are also above that of the α phase of PVDF. NAn-1 and -2 (negative surface charge) are sulfate salts with different melting temperatures above that of the α phase of PVDF. NAn-3 has a different counterion. Flavanthone is a high-melting nucleation agent with a neutral surface charge. To verify the surface charges of the various nucleation agents, we used a fluorescent labeling method. Figure 1 shows

FT-IR spectrometer. All spectra consist of 32 coadded scans each collected at 4 cm−1 resolution.

■

RESULTS AND DISCUSSION

Characterization of Surface Charge of Nucleating Agents. To show the effect of the surface charge of the nucleation agents on the crystallization kinetics and crystalline structure of PVDF, we chose nucleation agents with different surface charges (positive, negative, and neutral) and different melting temperatures. The chemical structures and melting temperatures of these candidates are shown in Scheme 1. We chose NAp-1, -2, and -3 for the following reasons: (1) they may 7381

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

Figure 1. Fluorescence optical micrographs of representative samples: (A) pure NAp-3; (B) NAp-3 after dipping into a cationic fluorescent molecule solution; (C) NAp-3 after dipping into an anionic fluorescent molecule solution; (D) pure NAn-1; (E) NAn-1 after dipping into a cationic fluorescent molecule solution; (F) NAn-1 after dipping into an anionic fluorescent molecule solution; (G) pure neutral; (H) neutral after dipping into a cationic fluorescent molecule solution; (I) neutral after dipping into an anionic fluorescent molecule solution.

fluorescence micrographs of pure nucleation agents and nucleation agents with adsorbed fluorescent molecules. Pure NAp-3 has a yellow color in the fluorescent micrograph of Figure 1A. In order to differentiate the surface charge of these nucleation agents, we chose fluorescent molecules that emit a different color from the NAp-3. Naphthol Yellow S and Basic Blue 3 are negative and positive fluorescent molecules, respectively. After dipping the slide with the NAp-3 crystallized on it into the anionic fluorescent molecule Naphthol Yellow S solution and washing thoroughly, it can be seen clearly in Figure 1C that the color changed from yellow to green. In contrast, after dipping the slide that has NAp-3 crystallized on it into the cationic fluorescent molecule Basic Blue 3 solution, after washing, its color continues to be yellow, as shown in Figure 1B. In this way we show that Naphthol Yellow S, an anionic fluorescent molecule, is adsorbed onto the surface of NAp-3, but Basic Blue 3, a cationic fluorescent molecule, is not. This indicates that the surface of NAp-3 is positively charged. Similarly, we show that the nucleation agents are positive (Figure 1A−C), negative (Figure 1D−F), or neutral (Figure 1G−I). Dispersion of the Nucleation Agents. The efficiency of nucleation agents depends on their dispersibility in the polymer melt. To elucidate the dispersion of nucleation agents in the PVDF, we conducted optical microscopy at 220 °C between cross polarizers. Here only unmelted nucleation agent particles contribute to the birefringence. The isotropic PVDF melt is seen as the dark regions. The micrographs obtained are shown in Figure 2. It is clear that unmelted particles of NAp-4 are seen in the PVDF-NAp-4 blend (Figure 2B), while good dispersion was achieved for other positive or negative nucleation agents (Figure 2A,C,D,E). For PVDF-neutral blends, some aggregates

were observed (Figure 2F). In order to confirm the dispersion of nucleating agents at a finer scale, small-angle X-ray scattering has also been employed. This technique has been used in the past to characterize the dispersion of nucleating agents and their effects on the polymer crystallization process.41−44 The scattered intensities observed for the various nucleating agents dispersed in PVDF as a function of the scattering vector are shown in Figure 3. It is clear when the nucleating agents are well dispersed, the overall scattered intensities match the one observed for neat PVDF (Figure 3A,B). The only exceptions are PVDF-NAp-4 and PVDF-neutral, which we have already established to be poorly dispersed. In addition, we found excess scattering for NAp-1 in PVDF (Figure 3D). Our X-ray data show that all nucleating agents, with noted exceptions, are homogeneously dispersed confirming the optical microscopic measurements (Figure 2A,C,D,E). The scattering intensity changes dramatically when the sample temperature is lowered to be below the crystallization temperature of the nucleating agent but still above the crystallization temperature of PVDF. This is shown in Figure 3C. The increase in the scattering can be ascribed to the formation of a suspension of randomly oriented nucleation agent aggregations.42 As an example, the SAXS patterns of PVDF-NAp-5 at 220 and 176 °C, respectively, are shown in Figure 3C. It is clear that the scattered intensity increased when the sample is cooled to 176 °C. Likewise, the scattered intensity also increased for the PVDF-NAn-1 cooled to 190 °C (Figure 3C, curves c and d), below the crystallization temperature of the nucleating agent but above the PVDF crystallization temperature. This scattering data confirms that for the nucleating agents of interest (NAp-3, -5, and -6), their homogeneous 7382

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

Figure 2. Optical micrographs of (A) PVDF-NAp-3, (B) PVDF-NAp-4, (C) PVDF-NAp-5, (D) PVDF-NAp-6, (E) PVDF-NAn-1, and (F) PVDFneutral. All micrographs were taken at 220 °C after annealing for 5 min; the concentration of the nucleation agents was 2% on the weight of PVDF.

the crystallization of PVDF significantly, represented by a higher Tc. In contrast, the nonisothermal crystallization temperature of PVDF with negative nucleation agents is also shown in Table 1. The Tc in this case decreases slightly (∼2 deg) with the addition of negative nucleation agents. Although these negative nucleation agents disperse well (Figures 2E and 3B), and thus, due to the effect of a large surface area, are expected to enhance crystallization, they in fact, suppress PVDF crystallization. The neutral nucleation agents increase Tc by 2 deg. The crystallization kinetics are similar to the negatively charged nucleation agents. As mentioned above, the dispersion of nucleation agents is an extremely important consideration. The chain configuration and nucleation agent concentration, specifically flavanthrone, can also alter the crystallization kinetics.49 The efficiency of nucleation agents can also be evaluated by using an expression developed previously.47,48 The nucleation efficiency is calculated as:

dispersion at the molecule level in PVDF melt has been established. Characterization of Crystallization Behavior. Several criteria can be used to evaluate the efficiency of various nucleation agents in influencing the crystallization behavior of PVDF. These include the following: (1) the onset of crystallization temperature during cooling from the melt; (2) characteristic half-time needed for crystallization to occur during isothermal crystallization; and (3) spherulite size measured as a function of time. Perhaps the determination of the crystallization temperature is the simplest, and of tremendous value in processing and various applications as shown in Figure 4 and summarized in Table 1. As shown in Figure 4A, the crystallization temperature (Tc) of the nonisothermal crystallization process shifts to higher temperatures for PVDF with positive nucleation agents except for NAp-1 and -4 as compared to neat PVDF (139 °C). It can be seen from Table 1 that the maximum shift in Tc is ∼8 deg as compared to PVDF. Other positive nucleation agents exhibit smaller shifts or none (NAp-1 and -4). The unchanged Tc for NAp-1 and -4 is attributed to aggregating of the NAp-4 (Figure 2B and Figure 3A,D) and a relatively poor dispersion of NAp-1 (Figure 3A,D) because of the high melting temperatures of NAp-1 as compared to PVDF; the melting point of NAp-1 is 212 °C as shown in Scheme 1. This indicates the addition of these positive nucleation agents lowers the free energy barrier of nucleation thus accelerates

efficiency (%) =100(ΔTc/ΔTc max) =100(Tc nucl − Tc1)/(Tc2 − Tc1)

where Tc nucl, Tc1, and Tc2 are crystallization temperatures of the nucleated, non-nucleated, and self-nucleated polymer, respectively. These values are defined because of the changing mechanism of crystallization as the sample temperature is lowered. This equation 7383

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

Figure 3. Small-angle X-ray scattering intensities of PVDF and PVDF with different nucleation agents: (A) PVDF with the positive nucleation agents at 220 °C; (B) PVDF with negative and neutral nucleation agents at 220 °C; (C) (a) PVDF-NAp-5 at 220 °C, (b) PVDF-NAp-5 at 176 °C, (c) PVDF-NAn-1 at 220 °C, and (d) PVDF-NAn-1 at 190 °C; and (D) PVDF, PVDF-NAp-1, and PVDF-NAp-4 (enhanced Y axis) at 220 °C. The concentration of the nucleation agents was 2% on the weight of PVDF.

Figure 4. DSC (A) nonisothermal crystallization curves and (B) subsequent melting curves of PVDF and PVDF with different positive nucleation agents.

of DSC, the crystallization kinetics obtained from integrated areas of the exothermic peaks are shown in Figure 5. The enthalpic changes for isothermal crystallization (ΔHc) represent the degree of crystallinity (χc) as a function of time. Figure 5A shows that the time needed to reach the final crystalline state is shortened for PVDF with positive nucleation agents. The crystalline fraction, χc, also increased compared to that of the neat PVDF except for PVDF-NAp-2 and -4. In fact, the crystallization rate is increased significantly, especially for the PVDF with NAp-3, -5, and -6. The higher crystallization rate is ascribed to the positive surface charge of the nucleation agents employed. The extreme electronegativity

was used to calculate the efficiency of each nucleation agent and tabulated in Table 2. The most efficient nucleation agents were found to be NAp-3 (E = 52%, Tc nucl = 147 °C at 2 wt %), NAp-5 (E = 52%, Tc nucl = 147 °C at 2 wt %), and NAp-6 (E = 40%, Tc nucl = 145 °C at 2 wt %). Table 2 also includes efficiency data as a function of concentration for NAp-3. The highest efficiency is 59% at a concentration of 0.5 wt %. Others either have shown no improvement or are, in fact, detrimental to the crystallization behavior of PVDF. Crystallization Kinetics. The crystallization kinetics are measured under isothermal crystallization conditions. With use 7384

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

of the fluorine atom (4.0) compared to that of the carbon atom (2.5) results in a strong polarized C−F bond. In comparison, the C−H bond is a weak dipole with the carbon and hydrogen electronegativity equal to 2.5 and 2.1, respectively.50 Therefore, the strong dipoles of the C−F bonds and the oppositely charged surface of nucleation agents exhibit a natural affinity through the ion−dipole interaction existing between nucleation agents and PVDF. This interaction induces polymer chains to be adsorbed onto the surface of positive nucleation agents, lowering the free energy barrier for nucleation and thus increasing the crystallization speed. Although NAp-1 and -7 also possess positive surfaces, they affect the crystallization behavior of PVDF to a much lesser degree. This is attributed to a relatively poor dispersion of these nucleation agents because of their high melting temperatures as

compared to PVDF. The melting temperatures of NAp-3, -5, and -6 are very close to that of the α phase of PVDF. In this case, the size of the positive nucleation agent nucleus is small, thus offering a substantial surface area for the PVDF chain to be adsorbed onto and facilitating the heterogeneous nucleation. Although, as seen in Scheme 1, the melting temperature of NAp-4 is also observed to be close to that of the α phase of PVDF, as shown in Figures 2B and 3A,D, this nucleation agent does not disperse well in PVDF, thus accounting for its lack of effectiveness. Our results have shown that the surface charge of the nucleation agents is important. In addition, dispersion is also important in order to achieve high efficiency. In comparison, the lack of efficiency for negative nucleation agents is independent of their ability to disperse, even for samples with a melting temperature very close to PVDF (Figure 5B). This is also true for the neutral nucleation agent. The concentration of nucleation agents does not need to be high in order to be effective. As seen in Figure 6, a very small

Table 1. DSC Parameters of Nonisothermal Crystallization and Subsequent Melting for the Pure PVDF and PVDF with Nucleation Agents samples

Tca (°C)

Tmb (°C)

ΔHmc (J/g)

χcd (%)

PVDF PVDF-NAp-1 PVDF-NAp-2 PVDF-NAp-3 PVDF-NAp-4 PVDF-NAp-5 PVDF-NAp-6 PVDF-NAp-7 PVDF-NAn-1 PVDF-NAn-2 PVDF-NAn-3 PVDF-neutral

139 138 146 147 137 147 145 141 137 137 134 141

168 174 175 176 172 176 177 177 172 172 170 171

48 54 52 57 51 55 54 53 48 49 50 48

46 52 50 54 49 53 52 51 46 47 48 46

Figure 6. Variations of Tm and Tc of PVDF-NAp-3 in the presence of differing amounts of NAp-3: (●) crystallization and (■) following melting.

a

Tc: nonisothermal crystallization temperature. bTm: the subsequent melting point. cΔHm: enthalpic change. dχc: degree of crystallinity.

Table 2. Efficiency of nucleation agents (2 wt %) for PVDF sample

NAp-1

Tc (°C) E (%)

138

Tc (°C) E (%)

NAp-2

NAp-3

NAp-4

NAp-5

146 46

147 52

137 ...

147 52

...

NAp-6 145 40 NAp-3

NAp-7

NAn-1

NAn-2

NAn-3

neutral

141 13

137

137 ...

134

141 13

...

...

0.05%

0.1%

0.5%

1%

1.5%

2%

4%

8%

136 ...

138

148 59

148 59

147 52

147 52

145 40

144 33

...

Figure 5. Integral isothermal crystallization DSC curves of PVDF and PVDF with different nucleation agents: (A) PVDF with the positive nucleation agents crystallized at ΔT = 20 °C(a) PVDF, (b) PVDF-NAp-1, (c) PVDF-NAp-2, (d) PVDF-NAp-3, (e) PVDF-NAp-4, (f) PVDF-NAp-5, (g) PVDF-NAp-6, and (h) PVDF-NAp-7; (B) PVDF with negative and neutral nucleation agents crystallized at ΔT = 20 °C(a): PVDF, (b) PVDF-NAn-1, (c) PVDF-NAn-2, (d) PVDF-NAn-3, and (e) PVDF-neutral. 7385

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

Figure 7. Polarized optical micrographs of PVDF and PVDF with nucleation agents melt at 220 °C for 5 min, and then cooled in room temperature: (A) neat PVDF; (B) PVDF-NAp-3; (C) PVDF-NAp-5; (D) PVDF-NAp-6; (E) PVDF-NAn-1; and (F) PVDF-neutral. All images were taken at room temperature, and the concentration of the nucleation agents was 2 wt %.

obtained on the addition of NAp-3. Similar results were also attained for NAp-5 and -6 (Figure 7C,D). The effectiveness of these nucleation agents is consistent with the DSC results obtained. By all criteria, the NAp-3, -5, and -6 are the most efficient nucleation agents. Characterization of the PVDF Structures Present. As shown in Figure 4B, Tm increased significantly upon addition of positive nucleation agents. The maximum increase in Tm (8 and 9 °C) was observed with 2 wt % of NAp-3, -5, -6, and -7. The higher Tm of PVDF in the blends has been attributed to the presence of the β and γ phase of PVDF as these crystalline phases are known to have higher melting temperatures.5,51 The crystalline phases of PVDF have been identified by using attenuated total reflectance infrared. Figure 8 shows representative infrared spectra of crystallized PVDF with positive nucleation agents, negative nucleation agents, neutral, and NAp-3 of different concentrations. The conformationally sensitive bands in the 400 to 1500 cm−1 are shown.52−54 The bands at 614, 763, 795, and 975 cm−1, characteristic of the α phase, are clearly observed in neat PVDF. The peaks at 1275 and 839 cm−1 indicate the presence of the β phase, and the peaks at 1231, 839, and 811 cm−1 are ascribed to the formation of the γ phase.52,54 Figure 8A shows that the α phase almost disappears in the samples containing 2 wt % positive nucleation agents. At the same time, strong β-phase and γ-phase bands at

amount can alter the crystallization behavior significantly. Beyond 0.5 wt % neither Tc nor the melting temperature (Tm) of crystallized PVDF is affected significantly by the concentration of the nucleation agents. For PVDF-NAp-3 the critical concentration needed is only 0.5 wt %. Even this small amount can alter the crystallization behavior of PVDF significantly. This is consistent with the efficiencies of the nucleation agents calculated above. Morphological Features of the Crystallized PVDF. In addition to the Tc and Tm for all samples, Table 1 also shows the associated endotherm and χc. The χc was calculated from the second melting endotherm by using the value of 104.6 J/g associated with the pure α phase of PVDF. The χc represented by the enthalpic change (ΔHm) increased significantly with the addition of positive nucleation agents. The maximum increase in χc is observed with NAp-3 and -5 with an incremental change of ∼8%. The optical micrographs obtained for PVDF with various nucleation agents are shown in Figure 7. It can be seen from Figure 7A that the pure PVDF formed well-defined crystalline spherulites with a size of about 20 μm, and spherulites filled the entire viewing field. As can be seen in Figure 7B−F, the PVDF spherulite size decreased and the number of spherulites increased dramatically with the addition of nucleation agents. The micrograph displayed in Figure 7B shows that very small and less undiscerned spherulites were 7386

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

Article

Figure 8. Infrared absorbance spectra of PVDF with (A) positive nucleation agents, (B) negative and neutral nucleation agents, and (C) NAp-3 of different concentration.

839, 1275 cm−1 and 839, 1231 cm−1 appear, indicating a change in the chain conformation (Figure 8A). In contrast, Figure 8B shows that the α-phase remains in the samples containing negative and neutral nucleation agents. In Figure 8C, pure PVDF shows α-phase absorption bands at 975, 795, 763, and 614 cm−1. As the concentration of the NAp-3 increases from 0.05 wt % up to 8 wt %, the peaks associated with the α phase decrease, replaced by the β bands at 1275 and 840 cm−1. Absorption bands attributed to γ-phase PVDF also occur at 1233 and 840 cm−1. Their intensity also increases as the concentration of NAp-3 increases. Due to its chain conformation the α phase of PVDF does not have large net dipoles to interact with the partial charges of the nucleation agents added. In contrast, the conformations of the β or the γ phase result in a much greater interaction with the positive nucleation agents. The ion−dipole interaction causes the polymer chain to align on the nucleation agent surface, favoring the preferential formation of β- or γ-chain conformation of PVDF (Figure 9) as compared to

is also an important factor in controlling the crystallization behavior in polymers with large dipoles, such as PVDF. The specific electrostatic interactions will induce more PVDF polymer chains to be adsorbed on the surface of nucleation agents, thus decreasing the free energy of nucleation. In the present study, the effect of surface charge of the nucleation agents on the crystallization behavior of PVDF has been investigated. It should be emphasized that all of our results reported in this study were obtained under identical experimental conditions with the same polymer. The results that we report truly represent the different effect of various nucleation agents on the crystallization behavior of PVDF under the same circumstances. The nucleation agents with positively charged surfaces have proven to be most effective. The dispersion is also important. This is characterized by the relative melting temperatures of the nucleating agents and PVDF. When uniformly dispersed, the closer the melting temperatures of the two components, the more effective they are. The surface charge was monitored by using a fluorescent labeling method. The detailed crystallization behavior and its kinetics, including the conformational changes of the PVDF chain during crystallization of neat PVDF and PVDF with nucleation agents, have been investigated by using DSC, ATR-IR, and polarized optical microscopy. The isothermal crystallization study demonstrated that the crystallization rate of PVDF was increased significantly by using the positive nucleation agents, due to the specific ion−partial dipole (C−F) interaction. Positive nucleation agents led to a remarkable increase in Tc of PVDF compared with that of neat PVDF. Tm and χc of PVDF also are enhanced with the addition of positive nucleation agents. PVDF formed β- and γ-phase crystals in the presence of positive nucleation agents. In contrast, only the α phase is formed in the absence of nucleation agents, and with negative or neutral nucleation agents. This study provides additional guidance in developing nucleation agents to either enhance the crystallization kinetics, crystallite size, or crystalline phase.

Figure 9. The schematic representation of electrostatic interaction between PVDF and positive nucleation agents.

the α conformation. In addition, it has been reported previously that the γ phase has a much slower crystallization rate as compared to the α phase.5,55 We are not aware of any relative crystallization kinetics for the β phase. Therefore, the preferential formation of extended chain conformation results from the presence of nucleation agents consistent with previous studies.5,20

■

CONCLUSIONS Undoubtedly, the efficiency of nucleation agents depends on the degree of dispersion in the polymer. However, we hypothesize that the surface charge of the nucleation agents 7387

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388

The Journal of Physical Chemistry B

■

Article

(30) Newman, B. A.; Scheinbeim, J. I. Macromolecules 1983, 16, 60− 68. (31) Scheinbeim, J. I.; Newman, B. A.; Sen, A. Macromolecules 1986, 19, 1454−1458. (32) Kawamoto, N.; Sakai, A.; Horikoshi, T.; Urushihara, T.; Tobita, E. J. Appl. Polym. Sci. 2007, 103, 198−203. (33) Liao, R. G.; Yang, B.; Yu, W.; Zhou, C. X. J. Appl. Polym. Sci. 2007, 104, 310−317. (34) Kalish, J. P.; Aou, K.; Yang, X. Z.; Hsu, S. L. Polymer 2011, 52, 814−821. (35) Aou, K. Effect of Molecular Structure on the Thermal Stability of Amorphous and Semicrystalline Poly(lactic acid). Ph.D. Dissertation, University of Massachusetts Amherst, Amherst, MA, 2007. (36) Ebengou, R. H. J. Polym. Sci., Part B: Polym. Phys. 1997, 35, 1333−1338. (37) Lauritzen, J. I.; Hoffman, J. D. J. Res. Natl. Bur. Stand. 1960, 64A, 73−102. (38) Noohom, W.; Jack, K. S.; Martin, D.; Trau, M. Biomed. Mater. 2009, 4, 015003. (39) Avella, M.; Cosco, S.; Di Lorenzo, M. L.; Di Pace, E.; Errico, M. E.; Gentile, G. Eur. Polym. J. 2006, 42, 1548−1557. (40) Shepard, T. A.; Delsorbo, C. R.; Louth, R. M.; Walborn, J. L.; Norman, D. A.; Harvey, N. G.; Spontak, R. J. J. Polym. Sci., Part B: Polym. Phys. 1997, 35, 2617−2628. (41) Gelfer, M. Y.; Song, H. H.; Liu, L. Z.; Hsiao, B. S.; Chu, B.; Rafailovich, M.; Si, M. Y.; Zaitsev, V. J. Polym. Sci., Part B: Polym. Phys. 2003, 41, 44−54. (42) Balzano, L.; Rastogi, S.; Peters, G. W. M. Macromolecules 2008, 41, 399−408. (43) Lipp, J.; Shuster, M.; Terry, A. E.; Cohen, Y. Langmuir 2006, 22, 6398−6402. (44) Balzano, L.; Portale, G.; Peters, G. W. M.; Rastogi, S. Macromolecules 2008, 41, 5350−5355. (45) Iwahashi, M.; Umehara, A.; Wakisaka, K.; Kasahara, Y.; Minami, H.; Matsuzawa, H.; Shinzawa, H.; Ozaki, Y.; Suzuki, M. J. Phys. Chem. B 2007, 111, 740−747. (46) Jiang, S.; Granick, S. Langmuir 2009, 25, 8915−8918. (47) Fillon, B.; Wittmann, J. C.; Lotz, B.; Thierry, A. J. Polym. Sci., Part B: Polym. Phys. 1993, 31, 1383−1393. (48) Fillon, B.; Lotz, B.; Thierry, A.; Wittmann, J. C. J. Polym. Sci., Part B: Polym. Phys. 1993, 31, 1395−1405. (49) Schneider, S.; Drujon, X.; Lotz, B.; Wittmann, J. C. Polymer 2001, 42, 8787−8798. (50) Jungnickel, B. J. Polymeric Materials Handbook; CRC Press: New York, 1996; Vol. 9, pp 7115−7122. (51) Buckley, J.; Cebe, P.; Cherdack, D.; Crawford, J.; Ince, B. S.; Jenkins, M.; Pan, J. J.; Reveley, M.; Washington, N.; Wolchover, N. Polymer 2006, 47, 2411−2422. (52) Kobayashi, M.; Tashiro, K.; Tadokoro, H. Macromolecules 1975, 8, 158−171. (53) Bachmann, M. A.; Gordon, W. L.; Koenig, J. L.; Lando, J. B. J. Appl. Phys. 1979, 50, 6106−6112. (54) Tashiro, K.; Kobayashi, M.; Tadokoro, H. Macromolecules 1981, 14, 1757−1764. (55) Silva, M. P.; Sencadas, V.; Botelho, G.; Machado, A. V.; Rolo, A. G.; Rocha, J. G.; Lanceros-Mendez, S. Mater. Chem. Phys. 2010, 122, 87−92.

AUTHOR INFORMATION

Corresponding Author

*Phone: 413-577-1411 (office). E-mail: [email protected]. edu. Present Address †

Formulation Scientist−Driveline Group, Afton Chemical Corporation, 500 Spring St., Richmond, VA 23219.

Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Zweifel, H. Plastics additives handbook; Hanser Gardner Publication: Cincinnati, OH, 2001; Vol. 6, pp 919−932. (2) Libster, D.; Aserin, A.; Garti, N. Polym. Adv. Technol. 2007, 18, 685−695. (3) Simoes, R. D.; Rodriguez-Perez, M. A.; de Saja, J. A.; Constantino, C. J. L. J. Therm. Anal. Calorim. 2010, 99, 621−629. (4) Qiao, X. Y.; Zhang, Y.; Zhang, Y. X.; Zhu, Y. T. J. Appl. Polym. Sci. 2003, 89, 513−520. (5) Ince-Gunduz, B. S.; Alpern, R.; Amare, D.; Crawford, J.; Dolan, B.; Jones, S.; Kobylarz, R.; Reveley, M.; Cebe, P. Polymer 2010, 51, 1485−1493. (6) Wunderlich, B. Macromolecular Physics: Crystal Nucleation, Growth, Annealing; Academic Press: New York, 1976; Vol. 2, pp 4−39. (7) Mercier, J. P. Polym. Eng. Sci. 1990, 30, 270−278. (8) Binsberg, F. L. J. Polym. Sci., Part B: Polym. Phys. 1973, 11, 117− 135. (9) Binsberg, F. L. Polymer 1970, 11, 253−267. (10) Priya, L.; Jog, J. P. J. Polym. Sci., Part B: Polym. Phys. 2003, 41, 31−38. (11) Ma, W. Z.; Wang, X. L.; Zhang, J. J. Therm. Anal. Calorim. 2010, 103, 319−327. (12) Kitamura, H.; Matsunaga., S.; Teramoto, Y. Polyvinylidene Fluoride Resin Composition, U.S. Patent 6610766 B1, 2003. (13) Judovits, L. H.; Rodak, N. J.; Zerafati, S.; Gupta, R. R. Polyvinylidene Fluoride Having an Increased Melting Point, U.S. Patent 2008/0281031 A1, 2008. (14) Abraham, F.; Schmidt, H. W. Polymer 2010, 51, 913−921. (15) Pillin, I.; Pimbert, S.; Levesque, G. Polym. Eng. Sci. 2002, 42, 2193−2201. (16) Benz, M.; Euler, W. B.; Gregory, O. J. Macromolecules 2002, 35, 2682−2688. (17) Yoon, S.; Prabu, A. A.; Kim, K. J.; Park, C. Macromol. Rapid Commun. 2008, 29, 1316−1321. (18) Ma, W. Z.; Wang, X. L.; Zhang, J. J. Polym. Sci., Part B: Polym. Phys. 2010, 48, 2154−2164. (19) Priya, L.; Jog, J. P. J. Polym. Sci., Part B: Polym. Phys. 2002, 40, 1682−1689. (20) Ramasundaram, S.; Yoon, S.; Kim, K. J.; Park, C. J. Polym. Sci., Part B: Polym. Phys. 2008, 46, 2173−2187. (21) Patro, T. U.; Mhalgi, M. V.; Khakhar, D. V.; Misra, A. Polymer 2008, 49, 3486−3499. (22) Levi, N.; Czerw, R.; Xing, S. Y.; Iyer, P.; Carroll, D. L. Nano Lett. 2004, 4, 1267−1271. (23) Ke, K.; Wen, R.; Wang, Y.; Yang, W.; Xie, B. H.; Yang, M. B. J. Mater. Sci. 2011, 46, 1542−1550. (24) Wittmann, J. C.; Lotz, B. Prog. Polym. Sci. 1990, 15, 909−948. (25) Mauritz, K. A.; Bare, E.; Hopfinger, A. J. J. Polym. Sci., Part D: Macromol. Rev. 1978, 13, 1−61. (26) Thierry, A.; Fillon, B.; Straupe, C.; Lotz, B.; Wittmann, J. C. Prog. Colloid Polym. Sci. 1992, 87, 28−31. (27) Li, L. Y.; Li, C. Y.; Ni, C. Y. J. Am. Chem. Soc. 2006, 128, 1692− 1699. (28) Xu, J. Z.; Chen, T.; Yang, C. L.; Li, Z. M.; Mao, Y. M.; Zeng, B. Q.; Hsiao, B. S. Macromolecules 2010, 43, 5000−5008. (29) Siripitayananon, J.; Wangsoub, S.; Olley, R. H.; Mitchell, G. R. Macromol. Rapid Commun. 2004, 25, 1365−1370. 7388

dx.doi.org/10.1021/jp3043494 | J. Phys. Chem. B 2012, 116, 7379−7388