Non-isothermal crystallization kinetics of UHMWPE composites filled by oligomer-modified CaCO3

The processability of ultrahigh molecular weight polyethylene (UHMWPE) improved by oligomer-modified calcium carbonate (CaCO3) was observed in our previous work. In order to understand the effect of oligomer-modified CaCO3 on the crystallization of UHMWPE, the non-isothermal crystallization behavior and crystallization kinetics of UHMWPE composites filled by oligomer-modified CaCO3 was studied by differential scanning calorimetry in this work. Jeziorny and Mo methods were used to describe the non-isothermal crystallization kinetics of UHMWPE composites. The effect of modified filler content and cooling rate on the crystallization temperature and crystallization rate was discussed. The heterogeneous nucleation of modified CaCO3 slightly increases the crystallization temperature of UHMWPE. The crystallization enthalpy of UHMWPE composites is significantly higher than that of UHMWPE. The crystallization rate of UHMWPE composites depends on the filler contents and cooling rate.

Sattari et al. [30] reported the non-isothermal crystallization and melting behavior of UHMWPE hybrid composites reinforced with short carbon fiber (SCF) and nano-SiO 2 particles. It is observed that adding SCF to UHMWPE increased the melting peak temperatures and the crystallinity of UHMWPE. The amounts of nano-SiO 2 had no obvious effect on melting temperatures of UHMWPE and decreased the crystallinity of UHMWPE. Ozawa-Avrami method can be used to describe the non-isothermal crystallization kinetics of the composites.
Liu et al. [31] prepared a novel antibacterial UHMWPE/ chlorhexidine acetate (CA)-montmorillonite (MMT) composites and investigated the crystallization kinetics of UHMWPE, UHMWPE/MMT, UHMWPE/CA and UHMWPE/CA-MMT. It is observed that the CA and MMT exhibited strong heterogeneous nucleation to increase the crystallization temperatures of UHMWPE. Addition of CA lowered the melting temperature of UHMWPE due to its plastization. The results of crystallization kinetics indicated that the addition of CA decreased the crystallization rate and broadened the range of crystal growth temperature, while MMT increased the crystallization rate due to its heterogeneous nucleation.
Zhang et al. [32,33] studied the non-isothermal crystallization kinetics of UHMWPE in liquid paraffin (LP) systems and used the Avrami method modified by Jeziorny and Mo to describe the influence of UHMWPE content and cooling rate on crystallization mechanism and spherulitic structure of UHMWPE. Addition of LP increased the crystallization rate of UHMWPE. The Avrami plots of UHMWPE and UHMWPE/LP blends show a good linearity. At the primary crystallization stage, it is found that the Avrami exponent n is variable around 5 and decreases slightly as the cooling rate decreases. High values of n 1 for UHMWPE and UHMWPE/ LP blends may be generated from their high viscosities yielding a more complicated crystallization mechanism. In addition, the extent of secondary crystallization increases with increasing cooling rate. The Avrami exponent n 2 , ranging from 0.80 to 1.96, indicated the simpler crystallization mode. Further, the value of F(T) in the Mo method increased with an increase in relative crystallinity and UHMWPE content in the blends.
Shen et al. [34] researched the influence of compatibilizer and dispersant on the non-isothermal crystallization behavior of HDPE/UHMWPE/hydroxyapatite blends. Analyzed by Jeziorny method, the value of Avrami exponent n is between 2.7 and 3.5. At the same temperature, adding compatibilizer and dispersant increased the crystallization rate, while decreased the half crystallization time and the activation energy of crystallization. In addition, the compatibilizer and dispersant also improved the mechanical properties of blends.
Except as mentioned above, the effect of UHMWPE on the crystallization behavior and crystallization kinetics of HDPE has been reported [13,35]  composites filled by inorganic particles. In our previous works, a new UHMWPE composite filled by oligomermodified CaCO 3 was prepared in our laboratory. It is found that our prepared oligomer-modified CaCO 3 can improve the processability of UHMWPE [36]. In order to understand the effect of oligomer-modified CaCO 3 on the crystallization of UHMWPE, the crystallization behavior of UHMWPE composites was investigated by DSC, while the nonisothermal crystallization kinetics was described by different kinetic models. The influence of the modified filler on the UHMWPE crystallization was discussed in this article.

Sample preparation
In order to obtain acrylic acid (AA)-modified calcium carbonate (AA-CaCO 3 ), 500 g CaCO 3 was added in 2.5 L ethyl alcohol at room temperature, then 40 g AA was added and stirred 20 min by mechanical stirring so that CaCO 3 reacted with AA thoroughly. Then, the solvent was volatilized at room temperature, and the product of AAmodified calcium carbonate (AA-CaCO 3 ) was dried in at 80°C for 1 h. The AA-CaCO 3 , PE wax (mass ratio 8:2) and 0.5 mass% BPO (relative to CaCO 3 ) as initiator were mixed in mixer chamber at 150°C for 10 min. The long chain-modified CaCO 3 (PEW-g-CaCO 3 ) was prepared due to reaction between PEW and AA. Composites of UHMWPE filled by different contents of PEW-g-CaCO 3 were prepared in mixer chamber at 190°C for 10 min.

Sample characterization
The melting and crystallization behavior of UHMWPE composites were measured by differential scanning calorimetry (DSC) at nitrogen atmosphere with the DSC-8500 (PE, America), calibrated with indium. About 2-3 mg sample was firstly heated up to 180°C at the rate of 100°C min -1 and kept this temperature for 3 min to eliminate the thermal history of sample. Then, the sample was cooled at the rate of 5, 10, 15, 20, 25°C min -1 from 180 to 60°C to obtain the non-isothermal crystallization curves. At last, the sample was reheated up to 180°C at the rate of 10°C min -1 to obtain the melting curves. The crystallinity of sample (X c ) is calculated by Eq. 1.
where DH m is the melting enthalpy of sample measured by DSC, DH 0 m is the melting enthalpy of 100% crystalline UHMWPE (290 J g -1 ) [30] and w is the mass content of UHMWPE in the composites.

Results and discussion
Effect of modified-CaCO 3 on non-isothermal crystallization behavior of UHMWPE Figure 1 shows the crystallization curves of UHMWPE composites cooled at different rates. The plots of crystallization peak temperatures and crystallization enthalpy versus cooling rate are shown in Fig. 2. It is observed that the crystallization peak of UHMWPE is shifted to the lower temperature and crystallization peak becomes wider with an increase in cooling rates. The crystallization peak temperature of UHMWPE composites is higher than that of UHMWPE, and the contents of PEW-g-CaCO 3 have little influence on the crystallization peak temperature of UHMWPE composites. The crystallization enthalpy of composites also decreases with an increase in cooling rate. At the same cooling rate, the crystallization enthalpy of composites is significantly higher than that of UHMWPE and increases with an increase in the PEW-g-CaCO 3 contents at cooling rate of 5°C min -1 . However, the contents of PEW-g-CaCO 3 have little influence on the crystallization enthalpy of UHMWPE composites at high cooling rate. The crystallization enthalpy of composites decreases with an increase in cooling rate, while crystallization peak becomes wider. It is suggested that the molecular chain of UHMWPE at higher cooling rate has poor mobility which is difficult to crystallize, resulting in a wider crystallization peak. At the same cooling rate, addition of modified filler increases the crystallization temperature of UHMWPE which indicates the presence of the heterogeneous nucleation of PEW-g-CaCO 3 . The higher crystallization enthalpy of UHMWPE is attributed to improved mobility of molecular chain of UHMWPE by oligomer-modified CaCO 3 .
In order to understand the effect of different modified-CaCO 3 on crystallization of UHMWPE, the non-isothermal crystallization behavior of UHMWPE composites filled by CaCO 3 , AA-CaCO 3 and PEW-g-CaCO 3 are investigated by DSC. The plots of crystallization peak temperature and crystallization enthalpy versus cooling rate are shown in Fig. 3. It can be observed from Fig. 3 that the crystallization peak temperatures from high to low are AA-CaCO 3 / UHMWPE [ PEW-g-CaCO 3 /UHMWPE & CaCO 3 /UHM WPE [ UHMWPE at the same cooling rate. It is indicted that the heterogeneous nucleation of AA-CaCO 3 , PEW-g-CaCO 3 and CaCO 3 can increase the crystallization peak temperature of UHMWPE and the heterogeneous nucleation of AA-CaCO 3 is higher than that of PEW-g-CaCO 3 and CaCO 3 . The crystallization enthalpy of UHMWPE composites is also higher than that of UHMWPE due to the heterogeneous nucleation of fillers. It can also be seen the crystallization enthalpy of PEW-g-CaCO 3 /UHMWPE is higher than that of AA-CaCO 3 /UHMWPE, CaCO 3 / UHMWPE and UHMWPE at cooling rate of 5°C min -1 . In our previous work [36], it is found that the oligomer-modified CaCO 3 can improve the processability of UHMWPE. It is suggested that the presence of oligomer in the surface of PEW-g-CaCO 3 is benefit to the mobility of molecular chain of UHMWPE to increase the degree of crystallization. However, the crystallization enthalpy of PEW-g-CaCO 3 / UHMWPE is lower than that of AA-CaCO 3 /UHMWPE and CaCO 3 /UHMWPE at high cooling rate above 5°C min -1 . It is considered that the mobility of molecular chain of UHMWPE is restricted in PEW-g-CaCO 3 /UHMWPE at high cooling rate and the heterogeneous nucleation of PEWg-CaCO 3 is lower than that of AA-CaCO 3 , resulting in the decrease in the crystallization enthalpy of PEW-g-CaCO 3 / UHMWPE with an increase in cooling rate.

Effect of modified-CaCO 3 on non-isothermal crystallization kinetics of UHMWPE
Some models were used to describe the non-isothermal crystallization kinetics of polymer and its blends and composites, including Ozawa, Jeziorny and Mo models, etc. In our work, effect of modified-CaCO 3 on nonisothermal crystallization kinetics of UHMWPE was investigated by the Avrami theory modified by Jeziorny and Mo models.
Based on the Avrami theory modified by Jeziorny [33,37,38], the non-isothermal crystallization kinetics can be described as follows: where n is the Avrami exponent, Z t and Z c are the Avrami and Jeziorny crystallization rate constants, respectively, R is the cooling rate, the time-dependent relative crystallinity, X t , can be calculated by Eqs. 5 and 6.
where T 0 and t 0 are the crystallization onset temperature and time, T and t are the arbitrary crystallization temperature and time, T ! and t ! are the ultimate crystallization temperature and time, respectively, X T is the temperaturedependent relative crystallinity. If lg[ln (1 -X t )] is linear on lgt, the Avrami exponent n and the crystallization rate constant Z t can be obtained by the slope and intercept. The crystallization rate constant Z c can be calculated by Eq. (4). Figures 4 and 5 are the representative curves of relative crystallinity of UHMWPE composites versus temperature or time, respectively. The plot of half crystallization time (t 1/2 ) ( Table 1) obtained from Fig. 5 versus cooling rate is shown in Fig. 6. It can be seen that the half crystallization time for all samples decreases with an increase in cooling rate, indicating that accelerating the cooling rate can increase the crystallization rate so that the crystallization  Fig. 7. A good linear relation means that Jeziorny method is suitable to describe the non-isothermal crystallization kinetics of UHMWPE and its composites. The Avrami exponent n and crystallization rate constant Z t obtained by the slope and intercept are shown in Table 1. It can be observed from Table 1 that the crystallization rate constant of all samples increases with accelerating the cooling rate from 5 to 10°C min -1 and then decreases with an increase in the cooling rate. The Avrami exponent n is about 5, which may be generated from their high viscosities, yielding a more complicated crystallization mechanism [32].
The relationship of crystallization rate constant of UHMWPE composites versus cooling rate is shown in Fig. 8. It can be seen that the crystallization rate constant of UHMWPE composites depends on the filler contents and cooling rate. At cooling rate of 10°C min -1 , UHMWPE and its composites exhibit maximum crystallization rate. At cooling rate above 10°C min -1 , the crystallization rate of UHMWPE decreases with an increase in cooling rate. At the filler content below 10 wt%, the crystallization rate of UHMWPE composites increases with an increase in filler contents. At the filler content above 10 wt%, the crystallization rate of UHMWPE composites decreases with an increase in filler contents. It is suggested that higher contents of filler and high cooling rate should hinder the For the verification of the kinetic calculation results, we compare the experiment results with the calculation results by kinetic constants, shown in Fig. 5c. It can be found that the experiment results are agreed with the kinetic calculation results at relative crystallinity range of 0-50%. Above 50% relative crystallinity, the experiment results deviated with the kinetic calculation results may be due to the effect of second crystallization.
Mo method is generally used to investigate the nonisothermal crystallization kinetics of polymer and its composites [33,37,38].
where the R is the cooling rate, F(T) = [k/Z t ] 1/m can be used to characterize the crystallization rate. The value of F(T) is bigger, the crystallization rate is slower. If lgR is linear on lgt, the values of F(T) can be calculated by the intercept. Figure 9 is the plots of lgR versus lgt for UHMWPE and its composites. The plots of lgR have a good linearity with lgt, which means that Mo method is suitable to describe the non-isothermal crystallization kinetics of UHMWPE and its composites. The plots of value of F(T) calculated by the intercept versus relative crystallinity is shown in Fig. 10. The value of F(T) increases with an increase in relative crystallinity, indicating that the crystallization rate becomes slower in high crystallinity. The F(T) valve of  pure UHMWPE is higher than those of all composites, also indicating that the PEW-g-CaCO 3 can accelerate the crystallization of UHMWPE. Besides, the lower F(T) of UHMWPE composites filled by 10-20% PEW-g-CaCO 3 also indicated the higher crystallization rate, conforming to the conclusion obtained by Jeziorny method.

Conclusions
In order to improve the properties of UHMWPE composites filled by CaCO 3 , oligomer-modified CaCO 3 was prepared and the effect of oligomer-modified CaCO 3 on the nonisothermal crystallization behavior and crystallization kinetics of UHMWPE was investigated by differential scanning calorimetry. The results indicated that the addition of oligomer-modified CaCO 3 increases the crystallization temperature and crystallization enthalpy of UHMWPE. It is attributed to the heterogeneous nucleation of oligomermodified CaCO 3 and the improved mobility of molecular chain of UHMWPE by oligomer-modified CaCO 3 . Jeziorny and Mo methods can be used to describe the non-isothermal crystallization kinetics of UHMWPE and its composites. The crystallization rate of UHMWPE composites depends on the filler contents and cooling rate. UHMWPE and its composites exhibit maximum crystallization rate at cooling rate of 10°C min -1 . The crystallization rate of UHMWPE decreases with an increase in cooling rate at cooling rate above 10°C min -1 . The crystallization rate of UHMWPE composites increases with an increase in filler contents at the filler content below 10 wt% and decreases with an increase in filler contents at the filler content above 10 wt%. The higher contents of filler and high cooling rate should hinder the movement of molecular chain to decrease the crystallization rate of UHMWPE.