Crystallization of poly(ε-caprolactone) in its immiscible blend with polylactide: insight into the role of annealing histories

Abstract

Polylactide/poly(ε-caprolactone) (PLA/PCL) is a very promising blend material with biodegradable characteristics and tailorable performance because of the good property complementarity between the two components. However, PLA and PCL are asymmetric thermodynamically: PLA has far higher melting point than PCL, and crystallization temperature of PCL is even lower than glass transition temperature of PLA. But this also provides good opportunity to control final structure and properties of PCL/PLA blends through thermal annealing. In this work, two annealing routes were designed to control supermolecular structure of discrete PLA phase in the PCL-rich blends, and the two systems, the blend with discrete amorphous PLA domains and the one with discrete crystallized PLA phase, were obtained. The interfacial property alteration and crystallization behavior of continuous PCL phase were then studied. The results are very interesting. Relative to the amorphous PLA phase, the crystallized PLA domains have better affinity to the continuous PCL phase, showing stronger nucleating effect to the crystallization of PCL and higher impeding effect on the shear flow of blend system. But the presence of discrete PLA phase, whether in crystallized or amorphous state, has no evident influence on the crystal structure and lamellar thickness of PCL. These effects on the crystallization of PCL make the mechanical properties of blends very sensitive to the annealing histories. The blend sample with the crystallized PLA domains shows higher modulus and strength than the one with the amorphous PLA domains, and the values of modulus and strength increase by about 130% and 43%, respectively, relative to the neat PCL. This work provides a facile and green approach to well tailor the supermolecular structure and mechanical properties of the PCL/PLA blends through the simple control of annealing process.

---

1. Introduction

Polymer blend technology has been studied extensively in the past several decades because mixing two or more immiscible polymers with different physical properties presents the possibility of enhancing the overall material performance by a synergistic combination of the desirable properties of each component in one system.¹ Among the numerous developed higher performance polymer blends, poly(ε-caprolactone) (PCL)/polylactide (PLA) blend system has attracted much attention recently.² This is not only because the two polymer components, PCL and PLA, are biodegradable and biocompatible, showing attractive environmental-friendly characteristics, but also because their physical properties are fully complementary. PLA with high degradation rate presents better tensile strength, while PCL with much slower degradation rate better toughness. Blending them together is hence an efficient way of fabricating new biomaterials with well-tailored properties and lifetime to meet different environmental or physiological requirements.

Up to now, this blend system has been studied extensively. The reported studies mainly focus on several important aspects around structure–property relations of this system: (1) design of phase morphology and its evolution control to obtain blend with special phase structure;^3–8 (2) improvement of immiscible phase morphology by reaction processing^9–11 or compatibilization technology using block copolymers;^12–19 (3) dependence of thermal and mechanical properties on the compositions and hierarchical system structure;^20–28 (4) flow behavior and viscoelasticity of system and their relations with thermodynamic and dynamic conditions;^29–33 (5) evaluation of system biodegradation^34–38 and fabrication of porous materials or nonwovens used for tissue substrates or drug delivery;^39–43 (6) property and structure design by controlling selective localization of nanoparticles.^44–53

Generally, the physical properties of the semicrystalline polymers are governed by their supermolecular structure, which in turn is dominated by crystallization histories. For a polymer blend system composed of two semicrystalline components, that means that its final structure and performance are dependent on the crystallization of each component strongly.⁵⁴ The crystallization behavior of immiscible PCL/PLA blend is very interesting because the two polymers are asymmetric thermodynamically. The melting point (T_m) and glass transient temperature (T_g) of PCL are far lower than those of PLA, and the crystallization temperature (T_c) of PCL is even lower than the T_g of PLA. It has been reported that the presence of minor PCL phase favors cold crystallization of PLA in their blend system because PCL is in its molten state during PLA crystallization, reducing system viscosity as a result,⁴⁷ or acts as additional substrates.²⁴ Similar trend has also been found by Urquijo et al.²⁵ The presence of minor PLA phase also promote crystallization of PCL because the solid-state PLA particles play the role of nucleating agents.⁴⁷ But Newman and coworkers³² reported a reverse trend because they found a decrease of melting and crystallization temperatures of PCL in the presence of PLA phase.

It is well known that the crystallization of PLA depends not only on its bulk D/L ratio,⁵⁵ but also highly on annealing process.^56,57 Its supermolecular structure, therefore, can be controlled through various annealing approaches. This means that discrete PLA domains can exist as amorphous phase or crystalline one in a PCL-rich blend. If so, are the crystallization of PCL dependent on the supermolecular structure of PLA phase? This is interesting and worthy of deep study. Therefore, in this work, two different annealing approaches were designed for control of the supermolecular morphology of discrete PLA phase in the PCL-rich blend. The effect of amorphous and crystalline PLA domains on the crystallization behavior of PCL was then detected by the nonisothermal method for a comparative study, aiming at finding the answer to question proposed above. The crystal structure and the mechanical properties of those blends with various annealing histories were explored as well, with the objective to establish hierarchical structure–annealing relationships of PCL/PLA blend materials.

2. Experimental part

2.1. Materials preparation

Poly(ε-caprolactone) (CAPA6500) is a commercial product purchased from Solvay Co. Ltd., Belgium, with –OH values lower than 2 mg KOH g⁻¹. Its number average molecular weight (M_n) is about 69 000 g mol⁻¹ and the melt index (MI) is about 7 g/10 min (160 °C/2.16 kg). Polylactide (2002D) was purchased from NatureWorks Co. Ltd., USA, which has a D content of 4.25 wt% and a residual monomer content of 0.3 wt%, with the density of 1.24 g cm⁻³. Its M_n is about 80 000 g mol⁻¹ and MI is about 8 g/10 min (190 °C/2.16 kg). The Newtonian viscosity ratio of these two polymers is about 1/16 (PCL/PLA, 180 °C).^17,47

The blends were prepared by melt mixing PLA with PCL (30/70 w/w) using a Haake polylab rheometer (Thermo Electron Co., USA) at 170 °C and 50 rpm for 6 min. The sheet samples for the morphological and rheological measurements were prepared by the compression molding at 170 °C and 10 MPa. For better property comparison, the neat PCL was also processed to keep the same thermal histories with those blend samples. All materials were dried under vacuum before using. The dog-bone shaped specimens (32 mm × 4 mm × 2 mm) for tensile tests were prepared by a Haake mini-jet (Thermo Scientific Co., USA). The injection molding was performed at the cylinder temperature of 180 °C, with the injection pressure 600 bar and holding pressure 500 bar.

2.2. Thermal property characterizations

The crystallization and melting behaviors of the samples were recorded using a differential scanning calorimeter (DSC, Netzsch DSC-204F1, Germany). Two kinds of annealing routes were employed here: the sheet samples (2.8–3.0 mg) were molten at 180 °C for 5 min to eliminate previous thermal histories, and then cooled to room temperature at the predetermined rates (3, 5, 7 and 8 °C min⁻¹). This is route I. In the subsequent heating process, the samples were heated to 130 °C at the rate of 5 °C min⁻¹, hold till the cold crystallization of PLA finished, and then cooled to room temperature again. This is route II. Detailed DSC sweep approaches of these two kinds of annealing routes are schematically shown in Fig. 1. All tests were carried out under nitrogen. The sample weight and thickness, and the cooling rates for the tests were determined according to the suggestion made by Vyazovkin et al.⁵⁸

Fig. 1 (a) DSC thermograms of PCL/PLA blend in the cooling and heating process at the rate of 5 °C min⁻¹ and the schematic annealing routes (I and II) for nonisothermal crystallization of PCL and (b) time–temperature profile of the DSC experiments for the two annealing routes (I and II).

2.3. Morphology and microstructure characterizations

The phase structure of PCL/PLA blend was investigated by a S-4800 field-emission scanning electron microscope (FE-SEM, Hitachi, Japan) with a 15 kV accelerating voltage. The brittle-fractured surface was coated with gold using an SPI sputter coater before observation. The spherulitic morphology was evaluated by a polarized optical microscope (POM, Leika DMLP, Germany) equipped with a hot stage (Linkam LTM350, England). The temperature ramps were the same as those in the DSC tests. The microstructure of blend system was evaluated by a Fourier transform infra-red spectrometer (TENSOR27 FT-IR, Bruker Co., Germany). The FT-IR spectra were obtained by coadding 128 scans and collected at room temperature with the reflection mode. The resolution is within 2 cm⁻¹. The crystal structure of PCL was detected by a D8 ADVANCE X-ray diffractometer (XRD, Bruker AXS, Germany) performed at room temperature with the Cu target and a rotating anode generator operated at 40 kV and 40 mA. The scanning rate was 2° min⁻¹ from 2° to 40°. The lamellar morphology was detected by a NanoStar small-angle X-ray scattering (SAXS) instrument (Bruker AXS GmbH, Germany) equipped with a Vantec-2000 2D detector. The measurements were also performed at room temperature, and the incident X-rays of CuKα radiation (1.54 Å) were monochromated by a cross-coupled Göbel mirror and passed through the sheet sample. The distance between the sample and detector was calibrated using silver behenate, giving the scattering vector q range from 0.07 to 2.3 nm⁻¹. The 1D SAXS profiles, shown as the normalized intensity I (arbitrary units) versus q (q = (4π/λ)sin θ, where λ is the wavelength of the X-rays and 2θ the scattering angle), were obtained by integration of the 2D pattern collected by the 2D detector.

2.4. Melt rheology measurements

The rheological tests were performed on a rotational rheometer (HAAKE RS600, Thermo Electron Co., USA) equipped with a parallel plate geometry (20 mm diameter plates). During steady shear flow, the stress and viscosity responses to the shear rates were recorded. In the linear dynamic rheological measurements, the frequency sweep was performed under at the strain level of 1%, which was predetermined by the strain sweep. First scan was performed at 80 °C. Then the blend sample was heated to 130 °C to let PLA phase fully cold-crystallized, and then cooled to 80 °C, experienced second scan.

2.5. Contact angle characterizations

Contact angle tests were performed with an OCA40 apparatus (Dataphysics Co Ltd., Germany). Static contact angles of two test liquids (distilled water and glycerol) were measured by depositing a drop of 3–5 μl on the sample surface and the values were estimated as the tangent normal to the drop at the intersection between the sessile drop and the surface. To avoid solvent evaporation, images were taken within 30 s of drop deposition. The optical images of droplet on the sheet samples are shown in Fig. S1 of the ESI.† The reported contact angle values here are the average of at least ten measurements at different spots of the surface. The surface parameters, including the surface energy (γ), and its dispersive component (γ^d) and polar one (γ^p), of the tested samples were then calculated according to Owens–Wendt method,⁵⁹ and the results are summarized in Table 1. The literature values of the two used liquids (H₂O: γ^p = 50.8 dyn cm⁻¹ and γ^d = 22.5 dyn cm⁻¹; C₃H₈O₃: γ^p = 37.0 dyn cm⁻¹ and γ^d = 26.4 dyn cm⁻¹) were used here.⁶⁰ The interfacial energy (γ₁₂) between two polymers is then calculated according to the geometric-mean equation⁶¹or to the harmonic-mean equationand the values are listed in Table S1 of the ESI.†

Table 1 The values of contact angles and surface parameters of polymers (20 °C)

Sample	Contact angle (°)		Surface energy (mN m^−1)
	H2O	C3H8O3	γ^ds	γ^ps	γ
a The amorphous PLA sheet was prepared by the compression molding with natural cooling.b The crystallized PLA sheet was obtained by the annealing of amorphous PLA sample at 130 °C heated from room temperature.
Neat PCL	87.2 ± 3.9	84.1 ± 3.0	7.5	12.8	20.3
Amorphous PLA^a	77.3 ± 3.7	65.7 ± 2.9	22.9	9.7	32.6
Crystallized PLA^b	84.5 ± 3.2	80.1 ± 2.4	9.3	13.3	22.6

---

2.6. Surface wetting experiments

The amorphous PLA fiber was cut from the sheet sample prepared by compression molding, which has typical amorphous morphology. Then, the fiber was heated to 120 °C using a hot stage, annealed for cold crystallization, and transformed to crystallized PLA fiber in this way. After natural cooling, tens of PCL fibers (cut from sheet sample) was put on the PLA fiber perpendicularly, and then heated to 70 °C to let PCL molten and fully relaxed. Finally, the samples were cooled to room temperature and observed using an optical microscope to record the wetted surface morphology.

2.7. Mechanical property characterizations

The tensile properties of the neat PCL and its blends with various annealing histories were determined by an Instron Mechanical Tester (ASTM D638) at a crosshead speed of 50 mm min⁻¹ at 25 °C using the dog-bone shaped specimens. Strength and modulus values reported here represent an average of the results for tests run on six specimens.

3. Results and discussion

3.1. Crystallization behavior of the PCL/PLA blends annealed with various routes

The immiscible phase morphology of current PCL/PLA blend are shown in Fig. 2. The blend presents typical matrix-dispersion phase structure, in which the discrete PLA domains with the average size of about 15 μm are dispersed in the continuous PCL phase. Detailed information on the morphological formation and its dependence on the PCL/PLA blending ratio and processing conditions can be found elsewhere.^17,31,32 Because the melting point of PLA are far higher than that of PCL, the discrete PLA domains are the solid particles during crystallization of PCL, rather than viscous droplets, as can be seen in Fig. 1a. After annealing with route I, the discrete PLA is amorphous because it is hard to be crystallized from molten state in cooling process, while keeps its crystalline structure as the system is annealed with route II, because it experiences the cold crystallization in the heating process, and followed by cooling process immediately before the temperature achieves up to its melting point, as indicated in Fig. 1.

Fig. 2 SEM image of the PCL/PLA blend with the scale bar of 50 μm. The inset graph is optical image (OM) of the sample in molten state (180 °C) with the scale bar of 20 μm.

Therefore, the discrete PLA phase has different supermolecular chain morphology in the continuous PCL matrix after the blend experiences various annealing processes. This is further confirmed by the optical microscope observation shown in Fig. 3. A spherulitic structure is seen clearly in the discrete PLA domains at the sampling point a in the cooling process of route II (indicated in Fig. 1), which is attributed to cold crystallization of PLA in the heating process (the upper limit temperature is 130 °C, lower than the melting point of PLA), instead of to melt crystallization of PLA in the following cooling process. But merely amorphous structure of PLA phase is observed at the same point of route I because the cold crystallization structure of PLA is molten in the heating process (the upper limit temperature is 170 °C, higher than the melting point of PLA), and the PLA used in this work cannot be crystallized from molten state during nonisothermal annealing.^56,57 At the sampling point b of two annealing routes, the continuous PCL phase is crystallized from molten state, showing grainy structure with the size out of the range for POM observation. This coarse crystalline structure of PCL has been widely reported, especially in the presence of additional nucleating agents.^62,63 The verification tests of POM agree with the DSC results well and is up to the experimental design. Hereafter these two systems are referred as to blend (route I) and blend (route II), respectively.

Fig. 3 Optical (OM) and polarized optical images (POM) for the blend sample at various stages of annealing process with the scale bar of 20 μm.

Fig. 4 gives the DSC traces of the neat PCL and two blend systems in the cooling and followed heating scans. The obtained calorimetric parameters are summarized in Table 2. Clearly, the presence of discrete PLA phase (actually the solid PLA particles) has strong nucleating effect on the PCL crystallization, leading to an evident increase of T_c by about 10 °C. Actually this trend can be seen on all cases at various scan rates, as shown in Fig. S2 of the ESI.† The heterogonous nucleation role of PLA has been widely reported in the PCL/PLA blend system.^23,24,47 However, it is notable that the blend (route II) has higher PCL T_c than the blend (route I). This is interesting because it indicates that the PLA particles with various annealing histories have different levels of nucleation to the PCL crystallization. Clearly, the crystallized PLA particles have better ability to act as additional active substrates relative to the amorphous ones. But the presence of PLA phase, whether crystallized or amorphous, results in the formation of the crystallite structure with thinner lamella or less ordered PCL crystals. Therefore, the two blend systems show coarse crystalline structure, with lower T_m and degree of crystallinity (X_c%) relative to the neat PCL (Table 2).

Fig. 4 DSC thermograms of the neat PCL and its blend samples annealed through various routes with the cooling rate of 7 °C min⁻¹ and the heating rate of 5 °C min⁻¹.

Table 2 Calorimetric data^a derived from the cooling and the second heating DSC scans (5 °C min⁻¹) on the neat PCL and its blend samples with PLA

Samples	Tc (°C)	ΔHc (J g^−1)	Tm (°C)	ΔHm (J g^−1)	Xc (%)
a Tc, crystallization temperature; Tm, melting point; ΔHc, enthalpy of the crystallization; ΔHm, enthalpy of the melting process; Xc = ΔHm/ΔH^0m (ΔH^0m = 139 J g^−1^64).
Neat PCL	26.6	79.0	57.6	71.2	51.2
Blend (route I)	34.5	51.2	56.7	53.3	38.3
Blend (route II)	35.8	58.0	56.8	55.7	40.1

---

3.2. Annealing history dependence of nucleation activity of discrete PLA phase

The nucleation activity of PLA particles in the PCL-based blends can be evaluated with a method suggested by Dobreva and Gutzow.^65,66 The nucleation activity (N_a) is a factor by which the work of three-dimensional nucleation decreases with the addition of a foreign substrate.^67,68 If the foreign substrate is extremely active, N_a approaches zero, while for inert particles it is unity. For nucleation from melts near their melting temperature, the cooling rate (ϕ) can be represented aswhere A is constant, the degree of supercooling ΔT_p = T_m − T_c. B is a thermodynamic parameterwhere V_m is the molar volume of the crystallization polymer, ΔS_m the entropy of melting, k the Boltzmann constant, σ the specific surface energy, ω a geometrical factor and n the Kolmogorov-Avrami exponent. N_a can then be calculated:where B* is the value of B when the polymer is filled and B⁰ when is unfilled, which can be obtained from the slope of a plot of log ϕ versus 1/ΔT_p², as shown in Fig. 5. The calculated values of N_a are listed in Table 3. Clearly, the blend (route II) shows the lower value of N_a as compared with the blend (route I). This well confirms that the crystallized PLA domains have higher nucleation activity than the amorphous ones.

Fig. 5 The Dobreva plots of log ϕ versus 1/ΔT_p² for the neat PCL and its blend samples.

Table 3 Kinetic parameters of the neat PCL and its blend samples with PLA

Samples	ϕ (°C min^−1)	t1/2 (min)	Na
Neat PCL	3	2.13	—
	5	1.47
	7	1.03
	8	0.97
Blend (route I)	3	1.58	0.83
	5	1.10
	7	0.74
	8	0.71
Blend (route II)	3	1.36	0.75
	5	0.86
	7	0.63
	8	0.60

---

Early in 1970s, Chattewee et al.⁶⁹ have found that the crystallization of a polymer substrate itself could affect the bulk crystallization of another polymer. According to their classification, the discrete PLA particles should be classified as type II, namely moderately active substrate to PCL, rather than very active substrate (type I), because there is no occurrence of transcrystallization in the PCL/PLA blend system. Although the detailed mechanism of heterogeneous nucleation in the polymers is not yet well understood, it is believed to arise from molecular interactions between the polymer and the surface of the nucleating agent.^70,71 This interaction results in a reduction in the interfacial free-energy barrier for stable nucleus formation.⁷² For the current PCL/PLA blend, however, it is hard to evaluate the alteration of folding surface free energy and nucleation constant of PCL in the presence of PLA particles (amorphous or cold-crystallized) using Lauritzen–Hoffman–Miller method because PCL commonly forms coarse crystalline structure during bulk crystallization, with the size out of the range for POM observation (growth rate of spherulite cannot be obtained in this case). But the rheology and surface tests can provide useful information.

Fig. 6 gives dynamic rheological responses of the neat PCL and its blend samples with amorphous and cold-crystallized PLA particles. All samples reveal the dynamic shear behavior dominated by the viscous flow, instead of elastic deformation, because the loss modulus is higher than the storage modulus in the low-frequency region. But the PCL/PLA sample looks more like a filled system rather than blend one at current testing temperature because there is no appearance of shape relaxation plateau in the middle-frequency region, which is caused by the viscoelastic deformation/recovery of phase interface.^17,30,31 This is reasonable because the discrete PLA phase is in the solid state at 80 °C, forming rigid interface with PCL phase. The low-frequency G′ of the blend shows no evident change after cold crystallization of PLA (indicated in Fig. 6), suggesting that rigidity alteration of the solid PLA particles nearly has no influence on the system elasticity. This is because the continuous flow phase is PCL and the PLA phase is discrete. However, it is notable that after cold crystallization of discrete PLA phase, the system viscosity increase remarkably (indicated by the arrow in Fig. 6), indicating an enhanced depression effect of discrete PLA particles on the flow of PCL. Because the cold crystallization of amorphous PLA phase is in situ, the size and location of those solid PLA particles are stable before the second scan. The tripled system viscosity, therefore, suggests that the crystallized PLA particles have much stronger phase interactions⁷³ with PCL than the amorphous ones.

Fig. 6 The dynamic rheological responses of (a) moduli and (b) viscosities for the neat PCL and its blend samples with amorphous (first scan) and cold-crystallized PLA particles (second scan).

This improved affinity is further confirmed by the surface characterizations. There is large difference in both the dispersive component and the polar one of surface energy between PCL and amorphous PLA (Table 1). The two polymers have a high level of interfacial energy, therefore (Table S1 of the ESI†). However, after cold crystallization, both the dispersive and polar component values of PLA approach to those of PCL, and in this case, crystallized PLA presents the surface property very close to the PCL. In other words, the annealing process with route II improves affinity of solid PLA phase to PCL matrix evidently in their blend. The FT-IR spectra shown in Fig. 7 reveal the same trend. The bands at 1457 cm⁻¹ and 1470 cm⁻¹ (region a) are assigned to the asymmetric deformation mode of CH₃ on the backbone of PLA⁷⁴ and PCL,⁷⁵ respectively. These two peaks are partially separated in the blend (route I), and similar trend is also observed on the bands around 1093 cm⁻¹ (PLA) and 1105 cm⁻¹ (PCL) (region b) caused by the symmetric stretching mode of C–O–C, and on the bands around 1040 cm⁻¹ (PLA) and 1046 cm⁻¹ (PCL) (region c) due to the stretching vibrations of C–CH₃ groups.^76,77 However, for the blend system (route II), the two adjacent peaks approach with each other, forming one broader or shouldered band in each region. This also indicates that the crystallized PLA phase has better affinity to PCL. The fiber surface wetting measurements reveal more intuitionistic evidence on this improved affinity, as shown in Fig. 8. Clearly, the surface of cold-crystallized PLA can be better wetted by the PCL droplets, relative to the amorphous PLA. It can hence be proposed that compared with the amorphous PLA domains, the discrete crystallized PLA particles have stronger nucleating effect to the crystallization of PCL matrix and higher impeding effect on the shear flow of blend system.

Fig. 7 FT-IR spectra of the neat PCL, PLA and their blends with various annealing histories.

Fig. 8 Optical images of the surface morphology of (a) cold-crystallized and (b) amorphous PLA fibers wetted by the PCL droplets with the scale bar of 20 μm. The inset graph is POM image of crystallized PLA fiber wetted by PCL droplets.

The nucleating effect of the discrete PLA domains, whether in the amorphous or crystalline state, can promote the overall crystallization kinetics of PCL. The values of half crystallization time (t_1/2) listed in Table 3 are obtained according to the curves of relative degree of crystallinity (X_t) shown in Fig. S3 of the ESI.† It is clear that the blend systems have the lower values of t_1/2 than the neat PCL, indicating that the presence of solid PLA particles accelerates PCL crystallization. At the same cooling rate, the system experienced annealing process of route II shows higher crystallization rates as compared with the one annealed with route I, while has higher values of crystallization activation energy (E_a), as can be seen in Fig. 9. The E_a values are calculated according to Friedman's equation⁷⁸

where dX_t/dt is the instantaneous crystallization rate as a function of time at a given degree of conversion, is the set of temperatures related to a given conversion X_t at different cooling rates (ϕ). R is the gas constant. This is reasonable because the blend (route II) has better affinity between two phases, which leads to an increased system viscosity and decreased chain mobility relative to the blend (route I). The reduced diffusion ability of PCL chain increases the overall activation energy of PCL crystallization. To the two blend systems, however, the nucleation dominates the overall kinetics of PCL crystallization, rather than the crystal growth, because of their higher crystallization rates than the neat PCL.

Fig. 9 The dependence of activation energy (E_a) on relative degree of crystallinity (X_t) for the neat PCL and its blend samples.

3.3. Effect of annealing histories on the crystal structure of PCL matrix

Although the presence of amorphous or crystallized PLA phase domains has large influence on the overall kinetics of PCL crystallization, it nearly does not change the crystal structure of PCL because all samples show the same three typical diffraction peaks, which correspond to (110), (111), and (200) planes of PCL crystallite,⁷⁹ as can be seen in Fig. 10a. Two new peaks at 16.9° and 19.1° for the blend annealed with route II are caused by the cold crystallization of PLA domains, which are attributed to diffraction of (110/200) and (203) planes of PLA crystallite (α form), respectively.^80,81 The crystallite size (L_hkl) can be calculated by Scherrer equation:⁸²where L_hkl is the crystallite dimension, or coherence length, perpendicular to the (hkl) plane, K the Scherrer constant, λ the wavelength of the X-rays and θ the Bragg angle. When β_hkl is the diffraction half-width, K takes a value of 0.9. The calculated values of L_hkl are listed in Table 4.

Fig. 10 (a) XRD and (b) SAXS patterns for the neat PCL and its blend samples.

Table 4 Crystallite size of the neat PCL and its blend samples with PLA

Crystallite size	Neat PCL	Blend (route I)	Blend (route II)
L110 (nm)	30.72	29.45	28.16
L111 (nm)	28.06	25.16	25.08
L200 (nm)	29.47	22.39	22.54

---

It is seen that the two blends have almost identical L_hkl values within experimental errors, indicating that the crystallite size of PCL in the blends is not sensitive to the annealing histories designed in this work. However, compared with that of the neat PCL, the L₂₀₀ values decrease evidently in the presence of discrete PLA domains, whether in amorphous or crystallized. This elucidates that the presence of immiscible PLA phase impedes growth of PCL crystallite along (200) plane, which is attributed to the nucleation effect of PLA domains.^47,83 Actually this is already indicated by the decreased intensity of the diffraction peak by (200) plane (see arrow in Fig. 10a). However, the presence of PLA phase does not change lamellar thickness because the primary peaks of the SAXS profiles of the blend samples, which are associated with interlamellar distance (d_lam = 2π/q) in the PCL lamellar stack, have no evident shift relative to that of the neat PCL sample, as can be seen in Fig. 10b. It indicates that all samples almost have identical values of long period of lamellae (16–17 nm). This is reasonable because PLA and PCL are fully immiscible and as a result merely form macrophase-separated structure, instead of microphase-separated one in their blends. In this case, the presence of PLA domains does not result in the confined nucleation or crystallization of PCL,¹⁰ which is commonly observed on the microphase-separated PCL based block copolymers.^83,84

3.4. Tensile properties of the PCL/PLA blends with various annealing histories

Since the crystallization of PCL and the phase adhesion between PCL and PLA have various dependences on the annealing routes, the final mechanical properties of the blends may also depend on the annealing histories of PLA phase. Fig. 11 shows the tensile behavior for the neat PCL and the two blend samples. It is seen that the values of yield strength and Young's modulus of the blend sample annealed with common way (route I) increase from 20.9 MPa to 25.3 MPa by about 21%, and from 412.5 MPa to 732.5 MPa by about 78%, respectively, relative to those of the neat PCL. This is caused by the presence of rigid PLA phase,^15,17,45 despite decreased degree of crystallinity of the PCL matrix (Table 2).

Fig. 11 Stress–strain curves for the neat PCL and its blend samples (σ is yield strength for the neat PCL and tensile strength for the blends).

However, the reinforcement role of PLA domains is not fully played for the blend (route I) because of poor interfacial adhesion between two phases.¹⁷ Since the cold crystallization of discrete PLA phase can improve its affinity to PCL (Fig. 8), the mechanical strength of the blend sample may further increase after annealing with route II. As expected, the sample (route II) shows the tensile strength of 29.8 MPa, increasing by about 18% relative to the sample (route I). Clearly, enhanced phase adhesion favors improvement of load transfer between two phases.⁴⁵ Besides, cold crystallization of discrete PLA phase has contribution to the increased strength, too, especially to the further increase of modulus of the blend. But the presence of rigid PLA domains and the existence of interfacial defects also result in disappearance of yield behavior of the blend system, accompanied by the sharply decreased elongation levels at break. Therefore, design of annealing process is a facile way to control the mechanical properties of thermodynamically asymmetric PCL/PLA blends.

4. Conclusions

PLA/PCL blends were prepared by melt mixing, and then experienced various routes of annealing to control the supermolecular structure of PLA phase. The presence of discrete PLA domains has stronger nucleating effect on crystallization of continuous PCL phase, leading to an evident increase of crystallization rates, despite increase of crystallization activation energy. But the nucleation highly depends on supermolecular structure of PLA phase, which is in turn controlled by the annealing routes. Relative to the amorphous PLA domains, the crystallized PLA ones has better affinity to PCL, and as a result, having stronger nucleating effect to the PCL crystallization. But the crystal structure and lamellar size of PCL does not depend on the presence of PLA domains and annealing histories. Therefore, the mechanical properties of PCL/PLA blends can be easily controlled by annealing. Compared with those of the neat PCL, the modulus and strength of the blend sample annealed with route II can increase by about 130% and 43%, respectively. The results obtained in this work can provides additional important information and possible way of property control and design of PCL/PLA bio-blend materials.

Acknowledgements

This work was supported by the research grants from the National Natural Science Foundation of China (51573156), the Prospective Joint Research Program of Jiangsu Province (BY2014117-01), and the Blue Project of Jiangsu Province.

References

1. D. R. Paul and C. B. Bucknall, Polymer Blends: Formulation & Performance, John Wiley & Sons, New York, 2000.
2. G. C. Eastmond, Adv. Polym. Sci., 2000, 149, 59–223.
3. R. Dell Erba, G. Groeninckx, G. Maglio, M. Malinconico and A. Migliozzi, Polymer, 2001, 42, 7831–7840.
4. X. Roy, P. Sarazin and B. D. Favis, Adv. Mater., 2006, 18, 1015–1019.
5. P. Sarazin, X. Roy and B. D. Favis, Biomaterials, 2004, 25, 5965–5978.
6. P. Sarazin, G. Li, W. J. Orts and B. D. Favis, Polymer, 2008, 49, 599–609.
7. V. Mittal, T. Akhtar, G. Luckachan and N. Matsko, Colloid Polym. Sci., 2015, 293, 573–585.
8. I. Fortelny, A. Ostafinska, D. Michalkova, J. Juza, J. Mikesova and M. Slouf, Polym. Bull., 2015, 72, 2931–2947.
9. L. Wang, W. Ma, R. A. Gross and S. P. McCarthy, Polym. Degrad. Stab., 1998, 59, 161–168.
10. K. Kawaguchi, H. Nakao, E. Masuda and Y. Tajima, J. Appl. Polym. Sci., 2008, 107, 1269–1279.
11. B. Y. Shin and D. H. Han, Radiat. Phys. Chem., 2013, 83, 98–104.
12. G. Maglio, A. Migliozzi, R. Palumbo, B. Immirzi and M. G. Volpe, Macromol. Rapid Commun., 1999, 20, 236–238.
13. C. H. Kim, K. Y. Cho, E. J. Choi and J. K. Park, J. Appl. Polym. Sci., 2000, 77, 226–231.
14. S. Aslan, L. Calandrelli, P. Laurienzo, M. Malinconico and C. Migliaresi, J. Mater. Sci., 2000, 35, 1615–1622.
15. Y. H. Na, Y. He, X. T. Shuai, Y. Kikkawa, Y. Doi and Y. Inoue, Biomacromolecules, 2002, 3, 1179–1186.
16. S. M. Li, L. J. Liu, H. Garreau and M. Vert, Biomacromolecules, 2003, 4, 372–377.
17. D. F. Wu, Y. S. Zhang, L. J. Yuan, M. Zhang and W. D. Zhou, J. Polym. Sci., Part B: Polym. Phys., 2010, 48, 756–765.
18. C. M. Zhang, T. L. Zhai, L. S. Turng and Y. Dan, Ind. Eng. Chem. Res., 2015, 54, 9505–9511.
19. T. F. Shen, M. G. Lu and L. Y. Liang, J. Macromol. Sci., Part A: Pure Appl. Chem., 2013, 50, 547–554.
20. M. E. Broz, D. L. Vander Hart and N. R. Washburn, Biomaterials, 2003, 24, 4181–4190.
21. H. Tsuji and G. Horikawa, Polym. Int., 2007, 56, 258–266.
22. L. Chen, J. H. Yang, K. Wang, F. Chen and Q. Fu, Polym. Int., 2010, 59, 1154–1161.
23. N. Lopez-Rodriguez, A. Lopez-Arraiza, E. Meaurio and J. R. Sarasua, Polym. Eng. Sci., 2006, 46, 1299–1308.
24. Y. Q. Zhang, Z. K. Wang, F. Jiang, J. Bai and Z. G. Wang, Soft Matter, 2013, 9, 5771–5778.
25. J. Urquijo, G. Guerrica-Echevarria and J. E. Eguiaz-abal, J. Appl. Polym. Sci., 2015, 132, 42641.
26. T. Patrício and P. Bártolo, Procedia Eng., 2013, 59, 292–297.
27. Z. N. Correa-Pacheco, J. L. Jimenez-Perez, M. A. Sabino, A. Cruz-Orea and M. Loaiza, Appl. Phys. A, 2015, 120, 1323–1329.
28. V. B. Carmona, A. C. Correa, J. M. Marconcini, L. Henrique and C. L. Mattoso, J. Polym. Environ., 2015, 23, 83–89.
29. S. Han, T. J. Moon, Y. C. Bae, S. Yi and S. H. Lee, Polymer, 1998, 39, l113–1117.
30. D. F. Wu, Y. S. Zhang, M. Zhang and W. D. Zhou, Eur. Polym. J., 2008, 44, 2171–2183.
31. Y. S. Zhang, D. F. Wu, M. Zhang, W. D. Zhou and C. C. Xu, Polym. Eng. Sci., 2009, 49, 2293–2300.
32. D. Newman, E. Laredo, A. Bello, A. Grillo, J. L. Feijoo and A. J. Muller, Macromolecules, 2009, 42, 5219–5225.
33. F. Cock, A. A. Cuadri, M. García-Morales and P. Partal, Polym. Test., 2013, 32, 716–723.
34. Z. Gan, D. Yu, Z. Zhong, Q. Liang and X. Jing, Polymer, 1999, 40, 2859–2862.
35. L. Liu, S. Li, H. Garreau and M. Vert, Biomacromolecules, 2000, 1, 350–359.
36. H. Tsuji and T. Ishizaka, Macromol. Biosci., 2001, 1, 59–65.
37. H. Tsuji, G. Horikawa and S. Itsuno, J. Appl. Polym. Sci., 2007, 104, 831–841.
38. G. Sivalingam, S. P. Vijayalakshmi and G. Madras, Ind. Eng. Chem. Res., 2004, 43, 7702–7709.
39. J. X. Chen, L. L. Lu, D. F. Wu, L. J. Yuan, M. Zhang, J. J. Hua and J. Xu, ACS Sustainable Chem. Eng., 2014, 2, 2102–2110.
40. H. J. Haroosh, Y. Dong and G. D. Ingram, J. Polym. Sci., Part B: Polym. Phys., 2013, 51, 1607–1617.
41. H. J. Haroosh, Y. Dong, D. S. Chaudhary, G. D. Ingram and S. Yusa, Appl. Phys. A, 2013, 110, 433–442.
42. X. F. Wang, M. R. Salick, X. D. Wang, T. Cordie, W. J. Han, Y. Y. Peng, Q. Li and L. S. Turng, Biomacromolecules, 2013, 14, 3557–3569.
43. H. J. Haroosh, Y. Dong and K. T. Lau, J. Mater. Sci., 2014, 49, 6270–6281.
44. A. Hasook, S. Tanoue, Y. Iemoto and T. Unryu, Polym. Eng. Sci., 2006, 46, 1001–1007.
45. D. F. Wu, Y. S. Zhang, M. Zhang and W. Yu, Biomacromolecules, 2009, 10, 417–424.
46. E. Laredo, M. Grimau, A. Bello, D. F. Wu, Y. S. Zhang and D. P. Lin, Biomacromolecules, 2010, 11, 1339–1347.
47. D. F. Wu, D. P. Lin, J. Zhang, W. D. Zhou, M. Zhang, Y. S. Zhang, D. M. Wang and B. L. Lin, Macromol. Chem. Phys., 2011, 212, 613–626.
48. A. L. Goffin, Y. Habibi, J. M. Raquez and P. Dubois, ACS Appl. Mater. Interfaces, 2012, 4, 3364–3371.
49. M. Forouharshad, L. Gardella, D. Furfaro, M. Galimberti and O. Monticelli, Eur. Polym. J., 2015, 70, 28–36.
50. S. C. Agwuncha, S. S. Ray, J. Jayaramudu, C. Khoathane and R. Sadiku, Macromol. Mater. Eng., 2015, 300, 31–47.
51. J. Rodenas-Rochina, A. Vidaurre, I. C. Cortazar and M. Lebourg, Polym. Degrad. Stab., 2015, 119, 121–131.
52. J. P. Mofokeng and A. S. Luyt, Polym. Test., 2015, 45, 93–100.
53. J. P. Mofokeng and A. S. Luyt, Thermochim. Acta, 2015, 613, 41–53.
54. M. L. Di Lorenzo, Prog. Polym. Sci., 2003, 28, 663–689.
55. R. M. Rasal, A. M. Janorkar and D. E. Hirt, Prog. Polym. Sci., 2010, 35, 338–356.
56. D. F. Wu, L. Wu, L. F. Wu, B. Xu and M. Zhang, J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 1100–1113.
57. D. F. Wu, L. Wu, L. F. Wu, B. Xu and M. Zhang, J. Nanosci. Nanotechnol., 2008, 8, 1658–1668.
58. S. Vyazovkin, K. Chrissafis, M. L. Di Lorenzo, N. Koga, M. Pijolat, B. Roduit, N. Sbirrazzuoli and J. J. Suñol, Thermochim. Acta, 2014, 590, 1–23.
59. D. K. Owens and R. C. Wendt, J. Appl. Polym. Sci., 1969, 13, 1741–1747.
60. E. N. Dalai, Langmuir, 1987, 3, 1009–1015.
61. M. Stamm, Polymer Surface and Interface: Characterization, Modification and Applications, Springer-Verlag, Berlin, 1st edn, 2008.
62. Q. L. Lv, D. F. Wu, Y. X. Qiu, J. X. Chen, X. Yao, K. S. Ding and N. X. Wei, Thermochim. Acta, 2015, 612, 25–33.
63. D. F. Wu, L. Wu, Y. R. Sun and M. Zhang, J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 3137–3147.
64. C. G. Pitt, F. L. Chasalow and Y. M. Hibionada, J. Appl. Polym. Sci., 1981, 26, 3776.
65. A. Dobreva and I. J. Gutzow, J. Non-Cryst. Solids, 1993, 162, 1–12.
66. A. Dobreva and I. J. Gutzow, J. Non-Cryst. Solids, 1993, 162, 13–25.
67. M. W. Alonso, J. I. Velasco and J. A. De Saja, Eur. Polym. J., 1997, 33, 255–262.
68. S. H. Kim, S. H. Ahn and T. Hirai, Polymer, 2003, 44, 5625–5634.
69. A. M. Chattewee and F. P. Price, J. Polym. Sci. Polym. Phys. Ed., 1975, 13, 2369–2383.
70. J. P. Mercier, Polym. Eng. Sci., 1990, 30, 270–278.
71. R. H. Ebengou, J. Polym. Sci., Part B: Polym. Phys., 1997, 35, 1333–1338.
72. J. W. Mullin, Crystallization, Butterworth-Heinemann, Oxford, 4th edn, 2001.
73. S. H. Wu, Polym. Eng. Sci., 1987, 27, 335–343.
74. J. M. Zhang, H. Tsuji, I. Noda and Y. Ozaki, Macromolecules, 2004, 37, 6433–6439.
75. J. Xu, B. H. Guo, R. Yang, Q. Wu, G. Q. Chen and Z. M. Zhang, Polymer, 2002, 43, 6893–6899.
76. S. K. Mallapragada and B. Narasimhan, Infrared Spectroscopy in Analysis of Polymer Crystallinity, John Wiley & Sons, Ltd, London, 2000.
77. S. H. Murphy, G. A. Leeke and M. J. Jenkins, J. Therm. Anal. Calorim., 2012, 107, 669–674.
78. H. Friedman, J. Polym. Sci. C, 1996, 6, 183–195.
79. H. Hu and D. L. Dorset, Macromolecules, 1990, 23, 4604–4607.
80. J. M. Sarasua, N. L. Rodriguez, A. L. Arraiza and E. Meaurio, Macromolecules, 2005, 38, 8362.
81. D. F. Wu, L. Wu, L. F. Wu, B. Xu and M. Zhang, J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 1100–1113.
82. M. Kakudo and N. Kasai, X-ray Diffraction by Polymers, Kodansha, Tokyo, 1972.
83. I. W. Hamley, P. Parras, V. Castelletto, R. V. Castillo, A. J. Muller, E. Pollet, P. Dubois and C. M. Martin, Macromol. Chem. Phys., 2006, 207, 941–953.
84. L. L. Chen, J. Jiang, L. Wei, X. L. Wang, G. Xue and D. S. Zhou, Macromolecules, 2015, 48, 1804–1812.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07752h

---