Pathway mediated microstructures and phase morphologies of asymmetric double crystalline co-oligomers

Abstract

Various microstructures and phase morphologies of an amphiphilic poly(ethylene oxide)-block-polyethylene (PEO-b-PE) co-oligomer, controlled by topological restriction of PE segments on the tethered PEO chains, were characterized by differential scanning calorimetry (DSC), polarized optical microscopy (POM), scanning electron microscopy (SEM), and synchrotron radiation wide-angle/small-angle X-ray scattering (WAXS/SAXS) in drop-cast films. The crystallization processes were mediated by two pathways, a one-step crystallization process (I) and a sequential crystallization process (II). Results show that the thermal procedures have great influence on the microstructures and phase morphologies of PEO-b-PE co-oligomer, e.g., negative spherulites with radial stripes were detected in the one-step crystallization process (I), while crystalline texture, which contains a large number of crystals with reduced sizes, formed in the sequential crystallization process (II). Based on our experimental data, the topological restriction effect encountered by PEO chains depends on the hard confinement of PE crystals and the soft confinement of amorphous PE in the two crystallization procedures. The formation mechanisms of the long-range order structures within the co-oligomer were elucidated through morphology models. These nano-patterned structures make the double crystalline block copolymers outstanding candidates for surface modification, micromolding, and optoelectronic devices in nanotechnological and biomedical applications.

---

1 Introduction

Asymmetric double crystalline copolymers, composed of two chemically joined crystalline repeat units (named blocks, or sequences) with unequal composition, are thought to form long-range crystals.^1,2 With the undisputed advantages in templates synthesis,^3,4 drug delivery systems,⁵ conformation and assembly control of proteins,^6–8 and adsorbed biomolecules orientation,^9,10 among others, the wide variety of block copolymers allows for fine-tuning of these systems to meet the requirements of specific applications. Meanwhile, the established hierarchies on different length scales are the most appealing templates to study crystal structure,¹¹ morphology,^12,13 and crystallization kinetics.¹⁴

The final confined morphology of block copolymers is shaped either by breaking through or by being restricted in the preformed crystalline structure and microdomains of the first crystallized block.¹¹ For poly(ethylene oxide)-block-polystyrene (PEO-b-PS) diblock copolymer, polystyrene (PS), with a glass transition temperature above the crystallization temperature of poly(ethylene oxide) (PEO), T_C,_(PEO), provides hard confinement to restrict the PEO chains within the microdomains, and causes a microphase-separation dominated superstructure. As a result, multilayer lamellar morphology can form in this system, and the amorphous phase is incorporated in the inter-lamellar PEO regions.¹⁵ For poly(ethylene oxide)-block-polybutadiene (PEO-b-PB) diblock copolymers, polybutadiene (PB), with a glass transition temperature below T_{C (PEO)}, offers soft confinement and allows the crystallization process of PEO to dominate the final crystal structure of this copolymer. Hong et al.^16,17 observed a lamellar morphology which consisted of alternating layers of PEO and PB, and proved its nonspherulitic crystalline texture (with the absence of the Maltese cross). In addition, the PEO lamellae became tortuous and highly interconnected after crystallization in the cylindrical and spherical morphology.¹⁸

Poly(ethylene oxide)-block-polyethylene (PEO-b-PE) copolymers consist of two components with excellent biostability that have been introduced to many biomedical application fields.¹⁹ Short-chain double-crystalline PEO-b-PE co-oligomers are known to form heterogeneous and amphiphilic surface nanopatterns. Due to the special molecular structures and crystalline properties of PEO-b-PE co-oligomers, these surface nanopatterns can be mediated by designing the thermal processes and the chemical natures of the substrate.²⁰ In the crystallization process of PEO-b-PE co-oligomer, the PE block functions both as a hard confinement and a soft confinement to restrict PEO crystallization. The hard confinement stems from crystalline PE lamellae, while the soft confinement comes from the amorphous PE.^18,20–22

Sun et al.^14,21 proposed that when both blocks are crystalline, a lamellar structure forms in the bulk of PEO-b-PE co-oligomers, and the PE and PEO blocks arrange themselves into extended-chain crystals with the PE chains tilted ∼22° from the lamellar normal. Later, using 2D-SAXS, Cao et al.^12,13,23 found ordered phase morphology transitions among lamella, perforated lamella, gyroid, cylinder and sphere structures in the non-isothermal crystallization process of PEO-b-PE co-oligomer as function of temperature. They attributed these morphological transitions to the thermally activated motions of two short chain segments of this diblock copolymer. Furthermore, in our previous work,²⁰ a novel approach was presented to create amphiphilic surface nanostructures with lateral dimensions of the order of ∼10 nm by employing perpendicular lamellar thin film morphology of PEO-b-PE co-oligomer. We concluded that chain realignment, morphology development and macroscopic dewetting are induced by a temperature promoted chain flexibility and take place on different time scales. However, there are still some unsolved problems concerning the regulation of microstructures and phase morphologies of this co-oligomer. Answers for these pendent questions, such as the formation mechanism of these nano-textures, the influence of thermal procedure on the crystalline morphologies of PEO-b-PE film, and the topological restriction effects encountered by PEO chains during the whole crystallization procedure, are imperative to fabricate specific nano-confined structures in block copolymers.

In the present work, two kinds of pathways (thermal procedures) were adopted to investigate the crystallization behavior of the PEO-b-PE co-oligomer and analyze their influence on the microstructures and phase morphologies of PEO-b-PE drop-cast films. The purpose of this paper is to present detailed experimental results about the thermal history dependence and topological restriction dependence of the PEO chains tethered on the PE segments, explain the crystallization mechanism, and suggest regulations to induce specific surface nanostructures of amphiphilic copolymers by adjusting thermal procedures. It may introduce a new way in surface modification; micro-fabrication and other correlating fields to obtain block copolymers thin films with diverse nano-patterned structures, which could be interesting as templates for various applications as, e.g., in the biomaterials field as well as for nanotechnology applications.

2 Experimental

2.1 Materials

Poly (ethylene oxide)-block-polyethylene (PEO-b-PE) co-oligomer with 20 wt% PE, was purchased from Aldrich Chemical Co. Inc. (St. Louis, MO). According to the catalogue of the company, the molecular weight of the copolymer is 2250 g mol⁻¹, the molecular weight distribution is 1.1. On the basis of the ¹H NMR measurement, the content of 1,2-butadiene segment corresponding to the branching of ethyl side groups is negligibly small, and the block lengths were calculated to be 15 PE and 47 PEO units, taking the end group into account. Accordingly, we refer to this diblock co-oligomer in the following as E15EO47.²⁰

2.2 Film preparation

Approximately 30 μm thick E15EO47 films were prepared by drop-casting 3% (wt/wt) copolymer–chloroform solution on clean sheet glasses. The solvent was firstly evaporated in the fume hood for 24 h, and then the polymer coated sheet glasses were transferred to a vacuum drying oven at 40 °C for 24 h to remove the residual solvent.

The one-step crystallization process (I) of E15EO47, shown in Fig. 1, was performed after melting at 120 °C for 5 min and rapid cooling to the crystallization temperature with a rate of 30 °C min⁻¹. The samples were crystallized in the time which is longer than the fully crystallization time measured by DSC. The sequential crystallization process (II) is a distinct thermal procedure from the one-step crystallization process (I). Firstly the sample was heated from room temperature to 120 °C and kept for 5 min to erase any thermal history, then rapidly cooled to 80 °C and kept at this temperature for 5 min to allow PE block crystallization. It is enough for the PE segment to crystallize sufficiently at 80 °C judging from the DSC measurement. Finally, the sample was cooled with a rate of 30 °C min⁻¹ to a chosen crystallization temperature of the PEO block. According to the study of a similar oligomer by Cao et al.,^12,13,23 the melt is homogeneous at 120 °C. Hence, 120 °C was chosen to be the starting melt temperature and to ensure the initial states for both one-step crystallization process (I) and sequential crystallization process (II) to be the same. The crystallization temperatures in isothermal crystalline process can be set above the crystalline start temperature in non-isothermal crystalline process.¹¹ And the crystallization time is counted from the point that temperature drops to the crystallization temperature of PEO block.

Fig. 1 Two crystallization processes employed: one-step crystallization process (I) and sequential crystallization process (II).

2.3 Characterization

Polarization microscopy. Polarized optical microscopy (POM) observations of the samples were performed with an Olympus optical microscope (Olympus, BX51), equipped with a hot stage (Linkam LTS350) and a Canon EOS40D camera system. The films were melted on a Linkam LTS350 hot stage, and then treated according to the one-step crystallization and the sequential crystallization programs. The Linkam LTS350 hot stage was equipped with a liquid-nitrogen cooling system (LNP). The maximum cooling rate is 30 °C min⁻¹. The processes were carried out in nitrogen atmosphere. The morphology changes were recorded at different crystallization temperatures (29–42 °C in steps of 1 °C).

Differential scanning calorimetry. Differential scanning calorimetry (DSC) was employed with a Mettler-Toledo DSC 822e. Samples with 3–5 mg in weight were encapsulated in aluminum pans. The calibration was performed with indium and hexatriacontane. An ultra-pure nitrogen atmosphere was employed as circulating atmosphere for all tests. The following measurements were performed by cooling and heating at the rate of 30 °C min⁻¹ as well as in isothermal mode. The applied temperature cycles were the same as in the POM study. The rapid cooling processes were controlled by a liquid-nitrogen cooling system (LNP). The heat fusion evolved during the isothermal crystallization was recorded as a function of time.

Scanning electron microscopy. Scanning electron microscope (SEM) images of all samples were collected through a field emission scanning electron microscopy (FE-SEM, Carl-Zeiss Company) at 20 kV. The samples were collected on aluminum foil and were sputter-coated by a K-550X sputter coater with Au at a thickness of 100 Å to reduce charging.

Synchrotron radiation WAXD. The structure of the sample was characterized by synchrotron radiation X-ray diffraction (XRD) using BL14B1 at the Shanghai Synchrotron Radiation Facility (SSRF) (Shanghai, China) and the wavelength of X-rays employed was 1.2398 Å. BL14B1 is a bending magnet beamline and the energy of the storage ring of SSRF is 3.5 GeV. The beam was first collimated using an Rh/Si mirror and then was monochromatized using a Si (111) double crystal monochromator. After that, the beam was further focused by an Rh/Si mirror to the size of 0.5 mm × 0.5 mm. Higher order harmonics were also rejected by the Rh/Si mirror. A NaI scintillation detector was used for data collection.²⁴ The sample-detector distance for the WAXS experiments was 140.7 mm. The duration of exposure was 60 s. The E15EO47 films were measured between semi-transparent polyimid films within the Linkam LTS350 hot stage with a liquid-nitrogen cooling system (LNP). Polyimid films will not influence the scattering results of E15EO47. Small glass boards with the same thickness were placed between polyimid films to make sure as-prepared samples had the same thickness.

Synchrotron radiation SAXS. The sample-detector distance for the Wide SAXS experiments was 1866.1 mm. The duration of exposure was 60 s. Intensity profiles were interpreted as the plot of scattering intensity, I, versus scattering vector, q. The length of the scattering vector, q, is defined as²⁵

q = (4π/λ)sin(θ) = 2π/d (1)

where 2θ is the scattering angle, λ = 1.54 Å is the X-ray wavelength, and d is the repeat spacing. Plots of Lorentz-corrected desmeared absolute intensities versus scattering vector, q, were obtained by multiplying the SAXS intensities by q² (“Lorentz” correction) after desmearing.

To fit the SAXS data of the self-assembled PE-b-PEO morphologies, the Lorentz peak function were employed in the block copolymer.²⁵

The fitting parameters are the offset x₀ and y₀, the area A of the Lorentz peak, the width w of the peak, its position x_c and y_c. For the experimentally obtained SAXS data, one can then use the following functions:²⁵

where (Iq²)_c and q_c are the coordinate of the peak value; w₁ is the half peak width; A > 0 is a constant. The background scattering was well fitted by the exponential decay function:²⁵where A_n and t_n are the parameters of background scattering.

3 Results and discussion

3.1 Thermal property and crystallization morphology

E15EO47 is a typical short-chain double-crystalline co-oligomer. To identify the crystallization and melting temperatures of PE and PEO blocks, a non-isthermal DSC program was applied. The sample was firstly heated to 120 °C and kept for 5 min to erase any thermal history, then cooled down to 0 °C at a rate of 10 °C min⁻¹ to allow both the PE and PEO fully crystallize. After that the temperature was increased to 120 °C at a rate of 10 °C min⁻¹ to melt the PE and PEO crystals. The DSC thermographs of E15EO47 during heating and cooling processes are shown in Fig. 2. Small exothermic offsets in the interval of 75 to 85 °C upon cooling and 90 to 100 °C upon heating are attributed to phase transitions in the minority PE phase. Sharp exothermic peaks are observed and assigned to the crystallization temperature (28.4 ± 0.2 °C) and melting temperature of PEO block (58.2 ± 0.2 °C).

Fig. 2 DSC thermographs of the E15EO47 obtained at a rate of 10 °C min⁻¹. The melting and crystallization temperatures of PEO and PE are denoted in the graph. The arrow direction in the left vertical axis is the ENDO direction.

Spherulites are a morphological feature of melt-crystallized polymers and consist of a large number of chain-folded lamellar crystallites, radiating in all directions from a central nucleus with molecular chains oriented tangentially.^25–27 Spherulites can be further classified as negative or positive according to the birefringent character. A spherulite with the higher refractive index for light vibrating perpendicular to the radial direction is called negative. So macroscopic birefringence may give qualitative information on crystal orientation as indicated by the colors of the quadrants, i.e., quadrants 1 and 3 are nearly red, while quadrants 2 and 4 are deep blue.²⁵ In order to observe the crystalline superstructure morphologies of E15EO47 thin films with different thermal histories, POM measurements with and without λ wave plate were carried out. The ultimate crystalline superstructure morphologies are shown in Fig. 3(a)–(c). Obvious granular spherulite morphologies with Maltese cross can be observed in E15EO47 thin film prepared in the one-step crystallization (I) at 0 °C. These granular spherulites are proven to be negative spherulites.

Fig. 3 Morphologies of E15EO47 granular spherulites formed at 0 °C and 31 °C in the one-step crystallization process (I) and 31 °C in the sequential crystallization process (II), denoted at top left. (a), (d), and (g) were taken by POM without λ wave plate; (b), (e), and (h) were taken by POM with a λ wave plate; (c), (f), and (i) represent spherulite patterns.

In the one-step crystallization process (I), when the crystallization temperature is set at 31 °C, though the crystallite boundaries of granular spherulites become obscure, the typical Maltese cross can still be observed in Fig. 3(d)–(f), which prove the existence of negative spherulite structure. In one-step crystallization the negative spherulite structure becomes more obvious as the crystalline temperature becomes lower. This is because at higher annealing temperature PE block has more time to nucleate and fold chains into crystals. The PEO crystal plates cannot pattern exactly perpendicular to propagating direction. When the annealing temperature is low enough, the negative spherulite can be seen, as shown in Fig. 9(a) and (b). While in sequential crystallization, only nonsperulite crystalline texture can be seen at all the annealing temperatures, which contains a large number of crystals with reduced sizes. The unusual optical sign of negative spherulite had also been observed in poly(L-lactide)-block-poly(ε-caprolactone) (PLLA-b-PCL)^25–27 and PEO-b-PB diblock copolymers.¹⁷ They attribute this phenomenon to the small PE crystalline lamellae and PEO spherulites in the one-step crystallization process (I), amorphous PE and early-formed small PE crystalline lamellae adjacent to PEO only slightly hinder the diffusion and forming of PEO chain crystalline lamellae, and the PEO crystallization continuously propagates across the microdomain of the amorphous PE phase to form a regular spherulitic structure. On the contrary, in the sequential crystallization process (II), characteristic Maltese cross and regular radial spherulitic textures have not been detected in Fig. 3(g)–(i), which confirms the formation of the crystalline texture. This is due to the interdigitated large PE and PEO chain crystalline lamellae, the rigid crystalline PE lamellae restrict the mobility of the PEO chains within the microdomains and hinder the formation of a regular spherulitic structure of PEO.¹⁶ Finally, in this thermal procedure, the PEO-b-PE thin film exhibits entirely different crystallization morphology.

The SEM micrographs of the etched spherulite surfaces at 31 °C in the two procedures are shown in Fig. 4. In the one-step crystallization process (I), the spherulite boundary and the radial stripes can be observed clearly (Fig. 4(a)), while in the sequential crystallization process (II), there are no obvious spherulite boundaries, but only random stripe segments exist (Fig. 4(b)). According to previous research,^18,20–22 these significant differences in morphology originate from the topological restriction effect of the PE crystals.

Fig. 4 SEM image of etched spherulite surfaces of E15EO47 crystallized at 31 °C in (a) the one-step crystallization process (I) and (b) the sequential crystallization process (II), with the magnification factor, 1.0k×.

3.2 Crystalline state phase behavior

The ultimate crystal structures of E15EO47 cast films with different thermal histories were also investigated using synchrotron radiation WAXS/SAXS. In Fig. 5, different diffraction peaks for both the PEO and PE blocks can be detected, which correspond to the (120), (032) planes of PEO crystallites and the (110), (200) planes of PE crystallites, respectively. The PE (110) and (200) reflections, being at q = 15.28 and 18.56 nm⁻¹ respectively, indicate the existence of a conventional orthorhombic structure. The PEO (120) and (032) reflections, being at q = 13.56 and 16.48 nm⁻¹ respectively, indicate the existence of a monoclinic phase. The four typical peaks suggest that both blocks are able to crystallize and form separated crystalline phases. Here we take PE (110) peak intensity as reference to evaluate the variation of PEO reflections with temperature. It can be seen that the PEO (120) and (032) peak intensities decrease with an increasing crystallization temperature in the one-step crystallization process (I), but remain at roughly the same value in the sequential crystallization process (II).

Fig. 5 Synchrotron radiation WAXD patterns of E15EO47 co-oligomer obtained in the one-step crystallization process (I) and the sequential crystallization process (II) at 37, 41, 47, and 51 °C, respectively.

The unit-cell structure parameters (a, b, c) can be easily determined through inserting the values of λ (1.542 Å) and the main peak positions of the reflection planes into the formula of the inter-planar spacing of the (hkl) reflection planes, which is given by³⁴

for PE belonging to the orthorhombic system, and for PEO belonging to the monoclinic system

In eqn (8), the value of β is equal to 125.4°.³⁴ So the unit-cell parameters of PE crystal are a = 6.62 Å, b = 4.96 Å; the unit-cell parameters of PEO crystal are a = 8.02 Å, b = 12.95 Å, c = 1.945 Å. It is found that the unit-cell parameters of PE and PEO crystals remain invariant in the two thermal procedures, indicating that the topological restriction of microphases may lead to imperfect lattice orientation, but their crystal forms remain identical with that of the PE and PEO homopolymers.

The ordered nanostructures in E15EO47 were investigated by synchrotron radiation SAXS. The SAXS profiles of the co-oligomer in different thermal procedures are given in Fig. 6(a). To verify that the domain coalescence is responsible for the first stage of the perturbation of SAXS profiles (i.e., smearing of the lamellar form factor), we calculated the maximum size of the nano-scale morphologies from the three- dimensional correlation function defined as³⁵

Fig. 6 (a) Lorentz-corrected synchrotron radiation SAXS profiles for E15EO47 co-oligomer. Insets show the enlarged areas of the main lines from q = 0.35–0.6 nm⁻¹ magnified by a factor of 10. The vertical bars in the insets denote the positions of the main peaks. (b) Correlation functions, γ₁(r), in the one-step crystallization process (I) and the sequential crystallization process (II) at 37, 41, 47, and 51 °C, respectively. The straight vertical line denotes the horizontal ordinate of the first peak at 15.9 nm.

The corresponding SAXS profiles show distinct results for the two different pathways. For the one step crystallization (I), the SAXS profiles change remarkably with crystallization temperature. A sharp peak (q = 0.38 nm⁻¹) is accompanied by weak peaks (shown in the four enlarged insets in Fig. 6(a)), the scattering peaks of the diblock co-oligomer situating at q/q* values of 1 : 2 : 3 : 4 (q* = 0.38 nm⁻¹).

This can be attributed to the super-lattice scattering peaks from lamellar microdomains.³⁶ The primary scattering peak of the sample treated in both processes does not shift to larger q value position with increasing crystallization temperature, but broadens to some extent, which indicates that the average distance between neighboring microdomains changes little with supercooling. The crystalline phase of PEO chains in the high temperature region are imperfect, judging from the relatively broad spectral bands. On the other hand, it becomes more regular in the high temperature region because the width of the spectral bands becomes narrow. As Sun et al.^14,37 and Schulze et al.²⁰ predicted, the extended chain length is

where n_E and n_EO are the numbers of E and EO mers and c_PE and c_PEO represent the length of the c-axes of the orthorhombic PE (0.2546 nm) and the monoclinic PEO (72 helix, 1.948 nm), respectively. Furthermore, ϕ_E is the average tilting angle of PE chains (22°) with respect to the lamellar normal. Thus, the thickness of any realized film is an integral multiple of the theoretical extended chain length of 16.6 nm, with relative length contributions of the PEO and the PE of 13.1 and 3.5 nm, respectively.²⁰

In our experiment, we measured the chain length in E15EO47 to be L = 15.9 ± 0.5 nm. The correlation functions corresponding to the two morphologies are shown in Fig. 6(b). The chain length value is close to our previous study.²⁰ It is noteworthy that the minor difference of the L values between the value in our study and ref. 20 stems from the differences of thermal treatment condition and residence time at elevated temperature for phase separation and crystallization in thermal processes.^12,13,23 Consequently, various spherulite patterns can be achieved through pathway mediation, such as multi-stage crystallization control.

3.3 Crystallization kinetics of granular spherulites in situ

To investigate the crystallization and morphology development of the E15EO47, sample films were studied by optical microscopy under cross-polar conditions in situ. Two representative crystal morphologies in the one step crystallization process (I) and the sequential crystallization process (II) at 33 and 41 °C are shown respectively in Fig. 7 and 8.

Fig. 7 Spherulite morphology at 33 °C with increasing time in the one-step crystallization process (I), and the sequential crystallization process (II).

Fig. 8 Spherulite morphology at 41 °C with increasing time in the one-step crystallization process (I), and the sequential crystallization process (II).

Tiny granular crystals observable in the dark molten regions are due to the PE crystals. The dark regions around the tiny granular crystals correspond to the PEO segments and part of the PE segments in the molten state. The granular bright regions contain crystalline PEO and appear to grow from a nucleus located at the center.²¹ Similar granular aggregates were also observed in poly(p-dioxanone)-block-poly(ε-caprolactone) (PPDX-b-PCL) diblock copolymers at low supercoolings.²⁴ Fig. 9 summarizes the kinetic behavior for different crystallization temperatures. It can be seen that the diameters of the granular spherulites are linearly increasing with time. They further suggest that the induction periods in the one-step crystallization (I) are longer than in the sequential crystallization (II) at the same T_C, taking the different time stamps given in the insets into account. The growth speed (increase of the diameter) of emerging spherulites is constant for both processes because it depends only on the PEO crystallization. At high supercoolings, the size depends on the existence of nuclei which is time dependent. While at low supercoolings, the size is confined by the process for sequential crystallization (II) because of afore crystallized large PE lamellae, so the growth speed slows down.

Fig. 9 Spherulite growth diameters increased linearly with time at a long temperature range both in the one-step crystallization process (I) and the sequential crystallization process (II).

The enthalpy derived from DSC thermograms was used to obtain the relative degree of crystallinity, X(t):^15,22

where dH denotes the measured enthalpy of crystallization during an infinitesimal time interval dt. The limits t and infinity are used to denote the elapsed time during the course of crystallization and at the end of the crystallization process, respectively. The calorimetric experiment was conducted by monitoring the endothermal peak of the PEO block. There is a progressive increase of ΔH_t with the increase of time and a saturation time (ΔH_∞). The relative degree of crystallinity is plotted in Fig. 10 and can be obtained from the Avrami equation:¹⁵

log{−ln[1 − x_c(t)]} = log k + n log t (13)

where x_c(t) is the normalized degree of crystallinity as a function of time t; n is the Avrami exponent relating to the nucleation mechanism as well as the growth geometry, and k, the overall crystallization rate constant containing contributions from both nucleation and growth is calculated by fitting the experimental data using the double logarithmic form of eqn (12).¹⁸

Fig. 10 Relative degree of crystallinity of the PEO block evolving with time at different crystallization temperatures: (a) one-step crystallization process (I); (b) sequential crystallization process (II) at 35 and 39 °C, respectively.

In the one-step crystallization process (I), a smooth sigmoidal shape is detected, and the Avrami index n is approximately 2, indicating the crystallization of PEO is a nucleation-determined procedure, as shown in Fig. 10. The PEO spherulite fronts break through the PE microdomains to form negative spherulites, meanwhile the only few PE crystals are constrained to the gaps between the PEO crystal stripes. A similar phenomenon was obtained for PPDX-b-PCL diblock copolymers^22,38 and PE-b-PS-b-PEO triblock copolymers.³⁹ However, in the sequential crystallization process (II), the Avrami index n changes to 1.5 in the later stage of crystallization. This suggests that the crystallization of PEO experiences a transition from a nucleation- to growth-determined procedure. At the initial stage, the crystallization conditions of process (I) and (II) are nearly the same. The crystallization of the PEO chains is impaired by energy consumption necessary to break out of the established microdomains and exclude the small crystalline lamellae. At a later stage, the crystallization condition for process (II) changes, part of the PEO chains are crystallizing between the large PE crystal lamellae, the PEO chains have to overcome more energy consumption to exclude large PE crystalline lamellae. The large PE crystalline lamellae will hinder the PEO spherulite formation. In contrast, the small PE crystallites will not hinder the PEO spherulite formation so much as in one step crystallization (I). So the topological restriction is stronger than that in the initial stage of sequential crystallization (II) as well as in one step crystallization (I).

Fig. 11 shows an in situ WAXS analysis of the time dependence of the scattering profiles of E15EO47 for the two isothermal crystallization processes. It can be observed in Fig. 11(a) that, in the one-step crystallization process (I), the (110) peak of the orthorhombic PE crystal starts to appear one minute after the isothermal crystallization of PEO starts, while the (120) and (032) peaks of the monoclinic PEO crystal begin to appear 3 minutes later. At the same time, with increasing crystallization time, tiny shifts in peak positions from q = 13.8 to q = 13.6 nm⁻¹ for PEO (120) and from q = 16.8 to q = 16.5 nm⁻¹ for PEO (032) occur simultaneously, pointing to a continuous increase of the a, b, and c lattice parameters of the PEO unit cell in the diblock co-oligomer during the crystallization process. In the sequential crystallization process (II), as shown in Fig. 11(b), PE chains begin to crystallize when the temperature decreases from 120 °C to 80 °C, and the intensity of (110) peak increases gradually in the first three minutes and remains constant in the following time, indicating that the PE segments crystallize sufficiently. The (120) and (032) peak of PEO crystals, however, shift to low q value in two minutes, indicating that a, b, and c dimensions of the PEO unit cell change rapidly in this process. The temperature dependence of lattice parameters is essentially due to some monomer segments are incorporated in the unit cells with increasing temperatures.⁴⁰

Fig. 11 Time-dependent WAXS profiles obtained by in situ analysis through (a) the one-step crystallization process (I), and (b) the sequential crystallization process (II) at 39 °C, respectively.

In one-step crystallization process, low q value of PE (110) peak can be observed at the first stage of the PEO crystallization. After 7 min, the PE (110) reflection (q = 14.8 nm⁻¹) transformed into a larger and a smaller peak reflections, then gradually changed to one higher reflection. This means that there exist crystal form transitions of PE crystals accompanying the PEO crystallization. While in sequential crystallization process, the PE (110) reflection (q = 14.8 nm⁻¹) directly shifts to a higher q value position, whose shift extent is smaller than that in one-step crystallization process. This means that PE crystals experienced a small structure transform.

The peak shift behavior observed in E15EO47 was also reported for the rotator phase of n-alkanes with 20 < n < 26 during cooling. While the n-alkanes are cooled from the melt, the hexagonal rotator phase (R II) initially appears, followed by the transition to the distorted orthorhombic rotator (R I) phase. The hexagonal RII phase is characterized by a single peak in this q range. The RI phase is characterized by two principal peaks in this range.⁴¹ Only one scattering peak (100 reflection) is observed in the R II phase, while two peaks (110 and 200 reflections) are observed in the R I phase, where the respective peaks of 110 and 200 reflections appear at smaller and wider scattering angle than that of 100 reflection in the hexagonal phase. The WAXS peak positions of the 110 and 200 reflections in the R I phase drastically shift in opposite directions during the cooling, and the lattice size difference between the a-axis and b-axis decreases, to optimize the lattice structure.⁴² Furthermore, the dynamics of E15EO47 in one-step crystallization is similar to that of the rotator phase in n-alkanes. Initial transient formation of the hexagonal phase, which is similar to the R II phase, is followed by gradual conversion to a distorted orthorhombic phase, which is similar to the R I phase, and then changed into more stable R II phase, to optimize the lattice structure.

Interestingly, in both two thermal procedures, a sharp reduction in the peak intensity of the crystalline PE (110) reflection can be seen as PEO crystallizes. The crystallization of PEO was followed by monitoring the intensity of the (120) and (032) reflection of the PEO monoclinic crystal. This suggests a reduction in crystallinity of PE as PEO crystallizes. It is possible that to accommodate the crystallization of PEO, some rearrangement of the PE chains is necessary, which is accomplished by local melting. Moreover PE lamellae will be destroyed by PEO crystallization which enables spherulite growth. In all these cases the lamellar period remains constant, only interdigitated extended chain lamellae can be formed. Nonetheless, the lateral dimensions of the PE lamellae are important for the confined spherulite growth which can be influenced by the crystallization processes. Moreover, the peak intensity of PE (110) at the beginning of the PEO crystallization in one-step crystallization process is higher than that in sequential crystallization process. At the same crystallization temperature, PE crystal lamellae thickness will be bigger with the enlargement of crystallization time. So the PE crystal lamellae thickness in sequential crystallization (II) is larger than that in one-step crystallization process (I).²⁴ Here we suggest that in the one step crystallization process (I), when the temperature rapidly cooled to the crystallization temperature of PEO from the melting temperature (120 °C), most of the ethylene segments did not have time to adjust themselves to a stable state and were thus fixed without well-regulated packing. Therefore, at the initial stage of the one step crystallization, the majority of PE crystals are small and imperfect. When PEO segments start to crystallize, PE segments exerted a confinement effect on the diffusion and formation of PEO crystalline lamellae.^17,28–33 In the sequential crystallization process (II), the majority of PE chains could crystallize. Thus, the crystallization process of PEO was dominated by more restricted confinement of PE crystalline lamellae.

The formation of alternative lamellar nanostructures and continuous growth of crystallization superstructures mean that, in this weakly segregated BCP, the breakout appears when the crystallization driving force is large enough to overcome the energy barrier, and breakout crystallization is the dominated mode.³⁵ Breakout crystallization corresponds to the other extreme where crystallization is able to disrupt the melt mesophase and transforms it into an extended crystalline lamellar morphology.^43,44 When the crystallization occurs, the majority PEO microdomains coalescent with each other, the mesophases are generally destroyed and the microdomain structures change from original nanometer length scale domains to lamellae containing alternating PE and PEO layers.^35,43 According to the different crystallization kinetics and morphologies formation in the two thermal procedures, the mainly restrictions of PE block reflect in lowering even inhibiting the coalescence process of PEO microdomains.

Based on the above results, two corresponding spherulite patterns of E15EO47 with the two different thermal histories are schematically summarized in Fig. 12. In this model, the spherulite patterns are depicted in different crystallization time, e.g., the initial crystallization time, t₀ (a, a′), several minutes later, t₁ (b, b′), middle stage of spherulite growth, t₂ (c, c′), and final stage of crystallization, t₃ (d, d′), respectively. In the one-step crystallization (I), the negative spherulite with regularly folded PEO lamellae (L = 15.9 nm) are observed (Fig. 12(a)). In contrast, crystalline texture with the same L value forms in the sequential crystallization process (II), shown in Fig. 12(b). In the two thermal procedures, according to the theoretical calculation results, the relative length contribution of the PE block is 3.5 nm.

Fig. 12 Model of the E15EO47 co-oligomer's microstructure evolvement at 39 °C in: (A) the one-step crystallization process (I); and (B) the sequential crystallization process (II). The spherulite patterns are depicted in different temperature, e.g., 120 °C (a) and 80 °C (a′), and different crystallization time, e.g., the initial crystallization time, t₀ (a, a′), several minutes later, t₁ (b, b′), middle stage of spherulite growth, t₂ (c, c′), and final stage of crystallization, t₃ (d, d′). Yellow and blue are for PE and PEO crystalline lamellae, respectively.

In the thermal procedures, the many small and the large PE chain crystalline lamellae play a different role in confining the PEO crystallization in the one-step crystallization (I) and the sequential crystallization (II), to decelerate the crystallization growth rate of PEO segments.⁴⁵ In the one-step crystallization process (I), at 120 °C, both PE and PEO blocks are amorphous, so the whole samples are homogeneous. (Fig. 12(a)) As the temperature intensively decreased to the crystallization temperature of PEO block, only small and thin PE crystalline lamellae form prior to PEO crystallization and part of PE is amorphous. (Fig. 12(b)) When the temperature comes to the crystallization temperature of PEO block, PEO chains start to nucleate and fold chains in the presence of as-formed PE crystals. (Fig. 12(c)) Therefore, when PEO starts to crystallize, the small and thin PE crystalline lamellae serve as weak topological confinement and can be excluded from the crystallization front of PEO by overcoming a weaker energy barrier. With increasing of crystallization time, the extensive continuity of PEO aggregation across the crystal lamellae boundaries results in the formation of regularly folded lamellar nanostructures. When the temperature is low enough (e.g. 0 °C), this phenomenon is more obvious and negative spherulites can form. (Fig. 12(d)) During the crystallization of PEO, residual amorphous PE crystallizes between the PEO crystal lamellae and part of the early-formed PE crystals melt to form more perfect crystalline lamellae.

In the sequential crystallization process (II), the PE chains have enough time to crystallize at 80 °C and a lot of large and thick PE crystalline lamellae have formed before the PEO starts to crystallize. (Fig. 12(a)) When the temperature arrives at the crystallization temperature of PEO block, some small and imperfect PE crystals formed. (Fig. 12(b)′) As a result, the PE crystalline lamellae serve as strong topological confinement to restrict PEO crystallization. The topological restriction encountered by PEO chains is strengthened compared to that in the one-step crystallization process (I). When PEO crystallizes, it is very difficult to exclude large PE crystalline lamellae from the crystallization front of PEO because disrupting the preexisting large PE crystalline lamellae consumes more energy. Since the driving force of PEO crystallization is not strong enough to allow the PEO lamellae to continuously propagate through the microdomains formed by the PE crystals, the extensive continuity of PEO aggregation across the crystal lamellae boundaries is destroyed and interdigitated PE and PEO crystalline lamellae form. (Fig. 12(b)) When the temperature comes to the crystallization temperature of PEO block, PEO chains start to nucleate and crystallize. (Fig. 12(c)′) When PEO crystallizes, it is very difficult to exclude large PE crystalline lamellae from the crystallization front of PEO because disrupting the preexisting large PE crystalline lamellae consumes more energy. Even at 0 °C the PEO crystallization cannot rewrite the microphase separation. (Fig. 12(d)′) Understanding the evolution of phase morphology and crystal structure as function of the thermal treatment in this co-oligomer system may help to create advanced amphiphilic nanopatterned surfaces interesting as templates for various applications as, e.g., in the biomaterials field as well as for nanotechnology applications.

4 Conclusions

In this work, two pathways were adopted to investigate the crystallization behavior of PEO-b-PE drop-cast films. With the help of DSC, POM, SEM and synchrotron radiation WAXD and SAXS, the evolvement of microstructure and phase morphology of this co-oligomer with different thermal histories were systematically elucidated. It was found that negative spherulite with radial stripes and crystalline texture with random stripe segments occurred in the one-step crystallization process (I) and the sequential crystallization process (II), respectively. Furthermore, in the one-step crystallization process (I), the crystallization of PEO is a nucleation determined procedure, accompanied by an increase in a, b, and c dimensions of the PEO unit cell corresponding to the increase of crystallization time. Several small PE crystalline lamellae slow down the PEO crystal growth speed. In contrast, in the sequential crystallization process (II), the crystallization of PEO is a nucleation-to-growth determined procedure, in which a, b, and c dimensions of the PEO unit cell shift faster to low q value with the increase of time than that in one-step crystallization process (I). Both large and small PE crystalline lamellae impair the PEO crystal growth. In addition, in both thermal procedures, to accommodate the crystallization of PEO some rearrangement of the PE chains is necessary, which is accomplished by local melting. Consequently, the newly formed PEO crystals overwrite parts of the preexisting or just formed PE crystals.

Based on the current findings, we conclude that the morphology evolvement and multi-scale structure formation of the asymmetric diblock copolymers are closely bound up with their temperature scales. Hence, the microstructures and phase morphologies of amphiphilic copolymers could be mediated by adjusting their thermal pathways. This work may introduce a new way in surface modification, micro-fabrication and other correlating fields to obtain block copolymers thin films with diverse nano-patterned structures, which could be used as templates for various biological applications.

Acknowledgements

The authors gratefully acknowledge research support from the Alexander von Humboldt Foundation, Bonn, Germany, through a Humboldt Research Fellowship for Experienced Researchers, Beijing New-Star Program of Science and Technology (2009B10), Fundamental Research Funds for the Central Universities (project no. ZZ1307).

Notes and references

1. F. S. Bates, Annu. Rev. Phys. Chem., 1990, 41, 525–527.
2. Z. G. Wang, Curr. Opin. Colloid Interface Sci., 1998, 3, 428–435.
3. G. V. Jensen, Q. Shi, G. R. Deen, K. Almdal and J. S. Pedersen, Macromolecules, 2012, 45, 430–440.
4. H. L. Chen, S. C. Hsiao, T. L. Lin, K. Yamauchi, H. Hasegawa and T. Hashimoto, Macromolecules, 2001, 34, 6154–6164.
5. X. Shuai, H. Ai, N. Nasongkla, S. Kim and J. Gao, J. Control. Release, 2004, 98(3), 415–426.
6. P. A. George, B. C. Donose and J. J. Cooper-White, Biomaterials, 2009, 30, 2449–2456.
7. T. F. Keller, M. Grosch and K. D. Jandt, Macromolecules, 2007, 40, 5812–5819.
8. T. F. Keller, J. Schoenfelder, J. Reichert, N. Tuccitto, A. Licciardello, G. M. L. Messina, G. Marletta and K. D. Jandt, ACS Nano, 2011, 5, 3120–3131.
9. T. F. Keller, M. Mueller, W. Ouyang, J.-T. Zhang and K. D. Jandt, Langmuir, 2010, 26, 18893–18901.
10. M. Omichi, K. Kadowaki, B. H. Kim, S. O. Kim, I. Maruyama and M. Akashi, Chem. Commun., 2010, 46, 1911–1913.
11. S. B. Myers and R. A. Register, Macromolecules, 2008, 41, 6773–6779.
12. W. Cao, K. Tashiro, M. Hanesaka, S. Takeda, H. Masunaga, S. Sasaki and M. Takata, J. Phys. Chem. B, 2009, 113, 8495–8504.
13. W. Cao, K. Tashiro, M. Hanesaka, S. Takeda, H. Masunaga, S. Sasaki and M. Takata, J. Phys. Chem. B, 2009, 113, 2338–2346.
14. L. Sun, Y. Liu, L. Zhu, B. S. Hsiao and C. A. Avila-Orta, Macromol. Rapid Commun., 2004, 25, 853–857.
15. G. Floudas and C. Tsitsilianis, Macromolecules, 1997, 30, 4381–4390.
16. S. Hong, W. J. MacKnight, T. P. Russell and S. P. Gido, Macromolecules, 2001, 34, 2876–2883.
17. S. Hong, L. Z. Yang, W. J. MacKnight and S. P. Gido, Macromolecules, 2001, 34, 7009–7016.
18. H. L. Chen, J. C. Wu, T. L. Lin and J. S. Lin, Macromolecules, 2001, 34, 6936–6944.
19. M. S. Shoichet, S. R. Winn, S. Athavale, J. M. Harris and F. T. Gentile, Biotechnol. Bioeng., 1994, 43, 563–572.
20. R. Schulze, M. M. L. Arras, G. Li Destri, M. Gottschaldt, J. Bossert, U. S. Schubert, G. Marletta, K. D. Jandt and T. F. Keller, Macromolecules, 2012, 45, 4740–4748.
21. J. Albuerne, L. Marquez and A. J. Muller, Macromolecules, 2003, 36, 1633–1644.
22. A. J. Muller, J. Albuerne, L. Marquez, J. M. Raquez, P. Degee, P. Dubois, J. Hobbs and I. W. Hamley, Faraday Discuss., 2005, 128, 231–252.
23. W. Cao, K. Tashiro, M. Hanesaka, S. Takeda, H. Masunaga, S. Sasaki and M. Takata, J. Phys.: Conf. Ser., 2009, 184, 1–4.
24. S. Z. D. Cheng, J. H. Chen, J. S. Barley, A. Q. Zhang, A. Habenschuss and P. R. Zschack, Macromolecules, 1992, 25, 1453–1460.
25. W. Shi, J. Yang, Y. Zhang, J. Luo, Y. Liang and C. C. Han, Macromolecules, 2012, 45, 941–950.
26. I. W. Hamley, V. Castelletto, R. V. Castillo, A. J. Muller, C. M. Martin, E. Pollet and Ph. Dubois, Macromolecules, 2005, 38, 463–472.
27. 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.
28. S. P. Gido and E. L. Thomas, Macromolecules, 1994, 27, 6137–6144.
29. S. P. Gido, J. Gunther, E. L. Thomas and D. Hoffman, Macromolecules, 1993, 26, 4506–4520.
30. M. C. Newstein, B. A. Garetz, H. J. Dai and N. P. Balsara, Macromolecules, 1995, 28, 4587–4597.
31. S. P. Gido, D. W. Schwark, E. L. Thomas and M. Goncalves, Macromolecules, 1993, 26, 2636–2640.
32. S. P. Gido and E. L. Thomas, Macromolecules, 1994, 27, 849–861.
33. E. Burgaz and S. P. Gido, Macromolecules, 2000, 33, 8739–8745.
34. C. I. Huang, S. H. Tsai and C. M. Chen, J Polym Sci Part B: Polym Phys, 2006, 44, 2438–2448.
35. J. Y. Hsu, I. F. Hsieh, B. Nandan, F. C. Chiu, J. H. Chen, U. S. Jeng and H. L. Chen, Macromolecules, 2007, 40, 5014–5022.
36. N. V. Salim, T. Hanley and Q. P. Guo, Macromolecules, 2010, 43, 7695–7704.
37. L. Sun, Y. Liu, L. Zhu, B. S. Hsiao and C. A. Avila-Orta, Polymer, 2004, 45, 8181–8193.
38. A. J. Muller, J. Albuerne, L. M. Esteves, L. Marquez, J. M. Raquez, P. Degee, P. Dubois, S. Collins and I. W. Hamley, Macromol. Symp., 2004, 215, 369–382.
39. A. Boschetti de Fierro, A. T. Lorenzo, A. J. Mueller, H. Schmalz and V. Abetz, Macromol. Chem. Phys., 2008, 209(5), 476–487.
40. Y. Mao, X. Li, C. Burger, B. S. Hsiao, A. K. Mehta and A. H. Tsou, Macromolecules, 2012, 45, 7061–7071.
41. H. Gang, O. Gang, H. H. Shao, X. Z. Wu, J. Patel, C. S. Hsu, M. Deutsch, B. M. Ocko and E. B. Sirota, J. Phys. Chem. B, 1998, 102, 2754–2758.
42. Y. Nozue, Y. Kawashima, S. Seno, T. Nagamatsu, S. Hosoda, E. B. Berda, G. Rojas, T. W. Baughman and K. B. Wagener, Macromolecules, 2011, 44, 4030–4034.
43. P. Zhang, H. Huang, D. Yan and T. He, Langmuir, 2012, 28(15), 6419–6427.
44. Y. Loo, R. A. Register and A. J. Ryan, Macromolecules, 2002, 35, 2365–2374.
45. M. C. Lin, H. L. Chen, W. B. Su, C. J. Su, U. S. Jeng, F. Y. Tzeng, J. Y. Wu, J. C. Tsai and T. Hashimoto, Macromolecules, 2012, 45, 5114–5127.

---

Footnote

† Present address: Deutsches Elektronen-Synchrotron DESY, Notkestr. 85, D-22607 Hamburg, Germany.

---