Abstract
Although a great deal of unique nanostructures were already obtained from polymer self-assemblies in terms of conventional parameters, the self-assembly under the confinement is still not well understood. Here, dissipative particle dynamics simulations were used to explore the self-assemble behaviors of linear ABC triblock copolymers under rigid spherical confinements. First several unusual morphologies, such as multilayer onion, coupled helix, and stacked lamella, were distinguished from the total 210 simulations. Second, the influences of three important parameters (block sequence, wall selectivity, and spherical radius) on the morphologies were discussed in detail. Finally, the dynamics evolution of several typical aggregates was examined. This simulation enriches micelle morphologies for the self-assembly of linear ABC triblock copolymers under rigid spherical confinements and is helpful to understand the formation of valuable nanostructures from linear ABC terpolymers.
1 Introduction
Rich nanostructures from the self-assembly of block copolymers (BCPs) have gained tremendous scientific interests because of their potential applications, such as in drug delivery, microelectronic materials, advanced plastics, and so on (1), (2), (3), (4). Recently, it was reported that the confined assembly of BCPs can provide a powerful strategy to manipulate the shape and internal structure of nanomaterials (5), (6), (7), (8), (9), (10), (11), (12), (13), (14), (15), (16), (17), (18). Considering the key role of shape and internal structure in the function of polymer particles, how to systemically control their morphologies is still of great significance. However, confined spaces unavoidably introduce new parameters, such as space size, shape, and wall selectivity, to influence the morphology of BCPs, which brings the tremendous challenge for experimental works. In fact, the simulation techniques are the good supplement. Recently, we have successfully used dissipative particle dynamics (DPD) simulations to explore the self-assemble behavior of linear ABC (l-ABC) triblock copolymers in the dilute solution and shear condition (19), (20). Here, we continually utilized DPD methods to investigate the self-assembly of linear ABC terpolymers in three-dimensional (3D) spherical confinements.
Then we briefly summarized recent experimental and simulation works on the assembly of BCPs in confined spaces. Experimentally, in the one-dimensional (1D) confinement, Ludwigs et al. (21) found highly ordered hexagonal perforated lamella from the self-assembly of poly(tert-butyl methacrylate)-poly(2-vinylpyridine)-polystyrene triblock copolymers. Tang et al. (22) found half-sphere, half-sphere plus whole sphere, half-sphere plus two whole sphere, and coexistence of regions of hexagonal and square packing from the triblock copolymer of poly(ethylene oxide)-poly(methyl methacrylate)-polystyrene. In the two-dimensional (2D) cylindrical confinement, Dobriyal et al. (23) observed concentric cylinders, single-, double-, triple-helical, and torus-like morphologies from polystyrene-polybutadiene diblock copolymers. In the 2D confinements (spherical or polyhedral), Arsenault et al. (24) investigated the self-assembly of polystyrene-poly(ferrocenylethylmethylsilane) diblock copolymer and predicted the lamellae orthogonal to the sphere surface because of the strong confined influence. Xu et al. (25) studied the self-assembly of polystyrene-polyisoprene-poly(2-vinylpyridine) terpolymers and found several novel nanostructures, including pupa-like, core-shell, and onion-like stuctures. The more works can refer to the review by Yabu and coworkers (26).
On the other hand, a series of theoretical methods, including Monte Carlo (MC), simulated annealing, self-consistent field theory (SCFT), DPD, and so on, were utilized to explore the self-assembly of BCPs. For example, Geisinger et al. (27) studied the self-assembly of symmetric diblock copolymers by using SCFT and MC simulations under 1D confinement. Yang et al. (28) used real-space SCFT to search possible morphology of an asymmetric diblock copolymer confined in two homogeneous hard walls and found several interesting structures, including parallel cylinder, perpendicular cylinder, flat lamellae, undulated cylinders, and lamellae. Ludwigs et al. (21), (29) coarsed polystyrene-poly(2-vinylpyridine)-poly(tertbutylmethacrylate) triblock copolymers as Gaussian chain of A3B4C12 and used SCFT methods to systematically explore the influence of film thickness, surface field, and interaction parameters between the different polymer components on the phase behavior. Chen and Fredrickson (30) also used the same method to investigate symmetric ABC terpolymers confined in thin film conditions and predicted parallel and perpendicular cylinders. Moreover, Li et al. (31) applied SCFT method to study the assembly of symmetric ABC star terpolymers in spherical cavity. Spherically concentric lamellae and ringlike structures were also observed in their study. Arai et al. (32) utilized DPD simulations to study the self-assembly of Janus nanoparticle of ABC terpolymers under nanotube confinement and gained double-helix and triple-helix nanostructures. Chi et al. (33) have systematically investigated the self-assembly of diblock copolymers under soft confinements by simulated annealing methods. Based on MC methods, Sheng et al. (34) studied the self-assembly of ABA terpolymer in 3D confinements and observed double-helixes, stacked rings, and cagelike morphologies. By DPD simulations, Chang et al. (35) studied the phase behaviors of confined linear triblock copolymers and found that helical nanostructures can be induced by blending non-helix-forming terpolymers with homopolymers or diblock copolymers.
Although a lot of experimental and theoretical works have been performed, it is clear that the self-assembly of linear ABC triblock copolymers in 3D rigid confinement (spherical cavity) is still rare. In our previous study, we explored the morphology of linear ABC triblock copolymers under the shear rate, solutions, and different block sequences by DPD simulations (19), (20), (36). Here, we further utilized DPD simulations to investigate the self-assembly of linear ABC triblock copolymers in rigid spherical confinements.
2 Simulation details
DPD is a coarse-grained simulation technique, such as the common molecular dynamics (MD), however, which allows the larger length and longer time scale (37). Coarse-grained particles in DPD represent a block or cluster of atoms or molecules, and their evolutions obey Newton’s equation of motion. The force between pair nonbonded DPD particles includes a conservative force
where rij =ri −rj , rij =|rij |, eij =rij /rij , and vij =vi −vj . ζij is a Gaussian random number with zero mean and unit variance. αij is the repulsion parameter between bead i and j, which reflects the chemical characteristics of interacting beads. γ is the friction constant, and σ is the noise strength. To ensure the simulation system to satisfy the fluctuation-dissipation theorem and the Gibbs Canonical ensemble, only one of the two weight functions wD and wR can be chosen arbitrarily and this choice fixes the other one. There is also a relation between amplitudes (σ and γ) and kBT. It is wD =(wR )2 and σ2=2γkBT, where kB is the Boltzmann constant and T is the temperature (38). Generally, the simple forms for wC =wD =(wR )2=(1−rij )2 and σ=3 (i.e. γ=4.5) are chosen, and Newton equations for all paritcles are integrated by a modified version of the velocity-Verlet algorithm with λ=0.65 (39). lij is the bond length between connected two bead i and j. Here, the spring coefficient kS =4 and the balance bond length l0=0 are chosen. For easy numerical handling, the cutoff radius (rc ), the bead mass (m), and the temperature (kBT) are chosen as the unit of the simulated system.
In this work, the morphology of linear ABC triblock copolymers confined in a 3D spherical cavity is investigated. As shown in Figure 1a, l-ABC, l-BAC, and l-ACB consist of A, B, and C blocks according to the different sequences, x, y, and z denote A block with x beads, B block with y beads, and C block with z beads, respectively. The cavity wall is also constructed by DPD beads (denoted by W), and the radius R of the spherical cavity is chosen as 60 and 80 in DPD unit (Figure 1b). In detail, the cavity wall is modeled by a set of fixed W beads, which are placed randomly with the high number density of ρ=6 for preventing the transmission of BCPs. Triblock copolymers are stuffed into the spherical cavity with random distribution at ρ=3. The repulsive parameters in Table 1 for the conservative force between DPD particles still follow those used by our previous works (19), (36), (40), which are defined to describe the self-assembly of the experimental miktoarm terpolymers (41). Note that A, B, and C-blocks are incompatible to one another and αiW depicts the interaction between bead i and the cavity wall (W). In the initial stage, B-blocks are absolutely abstracted by the wall, and C-blocks are strongly repelled from the wall. Afterward, the additional parameters of αAS=25 and 120, αBS=120, and αCS=25 are also chosen to check the influence of αiW on the morphologies of linear ABC terpolymers. In this work, the time step Δt=0.04 is chosen to obtain the better control of the temperature fluctuation of simulation systems, and a total of 2×106 DPD steps are carried out to guarantee the equilibration for each system.

The coarse-grained model of linear ABC triblock copolymers with different block sequences (a), the spherical cavity (b), and the simulation box with the initial constitution (c).
Repulsion parameters αij (DPD unit) in this work.
αij | A | B | C | W |
---|---|---|---|---|
A | 25 | |||
B | 45 | 25 | ||
C | 90 | 75 | 25 | |
W | 50 (25, 120) | 25 (120) | 120 (25) | 25 |
3 Results and discussions
Clearly, the morphology of linear ABC triblock copolymers in 3D rigid confinements (spherical cavity) is influenced by a series of controlling parameters, such as block sequences (ABC, BAC, and BCA), interactions between terpolymers and wall of cavity (αiW), block lengths (x, y, z), the radius of spherical cavity (R), and so on. Considering the combination of the different parameters, they constitute a tremendous parameter space. It is not possible to explore the effect of all the factors. Hence, we first fixed the interactions between different blocks, that is, αAB, αAC, and αBC (see Table 1). Then we also considered the variation of other parameters, including three block sequences (l-ABC, l-BAC, and l-BCA), seven block lengths (x-y-z is equal to 2-2-2, 2-2-8, 2-8-2, 2-2-8, 8-2-2, 8-2-8, and 8-8-2, respectively), two radii of rigid spherical cavity (R=60 and 80), and five sets of αiW (αAW=50, αBW=25, αCW=120; αAW=25, αBW=αCW=120; αBW=25, αAW=αCW=120; αCW=25, αAW=αBW=120; and αAW=αBW=αCW=120). In summary, there are a total of 210 simulations that are considered in this work.
3.1 Block sequence
Generally, the effect of block sequence on morphologies for multiblock copolymers is vital. Therefore, the morphology for linear ABC terpolymers with three difference sequences (A-B-C, B-A-C, and B-C-A) is examined under the following stationary conditions. Two sets of parameters for the selectivity of confined wall are fixed as αAW=50, αBW=25, αCW=120, and αAW=αBW=αCW=120, respectively. The interaction parameters between different blocks are listed in Table 1. The radius of spherical cavity is R=60. First Figure 2 gives a series of morphologies from symmetrical A2B2C2, B2A2C2, and B2C2A2 triblock copolymers at two sets of interaction parameters (αiW). As for αBW(25)<αAW(50)<αCW(120) (Figure 2a), the sequence of B-A-C, that is, B2A2C2, can result in the formation of a normal onion-like corona-shell-core (CSC) structure. In detail, blocks B, A, and C aggregate into the corona, shell, and core, respectively. After exchanging the position of blocks A and C, B2C2A2 still aggregates into the normal CSC structure. The only difference is that the porous shell is from block C and the salient core is from block A. When block B is placed in the middle, that is, A2B2C2, the CSC structure disappears which replaced by coupling striped patterns from blocks A and C. It is interesting that block B forms the small core away from the confined wall, although the wall is partial to block B. Then the parameters of αBW=αAW=αCW=120 (Figure 2b) are used to weaken the effect of wall selectivity, in this way, blocks A, B, and C have the equal chance to form the inner core, in order to avoid the contact with the wall. The diagonal of Figure 2b clearly shows the different morphologies of block A, that is, the irregular core for A2B2C2, the helical structure for B2A2C2, and the lamella for B2C2A2. Under the condition of αBW=αAW=αCW=120, the effect of wall selectivity is eliminated; therefore, it thoroughly shows the effect of block sequences. Outwardly, the block sequence of block B in one end (B-A-C and B-C-A) can result in the similar morphologies, as compared with the other sequence of B in the middle (A-B-C). It is worth to note that the interaction parameters between different blocks (αAB=45, αBC=75, and αAC=90) also enter for the characteristic of morphologies.

Simulation morphologies for symmetrical A2B2C2, B2A2C2, and B2C2A2 triblock copolymers under two sets of interaction parameters of αiW as a function of block sequences. Pink, green, and blue represent A-, B-, and C-rich domains, respectively.
For several asymmetric triblock copolymers, we also compared the variation of morphologies caused by block sequences, and the results are depicted in Figure 3. Similarly, the discussion still begins with the first parameters of αBW(25)<αAW(50)<αCW(120). In Figure 3a, A8B8C2 and B8A8C2, with the position exchange of blocks A and B, show an interesting phenomenon, that is, the morphology of the exchanged blocks A and B has hardly any variation, and that of block C changes from several inlaid small dots to a whole core. The similar phenomenon is also found for A2B8C8 and A2C8B8 with the position exchange of B and C. B2A8C8 and B2C8A8 exhibit a slightly difference after switching the place of A and C. Block B actually becomes a core (B2C8A8) from a corona (B2A8C8). However, blocks A and C still occur a visible change, that is, a cage of block A around a polyhedral core of block C in B2A8C8 and the coupled blocks A and C. So we can know that the hydrophilic B can bring complex variations if it is short and at the end.

Simulation morphologies for asymmetrical A8B8C2 and B8A8C2, A2B8C8 and A2C8B8, B2A8C8 and B2C8A8 triblock copolymers under two sets of αiW as a function of block sequences. Pink, green, and blue represent A-, B-, and C-rich domains, respectively.
The second parameters of αBW=αAW=αCW=120 (Figure 3b) are also chosen to remove the effect of wall selectivity. First by comparing Figure 3a and b, the variation of block sequences under this set of interaction parameters brings the entirely different morphological evolution. Taking A8B8C2 and B8A8C2 as an example, blocks A and B exchange the position and, meanwhile also switch their morphological structures. It is different with the above-mentioned situation that they retained the respective morphological characteristic after exchanging the position. If we neglect the small fluctuation of morphologies caused by the different block interactions (i.e. αAB, αBC, and αCA), the similar conclusion can be obtained from A2B8C8, A2C8B8, B2A8C8, and B2C8A8.
3.2 Wall selectivity
Besides block sequences, the selectivity of confined wall also plays a significant role for the self-assembly under confined spaces. Combined with the fixed other parameters, we then tuned the parameter of αiW to build the different wall selectivity, which includes αAW=25, αBW=αCW=120 (means the wall prefers block A to other two blocks, defined as AW), αBW=25 αAW=αCW=120 (liking block B, BW), and αCW=25 αAW=αBW=120 (liking block C, CW), as shown in Figures 4–6. In fact, the different block lengths are also considered when building the phase diagrams for all linear ABC triblock copolymers. Therefore, the following discussion also includes the effect of block lengths.

Morphologies for all l-ABC triblock copolymers at αAW=25, αBW=αCW=120 (defined as AW), αBW=25 αAW=αCW=120 (BW), and αCW=25 αAW=αBW=120 (CW). Pink, green, and blue represent A-, B-, and C-rich domains, respectively.

Morphologies for all l-BAC triblock copolymers at αAW=25, αBW=αCW=120 (AW), αBW=25 αAW=αCW=120 (BW), and αCW=25 αAW=αBW=120 (CW). Pink, green, and blue represent A-, B-, and C-rich domains, respectively.

Morphologies for all l-BCA triblock copolymers at αAW=25, αBW=αCW=120 (AW), αBW=25 αAW=αCW=120 (BW), and αCW=25 αAW=αBW=120 (CW). Pink, green, and blue represent A-, B-, and C-rich domains, respectively.
Figure 4 first shows morphologies of all l-ABC terpolymers (2-2-2, 2-2-8, 2-8-2, 2-8-8, 8-2-2, 8-2-8, 8-8-2) in the rigid spherical cavity with R=60. On the whole, l-ABC terpolymers prefer to form the CSC structure, except for those at the second selectivity of BW. In other words, if the block liked by the confined wall locates in the middle of ABC terpolymers, they would generally possess the larruping rule of morphological variation. For instance, the morphologies of l-A2B2C2, l-A2B8C2, and l-A8B2C8 at the first selectivity (AW) are similar to those at the third selectivity (CW), but very different to those at the second one (BW). In detail, for l-A8B2C8 of AW parameters, block A forms the multiporous corona, block B forms the stripped shell, and block C aggregates the irregular core. For the same polymer, the wall selectivity of CW parameters can simply exchange the morphologies of blocks A and C and hardly alter the morphology of B. However, for BW parameters, the morphology, a patch of block C is sewn on block A by a string of block B, is far away from the normal CSC structure. In fact, block B with the good wall selectivity does not form the corona because it is too short compared with the other two blocks. For l-A2B8C2 of BW parameters, the longer B block is easier to build the cagelike corona of imprisoning several small spheres such as “fish” formed by blocks A and C.
As for all l-ByAxCz and l-ByCzAx triblock copolymers, the influence of the different wall selectivity on the morphology is also investigated, as drawn in Figures 5 and 6, respectively. Theoretically, the law of the two types of terpolymers should be similar to that of l-AxByCz. In Figure 5, B2A2C2 assembles the normal CSC structure at both BW and CW parameters and the coupled double-helix at AW parameters. l-B8A2C8 aggregates two helical strips from the longer blocks B and C that are sewn by the shorter block A. This morphology is very different to the CSC structures with the distinctly corona and core, which corresponds to the different BW and CW parameters, respectively. In addition for l-B2A8C2, the selectivity of AW parameters also brings an A-cage entangling several B- and C-“fishes”. Here, we can ensure that the law of wall selectivity shown by l-ByAxCz is consistent with that by l-AxByCz. In Figure 6, for symmetric l-B2C2A2, the selectivity of AW and BW parameters only can exchange the corona and core of CSC structures. The selectivity of CW parameters results in a rambling CSC aggregate, which is different with the helix formed by l-A2B2C2 and l-B2A2C2. The reason may be the asymmetrical block interactions, and block C has the large repulsive characteristic with blocks A and B. However, for the asymmetric l-B2C8A2 and l-B8C2A8, the effect of asymmetrical block interactions seems lost, and we still find the similar morphological evolution with other two types of terpolymers. Therefore, we do not describe the morphological details again.
3.3 Spherical radius
In the previous sections, we examined the effect of block sequence and wall selectivity. Here, the effect of radius of confined spherical cavity on self-assembled morphologies is considered. Two kinds of radii are used, that is, R=60 and 80. The corresponding morphologies for R=60 have been seen in Figures 2 and 3, and those for R=80 are shown in Figures 7 and 8 for comparisons.

Morphologies of three symmetrical A2B2C2, B2A2C2, and B2C2A2 terpolymers at R=80. Pink, green, and blue represent A-, B-, and C-rich domains, respectively.

Morphologies of six asymmetrical A8B8C2, B8A8C2, A2B8C8, A2C8B8, B2A8C8, and B2C8A8 terpolymers at R=80. Pink, green, and blue represent A-, B-, and C-rich domains, respectively.
We first compared the morphological diversity of symmetric terpolymers in Figures 2 and 7, respectively. In R=60 (Figure 2a), l-B2A2C2 forms the normal CSC structure when αAW=50, αBW=25, and αCW=120. However, when spherical radius increases from 60 to 80, its morphology shows a slight difference, as shown in Figure 7a. The routine corona becomes more scattered, and extra core also becomes the bigger than that in R=60. Besides, the A-shell and C-core also grow many porous. The previous phenomenon is also found for l-B2C2A2. l-A2B2C2 for R=80 (Figure 7a) has the thinner coupled strips than that for R=60 (Figure 2a). In Figure 7b (αAW=αBW=αCW=120), all three symmetric terpolymers form the coupled and independent helix that look like lamella from the whole morphology. Although Figure 2b shows a real lamella for l-B2A2C2, and the thick strips with a little link for l-A2B2C2, from the angle of the whole morphology, we can still think that when αAW=αBW=αCW=120, the radius of spherical confinement does not obviously change the morphology of symmetrical ABC triblock copolymers.
Second we also analyze the morphological difference of asymmetric ABC triblock copolymers, as shown in Figures 3 and 8. When the wall selectivity is fixed at αAW=50, αBW=25, and αCW=120, for example, we can clearly find that the morphology of l-B8A8C2 in R=80 (Figure 8a) shows an obvious change by comparing that in R=60 (Figure 3a). In detail, l-B8A8C2 in R=60 forms the normal CSC structure with the B-corona, A-shell, and C-core. However, when the radius increases from 60 to 80, block B becomes the corona to the irregular core (like a four-way connection), and blocks A and C become the cage shape that corresponds to the corona and the shell, respectively. The terpolymers of l-A8B8C2, l-A2B8C8, and l-A2C8B8 also have the similar phenomena, as shown in Figure 8a. In addition, the morphology of l-B2C8A8 also shows the distinct variation only for blocks A and C, that is, in R=60 blocks A and C form the two occlusive strips; however, in R=80, block A becomes an irregular core with several hump and block C develops into a cage. Only l-B2A8C8 does not show the marked morphological variation. If the wall selectivity is fixed at αAW=αBW=αCW=120, l-A2B8C8, l-A2C8B8, and l-B2A8C8 express the similar laws with the previously mentioned findings from l-A8B8C2, l-B8A8C2, and l-A2B8C8. In detail, in R=80 (Figure 8b) middle blocks aggregate the cagelike structure that are different with those in R=60 (Figure 3b). Other three terpolymers (l-A8B8C2, l-B8A8C2, and l-B2C8A8) do not show the marked morphological variation under this repulsive wall selectivity and spherical radius. Again, we repeat our findings on the effect of spherical radius. The increase of radius of spherical confinement can significantly change the morphology of linear ABC triblock copolymers, especially for asymmetric block compositions and αAW=50, αBW=25, and αCW=120. Symmetrical block compositions are not easily influenced by the spherical radius, especially for the wall selectivity of αAW=αBW=αCW=120.
To further investigate the effect of spherical radius on the morphologies, we amply draw the inner structure of several CSC aggregates with similar appearance under the different spherical confinements by cutting them. The results are given in Figure 9a. As shown by ball-and-stick models in Figure 9a, l-B2A2C2 in R=80 forms the multilayer-onion structure with seven layers (that is BACABAC). However, the same terpolymer in R=60 forms one with five layers (that is BACAB). The l-B2C2A2 has the similar results, that is, it forms seven layers (BCACBCA) in R=80 and five layers (BCACB) in R=60, respectively. Obviously, the reason is that the fully expanded length of linear ABC triblock copolymers is still far less than spherical radius. Therefore, they cannot fill in the cavity of R=80 by one chain, and need the relay of three chains (see Figure 9b). When the cavity decreases to R=60, they only need the relay of two chain to fill in it (see Figure 9c). They clearly show the influence of the block length/spherical radius on the morphology. It is easy to think that increasing the block length can decrease the layers of multilayer onion aggregates. Hence, we also check the morphology of l-B2A8C8 with longer blocks A and C in R=60 and 80, respectively. The extended blocks A and C result in that the onion layers decrease from 7 to 5 in R=80 and 5 to 3 in R=60, respectively.

(a) Inner structures of several CSC aggregates with similar outward for the different spherical radius (R=60 and 80) and αAW=50, αBW=25, and αCW=120. To clearly expression, the left inner structures are shown by the ball-and-stick model; (b) and (c) sketch maps for the distribution of ABC blocks. Pink, green, and blue represent A-, B-, and C-rich domains, respectively.
3.4 Dynamic evolution
Up to now, we have known the effect of block sequence, wall selectivity, and spherical radius on the assembled structures of linear ABC terpolymers. Furthermore, we also found several typical morphologies, such as multilayer onion, lamella, helix, and so on. Then to seek the formation reason, we examine the dynamic formation pathways for these special aggregates.
For l-B2A2C2 in R=80 and αAW=50, αBW=25, and αCW=120, we have known its inner structure (Figure 9a). Here, we come to know the formation process. Figure 10 presents the separated dynamic evolution processes of A, B, and C domain, respectively. The microphase separation between blocks A, B, and C quickly appears at the primary stage of about t=1000 DPD steps and becomes more and more striking when the simulation time increases. Clearly, block C begins to form a core except for the corona. Both A and C form a shell with several porous and gradually close. Finally, block A becomes an imperforate shell and B becomes a multiparous shell. Other onion-like structures also have the similar evolution process. In addition, we also find that the formative process of onion-like structure is faster than others structures, and the time of self-assembly is generally less than t=10,000 DPD steps. For l-B2C2A2 in R=60 and αAW=αBW=αCW=120, the early phase separation results in the formation of several small and dispersed lumps. Then the lump from same blocks begin to fuse and become big continuous regions (t=4000 DPD steps). After t=20,000, three blocks have begun to form the alternate independent phase that is the embryonic form of lamella. For the same terpolymer, when the spherical radius increases to R=80, the formation pathway is very similar to the stacked lamellae structure, and the difference is that the last aggregate is helix, in detail, three coupled helixes. Under the condition of αAW=αBW=αCW=120, all formed lamella-like morphologies in Figures 2b and 7b also have the similar dynamic evolution process. For l-A2B2C2 in R=60 and αAW=50, αW =25, and αCW=120, block B first aggregates several small balls, then go through the fusion, and finally arrive at the morphology of a core plus the corona. Almost from the outset, block A forms the last phase that is an irregular lump, and the corresponding shape evolution always accompany the changeover of block C, which change from a cage to a broken cage.

Dynamic evolution of B2A2C2, B2C2A2, and A2B2C2 in the different parameter spaces. t is DPD steps. Pink, green, and blue represent A-, B-, and C-rich domains, respectively.
4 Conclusions
In summary, we utilize DPD methods to explore the confinement-induced self-assembled morphologies of linear ABC triblock copolymers under the rigid spherical cavity and their dynamic evolution pathways. The interaction parameters between different blocks are fixed at αAB=45, αAC=90, and αBC=75 to represent the famous μ-EOF triblock copolymers (41), (42), like our previous works (36), (40). Then large parameter spaces including three block sequences, seven block lengths, two radii of rigid spherical cavity, and five sets of αiW are considered. Except for several novel nanostructures, such as multilayer onion, coupled helix, stacked lamella, cage, and so on, the effects of block sequence, wall selectivity, and spherical radius on the morphologies variation are stressed. For symmetric terpolymers, such as l-A2B2C2, l-A2C2B2, and l-B2A2C2, only when block B locates at the middle, the morphology shows the significant difference no matter how the wall selectivity. For asymmetric terpolymers, such as l-A2B8C8, l-A2C8B8, l-B2A8C8, and l-B2C8A8, block sequences can provide the more effective influence on the morphological variety. In fact, the influence of wall selectivity is generally related to the position of blocks (i.e. block sequences). When the wall-preferred block is at the middle, the morphology of the corresponding terpolymer is very different to that formed by the wall-preferred block at the ends. At the same time, the block length and symmetry can also play a significant role. As for spherical radius, its influence on the morphology is clearly given in Figure 9. The main characteristic is that the spherical radius is larger than the block length; therefore, the triblock copolymers need alternatively fill the cavity, which results in the formation of multilayer onion-like structures. If the cavity wall repels any one block, the effects of spherical radius nearly disappear, and the terpolymers are inclined to form the lamella or the helix. As a whole, based on the dynamics evolution of morphologies, we can know that all factors (block sequence, wall selectivity, spherical radius, and block radio/length) exert the influence on the self-assembly and morphology of linear ABC triblock copolymer almost simultaneously. Combining previous crucial works on confined self-assembly (5), (6), (7), (8), (9), (10), (11), (12), (13), (14), (15), (16), (17), (18), (25), (26), (31), (35), the results in this work are helpful to understand the formation mechanism of valuable nanostructures from the self-assembly of linear ABC triblock copolymers under the rigid spherical confinement.
Acknowledgments
All the authors appreciate very much the financial support from CAEP Foundation (no. 2014-1-075), National Nature Sciences Foundation of China (no. 11402241), and Sichuan University of Science & Engineering Foundation (No. 2015RC28).
References
1. Hamley IW. Block copolymers in solution: fundamentals and applications. Hoboken, NJ: Wiley; 2005.10.1002/9780470016985Suche in Google Scholar
2. Savic MR, Luo L, Eisenberg A, Maysinger D. Micellar nanocontainers distribute to defined cytoplasmic organelles. Science 2003;300(5619):615–8.10.1126/science.1078192Suche in Google Scholar PubMed
3. Ruiz R, Kang H, Detcheverry FA, Dobisz E, Kercher DS, Albrecht TR, Pablo JJ, Nealey PF. Density multiplication and improved lithography by directed block copolymer assembly. Science 2008;321(5891):936–9.10.1126/science.1157626Suche in Google Scholar PubMed
4. Bates FS, Fredrickson GH, Hucul D, Hahn SF. PCHE-based pentablock copolymers: evolution of a new plastic. AIChE J. 2001;47(4):762–5.10.1002/aic.690470402Suche in Google Scholar
5. He X, Song M, Liang HJ, Pan CY. Self-assembly of the symmetric diblock copolymer in a confined state: Monte Carlo simulation. J Chem Phys. 2001;114(23):10510–9.10.1063/1.1372189Suche in Google Scholar
6. Zhang L, Liu C, Shang H, Cao X, Chai S, Chen Q, Wu L, Li H. Electrostatic tuning of block copolymer morphologies by inorganic macroions. Polymer 2016;106(5):53–61.10.1016/j.polymer.2016.10.057Suche in Google Scholar
7. Yu B, Li B, Jin Q, Ding D, Shi AC. Self-assembly of symmetric diblock copolymers confined in spherical nanopores. Macromolecules 2007;40(25):9133–42.10.1021/ma071624tSuche in Google Scholar
8. Zhu Y, Jiang W. Self-assembly of diblock copolymer mixtures in confined states: a monte carlo study. Macromolecules 2007;40(8):2872–81.10.1021/ma062022xSuche in Google Scholar
9. Chen P, Liang H, Shi AC. Microstructures of a cylinder-forming diblock copolymer under spherical confinement. Macromolecules 2008;41(22):8938–43.10.1021/ma800443hSuche in Google Scholar
10. Tanaka T, Saito N, Okubo M. Control of layer thickness of onionlike multilayered composite polymer particles prepared by the solvent evaporation method. Macromolecules 2009;42(19): 7423–29.10.1021/ma901100nSuche in Google Scholar
11. Yan N, Zhu Y, Jiang W. Self-assembly of AB diblock copolymer confined in a soft nano-droplet: a combination study by Monte Carlo simulation and experiment. J Phys Chem B. 2016;120(46):12023–9.10.1021/acs.jpcb.6b10170Suche in Google Scholar PubMed
12. Yan N, Liu H, Zhu Y, Jiang W, Dong Z. Entropy-driven hierarchical nanostructures from cooperative self-assembly of gold nanoparticles/block copolymers under three-dimensional confinement. Macromolecules 2015;48(16):5980–7.10.1021/acs.macromol.5b01219Suche in Google Scholar
13. Shi AC, Li B. Self-assembly of diblock copolymers under confinement. Soft Matter. 2013;9(5):1398–413.10.1039/C2SM27031ESuche in Google Scholar
14. Yu B, Deng J, Li B, Shi AC. Patchy nanoparticles self-assembled from linear triblock copolymers under spherical confinement: a simulated annealing study. Soft Matter. 2014;10(35):6831–43.10.1039/C4SM00967CSuche in Google Scholar PubMed
15. Sheng Y, An J, Zhu Y. Self-assembly of ABA triblock copolymers under soft confinement. Chem Phys. 2015;452:46–52.10.1016/j.chemphys.2015.02.019Suche in Google Scholar
16. Su YJ, Ma ZX, Huang JH. Simulation study on the assembly of rod-coil diblock copolymers within coil-selective nanoslits. e-Polymers 2016;16(4):343–9.10.1515/epoly-2015-0275Suche in Google Scholar
17. Yan N, Zhu Y, Jiang W. Self-assembly of ABC triblock copolymers under 3D soft confinement: a Monte Carlo study. Soft Matter. 2016;12:965–72.10.1039/C5SM02079DSuche in Google Scholar PubMed
18. Yu B, Sun P, Chen T, Jin Q, Ding D, Li B, Shi AC. Confinement-induced novel morphologies of block copolymers. Phys Rev Lett. 2016;96(13):138306.10.1103/PhysRevLett.96.138306Suche in Google Scholar PubMed
19. Zhou Y, Xia HG, Long XP, Xue XG, Qian W. Complex multicompartment micelles from simple ABC linear triblock copolymers in solution. Macromol Theory Simul. 2015;24(2):85–8.10.1002/mats.201400072Suche in Google Scholar
20. Zhou Y, Zhou C, He X, Xue XG, Qian W, Luo SK, Xia HG. Shear-induced self-assembly of linear ABC triblock copolymers in solution: creation of 1D cylindrical micellar structures. RSC Adv. 2016;6(7):5711–7.10.1039/C5RA23474CSuche in Google Scholar
21. Ludwigs S, Schmid K, Stafford CM, Amis EJ, Fasolka MJ, Karim A, Magerle R, Krausch G. Combinatorial mapping of the phase behavior of ABC triblock terpolymers in thin films: experiments. Macromolecules 2005;38(5):1850–8.10.1021/ma049048dSuche in Google Scholar
22. Tang CB, Bang J, Stein GE, Fredrickson GH, Hawker CJ, Kramer EJ, Sprung M, Wang J. Square packing and structural arrangement of ABC triblock copolymer spheres in thin films. Macromolecules 2008;41(12):4328–39.10.1021/ma800207nSuche in Google Scholar
23. Dobriyal P, Xiang HQ, Kazuyuki M, Chen JT, Jinnai H, Russell TP. Cylindrically confined diblock copolymers. Macromolecules 2009;42(22):9082–8.10.1021/ma901730aSuche in Google Scholar
24. Arsenault AC, Rider DA, Tetreault N, Chen JL, Coombs N, Ozin GA, Manners I. Block copolymers under periodic, strong three-dimensional confinement. J Am Chem Soc. 2005;127(28):9954–5.10.1021/ja052483iSuche in Google Scholar PubMed
25. Xu JP, Yang Y, Wang K, Li JY, Zhou HM, Xie X L, Zhu JT. Structural transformation of diblock copolymer/homopolymer assemblies by tuning cylindrical confinement and interfacial interactions. Langmuir 2015;31(45):12354–61.10.1021/acs.langmuir.5b03146Suche in Google Scholar PubMed
26. Yabu H, Higuchi T, Jinnai H. Frustrated phases: polymeric self-assemblies in a 3D confinement. Soft Matter. 2014;10(17):2919–31.10.1039/c3sm52821aSuche in Google Scholar PubMed
27. Geisinger T, Müller M, Binder K. Symmetric diblock copolymers in thin films. I. Phase stability in self-consistent field calculations and Monte Carlo simulations. J Chem Phys. 1999;111(11):5241–7.10.1063/1.479778Suche in Google Scholar
28. Yang YZ, Qiu F, Zhang HD, Yang YL. Cylindrical phase of diblock copolymers confined in thin films. A real-space self-consistent field theory study. Polymer 2006;47(6):2205–16.10.1016/j.polymer.2006.01.047Suche in Google Scholar
29. Ludwigs S, Krausch G, Magerle R, Zvelindovsky AV, Sevink GJA. Phase behavior of ABC triblock terpolymers in thin films: mesoscale simulations. Macromolecules 2005;38(5):1859–67.10.1021/ma049047lSuche in Google Scholar
30. Chen HY, Fredrickson GH. Morphologies of ABC triblock copolymer thin films. J Chem Phys. 2002;116(3):1137–6.10.1063/1.1426414Suche in Google Scholar
31. Li SB, Jiang Y, Chen JZY. Morphologies and phase diagrams of ABC star triblock copolymers confined in a spherical cavity. Soft Matter 2013;9(19):4843–54.10.1039/c3sm27770dSuche in Google Scholar
32. Arai N, Yausoka K, Zeng XC. Self-assembly of triblock Janus nanoparticle in nanotube. J Chem Theory Comput. 2013;9(1):179–87.10.1021/ct3007748Suche in Google Scholar PubMed
33. Chi P, Wang Z, Li BH, Shi AC. Soft confinement-induced morphologies of diblock copolymers. Langmuir 2011;27(18):11683–9.10.1021/la202448cSuche in Google Scholar PubMed
34. Sheng YP, An J, Zhu YT. Self-assembly of ABA triblock copolymers under soft confinement. Chem Phys. 2015;452(1):46–52.10.1016/j.chemphys.2015.02.019Suche in Google Scholar
35. Chang HY, Chen YF, Sheng YJ, Tsao HK. Blending-induced helical morphologies of confined linear triblock copolymers. J Taiwan Inst Chem E. 2015;56(5):196–200.10.1016/j.jtice.2015.05.008Suche in Google Scholar
36. Zhou Y, Long XP, Xue XG, Qian W, Zhang CY. Morphologies and dynamics of linear ABC triblock copolymers with different block sequences. RSC Adv. 2015;5(10):7661–4.10.1039/C4RA13814GSuche in Google Scholar
37. Hoogerbrug PJ, Koelman JMVA. Simulating microscopic hydrodynamic phenomena with dissipative particle dynamics. Europhys Lett. 1992;19(3):155–9.10.1209/0295-5075/19/3/001Suche in Google Scholar
38. Espanol P, Warren PB. Statistical mechanics of dissipative particle dynamics. Europhys Lett. 1995;30(4):191–9.10.1209/0295-5075/30/4/001Suche in Google Scholar
39. Groot RD, Warren PB. Dissipative particle dynamics: bridging the gap between atomistic and mesoscopic simulation. J Chem Phys. 1997;107(11):4423–36.10.1063/1.474784Suche in Google Scholar
40. Zhou C, Xia HG, Zhou Y, Xue XG, Luo SK. Dissipative particle dynamics simulations of the morphologies and dynamics of linear ABC triblock copolymers in solutions. RSC Adv. 2015;5(71):58024–31.10.1039/C5RA09661HSuche in Google Scholar
41. Li Z, Kesselman E, Talmon Y, Hillmyer MA, Lodge TP. Multicompartment micelles from ABC miktoarm stars in water. Science 2004;306(5963):98–101.10.1126/science.1103350Suche in Google Scholar PubMed
42. Xia J, Zhong C. Dissipative particle dynamics study of the formation of multicompartment micelles from ABC star triblock copolymers in water. Macromol Rapid Commun. 2006;27(14):1110–4.10.1002/marc.200600187Suche in Google Scholar
©2017 Walter de Gruyter GmbH, Berlin/Boston
This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Artikel in diesem Heft
- Frontmatter
- In this Issue
- Full length articles
- Ultralight sponges of poly(para-xylylene) by template-assisted chemical vapour deposition
- Cryostructuring of polymer systems. 44. Freeze-dried and then chemically cross-linked wide porous cryostructurates based on serum albumin
- Macroporous polymer beads derived from a novel coporogen of polyethylene/dichlorobenzene
- Gas separation properties of Troeger’s base-bridged polyamides
- Catalytic crosslinking of a regenerated hydrophobic benzylated cellulose and nano TiO2 composite for enhanced oil absorbency
- On orientation memory in high density polyethylene – carbon nanofibers composites
- Comparative investigation of physical, mechanical and thermomechanical characterization of dental composite filled with nanohydroxyapatite and mineral trioxide aggregate
- Dissipative particle dynamics simulation on the self-assembly of linear ABC triblock copolymers under rigid spherical confinements
- Synthesis and characterization of cellulose acetate naphthoate with good ultraviolet and chemical resistance
- Thermal characterization and flammability of polypropylene containing sepiolite-APP combinations
Artikel in diesem Heft
- Frontmatter
- In this Issue
- Full length articles
- Ultralight sponges of poly(para-xylylene) by template-assisted chemical vapour deposition
- Cryostructuring of polymer systems. 44. Freeze-dried and then chemically cross-linked wide porous cryostructurates based on serum albumin
- Macroporous polymer beads derived from a novel coporogen of polyethylene/dichlorobenzene
- Gas separation properties of Troeger’s base-bridged polyamides
- Catalytic crosslinking of a regenerated hydrophobic benzylated cellulose and nano TiO2 composite for enhanced oil absorbency
- On orientation memory in high density polyethylene – carbon nanofibers composites
- Comparative investigation of physical, mechanical and thermomechanical characterization of dental composite filled with nanohydroxyapatite and mineral trioxide aggregate
- Dissipative particle dynamics simulation on the self-assembly of linear ABC triblock copolymers under rigid spherical confinements
- Synthesis and characterization of cellulose acetate naphthoate with good ultraviolet and chemical resistance
- Thermal characterization and flammability of polypropylene containing sepiolite-APP combinations