Revisiting Vertical Orientation in Block Copolymer Complexes through Controlled Solvent Vapor Annealing

Abstract

Block copolymers (BCPs) are macromolecules made up of two or more chemically different polymer blocks that are covalently bound together. Their mutual incompatibility causes them to separate into nanoscale microphases, which allows the blocks to self-assemble into ordered shapes, such as lamellae, spheres, and cylinders. The ability to self-assemble offers a versatile route for forming well-ordered nanoscale architectures relevant to nanolithography, membranes, and functional materials. Incorporating hydrogen-bonding small molecules to form supramolecular BCP complexes expands structural diversity by altering domain size, interfacial energies, and chain mobility. These complexes expand the range of parameters for precise control over nanoscale morphology while also providing highly ordered nanostructures that can be used as soft templates for the formation of inorganic films and porous networks. However, the added complexity raises challenges in obtaining vertically oriented nanostructures, which are crucial in nanopatterning applications. In the case of solvent vapor annealed films, the effect of swelling parameters on vertical orientation has not been thoroughly investigated. Here, we demonstrate that swelling-induced surface neutrality promotes vertical orientation in PS-b-P4VP­(PDP) thin films, where P4VP­(PDP) is the majority block and PS the minority block. Morphological transitions were observed, including horizontal, mixed, and vertical orientations, as film thickness, PDP ratio, and swelling ratio were varied. A swelling ratio window ranging from ∼ 1.6 to 1.8 that reliably stabilizes vertical orientation was discovered. Significantly, vertical structures require a balance between selective swelling for surface energy neutrality and sufficient chain mobility for domain re-organization. These findings depart from previous reports where the formation of hierarchical lamellar structures within P4VP­(PDP) was described as a key requirement for vertical orientation. Overall, this study identified solvent and wetting parameters that influence the BCP orientation, thereby advancing supramolecular block copolymer thin films for nanopatterning and functional material applications.

Introduction

The self-assembly of block copolymer (BCP) thin films have emerged as a promising platform for next-generation nanomanufacturing owing to their ability to organize into periodic domains with a high degree of long-range order. These structures have been explored for applications, including lithographic masks, filtration membranes, and photonic devices. − The different parameters that govern the morphology of BCPs include the Flory–Huggins interaction parameter (χ), the degree of polymerization N, and the volume fraction f of each block.

Beyond this intrinsic tendency for microphase separation, their behavior can be tailored through external stimuli and supramolecular interactions where noncovalent interactions such as hydrogen bonding, ionic interactions, and metal–ligand coordination enable the incorporation of small guest molecules into one of the polymer domains. , This changes the relative size of the blocks, alters the interfacial energy balance, and strengthens the driving force for phase separation, thereby expanding the design space of block copolymers to access morphologies that would not form in neat polymers. − A well-studied example is poly­(styrene)-block-poly­(4-vinylpyridine) (PS-b-P4VP), in which the P4VP block is strongly Lewis basic and highly capable of hydrogen bonding, making it an ideal host for small molecules such as benzoic acid or 3-pentadecylphenol (PDP) (Figure ). − When these additives form hydrogen bonds with P4VP, they effectively swell the P4VP domain, shift the balance of volume fractions, and lower the glass transition temperature of the block copolymer, which improves chain mobility and enables the system to adapt more readily to thickness variations in thin films. − Due to this increased mobility, films near lamellae–cylinder phase boundaries often display mixed morphologies, stepped surfaces, and hierarchical patterns arising from local thickness fluctuations. ,− While such behavior highlights the role of mobility in supramolecular systems, achieving comparable levels of ordering in conventional block copolymer thin films typically requires significantly longer thermal or solvent annealing times, often due to kinetic trapping and slower chain dynamics. − In contrast, the incorporation of small molecules in supramolecular BCP complexes significantly accelerates self-assembly by plasticizing the interacting block, enabling rapid morphology evolution on much shorter time scales. − The PS-b-P4VP­(PDP) supramolecular complex has been established in several studies as a model system for understanding how small molecules influence the block copolymers’ morphology, revealing that hydrogen bonding between the pyridyl nitrogen of P4VP and the hydroxyl group of PDP drives the formation of a comb-like supramolecular structure in which the long PDP alkyl tails extend into the P4VP domain (Figure ). ,,,, Beyond enabling rapid self-assembly, these supramolecular complexes also provide a functional pathway for nanomaterial fabrication. The selective incorporation of small molecules within the P4VP domain creates chemically distinct regions that can direct the infiltration of inorganic precursors, allowing the formation of hybrid or nanoporous structures. − This capability makes supramolecular BCP systems particularly attractive as soft templates for advanced functional nanomaterials. In addition to structural diversity, complexation of BCPs with small molecules expands their versatility by mediating interfacial mismatch with nanoparticles to yield hierarchical nanocomposites, − and their temperature sensitivity to fabricate stimuli responsive materials. , Controlling the orientation of self-assembly in BCP complexes is challenging due to multicomponent surface energy contributions and asymmetric diffusion between blocks. ,, Interfacial interactions at the substrate and air interfaces play an important role in controlling BCP orientation. , Typically, the block with a stronger affinity for the substrate preferentially wets the bottom interface, while the block with lower surface energy tends to migrate toward the air interface. In PS-b-P4VP­(PDP), the P4VP­(PDP) block forms a secondary lamellar structure within the BCP lattice, and is presumed to dominate wetting at both the air–polymer and polymer–substrate interface. ,, The preference for P4VP­(PDP) lamellar hierarchical structures to orient parallel to the substrate has been proposed as a means to obtain vertical orientation. , Thus, previous studies suggest two key requirements for achieving vertical orientation. First, a high PDP grafting densities to promote the formation of hierarchical structures. Second, the use of non-oxygen-containing solvents for solvent vapor annealing (SVA) to minimize the disruption of PDP hydrogen bonding of PDP. However, we and others have shown that applying a high PDP ratio and nonpolar solvents can produce horizontal orientation, suggesting that other factors are at play. ,

1.

Schematic of hydrogen bonding between P4VP and PDP, forming comb-like supramolecular chains.

The combined effects of controlled SVA with precise tuning of swelling ratio and film thickness in PS-b-P4VP­(PDP) have yet to be systematically clarified, despite their strong influence on domain orientation and stability. SVA is a technique with growing significance for inducing self-assembly in BCP complexes due to the potential thermal sensitivity of the small molecules. ,, The absorption of solvent molecules can lead to screened surface energies, promoting vertical orientation. Furthermore, the film thickness relative to the BCP domain spacing (L₀) is crucial in determining the vertical orientation and lateral ordering of microdomains. When the film thickness does not match the natural domain spacing (L₀) of the microdomains, horizontally oriented (parallel) domains must stretch or compress to accommodate the confinement, resulting in an increased elastic free energy penalty. − In contrast, vertically oriented (perpendicular) domains can more easily adapt to this mismatch without substantial chain deformation, making them energetically favored under these frustrated conditions. Consequently, thickness incommensurability promotes reorganization toward structures that minimize the total free energy, frequently favoring perpendicular domain orientation. ,

Although early studies on PS-b-P4VP­(PDP) supramolecular systems established the fundamental phase behavior and highlighted the role of hydrogen-bonded additives during SVA, they largely focused on “jar annealing” and did not systematically control solvent activity. ,, As a result, there remains a knowledge gap in the relationship linking solvent activity, swelling ratio, and resulting morphology. In contrast, the present study advances this understanding by establishing a quantitative framework for the evolution of morphology under precisely controlled SVA conditions. By systematically tuning solvent activity (via swell ratio), film thickness, and PDP ratio, we identify distinct morphological regimes and explicitly map the transition pathway from horizontal to vertical and back to horizontal orientations. Accordingly, this study investigates PS-b-P4VP­(PDP) thin films, in which PS constitutes the minority block while the P4VP­(PDP) complex functions as the continuous phase. At a PDP ratio of r = 1.7, the supramolecular system resides in a cylinder-forming regime, with PS forming minority cylindrical domains within a continuous P4VP­(PDP) matrix. The corresponding domain spacing (L₀ ≈ 31 nm), obtained from GISAXS measurements, is used to define confinement effects throughout this study. We demonstrate that the vertical orientation of BCP complexes can be controlled by precisely regulating solvent vapor annealing activity in the absence of the hierarchical P4VP­(PDP) structure. This is achieved by selective swelling, which effectively neutralizes the surface energies of both blocks at the air–polymer interface. Swelling behavior of the polymer blocks is systematically evaluated using absorption–desorption isotherms. Vertical orientation is exclusively observed in the thin film regime (<2 L₀), where surface and confinement effects are dominant. A swelling window for vertical orientation is identified and attributed to a balance between maintaining selective swelling to minimize surface energy mismatch and achieving sufficient overall chain mobility for domain reorientation. Such domain-orientation control is critical for reliable nanostructure fabrication in applications such as nanopatterning and advanced lithography.

Methods

Materials

The polystyrene-block-poly­(4-vinylpyridine) (PS-b-P4VP), PS, and P4VP were sourced from Polymer Source Inc. (Canada) and used as received. PS (M _n = 20,000 g mol^–1, M _w = 22,500 g mol^–1, PDI = 1.13), P4VP (M _n = 7,000 g mol^–1, M _w = 7,300 g mol^–1, PDI = 1.04), and PS-b-P4VP (20k–5k; M _n = 25,000 g mol^–1, PDI = 1.02). The polymer purity and composition were further verified by proton nuclear magnetic resonance (¹H NMR) spectroscopy (Figure S1). The small-molecule, 3-pentadecylphenol (PDP, M = 304.5 g mol^–1) was obtained from TCI America, and chloroform (stabilized with amylene) was purchased from Sigma-Aldrich. All reagents were used as received. Prime silicon wafers were acquired from University Wafers.

Sample Preparation

Thin films of supramolecular complexes were prepared by dissolving polystyrene-block-poly­(4-vinylpyridine) (PS-b-P4VP) diblock copolymer and 3-pentadecylphenol (PDP) in chloroform. Chloroform was selected due to its good solubility for both PS and P4VP­(PDP) components and its moderate volatility that enables uniform film formation and effective solvent vapor annealing (SVA). The polymer solutions were filtered through a 0.45 μm pore-size syringe filter to remove contaminants, including dust and other airborne particulate impurities that may be introduced during sample preparation under noncleanroom conditions. Homogeneous casting solutions with polymer concentrations of 0.3, 0.5, and 1.0 wt %, which were chosen to systematically access different film thickness regimes relative to the natural domain spacing (L₀) and to probe confinement effects on morphology. Silicon wafer substrates (approximately 10 mm × 10 mm) were pretreated with atmospheric plasma at 100 mTorr and 20 W. The spin coating was performed using a Laurell 650 M series spin-coater at 2000 rpm for 45 s, resulting in uniform thin films with an average thickness of 32 ± 5 nm for 0.3 wt % solutions, 53 ± 2 nm for 0.5 wt % solutions, and 110 ± 5 nm for 1.0 wt % solutions, as determined by ellipsometry.

Thermal Solvent Vapor Annealing (Thermal-SVA)

Thermal solvent vapor annealing was performed using an Instec mk2000 thermoelectric controller and an MKS mass flow controller. The PS-b-P4VP­(PDP)_1.7 thin films (where 1.7 denotes the molar ratio of 3-pentadecylphenol (PDP) molecules to pyridine units in the P4VP block) with an average film thickness of 31 ± 2 nm were placed in the solvent vapor annealing chamber. Chloroform vapor was introduced through the MKS controller at a fixed flow rate, and the chamber was sealed to ensure a stable vapor environment. To initiate swelling, the temperature of the SVA chamber was programmed on the Instec mk2000 controller and ramped from 20 to 15 °C at 2.5 °C/min – 3.5 °C/min. The films swelled continuously throughout the temperature ramp under a constant chloroform vapor environment. As the temperature decreases, excess solvent vapor condenses into the film, increasing solvent uptake and swelling.

The thermal-SVA process involved a temperature ramp followed by solvent-induced swelling, and the films were annealed for approximately 2 min before quenching. This annealing time corresponds specifically to the thermal-SVA method, where swelling is dynamically controlled through temperature modulation under a constant solvent vapor environment. At the end of the ramp window, the films were automatically quenched by the system, which shut off the solvent flow and activated a nitrogen purge. The thermal-SVA setup automatically monitored and controlled the swell ratio throughout the temperature ramp, with real-time swollen thickness recorded using a Filmetrics spectral reflectance. A schematic of the Instec mk2000 thermoelectric temperature controller used for thermal solvent vapor annealing (thermal-SVA) is provided in the Supporting Information (Figures S2–S5). This thermal ramp-and-quench method enabled the films to reach swell ratios up to ∼3.5 before the onset of dewetting. In contrast to conventional SVA, thermal-SVA provides dynamic control over solvent uptake via temperature modulation in a constant solvent vapor environment, enabling rapid access to higher swelling ratios over short time scales. This approach facilitates transient nonequilibrium swelling conditions, which are critical for accessing otherwise inaccessible morphologies and significantly reducing annealing times.

Solvent Vapor Annealing (SVA)

For controlled solvent vapor annealing (SVA), a custom-built continuous flow chamber was used (Figure S6), designed for controlled solvent exposure. The chamber had a vacuum evacuation system, mass flow controls (Alicat Scientific), a mixing chamber, and a solvent reservoir. The mass flow controllers (MFCs) were set to deliver a total flow of 500 standard cubic centimeters per minute (sccm), which constitutes a mixture of nitrogen and chloroform vapor. The solvent activity was adjusted by varying the ratio of chloroform to nitrogen. The solvent activity (α) was used to measure the degree of solvent exposure and is defined as follows:

$$\alpha =\frac{\mathrm{solvent\; vapor\; flow\; rate}}{\mathrm{total\; flow\; rate}}$$

where α = 0 represents pure nitrogen (no solvent), and α = 1 corresponds to 100% chloroform vapor. All films were annealed using SVA at a solvent activity of α = 0.8 for 15 s, providing consistent swelling conditions for morphology analysis. The in situ swelling profile (Figure S7, Supporting Information) shows that the film reaches a constant swollen thickness within this time, with a slight decrease attributed to the endothermic cooling effect in the solvent bubbler. This behavior is consistent with prior studies on supramolecular block copolymer systems, where solvent-induced plasticization enables rapid equilibration on short time scales. These results confirm that morphology development occurs under stable and controlled swelling conditions within this short annealing window and is governed by swelling ratio rather than annealing time (Figure S8). Following each annealing period, the solvent flow was stopped immediately upon reaching the set time, and the chamber was evacuated under vacuum to rapidly quench and preserve the swollen structure. Conventional SVA enables precise control over solvent activity by independently tuning chloroform and nitrogen flow rates, enabling systematic investigation of swelling-dependent morphology transitions. However, achieving higher swelling ratios is often limited by equilibrium vapor conditions and may require longer exposure times.

Metal Salt Infiltration

The presence of vertically oriented structures was confirmed through a metal precursor infiltration experiment. The films were immersed in a sodium tetrachloroaurate­(III) dihydrate precursor solution at room temperature for 30 min to enable selective infiltration into the P4VP­(PDP) domains. Following removal from the precursor solution, the films were rinsed, dried under a nitrogen stream, and exposed to ultraviolet (UV) light at 254 nm for 2 h to induce photoreduction of the gold precursor and ozone etching of the polymer.

Grazing Incidence Small-Angle X-ray Scattering (GISAXS)

GISAXS was performed at Beamline 11-BM, National Synchrotron Light Source II, Brookhaven National Laboratory, using a 13.5 keV X-ray beam (λ = 0.918 Å). Data were collected at incident angles of 0.05°, 0.10°, and 0.13° with 30 s exposures, and two-dimensional scattering patterns were recorded at a calibrated sample detector distance of 5.05 m.

Surface Energy Contribution Estimation

Static contact angle measurements were performed using a contact angle goniometer equipped with DROPimage software (ramé-hart Instrument Co., version 2.10.04) for image capture, angle fitting and measurement. Deionized water and diiodomethane were used as probe liquids with a droplet volume of 2 μL. Thin films of pure PS, P4VP, P4VP­(PDP)_x, and PS-b-P4VP­(PDP)_1.7 with comparable thickness (32 ± 2 nm) were measured, and reported values represent averages of multiple measurements. While different PDP ratios were investigated to understand their influence on morphology and phase behavior (see Table ), the surface energy measurements were specifically performed at r = 1.7. This composition represents the primary system used to establish surface energy modulation and vertical orientation under controlled SVA conditions. Furthermore, although diiodomethane is a highly polarizable nonpolar liquid, it is important to note that the PDP molecules are hydrogen-bonded to the P4VP backbone and are not free to dissolve during the short measurement times. Accordingly, the surface energy values obtained here should be regarded as approximate, as minor interactions between diiodomethane and PDP may influence the quantitative accuracy of the measurements. Surface energy was evaluated using static contact angle measurements, which reflect the balance of interfacial tensions. This relationship is described by Young’s equation:

$${\sigma }_s={\sigma }_{sl}+{\sigma }_l\cos \theta$$ 1

where σ _s is the surface energy of the solid, σ _sl is the solid–liquid interfacial tension, σ _l is the surface tension of the liquid, and θ is the equilibrium contact angle. According to this expression, lower contact angles correspond to stronger solid–liquid interactions and higher surface wettability, while higher contact angles indicate weaker interactions and reduced wettability. The total surface energy of the solid (σ _s ) and liquid (σ _l ) was decomposed into dispersive (D) and polar (P) components: −

$${\sigma }_s={\sigma }_s^D+{\sigma }_s^P;{\sigma }_l={\sigma }_l^D+{\sigma }_l^P$$ 2

where σ _s and σ _s represent the dispersive and polar (dipole–dipole and hydrogen-bonding) contributions of the solid surface energy, respectively. Similarly, σ _l and σ _l denote the dispersive and polar components of the liquid surface tension. According to the Fowkes model, the interfacial work of adhesion between a solid and a liquid is expressed as the sum of dispersive and polar interactions. Combining this relationship with Young’s equation yields:

$$\sqrt{{\sigma }_l^D{\sigma }_s^D}+\sqrt{{\sigma }_l^P{\sigma }_s^P}=\frac{{\sigma }_l(1+\cos \theta )}{2}$$ 3

where θ is the measured static contact angle of the probe liquid on the solid surface, and σ _l is the total surface tension of the liquid. The dispersive component of the solid surface energy (σ _s ) was determined using diiodomethane, a nonpolar liquid with a negligible polar component (σ _l ≈ 0), which simplifies the above expression to

$${\sigma }_s^D=\frac{{\sigma }_l{(1+\cos \theta )}^2}{4}$$ 4

Once σ _s was obtained, the polar component of the solid surface energy (σ _s ) was calculated using water as the probe liquid, for which both dispersive and polar surface tension components are known. This two-liquid approach enables quantitative separation of dispersive and polar contributions to the solid surface energy.

2. Polymer Volume Fractions of PS-b-P4VP­(PDP) (20k-b-5k) Thin Films (30 nm) at Different PDP Ratios.

PDP Ratio	ΦPS	ΦP4VP(PDP)	Assigned Morphology
1.0	0.50	0.50	Parallel lamellae
1.5	0.42	0.58	Vertical cylinder
1.7	0.40	0.60	Vertical cylinder

Atomic Force Microscopy (AFM)

Block copolymer thin films were characterized using an Asylum Research MFP3D AFM and Cypher AFM ES VRS system. Bruker TESPA tapping-mode probes with resonance frequencies of 330–380 kHz were used for imaging. All AFM measurements were performed at room temperature.

Results and Discussion

Despite the wealth of experiments performed in the self-assembly of PS-b-P4VP­(PDP) using solvent vapor annealing (SVA), the complexity of the supramolecular system has led to conflicting reports regarding the wetting characteristics and swelling behavior of the P4VP­(PDP) block. For example, it was assumed that chloroform vapor, a commonly used solvent in SVA, selectively swells the PS block. , However, we have recently demonstrated that chloroform is highly selective for P4VP­(PDP), particularly at low solvent activity due to PDP-induced plasticization.

With regard to surface wetting, PDP is an amphiphile that hydrogen bonds with P4VP to form a polymer complex that could screen secondary interactions of the pyridine groups. When the PDP grafting density (PDP-pyridine molar ratio, r) is high, the complexed comb block is assumed to have a lower surface energy than PS (σ_P4VP(PDP) < σ_PS), thereby preferentially segregating to the top surface. The comb-like conformation in P4VP­(PDP) has also been hypothesized to segregate to the surface to minimize entropic loss at the interface.

Therefore, to quantitatively validate the surface-energy interpretation of surface wetting and solvent selectivity, we performed wetting studies on PS and P4VP­(PDP) films. We measured the surface energies of PS and P4VP­(PDP)_1.7 using water and diiodomethane contact angle measurements and analyzed the data with the Fowkes geometric mean method (see Table ). The use of these two probe liquids enables separation of dispersive and polar surface energy contributions, with diiodomethane isolating dispersive interactions and water capturing polar interactions, providing a clear basis for interpreting interfacial behavior. PS has a predominantly dispersive component as expected from its nonpolar structure and high water contact angle. However, P4VP­(PDP)_1.7 displayed a substantial polar surface energy, despite the presence of excess amounts of the nonpolar PDP, resulting in a higher overall surface energy compared to PS. To further evaluate the effect of PDP loading on interfacial properties, the surface energy of P4VP was analyzed as a function of PDP ratio (Figure ). The results show a pronounced decrease in surface energy with increasing PDP ratio, followed by stabilization at higher loadings, indicating progressive screening of the polar pyridine groups through hydrogen bonding with PDP. The sizable presence of the polar component suggests incomplete screening by the PDP and a significant contribution by the pyridine ring to the overall surface energy. Thus, we establish that the PS block is preferred at the air–polymer interface while the P4VP­(PDP) block has a stronger preference at the Si substrate.

1. Surface Energy Components Determined from Contact Angle Measurements .

Polymer	Water (°)	Diiodomethane (°)	σs D (mN/m)	Error (σ s )	σs P (mN/m)	Error (σ s )	σs (mN/m)	Error (σ s )
P4VP(PDP)1.7	61	42	38	0.3	12	0.5	51	0.6
PS	92	40	40	0.6	0	0	40	0.6

^aContact angles were experimentally measured in this study, while the dispersive and polar components of the probe liquids (water and diiodomethane) were obtained from standard literature values. − .

^bWater and diiodomethane contact angles were measured on polymer thin films.

^cThe dispersive surface energy component and its uncertainty were determined from diiodomethane contact angles using eq .

^dThe polar surface energy component and its uncertainty were extracted from water contact angles using eq together with the previously determined dispersive contribution.

^eThe total surface energy and its uncertainty were calculated as σ_s = σ_s + σ_s (eqs –). −

2.

Surface energy of P4VP as a function of PDP ratio, determined using the Fowkes method from contact angle measurements. Error bars represent the standard deviation from multiple independent measurements (n ≥ 3).

While prior studies have employed various nonpolar solvents, chloroform is used as the annealing vapor here due to its high vapor pressure, which allows for a wider range in solvent activity. Given that chloroform has a low surface energy (27.2 mN/m at 20 °C) , but selectively swells the P4VP­(PDP) block, we hypothesize that the SVA of PS-b-P4VP­(PDP) can tune the effective interfacial energies at both the air–polymer and polymer–substrate interfaces, thereby enabling perpendicular (vertical) orientation. In principle, the infiltration of chloroform vapor lowers the effective surface energy of the P4VP­(PDP) block, improving wetting at the air interface and reducing its dominance at the Si substrate.

Thus, solvent-induced neutrality in PS-b-P4VP­(PDP) complexes was investigated using monolayer (∼1L₀) thin films (∼30 nm) to emphasize the surface fields. PS-b-P4VP (20-b-5) was complexed with PDP to a pyridine molar ratio of 1.7 to obtain a well-known cylinder morphology with a large excess of small molecules, and predominantly adopts a horizontal orientation. The swelling of the PS-b-P4VP­(PDP)_1.7 thin films was controlled using a thermal solvent vapor annealing (SVA) method with in situ film thickness measurements during solvent vapor flow. Here, temperature and flow rate were adjusted based on the target swelling ratio. The samples were annealed in chloroform for 2 min, after which a quench step was automatically initiated, shutting off the solvent flow and purging the chamber with nitrogen to preserve the morphology. The resulting structures correspond to swelling-dependent morphologies retained upon quenching during the thermal SVA process. This thermal ramp–anneal–quench approach enabled the films to reliably reach swell ratios up to ∼3.5. As a result, we could observe the full range of swelling-driven morphological transitions discussed below. Figure shows that at low swell ratios (1.3–1.5), the films exhibit a horizontally oriented morphology with poor long-range order, but improved compared to the as-cast film. The morphology indicates that domain reorganization is constrained by low chain mobility, resulting from limited solvent uptake. In comparison, poorly ordered horizontal structures have been reported in related supramolecular block copolymer systems under low-swelling conditions; solvent selectivity preferentially swells one block but does not fully offset interfacial asymmetry, leading to incomplete domain reorientation. The morphology gradually changes from horizontally aligned to vertical orientation, as the swell ratio increases into the intermediate range of 1.6–1.8 (Figure ). Chloroform vapor softens the film, increases chain mobility, and reduces the surface energy imbalance between the two blocks. At higher swell ratios, between 2.0 and 3.5, the film reverts to a horizontal orientation. Further increasing the swelling ratio above 3.5 resulted in dewetting as AFM images revealed ruptures and morphological defects, indicating a loss of film stability as illustrated in Figure . This behavior is attributed to excessive solvent uptake, which significantly enhances chain mobility while weakening interchain entanglement and reducing adhesion to the substrate. As a result, the film becomes unstable and undergoes hole nucleation and growth, ultimately leading to rupture and dewetting. Similar solvent vapor annealing-induced dewetting pathways progressing from continuous films to hole formation and eventual breakup have been widely reported and are driven by solvent selectivity and dynamic changes in surface energy. , The observation of vertical orientation confirms that chloroform SVA can neutralize the surface energies of both blocks; however, these findings also reveal a crucial processing window for reliably preserving perpendicular orientation (Figure ).

3.

AFM phase images of PS-b-P4VP­(PDP) thin films (32 ± 2 nm, PDP ratio = 1.7) at selected swelling ratios. The inset shows the as-cast film prior to solvent vapor annealing, which exhibits a disordered morphology. Poorly ordered horizontal morphologies are present at low swelling ratios (1.3–1.5). Vertically oriented morphologies are stabilized at intermediate swelling ratios (1.6–1.8). Horizontal orientation is observed at higher swelling ratios (≥2.0). Excessive swelling (≥3.5) leads to dewetting of the film. Scale bars represent 200 nm.

4.

Swelling-induced morphology evolution in PS-b-P4VP­(PDP) thin films. Schematic illustration of nanostructure evolution with increasing swell ratio, progressing from as-cast disordered morphologies to short, disordered horizontal cylinders at low swelling, vertical cylinder orientation at intermediate swelling, ordered horizontal cylinders at higher swelling, and film dewetting under overswelling conditions (>3.5). This figure is an overview of the process to illustrate the effect of swell ratio and is based on experimental observation (Figure ).

Orientational shifts in BCP thin films are known to be solvent-sensitive, with a complex balance between the polymer–solvent interface, the substrate, and the air interface. To minimize interfacial free energy, the BCP chains should reorganize to selectively enrich P4VP­(PDP) at the substrate interface and PS at the air interface, according to their surface energies in Table . Water contact angle (WCA) measurements (Figure ) of the BCP complex films showed that the horizontal morphology (obtained from the high swelling ratio regime, >1.8) is more hydrophobic, ∼87 ± 2°, compared to the vertical orientation (swelling ratio 1.6), ∼73 ± 0.6°, indicating a greater exposure of the PS block in the former. This validates that the horizontal orientation where the PS wets the air interface and P4VP­(PDP) wets the substrate is the preferred morphology, and lowering the surface energy of the P4VP­(PDP) block by solvent annealing is required to stabilize vertical orientation. Interestingly, the horizontal morphology at the low swelling ratio regime has a lower WCA ∼ 78 ± 1° compared to the high swelling ratio regime. This could stem from poorly defined domain interfaces and PDP distribution with limited annealing.

5.

Water contact angles for 32 ± 2 nm films of pure P4VP­(PDP)_1.7, PS-b-P4VP­(PDP)_1.7 exhibiting vertical and horizontal morphologies, and pure PS. Mean ± standard deviation (°): P4VP­(PDP)_1.7 = 61 ± 1°, vertical morphology = 73 ± 0.6°, horizontal morphology = 87 ± 2.0°, PS = 92 ± 0.2°. Error bars represent the standard deviation obtained from multiple independent measurements (n ≥ 3) across different regions of each sample, reflecting measurement reproducibility and surface heterogeneity.

Solvent absorption studies were conducted to verify the selectivity of chloroform (Figure ). Here, the BCP and their respective homopolymer films were exposed to solvent vapor with increasing activity, and their resulting thickness was monitored in situ (swelling curves are provided in Figures S3–S5). The solvent uptake at low activity (swelling ratio <1.5) of the BCP closely resembles that of P4VP­(PDP)_1.7, indicating chloroform selectivity due to significant plasticization by the PDP. The solvent uptake masks the polar pyridine groups, thereby reducing the polar component of the P4VP­(PDP)₁.₇ surface energy, mediating the surface energy difference between the two blocks. However, the PS block is expected to have poor mobility and remain kinetically trapped under low solvent activity, which suppresses orientation rearrangement, thereby maintaining the horizontal morphology. To further clarify the role of supramolecular complexation, the solvent absorption behavior of neat PS-b-P4VP is included for comparison (Figure ). The neat PS-b-P4VP exhibits a swelling trend similar to that of PS across the activity range, indicating relatively weak solvent selectivity between domains, particularly at low swelling ratios. In contrast, PS-b-P4VP­(PDP)₁.₇ closely follows the swelling behavior of P4VP­(PDP)₁.₇ at low solvent activity, confirming preferential solvent uptake within the P4VP­(PDP) domain due to PDP-induced plasticization. As a result, the supramolecular system exhibits enhanced solvent selectivity at low swelling ratios compared to the neat block copolymer.

6.

Absorption isotherms of PS (M _n = 20 kg mol^–1), P4VP­(PDP)₁.7 (P4VP = 7 kg mol^–1; PDP = 304.5 g mol^–1), PS-b-P4VP, and the PS-b-P4VP­(PDP)₁.₇ supramolecule (PS = 20 kg mol^–1; P4VP = 5 kg mol^–1; PDP = 1.7). P4VP­(PDP) shows higher solvent uptake than PS over the full activity range, with the swelling ratio increasing from ∼ 1.0–1.6 at low activity, to ∼1.6–1.8 near the neutral-swelling region, and to 1.8–2.3 at high activity. The neat PS-b-P4VP exhibits a swelling trend similar to that of PS across the activity range and shows reduced solvent selectivity at low swelling ratios compared to the supramolecular system. In contrast, PS-b-P4VP­(PDP)₁.₇ follows the P4VP­(PDP) trend at low activity, indicating preferential solvent uptake within the P4VP­(PDP) domain. Error bars represent the standard deviation from repeated measurements.

At higher activities, 0.8–0.9 (swelling ratios 1.6–1.8), the PS domain gains significantly higher mobility while maintaining a near-neutral condition, stabilizing vertical orientation, albeit within a narrow range. As the activity increases further, solvent uptake becomes less selective, with swelling behavior of the BCP lying directly between both PS and P4VP­(PDP)₁.₇. More uniform swelling lowers the effective surface energy of PS and restores a surface energy difference between the two blocks, making the swollen PS more compatible with the air interface. Moreover, swelling of the PS chains could disrupt thin film confinement needed to maintain vertical orientation, causing the morphology to revert to a horizontal orientation. At very high activity, the excessive swelling weakens the film structure, eventually leading to dewetting.

Therefore, we attribute the vertical orientation observed in PS-b-P4VP­(PDP)₁.₇ thin films to the selective swelling of the P4VP­(PDP)₁.₇ block. This is driven by hydrogen-bonding complexation between PDP and P4VP, which enhances its affinity for chloroform relative to PS. The solvent-mediated modulation of surface energy provides an effective mechanism for stabilizing vertical orientation.

When considered alongside the swelling experiments, these results confirm that swelling-induced morphological transitions are directly coupled to changes in surface composition. In the intermediate swelling regime, stabilization of vertical domains is accompanied by increased solvent uptake, which enriches the film–air interface with the P4VP­(PDP)_1.7 matrix, and the resulting surface enrichment of P4VP­(PDP) reduces the contact angle. In swelling conditions that favor horizontal morphologies, lateral alignment of PS domains results in higher contact angles. Thus, contact angle measurements using reference films provide clear experimental validation that systematic and predictable changes in the top-surface composition accompany morphology evolution in PS-b-P4VP­(PDP)_1.7 thin films.

Validation of Vertical Cylinder Orientation via Selective Precursor Infiltration

Selective gold precursor infiltration was used to validate the vertically oriented cylinders. Inorganic salts are known to selectively infiltrate and coordinate to poly­(vinylpyridine). The experiment was carried out on thin films with a thickness of 20 nm (<1 L₀, where L₀ ≈ 31 nm) at a swelling ratio of 1.6. Figure shows AFM height images of the film before infiltration, after gold precursor infiltration, and after ozone etching. Before infiltration (Figure left), the film exhibits continuous height contrast, reflecting the block copolymer morphology. After immersion in the gold precursor solution for 30 min, the continuous phase representing P4VP­(PDP) appears swollen with more pronounced features surrounding the PS cylinders (Figure middle). Subsequently, UV ozone exposure at 254 nm for 2 h resulted in templated hole structures (Figure , right), formed by removal of the PS vertical cylinders. The AFM phase images are provided in the Supporting Information (Figure S9).

7.

AFM height images of solvent vapor annealed PS-b-P4VP­(PDP)_1.7 thin films. Before infiltration (left), the film exhibits a well-defined nanostructured morphology with clear contrast corresponding to the block copolymer domains. After 30 min precursor immersion (middle), increased surface roughness and contrast indicate selective uptake within the P4VP­(PDP) domains. Following UV/ozone treatment (right), removal of the polymer matrix reveals a nanoporous morphology that replicates the original microdomain structure, confirming successful and selective infiltration. Scale bars represent 200 nm.

Effect of Film Thickness

Orientation control was probed for film thicknesses above 1 L₀ (31 nm) at a swelling ratio of 1.6. Vertical structures were observed up to ∼ 54 nm (>1.5 L₀), where, despite a lower surface field, entropic confinement effects from incommensurability can still play a role in stabilizing vertical orientation. Thicker films (>90 nm, ∼ 3 L₀), however, exhibit horizontally oriented cylinders (Figure ). The transition to horizontal orientation with increasing film thickness is characteristic of a diminishing surface field, validating the role of swelling-induced surface neutrality.

8.

AFM phase images of PS-b-P4VP­(PDP)_1.7 thin films showing the effect of film thickness on morphology. Films with thicknesses of 116 and 90 nm exhibit horizontal cylinders, while films with thicknesses of 54 and 37 nm exhibit vertical orientation. Scale bars represent 200 nm.

Grazing-incidence small-angle X-ray scattering (GISAXS) was used to confirm the internal morphology of the BCP complex at various thicknesses (Figure ). The GISAXS patterns of the ∼ 54 nm and ∼ 33 nm films (Figure a-b) exhibit scattering rods consistent with vertical cylinders at q _x = 0.02 Å^–1 (obtained from a line cut at q _z = 0.05 Å^–1, Figure S10) and ∼ 31 nm lateral periodicity. The 54 nm film exhibited a reduced vertical streak intensity and a broadened peak (Figure a), which could be attributed to an increase in thickness but may also indicate mixed orientations. In contrast, the 33 nm film displays sharp, vertically elongated scattering rods (Figure b). Considering the GISAXS patterns above and the vertical orientation observed in films down to 15 nm (<0.5 L₀), we infer that the cylinders remain vertical throughout the film.

9.

AFM phase images (left) and their respective 2D GISAXS patterns (right) of PS-b-P4VP­(PDP) thin films. Films with thicknesses of (a) 54 and (b) 33 nm were prepared at a PDP ratio of 1.7, and (c) 33 nm film at a ratio of 1.5. Scale bars represent 200 nm.

At ∼ 109 nm, Debye–Scherrer ring-like features (Figure S11) suggest mixed orientation in agreement with the AFM micrograph at 90 nm (Figure ), where horizontal cylinders with vertical features in between were observed. Interestingly, dot-like patterns along q _x = 0.02 Å^–1, at q _z = 0.04 Å^–1 and q _z = 0.07 Å^–1 were obtained at ∼ 123 nm (∼4 L₀) (Figure S12). Their $\sqrt{3}$ relationship, when measured from the center of the Yoneda band, indicates highly oriented horizontal cylinders (Figure S13). The diffraction pattern is consistent with a 2D hexagonal cylinder with the (10) plane oriented parallel to the substrate. The absence of the $\sqrt{3}$ peak in the in-plane direction suggests limited lateral order in the lower layers of the films. Alternatively, it has been proposed that the diffracted signals in a BCP complex may further scatter the secondary lamellar structures from the P4VP­(PDP) block. This is supported by the emergence of peaks at 0.16 Å^–1 on the q _x -axis along the Yoneda band (q _z = 0.028 Å^–1) seen in horizontal features of both 109 and 123 nm films, which is characteristic of the 4 nm P4VP­(PDP) internal lamellar periodicity (Figure S11 and Figure S12). , Taken together, the hierarchical structure is known as a “lamellae within cylinder” morphology. The P4VP­(PDP) lamellar structure forms as a result of microphase separation between the crystallizing PDP side groups and P4VP backbone.

Effect of PDP Ratio

The PDP ratio, corresponding polymer volume fractions, and resulting morphologies are summarized in Table . The PDP loading in PS-b-P4VP­(PDP) supramolecular complexes plays a pivotal role in directing the resulting morphology by modulating the effective volume fraction of the P4VP domain, altering the strength of hydrogen bonding, and influencing interfacial interactions. Importantly, these effects are not independent but collectively determine both the accessible morphology and the solvent response of the system during SVA. The PDP ratio has been shown to be a significant factor in controlling surface preference, with a higher ratio promoting vertical orientation. This was attributed to the tendency of the PS cylinders to align perpendicular to the P4VP­(PDP) internal lamellae to maximize conformational entropy. Thus, when P4VP­(PDP) lamellar orients parallel to the substrate, it forces the PS cylinders to align in the perpendicular direction. However, this interpretation does not fully capture the behavior observed in this study. Our results do not show the secondary 4 nm lamellar in the vertically oriented films (Figure S14), indicating that the BCP complex can adopt perpendicular structures without the P4VP­(PDP) lamellar order. This suggests that vertical orientation is not directly governed by hierarchical lamellar ordering, but instead arises from solvent-mediated modulation of interfacial energies. Reducing the PDP ratio to 1.5 (Figure c) in fact produced sharper scattering rods, suggesting improved order. The lateral order of the vertical cylinders may be disrupted due to excess PDP acting as a solvent. Further reduction in the PDP ratio to 1.3 or lower results in a morphological shift from cylinders to horizontal lamellar hole-island features (Figure S15–S17). As summarized in Table , decreasing the PDP ratio systematically decreases the effective volume fraction of the P4VP­(PDP) domain (ϕ_P4VP(PDP)), shifting the system from an asymmetric cylinder-forming regime (ϕ ≈ 0.58–0.60 at r = 1.5–1.7) to a symmetric lamellar regime (ϕ ≈ 0.5 at r = 1.0).

Overall, these results demonstrate that the PDP ratio does not directly dictate vertical orientation, but instead governs morphology and solvent selectivity, which together define the narrow processing window for surface energy neutralization and vertical morphology evolution.

Terrace Formation in PS-b-P4VP­(PDP) Thin Films

At a near-monolayer film thicknesses (∼30 nm), PS-b-P4VP­(PDP) thin films with PDP loadings of 1.5 and 1.7 show clear terrace structures after chloroform solvent vapor annealing, as seen in Figure . These terrace features are consistently observed across multiple scanned regions (2 μm × 2 μm), confirming that they are representative of the overall film morphology rather than localized artifacts. These terraces appear as steps in height profiles across the film, ranging between 10 and 12 nm (Figure S18). Terrace formation is a common feature in block copolymer thin films when the film thickness does not match the spacing of the microdomains. , In these situations, the film reduces its free energy by forming regions of specific thicknesses rather than stretching or compressing the domains evenly. A related terrace behavior has been reported by van Zoelen et al. for PS-b-P4VP­(PDP) thin films, in which chloroform vapor annealing of thicker films (∼70–80 nm) over long annealing times (∼20 h) produced terraces associated with parallel microdomain orientations and hierarchical terrace-within-terrace structures. In contrast, the terraces observed here form at a near-monolayer thickness under short, controlled SVA conditions and occur within a vertical cylinder morphology, without hierarchical subterracing. Instead, they arise from discrete height accommodation of vertically oriented domains under thickness incommensurability. The appearance of terraces has only been reported for morphologies parallel to the substrate, where incommensurability results in sub-filled layers at the top of the film and vertical structures appear at the edges of the terraces. To the best of our knowledge, this is the first report of a vertical cylinder morphology with terraces that exhibit horizontal cylinders on the edges in BCP complexes. The morphology at separate steps appears identical, including incommensurate films below 1 L₀ (20 nm, Figure S19). The formation of terraces in a vertical orientation suggests that the BCP complex has a preference to discretize the height of the cylinders. This may be attributed to the PDP orientation in the BCP complex. While we do not observe the secondary lamellar structure associated with the P4VP­(PDP) block in the q _z direction, the grafted PDP could adopt a metastable nematic phase parallel to the substrate.

10.

AFM phase images of ultrathin PS-b-P4VP­(PDP) thin films (32 nm) after chloroform solvent vapor annealing, illustrating terrace formation and local domain reorientation. Step-like terrace boundaries are observed where vertically oriented cylinders at different levels are bridged by horizontal features. The images show representative terrace geometries consistent with terrace-mediated accommodation of noncommensurate film thickness. Scale bars: 400 nm.

Conclusions

The present study investigated the selective swelling-induced vertical orientation in PS-b-P4VP­(PDP) block copolymer complexes. The wetting properties and swelling behavior of PS-b-P4VP­(PDP) thin films were revisited, highlighting several discrepancies with longstanding assumptions surrounding this system. The role of surface neutrality in vertical orientation was revealed by probing the effects of swelling ratio, film thickness, and PDP loading on the orientation of the supramolecular PS-b-P4VP­(PDP) thin films. By varying each parameter, clear transitions were observed between horizontal orientations, mixed morphologies, and vertically oriented domains. Thermal solvent vapor annealing revealed three distinct swelling regimes, with a narrow neutral window in the intermediate regime. This window arises from a balance between selective swelling, which is favored at low solvent activity, and mobility of the polymer for self-assembly at higher solvent activity. In contrast to previous reports, which suggested that the P4VP­(PDP) secondary lamellar packing directs cylinders toward perpendicular alignment, these structures were absent in the vertically oriented films in question and were observed exclusively in the horizontally aligned cylinders. These findings offer insight into the swelling-induced neutrality for vertical orientation in supramolecular block copolymer self-assembly, supporting their application in areas such as next-generation nanolithography, templated nanomaterials, and soft functional devices.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmaterialsau.6c00018.

• Additional experimental details and supporting characterization: ¹H NMR spectra of P4VP, PS, PS-b-P4VP, and 3-pentadecylphenol; thermal solvent vapor annealing setup; in situ Filmetrics thickness evolution during solvent absorption of P4VP­(PDP)_1.7, PS, and PS-b-P4VP­(PDP)_1.7 thin films; schematic of the controlled solvent vapor annealing system; in situ swelling profile at α = 0.8; annealing-time-dependent AFM phase images; AFM phase images after selective precursor infiltration and polymer removal; GISAXS line cuts and two-dimensional scattering patterns for films of different thicknesses; PDP-ratio-dependent AFM height images; and terrace height and phase images of ultrathin PS-b-P4VP­(PDP)_1.7 films (PDF)

B.S.C. conceptualized the idea and oversaw the work. C.L.I conducted experiments, analyzed the data and prepared the figures. N.H., S.T.M. S.D., G.M., S.A., and G.D. conducted the experiments. All authors contributed to preparing this article.

The authors declare no competing financial interest.

Published as part of ACS Materials Au special issue “Design of Complex Materials”.