Influence of spin coating parameters on the fabrication of free standing porous and nonporous poly(ε-caprolactone) based films

Abstract

Spin coating is a widely used technique for producing uniform thin films with controlled thickness; however, its use in fabricating free-standing and porous films remains relatively underexplored. This study reports the successful fabrication of free-standing biodegradable PCL and PCL/PEO (50/50) blend films via spin coating by varying polymer concentration, spin speed, and the number of spinning cycles. The resulting films exhibited uniform thicknesses ranging from 15 to 141 μm for PCL and 24–228 μm for the PCL/PEO blends, with a milky, translucent appearance and occasional macroscopic dendritic structures. The incorporation of PEO increased the solution viscosity, leading to thicker films. SEM imaging revealed diverse surface morphologies, including spherulites, phase-separated PEO-rich protrusions, worm-like structures, and micropores. XRD analysis confirmed the crystalline structures of both PCL and PEO, with reduced peak intensities at higher polymer concentrations and spin speeds, suggesting increased lattice imperfections and amorphization. Selective leaching of PEO yielded white, opaque, and porous PCL films with thicknesses ranging from 8 to 90 μm and an average porosity of 47%. Water uptake ranged from 86 to 196%, correlating with film thickness and pore structure. XRD analysis confirmed complete removal of PEO while preserving the crystalline structure of PCL. Optimal porous morphologies were achieved at a polymer concentration of 160 mg/ml, with a spin speed of 6000 rpm further enhancing crystallinity. These tunable PCL-based films show promise for applications in biodegradable scaffolds, wound dressings, drug delivery systems, and filtration.

Introduction

Polymeric films are thin, continuous materials typically defined as having thicknesses below 200 μm, making them highly versatile for a wide range of industrial applications¹. Within this category, films with thicknesses up to a few micrometers are referred to as thin films², while those below 100 nm are classified as ultrathin films^3,4. These films have found significant utility in specialized applications such as electronics, optics, surface coatings, energy generation and storage, sensors, medical devices, and filtration². The thinness of these films imparts unique properties compared to bulk materials, including high surface-area-to-volume ratios, enhanced permeability, increased flexibility, improved transparency, and reduced weight all of which are critical in many high-tech and scientific innovations. Depending on the intended use, these films may be applied as surface coatings to modify or enhance the functionality of substrates^5,6, or fabricated as free-standing, self-supported materials^6,7. Thin films are primarily produced using physical or chemical deposition methods^2,8. While many techniques are available for coating thin films onto substrates, the fabrication of free-standing films remains more limited⁷. Industrially, methods such as extrusion, blowing, solution casting, and calendering are commonly employed to produce thicker, free-standing films¹. Physical stretching of these films can generate thinner films with enhanced tensile strength and modulus, attributed to the orientation of polymer chains induced during deformation⁹.

Among the various thin film fabrication techniques, spin coating stands out for its simplicity, rapid processing, low energy requirements, and cost-effectiveness. It is used to produce both coatings and free-standing films with uniform thickness and smooth surfaces, typically ranging from tens of nanometers to several micrometers^7,10. This technique allows precise control over film thickness and morphology by adjusting parameters such as spin speed and time, solution concentration, surface tension, and viscosity (influenced by polymer molecular weight), as well as solvent evaporation rate, dispensed solution volume, and the number of coating cycles.

Spin coating has occasionally been observed to induce porosity in films. The formation of pores is influenced by interactions among the polymer, solvent, and substrate, as well as by atmospheric humidity, solution viscosity, and post-treatment methods. One primary mechanism for creating porosity is known as the breath figures phenomenon. During spin coating, rapid evaporation of volatile solvents cools the surface, leading to the condensation of water vapor into droplets. As these droplets evaporate, they leave behind cavities, resulting in a porous film structure^11–13. Pore size can range from a few hundred nanometers to several micrometers, depending on factors such as humidity, solvent surface tension, interfacial tension between solvent and water, solvent evaporation rate, polymer molecular weight, solution concentration, and spin speed^11,13,14. Saleem et al.^15,16 employed a different mechanism to induce porosity in thin films. In their work, porosity in polyethylene and polypropylene films was attributed to the presence of amorphous, loosely connected polymer chains in the as-produced films. Annealing after spin coating increased crystallinity and mechanical strength, allowing the films to be peeled from the substrate; however, this treatment also reduced porosity. Additional heating beyond the melting point produced nonporous, free-standing films.

In another approach, porous polysulfone coatings for drug delivery applications were fabricated by selectively dissolving calcium carbonate nanoparticles embedded within the polymer matrix¹⁷. Additionally, phase separation in immiscible polymer blends combined with selective dissolution of a porogen has been employed to produce porous ultrathin films^18,19. Furthermore, Wang et al.²⁰ utilized a combination of phase inversion through the immersion-precipitation method and spin coating to create membranes with a variety of surface textures smooth, dense, porous, or nodular as well as cross-sectional structures ranging from macrovoid and disordered to dense or mixed morphologies, by carefully adjusting the spinning parameters. Another avenue was explored by Yuan et al.^21,22, where they developed a microporous through-hole membrane by spin coating uncured polydimethylsiloxane (PDMS) on a master mold featuring a micropillar array created via lithography. In this method, spin speed played a critical role in successful membrane formation.

To obtain free-standing films, both mechanical strength and thickness are critical factors, as the strain required for delamination increases with decreasing film thickness⁷. Delamination occurs spontaneously when the strain energy stored in the film exceeds the interfacial energy resisting its separation from the substrate^7,23. Mechanical peeling can assist in releasing thicker films. The use of low surface energy substrates, such as fluorinated polymers (e.g., Teflon) or siloxane-based polymers like PDMS²⁴, as well as surface-modified substrates, can further reduce adhesion and facilitate easier release²⁵. In some cases, spontaneous delamination is achieved through swelling-induced detachment, where immersion in a solvent bath causes the film to swell, weakening adhesion and enabling separation²³. In one case, poly(ε-caprolactone) (PCL) thin films were easily separated from a polyurethane acrylate (PUA) mold after being cooled to their vitrified state by immersion in liquid nitrogen²⁶. On the other hand, Saleem et al.^15,16 relied on strengthening the film through an annealing step to produce free-standing thin films (5 to 20 μm thick). For thinner films, more specialized detachment techniques are required to avoid damage during release. Common techniques include the use of a sacrificial underlayer between the film and substrate^18,27,28, or a supporting overlayer deposited on top of the film to assist in peeling^24,29. The sacrificial layer is later dissolved in a suitable solvent to release the film intact.

The main drawbacks of spin coating include its low material efficiency, as typically only 2–5% of the dispensed material is utilized, while the remaining 95–98% is discarded^7,10. Additionally, spin coating is limited to planar substrates and is constrained by substrate size; larger substrates pose challenges in achieving uniform film thinning. Despite these limitations, spin-coated free-standing films offer significant advantages for direct characterization and hold great potential in diverse applications, including biomedical uses (e.g., wound dressings, drug delivery systems, tissue engineering scaffolds), electronic devices (e.g., sensors, micromachines, solar cells), catalysis, and separation membranes^7,10.

Recent research has increasingly shifted toward biodegradable and bio-based materials as sustainable alternatives for various applications. This study reports on the production of biodegradable, PCL-based free-standing films, both nonporous and porous with porosity levels of ~ 50% and varying thicknesses. Poly(ethylene oxide) (PEO) is introduced as a biodegradable filler to modify the mechanical properties and hydrophilicity of PCL³⁰. PCL and PEO are both biodegradable yet immiscible polymers, leading to phase-separated morphologies when blended. This immiscibility enables selective removal of the water-soluble PEO phase, generating controlled porosity within the PCL matrix after spin coating. While several studies have explored the spin coating of semicrystalline PCL, PEO, and their individual blends and composites (as outlined in Table 1), only one recent study by Bauer et al.³¹ investigated the morphology of spin-coated PCL/PEO blends (100/0, 90/10, 80/20, 70/30) using selective enzymatic degradation and high-resolution SEM to reveal semicrystalline and phase-separated features. To the best of our knowledge, this is the first study to utilize this immiscible biodegradable polymer system in combination with the porogen leaching technique to fabricate porous PCL free-standing films via spin coating. This work serves as a proof of concept, systematically investigating the influence of three spin coating parameters on the fundamental process–structure relationships of the resulting films. The findings lay the foundation for future investigations into mechanical behavior, pore size distribution, and application-specific performance.

Table 1.

Summary of spin coating techniques for PCL- and PEO-based films and other solution-based fabrication methods for PCL/PEO blends from literature reports.

Ref.	Materials	Processing steps	Solvent(s)	Substrate(s)	Processing parameters and resultant thickness
9	PCL	spin coating – melt pressing – hot biaxial drawing – thermal annealing	dichloromethane (DCM)	metal foil	Mn = 80 kg/mol. concentration = 3 wt %. dispensed volume = 5 ml. substrate size = 20 cm2. spin speed = 500 rpm. spin time = 300 s. spin cast film thickness = 6.3 ± 1.4 μm. biaxial drawn film thickness = 1.2 ± 0.5 μm.
26	PCL	spin coating	toluene	nano-grooved polyurethane acrylate (PUA) mold	Mn = 50 kg/mol. concentration = 20 wt%. spin speed = 3500 rpm. spin time = 60 s. free-standing nanopatterned thin films thickness = 1.7 ± 0.3 μm.
32	PCL	spin coating –thermal annealing	cyclohexanone	glass	Mw = 8.4 kg/mol. concentration = 0.1–3.0 wt %. spin speed = 2000 rpm. spin time = 90 s. thickness = 18–105 nm.
33	PCL	spin coating –thermal annealing	toluene	mica	Mw = 22, 65, and 146 kg/mol. concentration = 2.87, 1.02, 0.52, 0.102, 0.052 wt%. thickness = 4–120 nm.
34	PCL	spin coating	tetrahydrofuran (THF)	silicon	Mn = 39 kg/mol. concentration = 1.0 mg/ml. spin speed = 3000 rpm. spin time = 60 s. Thickness = 6 ± 1 nm.
35	PCL	spin coating	tetrachloroethane [Cl2CHCHCl2]	glass	Mw = 80 kg/mol. concentration = 0.2, 0.4, 0.6, 1.0, 1.6 and 2.0 g/ml. spin speed = 1000, 2000, 4000, and 6000 rpm. thickness = 0.07–16.3 μm.
36	PCL	spin coating	toluene	mica	Mw = 22, 65, and 146 kg/mol. concentration = 0.05–4.0 wt%. thickness = 4–20 nm.
37	PCL, PCL/Poly(vinyl methyl ether) (PVME)	spin coating	toluene	silicon	Mw = 14 /90 kg/mol. concentration = 5–20 mg/ml. spin speed = 3000 rpm. spin time = 60 s. thickness = 30–120 nm.
38	PCL, PCL/PVC, PCL/SAN PCL/PC PCL/CPE, PCL/PVME, PCL/PB, PCL/PBA, and PCL/PBMA	spin coating	THF DCM toluene	silicon	Mw = 14/130, 180, 64, 190, 90, 3, 66, 320 kg/mol, respectively. concentration = 20 mg/ml. spin speed = 3000 rpm. spin time = 60 s. thickness = 120 ± 4 nm.
39	PCL, PCL/β-tricalcium phosphate (β-TCP) multilayer: PCL/TCP – PCL –PCL/hyaluronic acid (HA)	spin coating	2,2,2–Trifluoroethanol (TFE)	glass	Mw = 70–90 kg/mol. concentration = 10% w/v PCL; 10% w/v β-TCP; 10% w/v HA. spin speed = 1500 rpm. spin time = 15 s. drying for 900 s at room temperature between cycles. overall thickness of ~ 72 μm.
40	PCL, PCL/poly[2-(methacryloyloxy) ethyltrimethylammonium (PMTA), PCL/ gelatin bilayer: PCL/PMTA – PCL/Gel	spin coating combined with in-situ crosslinking polymerization	TFE	glass	Mw = 80 kg/mol. concentration = 7% w/w PCL; 8% w/w gelatin. spin speed = 800 rpm. spin time = 20 s. At the 8 s of coating time, PCL/Gel solution was instantaneously added on the already spread PCL/PMTA flat sheet.
41	PCL/chitosan (1:3, 1:1, 3:1 v/v)	spin coating	80% acetic acid aqueous solution	glass	Mw = 50 kg/mol. concentration = 1% w/v. dispensed volume = 100 µl. spin speed = 4000 rpm. spin time = 60 s. thickness = 10 μm.
42	PCL/chloramphenicol (CAM) antibiotic	spin coating	THF	glass	Mw = 130 kg/mol. concentration = 10 mg/ml. spin speed = 1000 rpm. spin time = 60 s. thickness = 800 nm.
43	PCL/alumina ( Al2O3), PCL/graphene, PCL/carbonated hydroxyapatite (cHAp), PCL/titania (TiO2) multilayer film	spin coating	Chloroform (CHCl3) – ethyl alcohol (2:1)	glass	Mw = 80 kg/mol. inorganic additive = 12 wt%. concentration = 8 wt%. dispensed volume = 2 ml. spin speed = 1000 rpm. spin time = 30 s. single layer thickness = 20 ± 1.5 μm. multilayer thickness = 50 ± 4 μm.
44	PCL/polyetherimide (PEI) (5/95, 10/90, 25/75, 50/50)	spin coating	DCM	glass	Mw = 65/30 kg/mol. concentration = 0.02 g/ml. substrate size = 25 mm. spin speed = 500 rpm. thickness = 300 nm.
45	PCL, PCL/polystyrene (PS) (90/10, 60/40, 50/50, 20/80)	spin coating	toluene	mica	Mw = 50/54 kg/mol. PCL concentration = 1 and 0.5 wt %. PS concentration = 0.1–1 wt %. spin speed = 3000 rpm. spin time = 30 s. pure PCL thickness = 30 ± 3 and 15 ± 3 nm. pure PS thickness = 30 ± 3 nm.
46	PCL/silica (SiO2) nanotubes (2.5 wt%), PCL/strontium hydroxyapatite (HAp) nanorods (2.5 wt%)	spin coating	CHCl3	glass	Mn = 65 kg/mol. concentration = 10 wt%. substrate size = 2 × 2 cm2.
47	PCL/amorphous carbonated calcium phosphate (caCP)	spin coating	DCM	Si3N4–3 wt% multiwalled carbon nanotube (MWCNT)	Mw = 80 kg/mol. concentration = pure 10% w/v PCL/DCM solution mixed in a 2:1 ratio with 5% w/v caCP/DCM suspension. substrate size = 5 × 5 × 3 mm3. spin speed = 300 rpm then 1000 rpm.
48	PCL/chlorinated polyethylene (CPE) (100/0, 90/10, 80/20, 70/30, 60/40)	spin coating	toluene	silicon	Mw = 14 kg/mol. spin speed = 3000 rpm. spin time = 60 s. 20 mg/ml concentration produced 120 ± 4 nm thickness.
49	curcumin-loaded PCL/polyvinyl alcohol-hydroxyapatite (Cur-loaded PCL/PVA-HAp	spin coating of emulsion	DCM for PCL solution Water for PVA-HAp slurry	titanium (Ti) plates	substrate size = 20 × 20 × 2 mm3 for the drug release tests or 10 × 10 × 2 mm3 for other tests. spin speed = 3000 rpm. spin time = 40 s. Single layer and multi-layer.
50	PCL/HAp gradient (50/50, 55/45, 60/40, 65/35, 70/30, 75/25, 80/20, 85/15, 90/10, 95/5)	layer-by-layer spin-coating. Film patterned with an array of funnel-shaped microchannels by laser machining	1,4-dioxane	silicon	Mn = 80 kg/mol. concentration = 6, 8, 10 wt% PCL. dispensed volume = 1.5 ml. substrate size = 4 inches. spin speed = 1460 rpm. spin time = 63 s. 20-layer film, two-layer spin coated from each composition. thickness = 24.38 ± 0.74 μm, 46.88 ± 0.83 μm, and 103.38 ± 1.06 μm.
51	PCL/PDMS (10:90, 25:75)	spin coating	CHCl3	PDMS/NdFeB	Mw = 14 and 80 kg/mol. substrate size = 60 mm. dispensed volume = 0.4 ml. spin speed = 500–2000 rpm. spin time = 10, 15, 20 s. thickness = 36 ± 3.8 − 159 ± 4.0 μm.
52	PEO	spin coating	CHCl3, methanol (MeOH), or THF	glass	Mn = 4, 10, 34 kg/mol. concentration = 10 wt%. dispensed volume = 200 µl. spin speed = 1500 rpm. spin time = 66 s.
53	PEO	spin coating – melt recrystallization	MeOH	glass	Mw = 4, 8, 20 kg/mol. concentration = 10, 20, 50 mg/ml. substrate size = 20 × 20 × 0.16 mm3. spin speed = 300 rpm for 5 s, then 1500 rpm for 5 s, then 3000 rpm for 30 s. thickness = 220 ± 30, 450 ± 50, and 1500 ± 200 nm.
54	PEO	spin coating at 15 °C	benzene	mica	Mw = 9–12.5 kg/mol. concentration = 0.04–0.4 mg/ml. spin speed = 4000–8000 rpm.
55	PEO/polyvinylidene fluoride (PVDF) (25/75)/ Tungsten oxide (WO2) nanoparticles (5, 10, 15, 20%)	spin coating	dimethylformamide (DMF)	glass	concentration = 25 mg/ml. substrate size = 2.5 × 2.5 cm2.
56	PEO/chitosan (0/100, 10/90, 20/80, 30/70, 40/60, 50/50, 100/0)	spin coating	acetic acid	silicon	Mw = 900/150 kg/mol. concentration = 1% w/w. dispensed volume = 0.5 ml. substrate size = 4 cm2. spin speed = 2500 rpm. spin time = 60 s. thickness = 32–74 nm.
57	PEO/CoCl2 (PCL5) (95/5 wt%)	spin coating	MeOH	glass, quartz, indium tin oxide coated glass	Mw = 5,000 kg/mol. spin speed = 3500 rpm. spin time = 40 s. thickness = 8.6–9.2 μm.
58	PEO/poly(methyl methacrylate) (PMMA) (30/70 − 50/50)	spin coating	1,2–dichloroethane	silicon	Mw = 101.2/ (6.88–101) kg/mol. concentration = 1.25 wt%. thickness = 120 nm.
59	PEO/PMMA/cloisite (montmorillonite) clay particles	spin coating	CHCl3	silicon	Mw = 150/7.3 kg/mol. concentration of 1 wt%. spin speed = 2000 rpm. Thickness = 200 nm.
31	PCL/PEO (100/0, 90/10, 80/20, 70/30)	spin coating	CHCl3	hydrophobic glass or silicon	Mn = 80/Mv = 600 kg/mol. concentration = 2 wt% and 5 wt% PCL. dispensed volume = 1–1.5 ml. substrate size = 1 × 1 cm2. spin speed = 2500 rpm. spin time = 60 s.
60	PCL/PEO	electrospinning	DCM – DMF (3:2)	aluminum foil	Mw= (70–90)/900 kg/mol.
61	PCL/PEO (100:0, 75:25, 50:50, 25:75, 85:15, 90:10, 5:95, 0:100)	electrospinning	acetic acid –formic acid (2:1)		Mn = 80/100 kg/mol.
62	PCL/PEO	electrospinning	DCM – DMF (3:2)		Mw= (70–90)/900 kg/mol. PEO concentration = 1.8 w/v%; PCL concentration = 14 w/v%. PEO concentration = 2.4 w/v%; PCL concentration = 12 w/v%.
63	PCL/PEO	electrospinning	DMF – CHCl3 (10:90)		Mw = 80/100 kg/mol. PCL concentration = 18 wt%. PEO concentration = 1 wt%.

Materials and methods

Materials

The PCL (CAPA6506) used in this study was kindly supplied by Perstorp UK limited, UK. Its reported properties include a molecular weight (Mn) of 50 kg/mol, a density of 1.14 g/cm³, and a melting temperature (T_m) of 60 °C. PEO was purchased from Sigma-Aldrich. Its reported properties include a molecular weight (Mv) of 100 kg/mol, a density of 1.13 g/cm³, and a T_m of 65 °C. Both polymers were in powder form. Chloroform, the solvent used, was purchased from the local market. The substrate was silicone rubber, sourced from a commercially available silicone rubber baking pan. The pan was cut into 50 mm circular sheets using a precision-machined circular punch obtained from a local workshop. This method ensured consistent and accurately shaped circular substrates (approximately 100% round) for spin coating.

Preliminary trials and experimental design

This study consisted of two main phases. In the initial phase, several preliminary trials were conducted to: (a) determine an appropriate solvent and substrate, (b) confirm the feasibility of spin coating pure PCL and a PCL/PEO (50/50 wt%) blend, and (c) establish feasible ranges for various processing parameters. These trials involved systematically adjusting the parameters, starting from values reported in the literature, to identify suitable levels for subsequent experiments. Following the preliminary trials, a general full factorial design, with only one replication was designed using Minitab to examine the effects of three selected factors and their interactions on the response variables, which were film thickness and morphology. The selected factors were concentration, spin speed, and the number of spinning cycles. Two concentration levels were selected (160 and 320 mg/ml), along with two levels of spin speed (3000 and 6000 rpm) and two levels for the number of cycles (1 and 3 spinning cycles).

Spin coating

To prepare films by spin coating, a solution of the required concentration was first prepared. Specific amounts of PCL or PCL/PEO were gradually dissolved and mixed in chloroform using a magnetic stirrer (C-MAG HS 7, IKA, Germany) at room temperature for 15 to 30 min, depending on the concentration, until a clear solution was obtained. The prepared solution was then spin-coated at room temperature using a VTC 100-A spin coater (MTI, USA), fitted with a 50 mm chuck, selected to accommodate the 50 mm silicon rubber substrate. Prior to spinning, the substrates were thoroughly cleaned with acetone to ensure a clean surface for coating. A constant volume (2 ml) of the prepared solution was dispensed onto the substrate for each spin cycle, regardless of the solution’s concentration. The solution was dispensed onto the substrate either statically or dynamically, depending on its viscosity. Static dispensing, where the solution is dispensed before spinning starts, was used for higher concentration solutions to reduce the risk of uneven coverage or material loss during the spin process. Conversely, dynamic dispensing, where the solution is dispensed while the chuck is spinning, ensured more uniform spreading for lower concentration solutions that would otherwise drip off the edges too quickly⁷.

The spinning cycle was carried out in two stages, each with distinct parameters for time and speed. The first stage, set at 500 rpm for 30 s, aimed to spread the solution evenly over the substrate. The second stage, intended to refine the film thickness and facilitate solvent drying, used higher speeds of 3000 or 6000 rpm for an additional 30 s. After spin coating, the films were allowed to air dry for several days to ensure complete evaporation of the remaining solvent. Once dry, the films typically detached themselves from the substrate and were gently removed using tweezers.

Dissolution of PEO

To create a porous film, air-dried PCL/PEO films were soaked in distilled water at room temperature for 4 days to allow the PEO component to completely dissolve. Due to the immiscibility of PCL and PEO, the blend exhibits a phase-separated morphology, and at a 50/50 composition, this typically forms an interconnected (co-continuous) structure. This co-continuous morphology facilitates the complete and uniform selective removal of PEO upon water immersion, resulting in a porous PCL film. After the dissolution process, the films were carefully removed from the water and dried in a Heratherm oven (ThermoFisher Scientific, Inc., USA) at 40 °C. The weights of the films were measured before and after leaching the PEO, (m_blend) and (m_dry), respectively, to determine the weight loss of PEO and, consequently, the porosity of the films, using the following equation:

1

The water uptake capacity of the films was further evaluated by comparing the wet weight (m_wet) of the sample after PEO leaching with the dry weight (m_dry) after drying in the oven accordingly, using the following equation:

2

Film thickness measurement

The thickness of the films was measured using a stereo microscope (SMZ745, Nikon Instruments, Inc., USA) equipped with an image analysis software. The circular films were cut through the center to assess thickness uniformity from one edge to the opposite edge. To ensure accurate measurements, the film sections were placed between acrylic strips to maintain the correct orientation for viewing the cross-section. A 0.1 mm thickness gauge was used to calibrate the image analysis software. Five measurements were taken at different locations along the film diameter, and the average thickness and standard deviation of the readings were calculated.

Surface imaging

The surface morphologies of the films were analyzed using a Vega 3 TESCAN scanning electron microscope (SEM) at an accelerating voltage of 10 kV. The samples were mounted on aluminum stubs with double-sided carbon tape and were sputter coated with platinum using an Emitech K550X sputter coating device from Quorum Technologies.

X-ray diffraction (XRD)

X-ray diffraction (XRD) patterns were collected at room temperature using a D2 Phaser (Bruker, Germany) powder diffractometer, employing Cu Kα radiation (1.54056 Å, 30 kV, 10 mA). Data were recorded from 15° to 28° 2θ at a scanning rate of 0.02°/sec. Samples included pure PCL, PCL/PEO blend, and porous PCL films, all prepared using three spinning cycles. To examine the PEO distribution, both faces of the PCL/PEO blend films were analyzed. Additionally, a PCL/PEO blend sample prepared using a single spinning cycle was tested to investigate differences in crystalline structure compared to the three-cycle samples. OriginPro 8.5 software (OriginLab Corporation, Northampton, MA, USA) was used to fit the peaks. The Scherrer formula was applied to estimate the average crystallite size for both PCL and PEO components:

3

where (K) is the crystal shape factor (taken as 0.9), (λ) is the wavelength, FWHM is the full width at half maximum of the peak, and (θ) is the peak position. For this analysis, only non-convoluted peaks were used. PCL exhibits two main high-intensity peaks at 2θ ≈ 21.3° and 23.7°, while PEO shows high-intensity peaks at 2θ ≈ 19.2° and 23.2°. The peaks around 23° overlap, making them unsuitable for separate analysis.

Results and discussion

In previous research^30,64, biodegradable PCL/PEO porous and nonporous films were fabricated using cryomilling to blend the immiscible polymers, followed by hot-pressing to form bulk films. At a 50/50 wt% composition known to produce co-continuous morphologies the selective leaching of PEO generated porous structures. The resulting films were approximately 150 μm thick and exhibited dense surface skins with lamellar porous internal structures. These films were examined for their potential in microfiltration applications³⁰. In the present study, spin coating is explored as an alternative fabrication technique to produce both porous and nonporous films. The objective is to assess whether spin coating can produce thinner, structurally controlled films that expand the functional and application potential of the PCL/PEO system. Three types of films were studied: pure PCL films, PCL/PEO (50/50) blend films, and porous PCL films obtained by selective leaching of PEO.

Preliminary trials

The study was conducted in two phases. In the first phase, preliminary trials were performed to identify a suitable solvent and substrate, assess the feasibility of spin coating pure PCL and PCL/PEO blends, and establish feasible ranges for key processing parameters. Insights from the literature guided the selection of solvents, substrates, and spin coating parameters, as summarized in Table 1.

Spin coating has been frequently used to fabricate both PCL- and PEO-based films. For PCL-based films, solvents such as dichloromethane (DCM), tetrahydrofuran (THF), toluene, and chloroform (CHCl₃) have been reported. Solution concentrations typically range from 0.05 wt% to 20 wt% or 1 mg/ml to 2000 mg/ml. Spin speeds vary between 500 and 6000 rpm, with spin times from 15 s to 5 min. These parameters yield film thicknesses from 4 nm to 16 μm for single-layer films and up to 72 μm for multilayer films, depending on the solvent, concentration, and spin parameters. Similarly, PEO-based films have been produced using solvents such as THF, CHCl₃, dimethylformamide (DMF), and methanol (MeOH), with concentrations ranging from 1 wt% to 10 wt% or 0.04 mg/ml to 50 mg/ml. These typically produce films with thicknesses between 32 nm and 9 μm. Most spin-coated films reported in the literature are not free-standing and predominantly fall within the nanometer thickness range. Despite the frequent use of spin coating for PCL- and PEO-based films, only one study has reported on spin coating PCL/PEO blends in chloroform³¹, focusing on morphological characterization across four different compositions (100/0, 90/10, 80/20, 70/30) and two solution concentrations (2 wt% and 5 wt% PCL). That study observed a transition in phase-separated structures from nucleation and growth to spinodal decomposition at the 70/30 composition. Additionally, it mapped the morphological features of the crystalline PCL phase using high resolution SEM. However, no measurements of film thickness or porosity were reported, and the influence of spin coating parameters on the resulting structures was not investigated. In contrast, electrospinning has been more commonly applied to PCL/PEO blends using various binary solvent systems, such as DCM-DMF and DMF-CHCl₃, offering useful insights into solvent combinations potentially suitable for spin coating this blend system.

Chloroform was identified as a suitable solvent for both pure PCL and the PCL/PEO blend, with preliminary tests confirming its feasibility for further experimentation. Its high volatility contributes to the rapid film formation needed for spin coating. However, due to its known toxicity and environmental impact, all handling was performed with proper safety protocols. Future studies may investigate more sustainable solvent options.

Various substrates, including glass, Teflon, and silicone rubber, were evaluated for their ability to produce free-standing films. However, strong adhesion of the polymers to glass and Teflon made them unsuitable for the study purpose, whereas silicone rubber proved to be the most practical option. When spin coating PCL onto silicone rubber, the film spontaneously detached upon drying (Fig. 1). This detachment likely resulted from the weak adhesion between hydrophobic PCL and the low surface energy silicone rubber, combined with internal stresses generated during solvent evaporation. As the solvent evaporated and the PCL film contracted, the lack of sufficient adhesive forces to counteract these stresses led to spontaneous detachment. Similarly, PCL/PEO films detached spontaneously under the same conditions. Interestingly, while a recent study has reported flaking and cracking of PCL coatings due to poor adhesion, stiffness mismatch, and glass-like behavior upon drying on an elastic PDMS/NdFeB composite substrate⁵¹, our results with cooking pan silicone rubber suggest that weak polymer–substrate interactions can be leveraged to facilitate clean film detachment.

Fig. 1.

Spin-coated films: (a) PCL film immediately after spin coating, (b) the same film after air drying, and (c) PCL/PEO film before (left) and after (right) PEO leaching, showing the resulting color change.

After PEO leaching in water, the PCL/PEO films changed in appearance, transitioning from a milky translucent appearance to an opaque white (Fig. 1(c)). Despite the removal of PEO, the films maintained their structural integrity, with complete films remaining intact.

Both syringes and pipettes were tested for dispensing the polymer solutions. Syringes were initially used in early trials, but as solution viscosity increased with higher concentrations and the addition of PEO, they became difficult to operate. Pipettes, on the other hand, provided more precise and manageable control, and were therefore used in all subsequent experiments. Various spin coating parameters, including dynamic and static dispensing modes, were systematically explored to optimize the process. Polymer solution concentrations ranging from 100 mg/ml to 500 mg/ml were tested, with dispensed volumes between 1 ml and 7 ml. Spinning was performed in two stages: the first stage employed speeds of 500, 1000, or 2000 rpm for 15–30 s, while the second stage used speeds between 500 and 8000 rpm for the same duration. Additionally, multi-cycle spin coating was explored as a strategy to increase film thickness.

The preliminary trials provided valuable insights into the spin coating behavior of PCL/PEO blends, key processing parameters, and potential defects. Based on these findings, a full factorial experiment with eight runs was designed to investigate the effects of three primary factors on film thickness, uniformity, and morphology. The selected factors were solution concentration, spin speed, and number of spinning cycles. Two concentration levels (160 and 320 mg/ml) were selected, as a higher concentration (480 mg/ml) resulted in solutions that were too viscous and difficult to dissolve. The second-stage spin speed was set at 3000 or 6000 rpm, while the number of spinning cycles was either 1 or 3. The first-stage spin speed was fixed at 500 rpm, with a duration of 30 s for each stage. A constant solution dispensing volume of 2 ml was used for all runs. The most suitable dispensing method dynamic or static was identified through preliminary testing.

Nonporous PCL films

Free-standing PCL films with diameters of 50 mm and average thicknesses ranging from 15 μm to 141 μm were successfully prepared by varying the polymer concentration, spin speed, and number of spinning cycles. Thickness measurements for the nonporous PCL films are summarized in Fig. 2. Film thickness generally decreased with increasing spin speed, while higher polymer concentrations and additional spinning cycles resulted in thicker films. These trends are consistent with previous findings for PCL³⁵ and HDPE¹⁵. The film thickness was generally uniform, with an average coefficient of variation of 16%. The solution dispensing step was identified as the most critical factor contributing to nonuniformity, automating this step could significantly improve film uniformity.

Fig. 2.

Average thickness of nonporous PCL films as a function of process parameters.

The films exhibited a milky translucent appearance. Films produced at a concentration of 160 mg/ml and a spin speed of 3000 rpm exhibited notable non-uniformity, characterized by bubble-like regions, some of which developed circumferential tears that separated them from the surrounding film. This phenomenon may be attributed to phase separation into solvent-rich and polymer-rich regions during solvent evaporation. In spin coating, rapid solvent evaporation can lead to localized concentration gradients, where the polymer-rich phase solidifies while the solvent-rich phase forms thinner or even void-like areas. This effect is more pronounced at lower polymer concentrations due to the higher solvent content, which makes the system more susceptible to phase separation. Film uniformity improved with increasing spin speed and/or concentration, likely due to enhanced solvent removal and more homogeneous solidification (Fig. 3). Some films also exhibited macroscopic snowflake-like dendritic structures, along with randomly distributed rimmed bubbles of varying sizes. These features were especially pronounced in three-layer films and those produced at the lower spin speed (Fig. 3(c) and (d)). Thermal annealing may help homogenize the film by promoting polymer chain relaxation and redistribution, thereby reducing defects such as dendritic structures, rimmed bubbles, and phase-separated regions.

Fig. 3.

PCL films spin-coated at 3000 rpm with different concentrations and spinning cycles: (a) 160 mg/ml and (b) 320 mg/ml, both fabricated using a single spinning cycle; (c) 320 mg/ml film fabricated using three spinning cycles; (d) magnified (4×) view of (c), highlighting surface features.

Several distinct surface morphologies were observed via SEM. In some regions, spherulites were visible, with smaller sizes noted at the higher polymer concentration (Fig. 4(a) and (d)). This observation is consistent with findings by Simon et al.³⁵, who reported that increased film thickness leads to reduced crystal sizes. Similar morphologies were reported by Bauer et al.³¹ for spin-coated PCL films, where spherulitic patterns and lamellar bundles became more evident after selective enzymatic degradation of the amorphous phase. In their as-spun films, the crystalline features were faintly visible but became pronounced upon degradation, highlighting the underlying lamellar and spherulitic structures.

Fig. 4.

SEM images of PCL films: (a) 160 mg/ml, 1 cycle, 6000 rpm; (b) 160 mg/ml, 3 cycles, 3000 rpm; (c) magnified view of (b); (d) 320 mg/ml, 1 cycle, 3000 rpm; (e) 320 mg/ml, 1 cycle, 6000 rpm; (f) magnified view of (e).

Other regions in the films exhibited networks of “worm-like” protrusions or surface wrinkles (Fig. 4(c),(e), and (f)). These features may arise from the complex crystallization behavior of PCL, in which lamellar bundles twist and intertwine to form fibrous networks. Internal stress generated during solidification can also cause surface buckling or wrinkling. Furthermore, rapid solvent evaporation during spin coating may induce residual stress fields that contribute to the formation of these textures. Additional morphological irregularities were noted, including sporadic particles and localized material buildup (Fig. 4(b)). These features likely result from uneven flow or insufficient leveling during the spin-coating process.

The XRD patterns of the three-layer PCL films exhibited two main characteristic peaks within the ranges of 21.4–21.8° 2θ and 23.7–24.1° 2θ, corresponding to the crystalline structure of PCL (Table 2). Increasing polymer concentration resulted in decreased peak intensity and noticeable peak broadening, which was further pronounced at higher spin speed (Fig. 5). This trend suggests a reduction in crystallite size or an increase in lattice imperfections. However, at the lower concentration, the higher spin speed produced better-defined peaks, indicating enhanced crystallinity and/or larger crystallite size. These observations are further supported by crystallite size and crystallinity calculations, as well as SEM imaging that revealed spherulites decreasing in size at the higher concentration. In contrast, raw PCL powder exhibited XRD peaks at approximately 21.4° 2θ and 23.8° 2θ, with an average crystallite size of 10 nm and crystallinity of 36%, suggesting that the spinning conditions enhanced PCL crystallization kinetics. Overall, these results highlight the complex interplay between processing parameters and the crystallization behavior and structure of PCL films.

Table 2.

XRD data for nonporous PCL films fabricated using three spinning cycles.

Concentration (mg/ml)	Spin Speed (rpm)	Thickness (µm)	Peak Position (° 2θ)	Crystallite Size (nm)	Crystallinity (%)
160	3000	41.5 ± 3.1	21.4, 23.7	21.8	51
160	6000	66.0 ± 19.8	21.8, 24.1	23.1	48
320	3000	140.9 ± 12.0	21.5, 23.9	16.0	47
320	6000	94.7 ± 7.3	21.6, 23.9	9.8	48

Fig. 5.

XRD profiles of nonporous PCL films fabricated using three spinning cycles at different polymer concentrations and spin speeds.

PCL/PEO (50/50) blend films

Free-standing PCL/PEO (50/50) blend films with a 50 mm diameter and an average thickness ranging from 24 μm to 228 μm were successfully prepared by adjusting the same three processing parameters. Figure 6 presents the thickness measurements for the nonporous PCL/PEO blend films. The incorporation of PEO increased the solution viscosity, resulting in thicker films compared to those obtained from pure PCL at the same concentration. The film thickness remained generally uniform, with an average coefficient of variation of 12%. Similar to pure PCL films, thickness trends followed expected patterns with spin speed, polymer concentration, and the number of spinning cycles. However, the higher solution viscosity made processing more challenging, as it can lead to uneven spreading and flow instabilities. Furthermore, phase separation between PCL and PEO during drying likely contributed to film nonuniformities and defects. The immiscibility of the two polymers and their differing crystallization rates can lead to the formation of microphase-separated domains, surface roughness, and incomplete or irregular film formation.

Fig. 6.

Nonporous PCL/PEO (50/50) blend film thickness as function of process parameters.

The films exhibited macroscopic features similar to those of nonporous PCL films, including a milky translucent appearance. Some films, particularly those formed with three spinning cycles or at low spin speed, developed macroscopic snowflake-like dendritic structures. Additionally, high-concentration films spun at high speeds displayed randomly distributed bubbles of varying sizes. SEM analysis revealed distinct surface morphologies between the two sides of the films. Films prepared from 160 mg/ml solutions at the low spin speed exhibited microscopic protrusions (< 5 μm) and randomly distributed pores (< 1 μm) on the air-facing side of the film (Fig. 7(a)). These features likely correspond to PEO-rich domains, as PEO, being more hydrophilic than PCL, may preferentially migrate toward the air interface during solvent evaporation. This interpretation is supported by SEM imaging after PEO leaching, where these protrusions disappeared, confirming their PEO origin. In contrast, the substrate-facing side displayed a smoother but phase-separated morphology with interconnected domains and randomly distributed pores or dents (< 2 μm) (Fig. 7(b)). The morphology suggests that PCL crystallization strongly influenced the phase separation process, with the crystallized PCL forming worm-like in-plane structures.

Fig. 7.

SEM images of PCL/PEO (50/50) blend films spin-coated from 160 mg/ml solutions at 3000 rpm. (a, A) show one side of the film, while (b, B) and (c, C) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

Occasionally, twisted fibrous spherulitic structures resembling snowflakes or branched stars were observed (Fig. 7(c)), which may correspond to the dendritic patterns visible to the naked eye. These findings point to localized crystallization occurring primarily on the substrate-facing side. Overall, the combination of phase separation, crystallization dynamics, and polymer-substrate interactions governed the formation of distinct topographies on the two surfaces of the films.

Three-layer films exhibited similar features to their single-layer counterparts (Fig. 7(A), (B), and (C)), with one distinct difference: the PEO-rich protrusions on the air-facing side appeared depressed. This depression is likely due to mechanical compression or solvent evaporation during the drying process, which may have caused reflow or flattening of the PEO-rich domains before complete solidification. As shown in Fig. 8, increasing the spin speed resulted in comparable surface textures, but with more pronounced depressions of the PEO protrusions, more distinct worm-like phase-separated domains, and a significant reduction in micropore density, leaving behind only occasional pore clusters. These pores (Fig. 8(B′)) may have originated from partial PEO leaching caused by ambient moisture exposure, similar to the changes observed in water-leached samples (discussed in the next section). Films prepared at higher polymer concentration showed a substantial reduction and flattening of surface features (Figs. 9 and 10). Some surfaces exhibited less distinct worm-like wrinkles (Fig. 9(B′) and Fig. 10(B′)), spherulites with circumferential cracking (Fig. 10(A′)), and isolated micropores (Fig. 9(a′), (b′), and Fig. 10(b′)). Additionally, some surfaces displayed visible cracks (Fig. 9(A′)), likely arising from film shrinkage during solvent evaporation and stresses induced by crystallization.

Fig. 8.

SEM images of PCL/PEO (50/50) blend films spin-coated from 160 mg/ml solutions at 6000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

Fig. 9.

SEM images of PCL/PEO (50/50) blend films spin-coated from 320 mg/ml solutions at 3000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

Fig. 10.

SEM images of PCL/PEO (50/50) blend films spin-coated from 320 mg/ml solutions at 6000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

The XRD patterns of the PCL/PEO blend films revealed characteristic diffraction peaks corresponding to the crystalline structures of both PCL and PEO (Table 3; Fig. 11). The PCL component displayed its main peak in the range of 21.1–21.8° 2θ, while the PEO component exhibited two distinct peaks at 18.9–19.6° 2θ and 23.1–23.8° 2θ, with the latter overlapping with the minor peak of PCL. At the higher polymer concentration, peak intensities were generally reduced, with further decreases observed at higher spin speeds— indicating increased lattice disorder or smaller crystallite size. In contrast, at the lower concentration, higher spin speed produced sharper peaks, suggesting enhanced crystallinity and/or larger crystallite size. These trends are consistent with SEM observations, where sharper phase-separated networks were seen under similar conditions. Overall, these results highlight the complex interplay of solution concentration and spin speed on the crystallization behavior of the blend films. Additionally, the incorporation of PEO appears to disrupt PCL crystallinity, as reflected by the reduced crystallite size compared to pure PCL films. These results align with Bauer et al.³¹, who observed that at higher PEO concentrations (30 wt%), phase separation in PCL/PEO blends transitions from nucleation and growth to spinodal decomposition, which significantly inhibits the spherulitic growth of PCL.

Table 3.

XRD data for PCL/PEO blend films. For each fabrication condition, data from both sides of the film are presented.

No. of cycles	Conc. (mg/ml)	Spin Speed (rpm)	Thickness (µm)	PCL		PEO		Crystallinity (%)
				Peak Position (° 2θ)	Crystal Size (nm)	Peak Position (° 2θ)	Crystal Size (nm)
3	160	3000	67.9 ± 10.5	21.8, 24.1	16.4	19.6, 23.7	20.0	55
				21.5, 23.8	8.5	19.2, 23.3	9.3	54
3	160	6000	88.1 ± 7.9	21.6, 23.8	11.8	19.4, 23.4	13.3	56
				21.6, 23.8	14.0	19.3, 23.4	15.2	50
3	320	3000	228.0 ± 48.6	21.8, 23.9	14.5	19.6, 23.8	9.0	47
				21.7, 23.7	15.1	19.5, 23.7	22.0	58
3	320	6000	89.4 ± 8.7	21.6, 23.9	15.7	19.3, 23.5	15.7	43
				21.1, 23.2	10.6	18.9, 23.1	10.9	50
1	160	6000	29.5 ± 4.6	21.4, 23.6	16.0	19.2, 23.4	14.1	67

Fig. 11.

XRD profiles of nonporous PCL/PEO (50/50) films fabricated using three spinning cycles at different polymer concentrations and spin speeds. For each fabrication condition, profiles from both sides of the film are shown.

A comparison of both film faces revealed similar XRD patterns, confirming the presence of PEO on both sides. This is supported by SEM images, which show porosity on both faces after PEO leaching, indicating that PEO was uniformly distributed and contributed to the final morphology. One film prepared using a single spinning cycle (160 mg/ml at 6000 rpm) was also examined to assess the effect of spinning cycles. The film exhibited significantly lower peak intensities compared to its three-cycle counterpart. This reduction is likely due to the limited material deposited in a single cycle, resulting in fewer crystallites and diminished crystalline order. In contrast, three-layer films provided greater material accumulation, facilitating more robust crystallite formation and higher overall crystallinity.

For reference, raw PEO powder exhibited very sharp and high intensity peaks at approximately 19.1° and 23.2° 2θ, corresponding to a crystallite size of 15.6 nm and 79% crystallinity. Toolan et al.⁵² reported that when spin-coated from non-polar solvents like chloroform, PEO crystallized into large, highly ordered spherulites consistent with Avrami kinetics. However, with increasing solvent polarity, the crystallinity of PEO diminished, and the resulting films contained mostly amorphous regions with small crystalline domains. In the present blend films, PEO peaks appeared weakened and broadened, suggesting that the presence of PCL disrupted PEO crystallization. This may be attributed to molecular-level interactions between the two polymers, kinetic limitations arising from phase separation, or spatial constraints imposed by the rapid solvent evaporation during spin coating.

Porous PCL films

Leaching PEO from the PCL/PEO blend films successfully yielded free-standing, porous PCL films with a diameter of 50 mm. Porosity, estimated from weight loss measurement, indicated an average porosity of approximately 47%, confirming the effective extraction of the PEO component (Table 4). Visually, the resulting films appeared opaque and white, in contrast to the translucent appearance of their nonporous PCL and PCL/PEO counterparts. Average film thickness ranged from approximately 8 μm to 90 μm (Fig. 12), with good uniformity (average coefficient of variation of 14%). As with the nonporous films, thickness decreased with increasing spin speed and increased with both higher polymer concentration and a greater number of spin-coating cycles. Notably, the porous films showed a substantial reduction in thickness compared to the original blend films. This reduction is primarily attributed to the removal of the PEO-rich protruding domains and the subsequent formation of a porous network. In many cases, film thickness decreased by half or more, depending on the initial blend composition and processing conditions. Additional thinning likely resulted from partial collapse or compaction of the porous network during drying, due to capillary forces and the loss of structural support from the leached material.

Table 4.

Porous PCL film thickness (average ± standard deviation), porosity, and water uptake.

No. of Spinning Cycles	Concentration (mg/ml)	Spin Speed (rpm)	Thickness (µm)	Porosity (%)	Water Uptake (%)
1	160	3000	9.5 ± 1.4	46.0%	89%
1	160	6000	7.7 ± 1.7	46.8%	86%
1	320	3000	38.9 ± 3.5	45.3%	170%
1	320	6000	31.3 ± 4.3	45.4%	166%
3	160	3000	74.8 ± 4.8	48.3%	98%
3	160	6000	57.3 ± 1.8	46.8%	101%
3	320	3000	89.4 ± 16.8	49.0%	186%
3	320	6000	72.2 ± 19.2	48.8%	196%

Fig. 12.

Porous PCL film thickness as a function of process parameters.

To further support the porosity and the effectiveness of PEO extraction, water absorption capacity was assessed. Water uptake ranged from 86% for single-layer films at 160 mg/ml concentration to 196% for three-layer films at 320 mg/ml concentration. The higher absorption in thicker films likely stems from increased internal volume and more extensive pore formation. Thicker films support the development of a more interconnected pore network, which enhances water retention through capillary action and facilitates diffusion throughout the structure. Variations in pore size, distribution, and connectivity governed by polymer concentration and spin-coating parameters also contribute significantly to water uptake. While high water absorption may be advantageous for applications that benefit from high hydration capacity (e.g., wound dressings or tissue engineering scaffolds), it may also reduce the mechanical integrity of the films during prolonged moisture exposure.

SEM imaging revealed several key structural changes following PEO extraction, confirming the role of phase separation in defining the final porous morphology. In films prepared from 160 mg/ml solutions at low spin speed, the air-facing side showed an irregular, interconnected network structure with microscopic pores (< 2 μm), suggesting that the PEO-rich protrusions at the air interface were embedded within the PCL matrix rather than existing solely as surface features (Fig. 13(a)). Their dissolution left behind pores, confirming their phase-separated origin. In contrast, the substrate-facing side of the film presented a flatter, more regular network, with smaller, elongated micropores (< 1 μm width), indicating a less continuous PEO phase on the surface (Fig. 13(b)). The higher continuity of PCL on the surface can be attributed to its superior film-forming ability, its higher compatibility with the substrate, greater solubility in chloroform, and lower viscosity compared to PEO. These factors likely facilitated the formation of a more continuous PCL matrix, while PEO phase-separated into discrete spherical or slightly elongated domains. Films spun at higher speed exhibited comparable textures (Fig. 14(a) and (b)).

Fig. 13.

SEM images of porous PCL films spin-coated from 160 mg/ml solutions at 3000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

Fig. 14.

SEM images of porous PCL films spin-coated from 160 mg/ml solutions at 6000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

Three-layer films showed similar features on the substrate-facing side (Fig. 13(B) and Fig. 14(B)). However, the air-facing side displayed distinctive differences (Fig. 13(A) and Fig. 14(A)). First, there was a noticeable increase in structural scale, likely due to a slower solvent evaporation rate caused by multiple spinning cycles. Second, the surface roughness increased, with numerous randomly distributed microscopic particles observed. These particles were likely PCL domains that had phase-separated within the PEO-rich regions and were deposited onto the surface upon PEO leaching. Higher spin speed resulted in a less porous surface, as rapid solvent evaporation limited the time available for domain coalescence.

At higher polymer concentrations, significant changes were observed (Figs. 15 and 16). On the substrate-facing side, the network structure was reduced and even densified. Both surfaces showed larger, spherical and elongated pores, likely resulting from shear-induced coalescence of the PEO domains during spin coating. The air-facing surface also exhibited an increased number of randomly distributed microscopic particles, again originating from PCL that had been trapped within the leached PEO regions. With increasing spin speed, the substrate-facing surface became denser, suggesting that rapid solvent evaporation reduced the time available for microdomain growth and coalescence. The three-layer films exhibited similar features but exhibited a greater degree of densification, reinforcing the idea that multiple coatings compress and influence the final phase-separated morphology.

Fig. 15.

SEM images of porous PCL films spin-coated from 320 mg/ml solutions at 3000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

Fig. 16.

SEM images of PCL/PEO (50/50) blend films spin-coated from 320 mg/ml solutions at 6000 rpm. (a, A) show one side of the film, while (b, B) show the opposite side. Lowercase letters (e.g., a, a') correspond to films prepared with one spinning cycle, uppercase letters (e.g., A, A') correspond to films prepared with three spinning cycles, and prime symbols (′) indicate magnified views of the respective images.

XRD analysis of the porous PCL films, following PEO leaching, revealed only the characteristic peaks associated with crystalline PCL, with no discernible peaks corresponding to PEO (Table 5; Fig. 17). This confirms the effective and complete removal of PEO, leaving behind a pure PCL structure. The diffraction peaks observed at approximately 21.4–21.7° 2θ and 23.7–24.0° 2θ are consistent with the typical crystalline diffraction patterns of PCL, indicating that PCL retained its crystalline integrity. The relative intensity and sharpness of these peaks followed the trend: 320 mg/ml at 3000 rpm > 160 mg/ml at 6000 rpm > 160 mg/ml at 3000 rpm > 320 mg/ml at 6000 rpm. This ordering suggests a complex interplay between polymer concentration and spin speed in determining final crystallinity. Specifically, higher concentration combined with high spin speed appears to inhibit crystallization likely due to rapid solvent evaporation and increased viscosity, which restrict polymer chain mobility and hinder orderly crystal growth. Conversely, at lower concentration, increased spin speed resulted in sharper and more intense peaks, indicating enhanced crystallinity potentially due to thinner films and faster solvent removal that promotes nucleation over disorder.

Table 5.

XRD data for porous PCL films fabricated using three spinning cycles.

Concentration (mg/ml)	Spin Speed (rpm)	Thickness (µm)	Peak Position (° 2θ)	Crystallite Size (nm)	Crystallinity (%)
160	3000	74.8 ± 4.8	21.7, 24.0	19.3	49
160	6000	57.3 ± 1.8	21.5, 23.8	24.0	47
320	3000	89.5 ± 16.8	21.4, 23.7	21.9	50
320	6000	72.2 ± 19.2	21.6, 23.9	20.2	37

Fig. 17.

XRD profiles of PCL films fabricated using three spinning cycles at different polymer concentrations and spin speeds. For each condition, the upper profile corresponds to nonporous films, and the lower profile corresponds to porous films.

Interestingly, when comparing these porous PCL films with their nonporous pure PCL counterparts, the porous films exhibited higher peak intensities and sharper peaks. This may be attributed to additional crystallization of PCL during the PEO leaching process and subsequent oven drying, as these conditions may promote crystal growth and enhance the overall crystalline order of the polymer.

Conclusions

Environmentally friendly, free-standing PCL-based films were successfully prepared by spin coating. Film average thickness ranged from approximately 15 μm for single-layer pure PCL films to 228 μm for three-layer nonporous PCL/PEO (50/50) blend films. Following PEO leaching, film thickness was nearly halved, with porous PCL films ranging from 8 μm to 90 μm, depending on the number of spin cycles. This reduction is attributed to the removal of PEO-rich protruding domains and partial collapse of the porous structure during drying. Film thickness generally decreased with increasing spin speed due to enhanced centrifugal spreading and faster solvent evaporation. In contrast, increasing polymer concentration and the number of spin cycles led to thicker films, as more material was deposited during processing. While overall film thickness was uniform, slight variations were observed in three-layer films, likely due to interactions between sequentially applied wet layers.

The inclusion of PEO disrupted PCL crystallinity and introduced distinct surface features, which were later transformed into pores upon leaching. SEM analysis confirmed well-defined phase-separated morphologies, with notable differences between the air-facing and substrate-facing surfaces. Processing conditions, especially spin speed and concentration, played a critical role in dictating phase separation, surface porosity, and crystallinity. Higher concentrations and speeds led to reduced surface features and crystallinity, while 160 mg/ml concentration and 6000 rpm spin speed yielded the most well-defined morphologies and enhanced crystalline features. XRD analysis confirmed that PEO leaching was complete and revealed that post-leaching crystallization during drying improved PCL crystallinity in porous films compared to their nonporous counterparts.

Importantly, the resulting films remained structurally intact and handled well throughout processing, indicating promising mechanical integrity despite their thinness and porosity. Overall, the ability to tailor film properties such as thickness, porosity, and crystallinity through simple adjustments in spin-coating parameters and selective leaching demonstrates the versatility of this approach. These tunable, free-standing porous films show strong potential for applications in biomedical membranes, drug delivery systems, and filtration, where precise control over structure and permeability is critical.

Author contributions

The author confirms sole responsibility for the study conception, design, methodology, analysis, and manuscript preparation.

Data availability

All data supporting the findings of this study are included in the article.

Declarations

Competing interests

The author declares no competing interests.