APDBD Plasma Polymerized PNIPAm Coatings with Controlled Thickness via Spin Coating Technique

Abstract

Thermosensitive Poly(N-isopropylacrylamide) (PNIPAm) films were synthesized via atmospheric pressure dielectric barrier discharge (APDBD) plasma polymerization. In order to control the thickness of the films, a spin coating technique was used during the deposition of N-isopropylacrylamide (NIPAM) monomer solution onto several glass substrates. We used the coefficient of determination (R-square value) in linear regression to investigate the significance and optimize spin coating parameters during the fabrication of NIPAM coatings before exposure to APDBD plasma to ensure reproducible and uniform film properties. The spin coating parameters investigated in this study include spin speed, spin time, and NIPAM solution concentration with R-square values of 0.978, 0.946, and 0.944, respectively. Also, as a result of the thermosensitive nature of NIPAM, the spin coating operating conditions of temperature and humidity were maintained at 39.0 °C and 15%, respectively. During the APDBD plasma polymerization, argon was used as the discharge gas, and the distance between the two parallel electrodes and plasma frequency were maintained at 5.0 mm and 17 kHz, respectively. The plasma exposure time required for polymerization of PNIPAm coatings was optimized to 60 s. Also, the results showed that a coating with minimal defects had an optimal thickness of 5.18 μm, fabricated under conditions of 90 wt.% NIPAM concentration, spin speed of 4000 rpm, and total spin time of 7 s.

1. Introduction

Poly(N-isopropylacrylamide) (PNIPAm) thin film coatings have attracted significant attention in biomedical and smart applications due to their reversible thermoresponsive behavior at a temperature near physiological conditions (32.0–34.0 °C) [1]. This phase-changing temperature is referred to as the lower critical solution temperature (LCST) for the case of PNIPAm. At temperatures below this LCST, amide groups (-CONH-) form strong hydrogen bonds with water molecules [2]. In contrast, at temperatures above the LCST, isopropyl groups (-CH(CH3)2) aggregate and the amide groups can no longer form hydrogen bonds with water molecules [3]. This makes PNIPAm exhibit reversible transitions that include hydrophilic to hydrophobic, soluble to insoluble, expansion to contraction behavior, and coiled to globular polymer chains conformation shifts, respectively [4]. In addition to temperature triggered transitions, PNIPAm is a biodegradable, easily tunable structure, non-toxic, and cost effective, rendering it ideal for biomedical, smart textiles and thermally adoptive surface applications [5].

Various fabrication techniques have been used to prepare PNIPAm film coatings [6,7,8]. These are generally divided into solution-based and vapor-based methods. Solution-based techniques involve the coating of precursor solutions onto surfaces of different substrates. These include methods such as dip coating, spin coating, spraying, and so on. Although these methods for PNIPAm coating synthesis are simple and inexpensive, toxic initiators, catalysts, or cross-linking agents are often employed during polymerization of precursor solutions [9,10,11]. On the other hand, vapor-based methods deposit coatings onto substrate surfaces through condensation (physical vapor deposition) or chemical reaction (chemical vapor deposition) with the precursor molecules in vapor state [12,13]. Vapor deposition methods are precise and produce ultra-thin coating on different substrate surfaces but also necessitates the use of very complicated and expensive equipment [14,15].

Atmospheric pressure dielectric barrier discharge (APDBD) plasma polymerization involves the formation of highly cross-linked polymer chains [16]. APDBD plasma set up consists of two parallel identical cylindrical or rectangular electrodes with at least one of them covered by a dielectric material such as ceramic, quartz, and so on [17]. When an electric field is induced between two parallel electrodes, free electrons in the flowing gas gain enough kinetic energy and accelerate at high velocities, causing random collision with other neutral gas molecules [18]. This initiates several processes including ionization, excitation, and dissociation, leading to the formation of plasma-active species such as ions, free radicals, excited atoms, and so on [19,20]. These active species then react with monomer molecules on the surface of the substrate, which is positioned between the two parallel electrodes in a process known as APDBD plasma polymerization.

Recently, APDBD plasma polymerization has been utilized by several scientists as a result of its advantages such as uniform surface changes, low cost, and environmental friendliness [21]. For example, Amorosi, C. et al. prepared polymer films from methacrylic acid (MAA) and ethylene glycol dimethacrylate (EGDMA) monomers under APDBD plasma conditions [22], whereas Molina R. et al. initiated in situ polymerization of NIPAM aqueous solution by APDBD plasma conditions [23]. Also, Tang X.-L. et al. prepared thermo-sensitive PNIPAm thin films under kHz DBD plasma conditions [24] and many other studies. Furthermore, current approaches report APDBD plasma treatment under optimal conditions to primarily induce chemical modifications without significantly altering physical properties such as roughness and thickness [25,26,27,28]. However, for applications such as biomedical engineering or cell sheet engineering, the thickness of PNIPAm coatings plays a critical role during temperature transitions below and above the LCST, and needs to be carefully controlled [29].

The primary objective of this study is to prepare PNIPAm coatings with controlled thickness using APDBD plasma. First, the nature and composition of argon APDBD plasma were characterized using a digital oscilloscope and emission spectral analysis. Subsequently, spin coating technique was employed to deposit NIPAM monomer solutions onto glass substrates prior to plasma exposure. The effects of the spin coating parameters—time, rotational speed, and monomer concentration on film thickness—were systematically investigated and analyzed with linear regression. Additionally, the surface morphology, chemical composition, and thermal behavior of PNIPAm coating synthesized under APDBD plasma was analyzed with scanning electron microscopy (SEM), Fourier transform infrared (FTIR) spectroscopy, and thermogravimetric analysis (TGA), respectively.

2. Materials and Methods

2.1. Materials

Glass wafers of dimensions 18 mm × 18 mm were used as substrates in this study; 98% NIPAM (Shanghai Shifeng Biotechnology Co., Ltd., Shanghai, China); Anhydrous ethanol; Deionized water; 99.999% pure argon gas (Shanghai Haoqi Gas Co., Ltd., Shanghai, China), all listed materials were used.

2.2. PNIPAm Films Preparation

Glass substrates were cleaned with anhydrous ethanol, immersed in deionized water, and then dried for 24 h in ambient air. NIPAM was dissolved in deionized water at concentrations ranging from 10 wt.% to 95 wt.%, weighed using the ME204E electronic balance (Mettler Toledo Instruments Shanghai Ltd., Shanghai, China), and heated to ensure complete dissolution in an HHS-1S electronic thermostatic stainless-steel water bath (Shanghai Yulong Instrument Equipment Co., Ltd., Shanghai, China). The temperature and humidity were then regulated to 39.0 °C and 15%, respectively, using dynamic temperature controller PRESTO A35 by ASYNT LTD, MS-990B Dehumudifier by MS SHIMEI Ltd., Suzhou, China, and Hygro Thermometer Model 445,703 by MRC LABORATORY INSTRUMENTS, Harlow, UK, to prevent structural changes in NIPAM that might alter film thickness during the spin-coating process under ambient conditions [30].

The spin coater (KW-4A Spin coater (Zhejiang Light Precision Instrument Co., Ltd., Yongkang, China)) used in this study was characterized by a two-stage spinning process: a low-speed first stage followed by a high-speed second stage. After identifying the minimum spin time and speed required to uniformly spread the NIPAM solution across the glass substrates (Figure 2a), the first stage parameters (4 s at 500 rpm) were kept constant throughout the optimization process. The second stage spin time and speed were then systematically varied before APDBD plasma was generated with CTP-2000K plasma generator (Nanjing Suman Electronics Co., Ltd., Nanjing, China). Second stage spin time ranged from 0 to 9 s (Figure 2b), and spin speed from 500 to 8000 rpm (Figure 2c). The total spin time shown in Figure 2b represents the sum of the first stage (fixed at 4 s) and second stage (variable) spin times. Finally, the thickness of PNIPAm coatings was measured using the MSD105 optical microscope (Murzider (Dongguan) Science and Technology Co., Ltd., Dongguan, China) on manually cut sample cross-sections. After calibrating the microscope software with a stage micrometer, the coating thickness at four equally spaced positions per sample was determined by measuring the vertical distance between the film-substrate and film-air interfaces using the software’s line measurement tool. This procedure was repeated for three independently fabricated samples. The accuracy of this optical method was confirmed by comparing measurements to SEM on randomly selected samples with known thicknesses.

2.3. Characterization Methods

Attenuated total reflection Fourier transform infrared spectroscopy (FTIR) was performed using NICOLET 6700 spectrometer (Thermo Fischer Scientific Instruments, Waltham, MA, USA) in the spectral range of 1750–900 cm⁻¹ to investigate chemical changes induced by APDBD plasma polymerization on substrates treated for varying durations. Each measurement included 16 scans at a resolution of 4 cm⁻¹. X-ray photoelectron spectroscopy (XPS) analysis was performed using an ESCALAB 250 Xi instrument (Thermo Fisher Scientific, Waltham, MA, USA), equipped with an Al Kα X-ray source and a hemispherical electron energy analyzer. The system operated at 12.5 kV accelerating voltage and 250 W power under a vacuum chamber pressure of 10⁻⁶ Torr. Also, to analyze the effect of APDBD plasma treatment time and coating thickness, the surface morphology of PNIPAM coatings was examined using an environmental scanning electron microscope (SEM, Quanta 250, FEI Company, Eindhoven, The Netherlands).

Thermogravimetric analysis (TGA; TGA 5500, TA Equipment, New Castle, DE, USA) was used to evaluate the thermal stability and decomposition behavior of PNIPAm coatings. The heating rate was set to 20 °C/min, using a sample weight of 4.9 mg and nitrogen as the purge gas in temperature of 50–600 °C. Differential scanning calorimetry (DSC) and UV-Vis-NIR spectrophotometer (UV-3600, Shimadzu, Kyoto, Japan) techniques were used to investigate the LCST for the APDBD plasma polymerized PNIPAm coatings. DSC measurements were conducted under nitrogen atmosphere with the sample heated at 3 °C/min, in the range between 20 °C and 50 °C. The phase transition temperature of a 1.0 wt.% PNIPAm aqueous solution was determined via turbidimetry using a UV-Vis-NIR spectrophotometer (UV-3600, Shimadzu, Japan). Measurements were performed by monitoring optical absorbance at wavelength of 500 nm in the temperature range of 20–45 °C. Several PNIPAm coatings, with thickness of 5.18 μm synthesized under optimal APDBD plasma conditions were scrapped off from glass substrates for DSC and UV analyses.

3. Results and Discussion

3.1. Plasma Diagnostics

The structure, morphology, and composition of APDBD argon plasma are as shown in Figure 1. The discharge voltage characteristics were captured using a P6015A high-voltage probe (Tektronix, Beaverton, OR, USA) connected to a UTD2102CEX-II digital oscilloscope (Unilead Technology Co., Ltd., Punjab, India), revealing a sinusoidal waveform with multiple discharge peaks within each half-cycle and excellent repetition (Figure 1b). These observations strongly suggest that plasma generated by APDBD exhibits uniformity, diffusion, and stability. This ensures a continuous generation of active species and consistency in their interactions with the substrate surface. Chemical composition analysis of APDBD plasma was performed using an HR4000 spectrometer (Ocean Optics, Orlando, FL, USA) prior to polymerization. Emission peaks corresponding to OH radicals were identified within the wavelength range of 296 nm to 309 nm. Additionally, transition peaks associated with nitrogen were observed at wavelengths of 315.44 nm, 336.38 nm, 357.27 nm, 380.07 nm, 399.25 nm, and 405.51 nm (Figure 1c), consistent with findings from previous studies [31,32]. Furthermore, characteristic lines of argon (Ar) atoms were observed within the wavelength range of 680 to 870 nm [33], as depicted in Figure 1d.

3.2. Influence of Spin Coating Parameters on PNIPAm Film Thickness and Surface Morphology

Secondly, we studied the effect of NIPAM concentration in the solution, spin speed, and spin time on the thickness of PNIPAm coatings. High NIPAM concentration in the solution produced thicker films in contrast with spin speed and time that demonstrated an inverse relationship with coating thickness. Centrifugal, cohesive, and shear forces are responsible for these variations [34].

When a precursor solution is dispensed onto the substrate material, surface tension arises as a result of cohesive forces between neighboring molecules, holding the solution together, and resisting its spread across the substrate surface [35]. And as spinning is initiated, centrifugal force act on the precursor solution, pushing it outward and away from the center of rotation [36]. This outward movement is particularly strong at the edges of the substrate due to higher tangential velocity [37]. This combination of surface tension and centrifugal forces creates shear forces within the solution, especially near the substrate surface. These shear forces help to overcome the cohesive forces within the precursor solution, causing excess solution to be ejected off the substrate surface [38].

Therefore, increasing NIPAM concentration results in formation of a solution with higher viscosity, which causes greater resistance to the flow due to increased molecular interactions and entanglements. This limits the effectiveness of shear forces in causing deformation or flow within the solution (Figure 2b). However, higher spin speeds generate stronger centrifugal forces, aiding in the spreading of the solution across the substrate surface and eases the ejection of excess solution (Figure 2d). Additionally, extended spin times allow for more thorough spreading and flow of the solution, compensating for slower flow rates associated with this higher viscosity (Figure 2c).

As shown in Figure 2b, lower NIPAM concentrations yield thinner coatings with more defects, which diminish significantly at 90 wt.%. Conversely, insufficient spin time produces uneven coatings with defects (Figure 2c). However, these defects are minimum at 7 s but reappear with longer spin times, accompanied by a decrease in coating thickness. Similarly, defects peak at low spin speeds, decrease between 3000 and 5000 rpm, and intensify at higher spin speeds (Figure 2d).

Using the plotted data in Figure 2b–d and applying linear regression analysis, the relationship between spin coating parameters and coating thickness is illustrated in

Table 1. Significance of spin coating parameters.

Spin Coating Parameter	R-Square Value
NIPAM concentration	0.978
Spin speed	0.946
Spin time	0.944

.

These results show that NIPAM concentration in the solution has a stronger relationship with film thickness as compared to other parameters. This is followed by spin speed and spin time, respectively. This is because longer spinning times and speeds allow more solvent evaporation causing non-uniform crystal regions that affect the smooth spreading of the solution onto the glass wafers.

The surface morphology of several PNIPAm coatings was observed as shown in Figure 3. At optimal APDBD plasma exposure time of 60 s a smooth coating was obtained as in Figure 3b, however prolonged exposure to plasma increased surface roughness due to the etching process caused by the increased number of active species bombarding the substrate surface Figure 3c. Additionally, inadequate control of spin coating parameters resulted in poor spreading of NIPAM solution. This caused uneven coating thickness across the substrate surfaces where regions with large coating thickness formed white solid regions upon drying as in Figure 3d,e. However, these regions were reduced through optimization spin coating parameters as in Figure 3f.

3.3. Chemical Composition and APDBD Plasma Treatment Time

Figure 4 represents FTIR spectra of NIPAM monomers and PNIPAm coatings with optimum thickness (5.18 μm) plasma polymerized under APDBD plasma conditions at different intervals. The NIPAM monomer spectrum exhibits a prominent conjugated C=C vinyl stretch at 1620 cm⁻¹, amide I carbonyl stretch at 1660 cm⁻¹, amide II (N-H bend/C-N stretch) at 1550 cm⁻¹ [29], vinyl CH₂ scissoring at 1420 cm⁻¹, C-N stretching, and -CH₃ bending vibrations corresponding to 1366 and 1387 cm⁻¹ peaks, respectively [39]. Also, the vinyl group peaks at 993, 966, and 918 cm⁻¹ of CH₂=CH and CH₂= [40]. However, upon exposure to APDBD plasma, critical spectral changes confirm the formation of PNIPAm. The disappearance of vinyl groups at (1620, 1420, and 993–918 cm⁻¹) confirms the radical addition via π-bond cleavage by plasma generated OH radicals (Figure 1c) [41,42]. In addition, the distinct monomeric peaks at 1620 cm⁻¹ (C=C) and 1660 cm⁻¹ (C=O) coalesce into a broad amide I band centered at 1650 cm⁻¹ with a shoulder near 1630 cm⁻¹, reflecting loss of a conjugated vinyl C=C stretch. Furthermore, no new peaks emerged after APDBD plasma polymerization, indicating no secondary molecular structures formed during the process. Also, extending APDBD plasma treatment beyond 60 s resulted in a pronounced increase in surface roughness and an emergency of hierarchical microstructures including porous domain as a result of etching (Figure 3c).

XPS spectral analysis of PNIPAm coating is shown in Figure 5. Wide-scan spectrum of glass substrate coated with PNIPAm (Figure 5a) indicates peaks of carbon at 285 eV, nitrogen at 399 eV, and oxygen at 532 eV [43]. The presence of Si2s and Si2p is as result of silicon elements in the glass substrate. Also, as observed in Figure 5b, functional groups corresponding to C-C, C-N, C-O-C, and C=O have peaks at 285 eV, 286 eV, 288 eV, and 289, respectively [44].

3.4. Thermal Stability

The thermal stability of PNIPAm was carried out by TGA as shown in Figure 6. Several coatings, each with a thickness of 5.18 μm, prepared under optimal APDBD plasma conditions, were scraped off from the glass substrate surface to attain about 4.9 mg. The TGA profile of PNIPAM reveals three distinct degradation stages. In the range between 30 and 100 °C, a minor weight loss of 6.11%, which corresponds to the desorption of residual moisture or other volatile compounds from the PNIPAm sample, is observed. As the temperature increases to 350 °C, there is a weight loss of 23.46%, attributed to degradation of the polymer side chains. Finally, in the range of 350–410 °C, there is a substantial weight loss of 63.67% as a result of breaking down polymer carbon–carbon backbone into smaller fragments to complete decomposition [45].

3.5. Phase Transition Temperature

Several PNIPAm coatings, with thickness of 5.18 μm, synthesized under optimal APDBD plasma conditions were dissolved in water to prepare a 1.0 wt.% aqueous solution. The visible light absorbance spectrum was taken at a fixed wavelength of 500 nm while heating the solution at a rate of 0.5 °C/min. At temperatures below 32.6 °C, the polymer chains remain hydrated and soluble, resulting in low absorbance due to minimal light scattering. As the temperature increases beyond 32.6 °C, a sharp rise in absorbance is observed, corresponding to the lower critical solution temperature (LCST) as in Figure 7. At this temperature, polymer chains collapse into hydrophobic aggregates, drastically increasing turbidity and light scattering [46]. The narrow temperature range of the absorbance shift (32–33 °C) confirms the homogeneity of the synthesized PNIPAm and its well-defined thermoresponsive behavior.

Figure 8 shows a DCS thermogram in the range of 20–40 °C PNIPAm coating synthesized via APDBD plasma polymerization. A strong endothermic peak in the range 32–33 °C is observed. According to previous reports, DSC endothermic peaks for PNIPAm are due to the dehydration of the C=O and N-H polymer side chains [23,47]. Below 32.6 °C, these polymer chains form bonds with water molecules thereby remaining hydrated, swollen, and hydrophilic. Increasing temperatures beyond 32.6 °C breaks these hydrogen bonds, polymer chains become dehydrated, and dominated by hydrophobic aggregation [48].

4. Conclusions

PNIPAm film coatings with thicknesses ranging from 1.34 to 8.76 μm were successfully deposited on several glass wafers via the spin coating followed by APDBD plasma polymerization. NIPAM solution concentration dominated thickness control compared to other spin coating parameters. While higher spin speeds and times reduced thickness, prolonged spinning above 5000 rpm and 7 s induced irregularities through non-uniform NIPAM solution spreading. Optimal parameters (90 wt.% NIPAM, 4000 rpm, 7 s) yielded 5.18 μm uniform coatings with minimal defects. FTIR analysis confirmed polymerization via disappearance of vinyl signatures at 1620 cm⁻¹, 1420 cm⁻¹, and modes at 993–918 cm⁻¹, consistent with free radical polymerization initiated by plasma generated OH radicals. The coatings exhibited a phase transition temperature of 32.6 °C, near-physiological relevance. This spin coating-assisted APDBD plasma polymerization approach provides a low-cost, environmentally friendly route to tunable thickness PNIPAm films with high reproducibility. Future work will employ machine learning to predict coating properties from process parameters for bioengineering and responsive material applications.