Experimental study of the process of inner treatment of porous polylactide in a pulsed DBD

Plasma treatment of polymeric materials is a well-known tool for modifying their surface properties, for example, for increasing hydrophilicity. Plasma hydrophilization is especially important for biodegradative materials and materials that are used for medical applications, such as polylactide. Plasma treatment of polylactide was intensively studied for the modification of the surface of polylactide.^1,2 However, uniform plasma treatment of porous polymer materials inside inner layers is still a significant challenge for some technology applications.

Plasma treatment of cavities in polymeric materials involves the ignition of plasma inside these cavities or, in other words, the electrical breakdown of a gas within a limited gap or a group of serial gaps. The conditions for gas breakdown between two electrodes are described by Paschen’s law in 1889;³ however, the scatter of experimental data in the area close to curve minimum is very large.^4,5 The problem is that Paschen’s law contains a parameter that is determined by the secondary electron emission from the surface of the gap boundary and can depend on the material of the gap boundary. In this case, the coefficients of secondary electron emission can themselves depend on the applied field.⁶ Changes in the parameters of Paschen’s curve in micrometer gaps between dielectrics were identified and studied in Refs. 7–9. However, the results obtained in these studies do not provide the necessary information on the breakdown conditions in the area of the minimum of Paschen’s curve, which are characteristic of breakdown in a microporous dielectric. Such studies were presented in Refs. 10 and 11.

This work is devoted to the study of the conditions for the ignition and development of breakdown in micro-cavities inside polymeric materials using the example of polylactide.

To prepare experimental samples, we used biopolymer kind 4032D, which is poly-L-lactide manufactured by NatureWorks LLC, USA. The solution of polylactide 3% (w/w) in dioxane was initially frozen in a refrigerator chamber for 24 h. Then, porous samples were prepared by freeze-drying on a Martin Christ Alpha 2-4LSC. Drying was carried out for 24 h with a vacuum depth of 0.250 mbar in the “main-drying” mode. Before ejecting the finished products, we have used the “final drying” mode with a vacuum depth of 0.001 mbar for 2 h. For spectroscopic studies, film samples with a thickness of 300 ± 20 µm were used. For film preparation, PLA polymer granules were heated in a frame up to 200 °C and cooled in press under a pressure of 10 MPa for 8 h.

The experiments were carried out in a discharge chamber with a gas input and output, which consists of a transparent upper half and a grounded metal bottom with a silicone rubber sealing ring (Fig. 1).

The electrical schematic of the experimental setup is shown in Fig. 2.

To study the breakdown process in a porous material located in the gap, electric discharge voltage and current waveforms were recorded. Typical waveforms of the voltage and current between the electrodes without a test sample and with a porous polylactide sample are shown in Fig. 3. All experiments were performed at a pulse repetition frequency of 1 kHz.

A comparative analysis of the discharge characteristics with and without a sample allows us to determine the ignition conditions and discharge parameters in a porous material in comparison with a discharge in an air gap of the same thickness.

During these experiments, a disk-shaped sample with a side of 2 cm and a thickness of 2.5 mm was used. The diameter of the glass electrode with the liquid electrode was also 2 cm.

To calculate the discharge parameters, the discharge system is convenient to be presented as series-connected flat capacitors, one of which is a capacitor with a glass barrier and the other is a capacitor with an air gap or a gap filled with a porous dielectric.

The capacitive current was calculated as Ic = c * U′, where Ic is the capacitive current, c is the electrical capacitance of the electrode system, and U′ is the time derivative of voltage. The discharge current was calculated as Idc = I − Ic = c * U′, where Idc is the capacitive discharge current and I is the measured current.

The calculation of the discharge current, which is the measured current minus the capacitive current, allows us to determine more accurately the breakdown moment and voltage. In our case, the breakdown voltage was taken as the voltage at which the current reached 20% of the maximum.

In all waveforms, the pulses of current correspond to the breakdown of the discharge gap with or without a porous dielectric. There are two impulses—the first corresponds to the moment when the voltage rises and the second when it falls. The reason for the occurrence of two breakdowns is the electric charge on the surface of the glass dielectric,¹² which decreases the voltage across the discharge gap, as one can see from Fig. 4.

The passed electric charge was calculated as the integral Idc * dt, and the voltage across the discharge gap Udg was calculated as the difference between the total voltage U across the electrodes and the voltage drop across the glass dielectric Ug. Udg = U − Ug = U − q/Cg, where Cg is the capacitance of a capacitor with a glass dielectric.

The average electric field strength in the discharge gap Edg was calculated as the ratio of the voltage across the discharge gap Udg to the gap thickness.

Figure 5 shows the dependence of the electric power of the discharge and the energy applied during one pulse into the discharge gap.

The discharge power was calculated as the product of the discharge current and voltage, and the energy was calculated as the time integral of the power. The energy applied to the discharge in a porous dielectric and the passed electric charge are very important parameters for understanding the processes of plasma treatment of a porous dielectric. For better understanding of the relationship between these parameters and the fraction of cavities of the dielectric in which the breakdown occurred, a simple model of breakdown in a porous dielectric, presented below, was proposed.

As we can see from Fig. 3, the difference between the results of experiments with an empty gap and a gap filled with a porous dielectric is a significant difference, despite the fact that the porosity coefficient of the material is more than 95%. Among the differences, one can single out a significantly higher breakdown voltage in a porous material and twice the value of the effective electric capacity of the discharge gap with porous polylactide than without it. Another significant difference is that the passed charge does not return to zero after the stage of secondary breakdown of the discharge gap in the opposite direction when the voltage drops. This is due to the electret properties of polylactide or in another words ability to keep residual polarization after a drop in the electric field. The electret properties of polylactide are significant even at such high porosity coefficients when the effect of the dielectric constant of the porous material on the capacity of the discharge gap is negligible at low applied voltage conditions.

Preliminary experiments on processing a smooth sample have already demonstrated a significant effect of discharge treatment on the water wetting contact angle. However, this approach is impossible when studying the process of processing a porous material. To characterize the degree of cavity treatment in porous polylactide, an experiment was carried out to change the water absorption of the material during treatment.

The processing was carried out in a discharge chamber under the conditions described above at different processing times and gas pressures. The analysis of the waveforms described above allowed us to calculate the important parameters of the treatment process—the discharge ignition voltage, the average electric field, the passed charge, the electric power, and the total energy input during the processing.

The sample was weighed before treatment. After treatment, the sample was placed in water for 5 s and then weighed again. The mass of water absorbed by the treated sample in relation to the mass of water absorbed by the untreated sample allows us to calculate the change in water absorption obtained during the treatment. The results are convenient to present as the degree of filling the cavities with water (100%—the sample is completely filled with water and 0%—dry sample). The results are shown in Fig. 6.

For better understanding of the mechanism of hydrophilization of porous polylactide due to its treatment in a barrier discharge solid, a 0.3 mm polylactide film was treated with the same parameters that are used for porous sample treatment for 30 s at 170 Torr. The IR spectra of PLA films before and after treatment were recorded, and the water contact wetting angle was measured. The treatment of the film caused a change in the water contact wetting angle (72.9° to 54.1°). The studies were carried out using a Thermo Scientific Nicolet iS5 FT-IR spectrometer with iD5 ATR and iD1 transmission accessories. The spectral resolution was 4 cm⁻¹, and the number of scans was 32. The recording of the ATR and transmission spectra was performed in a range of 4000–525 and 4000–400 cm⁻¹, respectively. The obtained spectra are shown in Fig. 7.

The IR spectra of the samples recorded in the ATR mode (which characterize only the near-surface layer with a depth up to 2–3 μm) after plasma treatment show the appearance of OH-groups (a broad peak with a maximum in the region of about 3300 cm⁻¹), a change in the absorption of C–H bonds (3000–2800 cm⁻¹), and the appearance of a wide shoulder in the region of 1700–1640 cm⁻¹ (C=O groups). At the same time, the spectra recorded in the transmission mode (from the entire thickness of the sample) do not show significant differences between the initial and treated samples (the main absorption bands go off-scale due to the large film thickness).

In a real porous dielectric, the cavity sizes are not the same. For example, an image of porous polystyrene (Fig. 8) shows cavities of different sizes, fluctuating around a certain average value. Both the average value and the width of the cavity size distribution are essential for analyzing the breakdown process and can be determined using a micrograph of the material.

For further analysis, we need the cavity size distribution in polylactide. In Fig. 8, a photomicrograph of a section of a polylactide sample with a wire of 0.28 mm in diameter for scale is shown.

To estimate the breakdown conditions in a porous material, Paschen’s law can be used as a convenient approximation,

Paschen’s curve extrapolation parameters for air are given as follows:¹³ 

Parameter γ was taken equal to 0.01 in this work.

However, as shown in Refs. 10 and 11, the best approximation to the experimental results is given by the variable value γ that decreases with increasing Pd reversely proportional to the square of Pd,

For the ideal case, when all the cavities are connected and have the same size, the breakdown occurs simultaneously when the breakdown voltage corresponding to the cavity size Upor is reached. This is the breakdown of the entire discharge gap. In this case, in Paschen’s law, we use the cavity size as the gap thickness Dpor and calculate the total voltage U across the discharge gap Dgap as U = Upor * Dgap/Dpor.

Such dependences of the breakdown voltage in a discharge gap filled with a porous dielectric for different cavity sizes are shown in Fig. 9.

This figure shows that for a voltage of 10 kV applied under experimental conditions, the breakdown conditions are achieved in all cavities with dimensions larger than even ∼0.15 mm at a pressure of 80 Torr and in cavities larger than ∼0.3 mm at a pressure of 170 Torr. Cavity size distribution provides an estimate of the volume fraction of treated cavities and correlates with the results shown in Fig. 6, although they are slightly overestimated, which will be explained in the last paragraph of Sec. VI.

Counting the number of cavities of different sizes on a cut of the material gives after averaging the approximate distribution of the number of cavities by size in the section, which can be easily converted into the distribution of the volume of cavities of a given size in the thickness of the material. This distribution is shown in Fig. 10.

So, a simple and visual approach is convenient to use for understanding the physical mechanisms of breakdown in a porous material; however, it has clear deficiencies in a real situation in the presence of cavities of different sizes.

For a more detailed description of the breakdown process and quantitative estimates of its parameters, a simple semi-empirical model was developed, which describes the breakdown of a porous dielectric. For this, a model porous material with a Gaussian cavity size distribution was chosen, the parameters of which were selected for the closest match with those measured using a micrograph. Cavities of different sizes form vertical columns. Breakdown development in each column was calculated using the following algorithm.

First, the voltage at each cavity in the column was calculated in accordance with its size and, accordingly, the electrical capacity. Next, cavities were selected to comply with breakdown conditions as per Paschen’s method described above (with variable γ). Their voltage was taken as zero. After this, the stresses on the remaining cavities of the column were recalculated, newly broken cavities were selected, and the cycle was repeated until the moment when newly broken cavities no longer appeared. The smallest cavities in which the breakdown did not occur were taken into account in further calculations. This procedure was carried out with all the columns of the model porous material.

The simulation results are shown in Figs. 11–13.

An important conclusion can be drawn from the calculation results. First, the threshold voltage corresponding to a significant (more than 20%) increase in the percentage of broken cavities differs from the threshold voltage for the growth of the passed electric charge and the input energy by ∼0.5 kV. This is due to the fact that the capacitance of a column of cavities (air condensers) becomes comparable to the capacitance of a glass dielectric barrier of the same area only when most of the cavities have already been broken.

As a result of this, we record the breakdown of the dielectric (the beginning of a rapid increase in the current) only at the moment when the breakdown has already occurred in most of the volume of the porous material. The main energy is also applied with a delay to the breakdown of most of the cavity volume, which explains some discrepancy between the results shown in Figs. 6 and 10.

The experiments resulted in determining the conditions for the ignition of a discharge and the characteristics of the discharge in samples of porous polylactide at different air pressures.

The findings show that the breaking voltage in a porous dielectric differs significantly (by several times) from the breaking voltage in air, despite a high coefficient of porousness of the dielectric, which is about 95%.

The experimental results also show that the effectiveness of the treatment (hydrophilization) of the cavities of polylactide samples depends strongly on the pressure applied.

The infrared spectroscopic study of the treated and untreated samples demonstrated the impact of the treatment on forming functional groups on the polymer surface and on decreasing the contact angle.

According to the experimental results, the electret properties of polylactide influence the parameters of the electric discharge in its pores, for instance, the effective absolute permittivity of the discharge gap.

This study offers a convenient semi-empirical model of the development of breakdown in a porous dielectric. Despite the simplicity of the model, good agreement was obtained between the calculated and experimentally measured values of the passed charge and the deposited energy.

The IR study of the samples was carried out using the equipment of the “Optics” resource center of the NRC “Kurchatov Institute.” Polymer samples were obtained from the laboratory of polymer materials of the National Research Center “Kurchatov Institute” under the thematic plan investigation.

This work was supported by the NRC “Kurchatov Institute” (Order No. 2073 from 9 October 2020).

The data that support the findings of this study are available from the corresponding author upon reasonable request.