Polydimethylsiloxane (PDMS) surface was treated by nitrogen ions of 20 keV energy with 1016 ions/cm2 fluence. The surface of modified PDMS is characterized by ceramiclike structures. The treated PDMS contains free radicals. The wettability and surface energy change significantly immediately after the treatment and recover in a few days to the values of silicon dioxide or glass. Complex kinetics of growing carbonyl and hydroxyl groups and decaying silane groups is observed. The cage structure of the silicon oxide in the surface ceramiclike layer remains stable, while network structures of the Si–O increase and suboxides of the silicon decrease with storage time. The presence of not-cross-linked low molecular fractions in PDMS gives a fast recovery of the wettability and surface energy to untreated PDMS values in a few hours. This effect of low molecular fractions on the wettability and surface energy was avoided by washing out the cured PDMS in a solvent before the treatment.

Polydimethylsiloxane (PDMS) is an elastomeric material with high chemical stability and good mechanical properties. PDMS is used in a number of modern technologies, including biomedical technologies, for example, as a material for implants in the human body.1–3 However, the inertness of the PDMS surface can be an obstacle for a number of applications.4 In some cases, the PDMS surface is modified by various methods in order to change only a thin surface layer to obtain the desired properties, while the bulk of the PDMS product remains unchanged. In particular, the low surface energy and poor wetting of the PDMS surface by water solutions, the associated adsorption and growth of bacteria on the PDMS surface, changes in the conformation and biological functionality of adsorbed proteins on PDMS, and the mechanically weak surface layer of PDMS can be a problem for some applications.5–8 

One of the useful methods for solving these problems is the plasma treatment of the PDMS surface and, in particular, ion beam treatment also known as ion implantation, ion irradiation, and plasma immersion ion implantation.9–15 In these methods, the surface of the PDMS is bombarded with different particles, energies, and fluences. Bombardment with ions and electrons causes deep structural transformations in the thin PDMS layer and changes the surface properties.

For example, in Vladkova et al.,16 the PDMS surface was treated with low-energy argon plasma. The contact angle of the PDMS surface with water decreased from 101° for the untreated PDMS to 60° for the treated surface. The total surface energy increased from 22.9 to 45.6 mJ/m2, respectively. In Tsuji et al.,17 medical grade PDMS was treated with negative carbon ions. The ion energy was in the range of 5–30 keV. The water contact angle of the PDMS surface decreased from 100° for the untreated surface to 83° for the surface treated with a fluence of 1016 ions/cm2. In Zheng et al.,18 the PDMS surface was treated with carbon and oxygen ions with energies of 10–20 keV. The contact angle has changed from 117.6° for the untreated surface to a minimum value of 99.3° after the treatment.

In addition, it has been found that the water contact angle of PDMS does not remain constant with time after treatment but can vary significantly. For example, in Mata et al.19 and McDonald et al.,20 PDMS was treated with oxygen plasma. The water contact angle decreased from 108.9° for the untreated surface to 60° after 5 min of turning off the plasma discharge. Then, the water contact angle increased to 84° during 30 min of exposure to air after the plasma treatment. Recovery of the high contact angle of PDMS after plasma treatment was also observed in Refs. 21–23. The recovery of the hydrophobicity of the PDMS surface was explained by the diffusion of low molecular weight components of PDMS from the bulk to the surface and the flipping of polar silanol and hydroxyl groups deep into the samples.

In Tong et al.,24 the PDMS surface was treated by oxygen ions with an energy of 10 keV. The contact angle changed from 110° for the untreated surface to 14.8° for the freshly treated surface. Then, the contact angle increased to 68.2° after 48 h and to 79.8° after 120 h of exposure to air. In Bodas and Khan-Malek,25 PDMS was treated in low-energy oxygen plasma. The water contact angle of the PDMS surface immediately after treatment decreased to 8°–18° and then recovered in some minutes. The recovery is due to low molecular fractions generated by plasma diffused from the surface into bulk layers. The disappearance of –OH groups with the storage of some treated PDMS products is observed by Fourier transform infrared attenuated total reflection (FTIR ATR) spectra.

The results of PDMS surface wettability after treatment can also be reversed. For example, in Cutroneo et al.,26 the contact angle of untreated PDMS was 42.6°, whereas after treatment with carbon ions with an energy of 6 MeV, the water contact angle increased to 94.4°.

The chemical structure of PDMS is changed under the plasma and ion beam. In Tsuji et al.,17 PDMS was treated by carbon ions with 10 keV energy. The destruction of the C–H bond and sputtering carbon atom was observed with the formation of –OH groups in the surface layer. In Tong et al.,24 PDMS was treated by PIII O+ ions with 10 keV energy. XPS spectra showed a loss of –CH3 groups and the appearance of Si–OH and Si = O groups. The new –OH groups in the PDMS surface layer treated by low-energy oxygen plasma were observed by FTIR ATR and XPS spectra.25 

Due to these contradictory results, the wettability and structure changes in PDMS after plasma and ion beam treatment remain unclear and should be investigated. The purpose of this work was to investigate the wettability of the PDMS surface before and after the ion beam treatment, as well as to investigate the kinetics of wetting change and to consider the mechanism of wetting change as a result of treatment and storage. However, at the very beginning of the study, we were faced with the problem of low molecular weight fractions in PDMS, which remain in the polymer despite the complete curing reaction. Therefore, the goal was finally extended to analyze the effect of low molecular weight fractions in PDMS on the result of changing the PDMS surface after ion beam treatment. Since all research in the above references, as well as the production of medical devices from PDMS, is carried out using the technology of curing (casting) from a two-component composition, we believe that these results can also be potentially useful for the general circle of researchers who work with PDMS in science and in industry.

The implantable grade PDMS MED-4860 (NuSil, USA) was used in experiments. The PDMS bulk samples were cured from two components: silicon resin and hardener. The mixed components were cast into preform and cured 3 h at 120 °C as prescribed.

The thin film samples were made from a solution. The components of MED-4860 were dissolved in toluene (1:20 wt./vol.), and mixed and cast in a Teflon cap. The solvent was evaporated for 1 h and then cured 3 h at 120 °C. Samples were cooled in the Teflon cap and then were peeled off.

Some cured PDMS samples were washed in n-heptane and toluene (1/100 wt./vol.) for 1 h to remove unbonded PDMS oligomers and then dried at room temperature overnight. The thickness of the cured and washed PDMS films was in the range of 0.2–0.3 mm. The residuals of n-heptane and toluene in the finally dried PDMS samples were monitored by FTIR spectra.

The PDMS samples were treated by 20 keV energy nitrogen ions in two plasma systems. First system VSIO-20KV-100NS (Imbiocom Ltd., Perm, Russia) is a vacuum chamber equipped with a dry spiral pump (Vacuummash, Kazan, Russia) and a turbomolecular pump (Izmeritel, St. Petersburg, Russia). The pumps provided a basic pressure of 10−2 Pa. High purity nitrogen was used to get the 5 × 10−1 Pa operating pressure in the chamber. The pressure was monitored with a digital vacuum meter (Meradat, Perm). Rectangular shape negative pulses with a voltage amplitude of 20 kV were applied to the high voltage electrode from a pulsed power generator (Imbiocom Ltd., MEC Ltd., ARST Ltd., Perm, Russia) against the grounded vacuum chamber. The pulse length was 20 μs. The pulse rise front was 100 ns in length. The pulse frequency of 100 Hz. The samples were placed on the high voltage stainless-steel flat electrode of 150 mm diameter and covered with stainless-steel mesh. The distance between the mesh and the sample surface was 45 mm. The voltage and current were monitored and recorded with the digital oscilloscope.

Other samples were treated in the plasma immersion ion implantation system with an inductively coupled radio-frequency plasma and a high voltage pulse generator. The base pressure was 10−4 Pa, and the pressure of high purity nitrogen during implantation was 4.4 × 10−2 Pa. The 13.56 MHz plasma power was 100 W with a reverse power of 12 W when matched. The plasma density during treatment was measured by a Langmuir probe. Acceleration of ions extracted from the plasma was achieved by the application of high voltage 20 kV bias pulses of 20 μs duration to the sample holder at a frequency of 50 Hz. The samples were mounted on a stainless-steel holder, with a stainless-steel mesh of 150 mm diameter, electrically connected to the holder, and placed 45 mm in front of the sample surface.

The ion fluence was calculated from the number of high voltage pulses multiplied by the fluence corresponding to one pulse. The fluence of one high voltage pulse was calculated by comparing UV transmission spectra of polyethylene films implanted with known fluence as described before in a study by Kondyurin and Bilek.15 The fluence of one pulse was calibrated in both systems. In both systems, the samples were treated with an ion fluence of 1016 ions/cm2. The uniform distribution of the ion fluence over the high voltage electrode area was monitored by the UV spectra of polyethylene. The treated PDMS samples were stored in closed polystyrene containers in darkness. The modified surface was not in contact with the container walls. The storage environment corresponded to the laboratory conditions: temperature 23–25 °C, humidity 60%–70%.

The electron spin resonance (ESR) spectra were recorded with a Bruker Elexsys 500 spectrometer. The Bruker Xepr software was used for the spectra analysis. The spectrometer was calibrated with standard pitches. The samples were cut to a width of 20 mm and a length of 40 mm, rolled along their width, and inserted into a glass sample tube. The empty glass tube without a sample was tested for a background signal before the sample measurements. Ten scans per sample were taken. Raw signals were transformed to g-factor using Bruker Xepr software. Peak fitting of the integrated signal was conducted using Grams software using a Lorentzian peak type.

The wettability of PDMS was measured using the sessile drop method. Kruss contact angle equipment DS10 was used. De-ionized water and diiodomethane were dropped on the sample, and the image of the drop was recorded. The drop shape was analyzed with DS10 software. The angle between the edge of the drop and the surface was calculated from the drop shape analysis. Surface energy and its components (polar and dispersic parts) were calculated using the Rabel model.

UV-Vis transmission spectra of the polyethylene films were recorded on a Specord 40 spectrophotometer in 250–800 nm region with 1 nm step scan. FTIR ATR spectra from the PDMS samples were recorded using a Digilab FTS7000 FTIR spectrometer (Digilab, USA) and Excalibur FTS3000MX FTIR spectrometer (Bio-Rad, USA). The spectrometers were fitted with an ATR accessory (Harrick, USA) with trapezium Germanium crystal and an incidence angle of 45°. The 500 scan spectra of background and samples with a resolution of 4 cm−1 were used to obtain a sufficient signal/noise ratio. To exclude the influence of ATR crystal and wetting liquids on measurements of the kinetics study, every new experimental point was carried out for a new sample at every measurement.

Optical images of the PDMS surface were obtained using a Hirox microscope (Hirox Co Ltd., Tokyo, Japan). The PDMS surface was measured using the atomic force microscope Ntegra Prima (NT-MDT Ltd., Zelenograd, Russia) with the FMG01 (Tip-sNano). The probes with a radius of the tip of 13 nm and 4.5 N/m stiffness cantilever were used. In the semicontact mode, an area of 20 × 20 μm2 was measured. The scanning step was 40 nm. The dynamic indentation mode Hybrid 3.0 was used in an area of 10 × 10 μm2. 144 force curves were obtained and averaged. The vertical component of the scanner velocity was equal to a constant value of 40 nm/s. The Hertz model was used to calculate the elastic modulus of the surface layer.

An analysis of the completely cured PDMS showed that the cured polymer contained uncross-linked low molecular weight fractions. These fractions remain on the ATR crystal after contact of the crystal with PDMS. The spectra of the residual fractions on the ATR crystal and the spectrum of the initial PDMS are shown in Fig. 1.

FIG. 1.

FTIR ATR spectra of PDMS. From bottom to top: unwashed PDMS (a), PDMS washed in toluene (b), residuals after untreated unwashed PDMS (c), residuals after treated unwashed PDMS (d), residuals from washed PDMS (f).

FIG. 1.

FTIR ATR spectra of PDMS. From bottom to top: unwashed PDMS (a), PDMS washed in toluene (b), residuals after untreated unwashed PDMS (c), residuals after treated unwashed PDMS (d), residuals from washed PDMS (f).

Close modal

The spectrum of the original PDMS has intense lines at 1080 and 1017 cm−1, related to Si–O vibrations, and a line at 1260 cm−1, related to Si–C vibrations in accordance with the literature data.27–30 The lines in the region of 750–900 cm−1 refer to bending vibrations of the Si–O–Si group. The SiO2 filler, which is introduced to improve the mechanical properties of PDMS, also contributes to the overall contour of the 1080–1017 cm−1 lines. In addition to intense lines, the spectrum contains a number of lines of much lower intensity, which can be seen when the spectrum is zoomed. Lines 2963 and 2905 cm−1 refer to stretching vibrations ν(C–H), lines 1446, 1412, and 1400 (in shoulder) cm−1 refer to bending vibrations δ(C–H), and line 1612 cm−1 refers to stretching vibrations ν(C = C) in the vinyl group, the reaction of which with the hardener provides cross-linking of PDMS macromolecules during polymer synthesis.

The spectrum of PDMS fractions remaining on the crystal has similar lines. Si–O vibrations are observed in the region of 1094 and 1022 cm−1. The normalized intensity of these lines to the intensity of the 1260 cm−1 line is lower in the spectrum of the remaining fractions compared to the spectrum of the original PDMS. In addition, the intensity of the high-frequency line wing at 1094 cm−1 has a lower intensity. This indicates the prevailing presence of polymeric macromolecules and the lower presence of the filler SiO2 in the residues on the ATR crystal. The line at 1612 cm−1 is not observed in the spectrum of residues on the crystal. Thus, the spectra of residues on the crystal show that the PDMS macromolecules in these fractions remaining on the ATR crystal do not contain vinyl groups and could not be cross-linked during the curing reaction.

Washing out of PDMS in a large amount of n-heptane and toluene (1/100 g/ml) and subsequent drying showed that low molecular weight fractions are completely removed from the cured PDMS. No fraction remains on the ATR crystal after the contact of the crystal with PDMS. The percentage of mass removed from each sample was on average 3.4% for n-heptane and 2.5% for toluene. Repeated washing of the samples showed no weight loss. Since the literature does not describe the washing of PDMS before its use as medical devices, further studies were carried out with both washed and unwashed PDMS.

The ion irradiation of PDMS leads to a significant change in the wettability of the surface. The result of the surface change is clearly visible by the difference in the shape of the water drop on the surface of the freshly treated PDMS sample (Fig. 2). The water contact angle of untreated PDMS is about 112°. For a freshly treated PDMS sample, the contact angle measurement is so strong that it presents a measurement problem. A drop of water widely spreads over the surface of the treated sample. Measurement of the water contact angle with the drop shape gave an average value of 2.2° for the sample within 5 min after turning off the plasma power in the chamber. This was the minimal time after the plasma was turned off. This value refers to washed PDMS.

FIG. 2.

(a) Water drops on untreated and ion irradiated washed PDMS. (b) Water (blue cubic) and CH2I2 (red triangle) contact angles for untreated and ion irradiated PDMS. PDMS samples were washed (empty signs) and unwashed (full signs) before the treatment. The water contact angle for glass slide is shown for comparison. (c) Total (black circle) and polar (red triangle) surface energy of untreated and ion irradiated PDMS with storage time. PDMS samples were washed (empty signs) and unwashed (full signs) before the treatment. The curves are the fitting results.

FIG. 2.

(a) Water drops on untreated and ion irradiated washed PDMS. (b) Water (blue cubic) and CH2I2 (red triangle) contact angles for untreated and ion irradiated PDMS. PDMS samples were washed (empty signs) and unwashed (full signs) before the treatment. The water contact angle for glass slide is shown for comparison. (c) Total (black circle) and polar (red triangle) surface energy of untreated and ion irradiated PDMS with storage time. PDMS samples were washed (empty signs) and unwashed (full signs) before the treatment. The curves are the fitting results.

Close modal

The contact angle does not remain constant. With time of storage of PDMS after treatment, the contact angle increases (Fig. 2). The change in the water contact angle occurs slowly along the asymptotic curve to the values close to the wetting angle of glass (Fig. 2). One week after treatment, the contact angle with water reaches 40°, and two weeks later, the contact angle becomes 50°. This angle is close to the wetting angle for an untreated glass or SiO2, which is approximately 50°–60° according to various literature data.31–34 This value of the wetting angle is related to the long storage time of the glass or quartz with adsorbed hydrocarbons from air. The contact angle curve shows that during further storage of the treated PDMS, the water contact angle asymptotically approaches this value of the contact angle with the long storage time.

The diiodomethane contact angle decreased from 95° for untreated washed PDMS to 41° immediately after the treatment. With storage time, the diiodomethane contact angle also slowly increases and reaches values of about 50° after one week of storage. No further change in the diiodomethane wetting angle within the standard deviation was observed.

Changes in the wetting angle of unwashed PDMS have a different character. The water contact angle is 48° after 38 min of plasma power is turned off. Furthermore, the wetting angle increases rather rapidly. After 32 h after treatment, the surface becomes hydrophobic with a water contact angle of 92°. Further storage of the treated unwashed PDMS leads to a further increase in the water contact angle close to that of the untreated PDMS. The diiodomethane contact angle of unwashed PDMS also changes slightly after treatment: from 78° for untreated PDMS to 52°for treated PDMS. After 32 h storage after the treatment, the diiodomethane contact angle becomes 68°.

Using water and diiodomethane contact angle data, the surface energy of PDMS and its polar and dispersion components were calculated. The surface energy of untreated PDMS is observed as low as 18 mJ/m2 for unwashed PDMS and 12 mJ/m2 for washed PDMS (Fig. 2). The main contribution to the surface energy comes from the dispersion component for both types of PDMS. Immediately after treatment with nitrogen ions, the surface energy increases sharply to values of 57 mJ/m2 for unwashed PDMS and 77 mJ/m2 for washed PDMS. An increase in the total surface energy is observed due to a sharp increase in both the dispersion and polar components of the surface energy.

With the storage time of the treated PDMS samples, the values of the surface energy and its components change. The dispersion component of the surface energy decreases slightly from 33 mJ/m2 immediately after treatment to 24 mJ/m2 after 32 h of storage for unwashed PDMS. Similarly, the dispersion component drops from 39 mJ/m2 immediately after treatment to 34 mJ/m2 after 2 weeks of storage for washed PDMS. The polar component changes more significantly with storage time. Thus, for example, the polar component drops from 23 mJ/m2 immediately after treatment to 3 mJ/m2 after 32 h of storage for unwashed PDMS and from 38 mJ/m2 immediately after treatment to 22 mJ/m2 after 2 weeks of storage for washed PDMS.

The experimental results of the surface energy and its polar component with storage time were fitted by the function,
σ = σ 1 EXP ( t / t 0 ) + σ 2 ,
(1)
where (σ1 + σ2) is the surface energy of PDMS immediately after the treatment, σ2 is the surface energy of PDMS after an infinite storage time, and t0 is the characteristic time of surface energy change that is inversely proportional to the rate of its change. The fitting results are shown in Fig. 2. Good agreement is observed between the experimental and theoretical values of the total surface energy and its polar part.

The results of fitting the experimental data by function (1) show that at the initial time immediately after treatment, the total surface energy reaches 64 mJ/m2 value for the unwashed PDMS and 76 mJ/m2 value for washed PDMS. The characteristic time of surface energy change for unwashed PDMS is 170 min and for the washed PDMS is 4000 min.

The EPR spectrum of the untreated PDMS showed no unpaired electrons within the noise level of the spectrum (Fig. 3). The EPR spectrum of the treated PDMS showed the presence of unpaired electrons, which is consistent with the presence of free radicals in the treated PDMS. The intensity of the line of unpaired electrons is much higher than the noise level. This spectrum was recorded from a sample stored for 24 h in air under laboratory conditions at room temperature after the treatment.

FIG. 3.

(a) ESR spectra of untreated and ion irradiated PDMS. (b) Fitting of ESR spectra of ion irradiated PDMS with three Lorentz functions. Black curve is experimental. Gray curve is a sum of fitting functions.

FIG. 3.

(a) ESR spectra of untreated and ion irradiated PDMS. (b) Fitting of ESR spectra of ion irradiated PDMS with three Lorentz functions. Black curve is experimental. Gray curve is a sum of fitting functions.

Close modal
The line shape of the EPR spectrum is complex and contains a contribution of at least three individual components. To analyze individual components, the experimental spectrum was approximated by a set of Lorentz functions (Fig. 3). A similar attempt to approximate the spectrum line with Gaussian functions did not give a satisfactory result, since the line wings clearly do not correspond to the shape of the Gaussian function. The position of the maxima of individual components corresponded to different g-factors: 2.0006, 2.0024, and 2.0085. The intensity and half-width of the individual components of the spectrum differ significantly from each other. The line with g-factor 2.0006 is interpreted as the Pb center at the silicon atom in the structure,
(2)
and the line with a g-factor of 2.0085 is interpreted as E’ center in the structure
(3)
in SiO2 glass.35 Such lines are observed at the interface of Si/SiO2 structures and are fairly well studied.35–38 The line with g-factor 2.0024 is interpreted as a free radical at the carbon atom.39 

The half-width and amplitude of the Lorentz functions were used to determine the contribution of each line to the overall shape of the spectrum line. The contribution of the line with a g-factor of 2.0006 is 26%. This line is the narrowest. Its half-width is about half of the half-width of the other two lines. The contribution of the line with a g-factor of 2.0024 is 51%. The contribution of the line with a g-factor of 2.0085 is 23%.

FTIR ATR spectra of unwashed PDMS samples treated by ions and stored at different times after the treatment are shown in Fig. 4. A number of intensive lines are observed as described above. The ratio of the peak intensities of the doublet lines at 1080 and 1017 cm−1 changed after ion treatment: the intensity of the 1017 cm−1 line is slightly higher in the spectrum of the treated samples than in the untreated sample. Other differences in the spectra are not observed. The penetration depth of ions is less than 150 nm, while the FTIR ATR spectra are recorded from the thicker surface layer of PDMS more than 1000 nm. The similarity of the treated and untreated PDMS spectra indicates that the bulk layer does not change during the treatment. The changes can only be observed in the thin surface layer of PDMS.

FIG. 4.

(a) FTIR ATR spectra of unwashed PDMS from bottom to top: untreated, and irradiated with increasing storage times of 30, 72, 93, 144, 210, 321, and 1500 min after the treatment. (b) Differential FTIR ATR spectra of unwashed PDMS with increasing storage time (from bottom to top 30, 72, 93, 144, 210, 321, and 1500 min) after the treatment. The spectrum of untreated PDMS is subtracted. (c) Differential FTIR ATR spectra of washed PDMS with increasing storage time (from bottom to top: 5, 15, 60, 240, 1440, and 11520 min) after the treatment. The spectrum of untreated PDMS is subtracted.

FIG. 4.

(a) FTIR ATR spectra of unwashed PDMS from bottom to top: untreated, and irradiated with increasing storage times of 30, 72, 93, 144, 210, 321, and 1500 min after the treatment. (b) Differential FTIR ATR spectra of unwashed PDMS with increasing storage time (from bottom to top 30, 72, 93, 144, 210, 321, and 1500 min) after the treatment. The spectrum of untreated PDMS is subtracted. (c) Differential FTIR ATR spectra of washed PDMS with increasing storage time (from bottom to top: 5, 15, 60, 240, 1440, and 11520 min) after the treatment. The spectrum of untreated PDMS is subtracted.

Close modal

Therefore, for a more detailed consideration, the differential spectra were obtained by subtracting the spectrum of the untreated polymer from the spectra of the treated polymer. In the differential FTIR ATR spectra of treated unwashed PDMS (Fig. 4), the following additional lines are observed: broad overlapping lines with a maximum of about 3380 cm−1 related to O–H stretching vibrations, overlapping lines in the region of 2000–2300 cm−1 with a sharp peak at 2120 cm−1 related to Si–H vibrations, overlapping lines in the 1550–1750 cm−1 region related to C = O, C = C and C = N vibrations, as well as a number of weak lines in the 1500–1350 cm−1 region with peaks at 1508, 1458, 1427, 1379, and 1356 cm−1. All these lines were not observed in the spectrum of the untreated unwashed PDMS. Moreover, the intensity of these lines varies with the storage time of PDMS after ion irradiation.

The differential FTIR ATR spectra of the washed PDMS show similar lines (Fig. 4): broad overlapping lines with a maximum at about 3400 cm−1 related to O–H stretching vibrations, overlapping lines in the 2000–2300 cm−1 region with a narrow maximum at 2160 cm−1 related to Si–H vibrations, overlapping lines in the 1550–1750 cm−1 region related to C = O, C = C, and C = N vibrations, as well as a number of weak lines in the 1500–1350 cm−1 region with peaks at 1462, 1407, 1380, and 1356 cm−1. These lines were not observed in the spectrum of untreated washed PDMS. The intensity of these lines also changes with the storage time of PDMS after ion irradiation.

The intensity of some lines changes significantly with the storage time after the ion treatment. For example, the intensity of the ν(C = O) line of carbonyl groups with a maximum at 1717 cm−1 increases with storage time (Fig. 5). The intensity increase occurs along an asymptotic curve with saturation. The intensity of the carbonyl group line in the spectrum of washed PDMS is higher than in the spectrum of unwashed PDMS at the same storage time. For unwashed PDMS, the intensity of the carbonyl group line stabilizes at a certain value after a storage time of 300 min. For washed PDMS, the increase in the intensity continues for at least 2 months of measurements and does not achieve saturation.

FIG. 5.

(a) Normalized absorbance of ν(C = O) vibration (1717 cm−1) line on the 795 cm−1 line absorbance for ion irradiated PDMS with time after ion irradiation. PDMS samples were washed (empty signs) and unwashed (full signs) before the ion treatment. (b) Normalized absorbance of ν(Si–H) vibration (2120 cm−1) line on 795 cm−1 line absorbance for ion irradiated PDMS with time after the treatment. PDMS samples were washed (empty signs) and unwashed (full signs) before the ion treatment. (c) Normalized absorbance of ν(O–H) vibration (3380 cm−1) line on 795 cm−1 line absorbance for ion irradiated PDMS with time after the treatment. PDMS samples were washed (empty signs) and unwashed (full signs) before the ion treatment.

FIG. 5.

(a) Normalized absorbance of ν(C = O) vibration (1717 cm−1) line on the 795 cm−1 line absorbance for ion irradiated PDMS with time after ion irradiation. PDMS samples were washed (empty signs) and unwashed (full signs) before the ion treatment. (b) Normalized absorbance of ν(Si–H) vibration (2120 cm−1) line on 795 cm−1 line absorbance for ion irradiated PDMS with time after the treatment. PDMS samples were washed (empty signs) and unwashed (full signs) before the ion treatment. (c) Normalized absorbance of ν(O–H) vibration (3380 cm−1) line on 795 cm−1 line absorbance for ion irradiated PDMS with time after the treatment. PDMS samples were washed (empty signs) and unwashed (full signs) before the ion treatment.

Close modal

The ν(Si–H) vibration line in the spectrum of untreated PDMS is observed at 2160 cm−1 and corresponds to the residual end groups of the macromolecule. The spectrum of treated PDMS in this region contains at least three lines with maxima at 2253, 2160, and 2120 cm−1. The intensity of the first two lines varies little, while the intensity of the last line at 2120 cm−1 decreases with storage time (Fig. 5). The decrease in the intensity is more noticeable in the spectrum of washed PDMS, although in the spectrum of unwashed PDMS, the relative intensity of this line is somewhat higher under the same ion treatment conditions. The variation in the intensity in the spectra of unwashed PDMS between the samples is high.

The intensity of the ν(OH) line increases with the storage time after treatment for all samples (Fig. 5). In the first period of time, the intensity of this line increases rapidly with storage time. Then, the increase in the intensity slows down. The intensity in the spectra of unwashed samples reaches saturation at a storage time of more than 300 min. In the spectrum of washed PDMS, the intensity of this line is higher than in the spectrum of unwashed PDMS under the same treatment and storage conditions. The intensity increase in the spectrum of washed PDMS takes a much longer time. In our measurements, the increase in the intensity continued for 2 months after treatment and did not reach saturation.

The shape of this line is complex and can be interpreted as a superposition of several individual vibration lines ν(OH) in OH groups and ν(CH) in –CH3 and –CH2– groups. The shape of the line also changes with the storage time of the samples after ion treatment. For a detailed analysis of the changes, the spectrum line was fitted with individual components. First of all, the number of lines and their position was found. The spectrum deconvolution procedures and the second derivative of the spectrum were used.40 The spectrum deconvolution and the second derivative showed that the number of lines does not change with the storage time after ion treatment for all samples including washed and unwashed PDMS. The position of the lines also slightly changes with the storage time and sample prewashing. This facilitated the task of fitting the line with individual components and their subsequent analysis. The approximation of the experimental shape of the lines was performed by Gaussian functions. Examples of line fitting are shown in Fig. 6 for unwashed PDMS.

FIG. 6.

FTIR ATR spectra of treated unwashed PDMS (a) and (b). The spectra were recorded in 30 min (a) and 7 days (b) after the treatment. Wavenumbers are noted in cm−1. (c) and (d) Normalized integral absorbance of ν(OH) line components in FTIR ATR spectra of treated unwashed PDMS with time after treatment. The absorbance is normalized on integral ν(CH) line absorbance. Signs are the experimental points. The lines are fitting curves.

FIG. 6.

FTIR ATR spectra of treated unwashed PDMS (a) and (b). The spectra were recorded in 30 min (a) and 7 days (b) after the treatment. Wavenumbers are noted in cm−1. (c) and (d) Normalized integral absorbance of ν(OH) line components in FTIR ATR spectra of treated unwashed PDMS with time after treatment. The absorbance is normalized on integral ν(CH) line absorbance. Signs are the experimental points. The lines are fitting curves.

Close modal

In the spectrum of the unwashed PDMS sample, immediately after ion treatment, narrow lines of ν(CH) vibrations at 2880, 2922, and 2950 cm−1 are observed. These lines refer to vibrations in –CH3, –CH2– and > CH– groups in the treated layer of PDMS. The intensity of the lines of ν(CH) vibrations does not change with the storage time of the samples after ion treatment. Therefore, these lines were used as an internal standard.

The fitted spectrum in the ν(OH) vibration range shows broad lines with maxima at 3034, 3194, 3357, 3520, and 3623 cm−1. The integrated intensity of the individual ν(OH) lines was normalized to the sum of the integrated intensity of the ν(CH) lines. The results of the fitted individual component intensities for unwashed PDMS with the storage time after ion treatment are shown in Fig. 6.

With a storage time of a week, the position of the lines in the unwashed PDMS spectrum remains unchanged, while the intensity of some lines increases. The increase of the 3034, 3357, and 3520 cm−1 lines intensity is observed up to 200–300 min of storage time and then saturated. The intensity of the 3194 cm−1 line continues further, although the growth rate slows down. The intensity of the 3623 cm−1 line does not increase with the storage time.

Similarly, the analysis of the spectra was made for the washed PDMS after ion treatment. Fitting of the spectrum by individual components showed similar vibration lines of ν(OH) and ν(CH) (Fig. 7). The intensities of the individual lines in dependence on the storage time are shown in Fig. 7. The intensity of individual ν(OH) lines was normalized to the sum of the integrated intensity of ν(CH) lines. The intensity of the lines at 3207 and 3373 cm−1 increases with time and does not stop increasing in 8 days after ion treatment. The intensity of the lines at 3061 and 3536 cm−1 increases immediately after treatment, then slows down, and is saturated after storage for more than one day. The intensity of the line at 3630 cm−1 does not change after the appearance of this line in the spectrum immediately after ion treatment.

FIG. 7.

FTIR ATR spectra of treated washed PDMS (a) and (b). The spectra were recorded in 5 min (a) and 8 days (b) after treatment. (c) and (d) Normalized integral absorbance of ν(OH) line components in FTIR ATR spectra of treated washed PDMS with time after treatment. The absorbance is normalized on integral ν(CH) line absorbance. Signs are the experimental points. The lines are fitting curves.

FIG. 7.

FTIR ATR spectra of treated washed PDMS (a) and (b). The spectra were recorded in 5 min (a) and 8 days (b) after treatment. (c) and (d) Normalized integral absorbance of ν(OH) line components in FTIR ATR spectra of treated washed PDMS with time after treatment. The absorbance is normalized on integral ν(CH) line absorbance. Signs are the experimental points. The lines are fitting curves.

Close modal

In the differential spectra in the region of 1000–1200 cm−1, a strong line of Si–O vibrations is observed (Fig. 8). Since the intensity of the initial PDMS lines is completely compensated upon subtraction, which can be seen from the almost complete disappearance of the 1260 cm−1 line of vibrations in the Si–CH3 group of PDMS, the observed lines correspond to the new structure of the surface layer formed as a result of ion treatment. The shape of this line is complex and consists of several components. With the storage time of samples after ion treatment, the shape of this line noticeably changes: the high-frequency wing of this line increases with respect to the low-frequency wing of the line.

FIG. 8.

(a) Differential FTIR ATR spectra (Si–O vibrations region) of washed PDMS with increasing storage time (from bottom to top: 5, 15, 60, 240, 1440, and 11520 min) after treatment. The spectrum of untreated PDMS is subtracted. (b) FTIR ATR differential spectra of treated washed PDMS in 5 min after treatment, its second derivative, and deconvoluted spectrum.

FIG. 8.

(a) Differential FTIR ATR spectra (Si–O vibrations region) of washed PDMS with increasing storage time (from bottom to top: 5, 15, 60, 240, 1440, and 11520 min) after treatment. The spectrum of untreated PDMS is subtracted. (b) FTIR ATR differential spectra of treated washed PDMS in 5 min after treatment, its second derivative, and deconvoluted spectrum.

Close modal

A fitting of the experimental spectra on individual components was done for the detailed analysis of this region of the spectra. A number of individual components and their position were determined with the spectrum deconvolution procedure according to the method40 and the method of the second derivative (p. 9). The results of deconvolution and the second derivative of the spectrum are shown in the example of the experimental spectrum of the sample immediately after ion treatment. The spectrum after deconvolution shows a number of narrow peaks at 1006, 1043, 1065, 1106, and 1141 cm−1. The spectrum of the second derivative shows the same peaks but with a negative sign. Similar peaks were obtained for other spectra. The resulting number of peaks and the position of their maxima were used for the first approximation of the fitting procedure. Note that the intensity of the peaks in the spectrum after deconvolution and in the spectrum of the second derivative showed that the half-width of the peaks is not the same for all individual components. This can be seen from the high peak at 1006 cm−1, while there is no strong absorption in this region in the original spectrum. That is, the height of the peak in the spectrum after deconvolution and in the spectrum of the second derivative is due to the small half-width of the individual component compared to other components. Therefore, when fitting the line, the half-width of the individual components was not limited.

The fitting results of the spectrum with individual components are shown in Fig. 9. For example, two spectra of washed samples immediately after treatment with ions and after 8 day storage are shown. The spectrum line of the freshly treated sample is fitted with five Gaussian functions with maxima at 1006, 1015, 1061, 1115, and 1140 cm−1. The spectrum of the sample after 8 day storage is also decomposed into five lines at approximately the same positions of the maxima. The spectra of other samples after different storage times show similar results.

FIG. 9.

FTIR ATR differential spectra of treated washed PDMS in 5 min (a) and in 8 days (b) after treatment. The spectrum of untreated PDMS is subtracted. The spectra were fitted with the Gauss function: signs are experimental data. The lines are fitting curves. (c) and (d) Integral intensity of the lines (noted in cm−1) in FTIR ATR spectra of treated washed PDMS with storage time after treatment. The line parameters received from the fitting curves were used.

FIG. 9.

FTIR ATR differential spectra of treated washed PDMS in 5 min (a) and in 8 days (b) after treatment. The spectrum of untreated PDMS is subtracted. The spectra were fitted with the Gauss function: signs are experimental data. The lines are fitting curves. (c) and (d) Integral intensity of the lines (noted in cm−1) in FTIR ATR spectra of treated washed PDMS with storage time after treatment. The line parameters received from the fitting curves were used.

Close modal

The position of these lines also coincides with the literature data for Si–O vibrations in various structures of silicon oxide.41–43 These lines are broad corresponding to the inorganic nature of the absorbing group. The lines at 1140 and 1115 cm−1 are typical for the cage structure of silicon oxide. The line at 1061 cm−1 corresponds to the network structure of the Si–O bonds in the silicon oxide. The lines at 1015 and 1006 cm−1 are laying in the range corresponding to the suboxide of the silicon. These structures are characterized by dangled bonds, traps, and silanol groups.

The integrated intensity of the fitting individual line components is shown in Fig. 9. The intensity of the lines at 1006 and 1115 cm−1 does not change with the storage time of the sample. The intensity of the 1015 cm−1 line decreases with the storage time of the sample after ion treatment during the first day and then stabilizes at a certain value. The intensity of the 1140 cm−1 lines increases with storage time during the day after ion treatment and then stabilizes. The intensity of the 1061 cm−1 line increases sharply during 1 day of storage, and then, its growth slows down but does not stop for 8 days of the storage time.

Optical microscopy showed no significant surface changes with treatment [Figs. 10(a) and 10(b)]. The untreated PDMS surface is smooth and uniform. A weak wave structure and rare cracks are observed on the treated surface.

FIG. 10.

Optical microphotos (a) and (b) (100 × 100 μm2) and AFM image (c) and (d) (20 × 20 μm2) of PDMS untreated (a) and (c) and treated (b) and (d).

FIG. 10.

Optical microphotos (a) and (b) (100 × 100 μm2) and AFM image (c) and (d) (20 × 20 μm2) of PDMS untreated (a) and (c) and treated (b) and (d).

Close modal

Similar results were obtained using atomic force microscopy [Figs. 10(c) and 10(d)]. The surface of the treated PDMS is characterized by a certain wave structure. The direction of the waves is irregular. The waves are small in amplitude in comparison with the roughness of the surface.

Contact measurement of the PDMS surface with an atomic force microscope probe showed a significant change in the surface elasticity after treatment. The distribution of the modulus of the untreated PDMS showed a set of surface areas with different moduli of elasticity (Fig. 11). The modulus of elasticity of untreated PDMS lies in the range from 2.2 to 9.3 MPa. The main part of the surface has a modulus between 2.2 and 3.8 MPa. Such a spread in the values of the elastic modulus over the surface of untreated PDMS corresponds to its heterogeneous structure, which includes a soft phase of the polymer PDMS itself and inclusions of silicon dioxide particles.

FIG. 11.

Modulus distribution (a) and (b) on the PDMS surface from AFM measurements (Hertz model): (a) is untreated PDMS and (b) is treated PDMS. Signs are the experimental points. Lines are Gauss fitting curves. Positions of the fitting components are noted in MPa. (c) Adhesion distribution on untreated and treated PDMS surfaces from AFM measurements. (d) Distribution of hydrogen, carbon, silicon, and oxygen vacancies in PDMS after nitrogen ion of 20 keV penetration. SRIM calculations.

FIG. 11.

Modulus distribution (a) and (b) on the PDMS surface from AFM measurements (Hertz model): (a) is untreated PDMS and (b) is treated PDMS. Signs are the experimental points. Lines are Gauss fitting curves. Positions of the fitting components are noted in MPa. (c) Adhesion distribution on untreated and treated PDMS surfaces from AFM measurements. (d) Distribution of hydrogen, carbon, silicon, and oxygen vacancies in PDMS after nitrogen ion of 20 keV penetration. SRIM calculations.

Close modal

After ion treatment, the surface becomes much harder. The modulus of elasticity lies in the range from 198 to 350 MPa. The main part of the surface has a modulus of 255 MPa.

The measurements also showed a significant change in the adhesion force of the AFM probe to the PDMS surface (Fig. 11). The adhesion of the AFM probe to the untreated PDMS is averaged about 18 relative units. The distribution of the adhesion force over the polymer surface is quite wide. After processing, the adhesion force is averaged about 5 relative units. The distribution of adhesion force over the surface is much narrower compared to the untreated polymer.

As a result of ion irradiation, a strong violation of the structure in the surface layer of PDMS occurs. This effect is due to several processes. The first of the processes is a transfer of ion energy to atoms of the PDMS macromolecule. Transfer occurs as a result of the interaction of the penetrating ion with an atom of the macromolecule. This process can be calculated based on the equations,44 which are implemented in the SRIM computer code. As a result of the calculation, the profiles of the ion energy loss due to electronic excitations, nuclear excitations, and phonon excitations are obtained. The result of calculating the distribution of vacancies of PDMS atoms after the passage of nitrogen ion with an energy of 20 keV is shown in Fig. 11.

The nitrogen ion, penetrating into the polymer, knocks out a number of atoms of macromolecules, creating a region of broken bonds in the macromolecule and moving the knocked-out atoms with broken bonds away from the parent macromolecule. The total number of vacancies of knocked-out polymer atoms is about 214 per incoming nitrogen ion. Thus, a region with a high content of broken chemical bonds is created in a relatively thin surface layer of PDMS up to 150 nm depth. Note that knocking out one hydrogen atom from the macromolecule results in two broken bonds, knocking out one oxygen atom leads to four broken bonds, and knocking out one silicon or carbon atom leads to eight broken bonds. Thus, integrating the distribution of vacancies with allowance for the number of broken bonds for each element, we obtain that one nitrogen ion with an energy of 20 keV generates about 875 broken bonds in a layer of about 150 nm.

Fragments of macromolecules and atoms with such broken bonds, otherwise, called free radicals have high chemical activity. As a result of such free radical formation, free radical reactions are triggered, leading to the formation of new chemical structures. Due to the high chemical activity of free radicals, the formation of such structures is largely random and fast. In this case, the knocked-out hydrogen atoms are most likely able to fly out of the surface layer of the polymer and pass into the vacuum of the chamber. Carbon, oxygen, and silicon atoms will basically form new structures. Within a short time after nitrogen ion passed, the new structures containing Si–O, Si–C, and C–O bonds, as well as multiple bonds of these atoms, appear.

Usually, due to high reactivity, free radicals react quickly and disappear in a short time. However, some free radicals cannot react due to geometric difficulties and low macromolecule mobility. The result is some broken bonds remain. This is observed from the lines of unpaired electrons in the EPR spectra of PDMS after 24 h of storage in air at room temperature. In the spectrum of PDMS treated with nitrogen ions, these lines are observed with g-factors of 2.0006 and 2.0085 related to the radical at the silicon atom and with a g-factor of 2.0024 related to the radical at the carbon atom.

Similar EPR spectra of free radicals in PDMS were observed earlier. For example, in a study by Menhofer and Heusinger,45 the following free radicals
(4)
were observed in PDMS treated by gamma irradiation from Co60 isotope in a wide temperature range from 77 to 293 K. It was noted that •CH3 free radicals disappeared at room temperature, while the radicals at the silicon atom remained stable.

The presence of free radicals in other polymers after ion beam treatment was also observed in the EPR spectra. In a study by Kondyurin et al.,46 low density polyethylene was treated with high energy nitrogen ions and the EPR spectrum showed the presence of free radicals at room temperature. Moreover, free radicals in the surface layer can interact with the external environment.47 In a study by Kosobrodova et al.,48 it has been shown that the active free radicals remain in nitrogen ion treated polystyrene for at least 3 weeks after treatment when stored in air at room temperature.

Note that the peak and total integral intensity of the free radical line in the EPR spectra of the treated PDMS is much lower than in the EPR spectra of other polymers treated with an ion beam under the same conditions. It is possible that the low concentration of remaining free radicals in PDMS is due to the fact that silicon structures do not stabilize free radicals, as happens in carbon chain polymers, where condensed aromatic structures are formed and they stabilize the free radicals.49 However, we did not find references where the kinetics of consumption or transformation of free radicals in PDMS after ion irradiation under the same conditions as for carbon chain polymers have been investigated. Therefore, this assumption has not yet been experimentally confirmed. On the other hand, the chemistry of free radical transformations in PDMS has been studied much less than the chemistry of free radicals in carbon chain polymers.39,50

Thus, the presence of free radicals in the surface layer of treated PDMS cannot be neglected. It is also obvious that free radicals do not collapse rapidly, but, due to apparently steric hindrances, are able to remain in the polymer for some time. Accordingly, their presence in the polymer and their effect on the properties of the treated polymer should be taken into account.

From this point of view, we can consider changes in the FTIR ATR spectra of PDMS with storage time after ion treatment. There is no carbonyl group in the original PDMS. After treatment, the line of carbonyl groups appears in the spectrum and its intensity increases with the time of storage of PDMS samples after treatment. The line of hydroxyl groups also appears, and its intensity increases with the storage time. The increase in the intensity of these lines occurs during the entire observation time (2 months). These spectral changes show that carbonyl and hydroxyl groups appear in the polymer and their concentration increases with time for at least a month. The appearance and increase in the concentration of such groups can only be interpreted as a result of free radical reactions in the surface layer of PDMS. Note that these free radical reactions take quite a long time. In the experiment, this is observed for at least 2 months after treatment.

The fitting of the hydroxyl group line with individual components showed that the kinetics of the various hydroxyl groups is complex and can radically differ from each other. At this stage, it is not yet possible to say definitely which reactions lead to the appearance of particular hydroxyl groups. However, some interpretation can be provided based on the comparison of the obtained results with known literature data. The line at 3623 cm−1 is interpreted as Si–OH groups not connected by a hydrogen bond.51 The lines at 3520 and 3357 cm−1 are interpreted as hydroxyl groups involved in a single hydrogen bond, while the lower frequency lines at 3194 and 3034 cm−1 correspond to hydroxyl groups in network hydrogen bonds with some proton-acceptor groups.52–56 It is hardly possible that these lines refer to adsorbed water in the surface layer of PDMS. The growth of the intensities of these lines continues for a month after ion treatment. Water adsorption on the surface is a rather fast process and cannot last slowly for 2 months in the 150 nm layer. Possibly, during this time, hydroxyl groups are formed in the surface layer of PDMS, so that adsorbed water molecules acquire the ability to form complex hydrogen bond networks with the inclusion of newly formed groups in the surface layer of PDMS.

One of the possibilities for the formation of proton-acceptor groups is observed by a decrease in the intensity of the 2120 cm−1 line of vibrations of Si–H groups in the spectrum. Oxidation of this group leads to the formation of a Si–OH group, which is able to participate in hydrogen bonding with water molecules.

Similar instability of the surface layer with the storage time can be seen from the spectrum in the region of Si–O bonds vibrations. The line intensity for cage and network structures in silicon oxide is growing with the storage time. These structures are regular and stable. The line intensity of suboxide at 1015 cm−1 decays with the storage time that shows a restructuring, when an unstable unsaturated structure of silicon oxide is transformed into a stable saturated structure.

Similar changes in the FTIR spectra of PDMS after plasma and ion treatment were also observed earlier. In Husein et al.,57 PDMS was treated with nitrogen ions with energies of 4 and 8 keV. The appearance of lines of the silanol group and changes in the region of vibrations of Si–O bonds were observed in the surface spectra. In a number of studies, see, for example, Toth et al.,58 Marletta et al.,59 Vladkova et al.,60 Zheng et al.,61 and Cutroneo et al.,62 a conclusion was made about the transformation of the PDMS structure under ion beams with different energies, compositions, and charges into ceramic structures such as SiOxCyHz, SiC, and SiOx types. Interestingly that the transformation of PDMS macromolecules into silicalike structures is also observed during plasma treatment at a much lower energy.63 

The structure like silica or ceramic is hard. When the AFM tip touches the hard surface layer, the adhesion force is determined by an interaction between the AFM tip and the surface in a small contact area. When the AFM tip touches the soft surface of untreated PDMS, the tip sinks into the PDMS layer. The contact area of the tip is much larger than that of the treated. As a result, the force to disjoin the tip from the PDMS is much higher. The sinking process is relatively random for the untreated PDMS surface, resulting in a wide distribution of the adhesion force. In the case of the ceramiclike surface of the treated PDMS, the contact area is small with a narrow distribution of the adhesion force over the surface. Therefore, the adhesion force result is an indicator of the softness of the PDMS surface layer in such experiments.

The formation of new structures in the surface layer of PDMS should undoubtedly affect the surface properties such as wettability and surface energy. Immediately after treatment, the surface is characterized by small contact angles and high surface energy. Moreover, the main contribution to the increase in surface energy as a result of treatment falls on both the polar and dispersive components of the surface energy. This distinguishes the nature of the change in the surface energy of PDMS from other polymers, where the change in the polar component of the surface energy as a result of ion treatment prevails.15 However, with the storage time of PDMS samples after treatment, the dispersion component of the surface energy changes little, as well as for other polymers.

The maximum value of the surface energy of PDMS immediately after treatment, calculated from the fitting experimental data, is 76 mJ/m2 for washed PDMS. This value is close to the surface energy of the hydroxylated quartz surface,64 while the fracture surface energy of the quartz is about 400–1150 mJ/m2. Similar values are observed for a surface of fused quartz after annealing at 240 and 1000 °C in air, which are 73 mJ/m2 for 240 °C annealing and 63–66 mJ/m2 for 1000 °C annealing.31 

With storage time, the surface energy reaches 55 mJ/m2 value. The contributions of the polar and dispersive components of the surface energy become approximately equal. This surface energy value of PDMS is close to the surface energy values65 for plasma-cleaned glass in the range of 40–60 mJ/m2. Such a glass surface is characterized by a large number of hydroxyl groups on the glass surface and ensures the hydrophilic nature of the interaction. A similar increase in hydrophilicity is observed when PDMS is treated with low-energy plasma66 and UV treatment.67 There is also an increase in the contact angle with water with time after plasma treatment, which is presumably associated with the diffusion of low molecular weight fractions of PDMS to the surface with the storage time of samples after treatment.21–23 

In our experiments with unwashed PDMS after ion treatment, we also observed the recovery of the contact angle to the values for untreated PDMS. This is also explained by the diffusion of low molecular weight non-cross-linked PDMS components onto the surface. That is, the PDMS surface actually modified by the ions is covered with unchanged low molecular weight fractions that have a low surface energy with an almost zero polar component of the surface energy. The persistence of such fractions on the surface of unwashed modified PDMS is observed from the FTIR spectra shown above.

However, washed PDMS does not contain low molecular weight fractions. A number of questions arise. First, if PDMS washed from low molecular weight fractions is treated, then the angle should not recover after processing with storage time. If the angle recovers, then the reason for the recovery of the contact angle on the washed PDMS is not the diffusion of low molecular weight fractions to the surface, but something else.

Second, in the literature, the low contact angle on the treated PDMS surface and treated glass is attributed to the presence of hydrophilic silanol, hydroxyl, and carboxyl groups on the surface.60,61,68 However, this contradicts the kinetics of accumulation of hydroxyl groups according to FTIR spectra. With the time of storage, the concentration of hydroxyl groups on the surface increases, and the hydrophilicity of the surface, observed by the contact angle, decreases.

Thirdly, in the literature, one can also find an explanation for the decrease in the hydrophobicity of the treated polymer surface with storage time due to the reorientation of macromolecular chains so that the surface is covered with hydrophobic groups, while the hydrophilic groups turn down.69–70 However, the treated surface of PDMS is characterized up to the rigid ceramiclike structure, and the rotation of any groups inside such a material is very difficult or impossible.

There is another explanation for the decrease in surface hydrophilicity after treatment: adsorption of polluting hydrophobic components from the atmosphere.71 However, the repeated wetting results with different polymers after treatment obtained in different laboratories by different researchers cannot be explained by the presence of random contaminants of the same concentration and nature. Therefore, the reason for the decrease in surface hydrophilicity with storage time must be associated with chemical processes in the treated polymer layer itself.

We explained this change in wetting and surface energy as a result of free radical reactions in polymers after ion treatment. Immediately after treatment, the polymer surface is saturated with broken bonds; that is, the concentration of free radicals is maximum. Taking into account that one nitrogen ion forms 875 broken bonds, then with a fluence of 1016 ions/cm2, about 1019 broken bonds per 1 cm2 are obtained. Provided that all broken bonds are in a layer 150 nm thick, then approximately every second monomer of the silicone macromolecule in this layer should, on average, have at least one broken bond. This calculation is made under the assumption of the conditions of the TRIM program44 where it is assumed that each following ion penetrates again into the unchanged polymer, which, in the case of high fluences, may not be entirely correct. Therefore, this result should be considered an approximate estimate. However, despite this simplification of the calculation, the calculated number of free radicals on the surface is very large.

Free radicals are characterized by an unpaired electron in the valence shell of the atom. Such an unpaired electron has a large energy of interaction. Even a small concentration of such unpaired electrons at atoms in the surface layer of the polymer can give a much higher surface energy value than the presence of any polar groups with saturated valences.46,72

With storage time, the free radicals in the treated surface are consumed as a result of free radical reactions, forming stable groups with filled valences. At the same time, the concentration of such new stable groups increases, which is observed in the experiment. The value of the surface energy of the polymer drops due to a decrease in the concentration of free radicals. The value of the surface energy of the polymer after long-term storage and the consumption of all free radicals is determined by new stable groups with filled valences.

Similar changes in the surface energy with the storage time and the increase in the new polar group concentration are observed in the case of PDMS. Therefore, it can be assumed that the high value of the surface energy of the treated PDMS is associated with the presence of free radicals on its surface, formed as a result of ion bombardment.

Ion irradiation creates a ceramiclike layer on the surface of PDMS characterized by the destruction of PDMS macromolecules and the formation of silicon oxide-type structures with inclusions of irregular structures such as suboxides and carbon islands. Immediately after the treatment, the surface contains free radicals, which provide good wettability and high surface energy. With time, free radical reactions proceed, leading to the formation of stable groups such as hydroxyl, carbonyl, silanol, and others, and the disappearance of free radicals. The wettability and surface energy of the treated PDMS tend to values corresponding to hydrated silica and glass.

The presence of low molecular weight non-cross-linked fractions in PDMS significantly changes the kinetics of surface transformations. The diffusion of such fractions onto the surface recovers the wettability and surface energy of PDMS close to that of untreated PDMS. Washing off cross-linked PDMS from non-cross-linked low molecular weight fractions before ion irradiation treatment removes this effect.

The study was supported by the Government of the Perm Krai, Research Project No. C-26/875. Vyacheslav Chudinov thanks Professor M. Bilek and Professor D. McKenzie for support and permission to do a part this study in their laboratory.

The authors have no conflicts to disclose.

Vyacheslav Chudinov: Data curation (equal); Investigation (equal); Writing – original draft (equal); Writing – review & editing (equal). Igor N. Shardakov: Supervision (supporting). Ilya A. Morozov: Investigation (supporting). Irina V. Kondyurina: Investigation (supporting). Alexey Kondyurin: Conceptualization (lead); Data curation (lead); Formal analysis (lead); Investigation (lead); Methodology (lead); Project administration (lead); Supervision (lead); Writing – original draft (lead); Writing – review & editing (lead).

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

1.
I.
Miranda
,
A.
Souza
,
P.
Sousa
,
J.
Ribeiro
,
E. M. S.
Castanheira
,
R.
Lima
, and
G.
Minas
,
J. Funct. Biomater.
13
,
2
(
2022
).
2.
Q.
Alkhalaf
,
S.
Pande
, and
R. R.
Palkar
, “
Review of polydimethylsiloxane (PDMS) as a material for additive manufacturing
,” in
Innovative Design, Analysis and Development Practices in Aerospace and Automotive Engineering. Lecture Notes in Mechanical Engineering
, edited by
N.
Gascoin
and
E.
Balasubramanian
(
Springer
,
Singapore
,
2021
).
3.
R.
Ariati
,
F.
Sales
,
A.
Souza
,
R. A.
Lima
, and
J.
Ribeiro
,
Polymers
13
,
4258
(
2021
).
4.
F.
Akther
,
S. B.
Yakob
,
N. T.
Nguyen
, and
H. T.
Ta
,
Biosensors
10
,
182
(
2020
).
5.
M.
Dardouri
et al,
Mater. Sci. Eng. C
134
,
112563
(
2021
).
6.
J. D. P.
Valentin
,
X.-H.
Qin
,
C.
Fessele
,
H.
Straub
,
H. C.
van der Mei
,
M. T.
Buhmann
,
K.
Maniura-Weber
, and
Q.
Ren
,
J. Colloid Interface Sci.
552
,
247
(
2019
).
7.
M. S.
Birajdar
,
B. H.
Kim
,
C.
Sutthiwanjampa
,
S. H.
Kang
,
C. Y.
Heo
, and
H.
Park
,
J. Ind. Eng. Chem.
89
,
128
(
2020
).
8.
H.
Tsuji
,
M.
Izukawa
,
R.
Ikeguchi
,
R.
Kakinoki
,
H.
Sato
,
Y.
Gotoh
, and
J.
Ishikawa
,
Nucl. Instrum. Methods Phys. Res., Sect. B
206
,
507
(
2003
).
9.
Y.
Suzuki
,
M.
Kusakabe
,
J.-S.
Lee
,
M.
Kaibara
,
M.
Iwaki
, and
H.
Sasabe
,
Nucl. Instrum. Methods Phys. Res., Sect. B
65
,
142
(
1992
).
10.
Y.
Suzuki
,
Nucl. Instrum. Methods Phys. Res., Sect. B
206
,
501
(
2003
).
11.
M.
Ionescu
,
B.
Winton
,
D.
Wexler
,
R.
Siegele
,
A.
Deslantes
,
E.
Stelcer
,
A.
Atanacio
, and
D. D.
Cohen
,
Nucl. Instrum. Methods Phys. Res., Sect. B
273
,
161
(
2012
).
12.
H.
Tsuji
,
M.
Izukawa
,
R.
Ikeguchi
,
R.
Kakinoki
,
H.
Sato
,
Y.
Gotoh
, and
J.
Ishikawa
,
Nucl. Instrum. Methods Phys. Res., Sect. B
206
,
507
(
2003
).
13.
D.
Fink
,
Fundamentals of Ion-Irradiated Polymers
(
Springer
,
Berlin
,
2004
).
14.
V. B.
Odzhaev
,
I. P.
Kozlov
,
V. N.
Popok
, and
D. B.
Sviridov
,
Ion Implantation of Polymers
(
Belorussian State University
,
Minsk
,
1998
).
15.
A.
Kondyurin
and
M.
Bilek
,
Ion Beam Treatment of Polymers
(
Elsevier
,
Oxford
,
2008
).
16.
T. G.
Vladkova
,
I. L.
Keranov
,
P. D.
Dineff
,
S. Y.
Youroukov
,
I. A.
Avramova
,
N.
Krasteva
, and
G. P.
Altankov
,
Nucl. Instrum. Methods Phys. Res., Sect. B
236
,
552
(
2005
).
17.
H.
Tsuji
,
M.
Izukawa
,
R.
Ikeguchi
,
R.
Kakinoki
,
H.
Sato
,
Y.
Gotoh
, and
J.
Ishikawa
,
Appl. Surf. Sci.
235
,
182
(
2004
).
18.
C.
Zheng
,
G.
Wang
,
Y.
Chu
,
Y.
Xu
,
M.
Qiu
, and
M.
Xu
,
Nucl. Instrum. Methods Phys. Res., Sect. B
370
,
73
(
2016
).
19.
A.
Mata
,
A. J.
Fleischman
, and
S.
Roy
,
Biomed. Microdevices
7
,
281
(
2005
).
20.
J. C.
McDonald
,
D. C.
Duffy
,
J. R.
Anderson
,
D. T.
Chiu
,
H.
Wu
,
O. J. A.
Schueller
, and
G. M.
Whitesides
,
Electrophoresis
21
,
27
(
2000
).
21.
Y.
Berdichevsky
,
J.
Khandurina
,
A.
Guttman
, and
Y.-H.
Lo
,
Sens. Actuators, B
97
,
402
(
2004
).
22.
J. L.
Fritz
and
M. J.
Owen
,
J. Adhes.
54
,
33
(
1995
).
23.
H.
Hillborg
and
U. W.
Gedde
,
Polymer
39
,
1991
(
1998
).
24.
L.
Tong
,
W.
Zhou
,
Y.
Zhao
,
X.
Yu
,
H.
Wang
, and
P. K.
Chu
,
Colloids Surf., B
148
,
139
(
2016
).
25.
D.
Bodas
and
C.
Khan-Malek
,
Microelectron. Eng.
83
,
1277
(
2006
).
26.
M.
Cutroneo
,
V.
Havranek
,
P.
Malinsky
,
A.
Mackova
,
A.
Torrisi
,
J.
Flaks
,
P.
Slepicka
, and
L.
Torrisi
,
Nucl. Instrum. Methods Phys. Res., Sect. B
459
,
137
(
2019
).
27.
G.-M.
Kim
,
S.-J.
Lee
, and
C.-L.
Kim
,
Materials
14
,
4489
(
2021
).
28.
L. J.
Bellamy
,
The Infrared Spectra of Complex Molecules
, 3rd ed. (
Chapman and Hall
,
London
,
1975
).
29.
A.
Lee Smith
,
Spectrochim. Acta
16
,
87
(
1960
).
30.
L.
Bistricic
,
V.
Borjanovic
,
L.
Mikac
, and
V.
Dananic
,
Vib. Spectrosc.
68
,
1
(
2013
).
31.
I.
Zgura
,
R.
Moldovan
,
C. C.
Negrila
,
S.
Frunza
,
V. F.
Cotorobai
, and
L.
Frunza
,
J. Optoelectron. Adv. Mater.
15
,
627
(
2013
).
32.
R. M.
Pashley
and
J. A.
Kitchener
,
J. Colloid Interface Sci.
71
,
491
(
1979
).
33.
A. K.
Helmy
,
S. G.
de Bussetti
, and
E. A.
Ferreiro
,
Appl. Surf. Sci.
253
,
6878
(
2007
).
34.
B.
Jaňczuk
,
E.
Chibowski
, and
T.
Bialopiotrowicz
,
Chem Papers
40
,
349
(
1986
).
35.
P. M.
Lenahan
and
J. F.
Conley
, Jr.
,
J. Vac. Sci. Technol. B
16
,
2134
(
1998
).
36.
Y.
Nishi
,
Jpn. J. Appl. Phys.
10
,
52
(
1971
).
37.
P. J.
Caplan
,
E. H.
Poindexter
,
B. E.
Deal
, and
R. R.
Razouk
,
J. Appl. Phys.
50
,
5847
(
1979
).
38.
E. H.
Poindexter
,
P. J.
Caplan
,
B. E.
Deal
, and
R. R.
Razouk
,
J. Appl. Phys.
52
,
879
(
1981
).
39.
N. M.
Emanuel
and
A. L.
Buchachenko
,
Chemical Physics of Polymer Degradation and Stabilization
(
Brill Academic
,
Leiden
,
1987
).
40.
J. K.
Kauppinen
,
D. J.
Moffatt
,
H. H.
Mantsch
, and
D. G.
Cameron
,
Appl. Spectrosc.
35
,
271
(
1981
).
41.
A.
Goullet
,
C.
Charles
,
P.
Garcia
, and
G.
Turban
,
J. Appl. Phys.
74
,
6876
(
1993
).
42.
K. T.
Queeney
,
M. K.
Weldon
,
J. P.
Chang
,
Y. J.
Chabal
,
A. B.
Gurevich
,
J.
Sapjeta
, and
R. L.
Opila
,
J. Appl. Phys.
87
,
1322
(
2000
).
43.
G.
Spiekermann
,
M.
Steele-MacInnis
,
C.
Schmidt
, and
Sandro
Jahn
,
J. Chem. Phys.
136
,
154501
(
2012
).
44.
J. F.
Ziegler
,
J. P.
Biersack
, and
U.
Littmark
,
The Stopping and Range of Ions in Solids, vol. 1 of Series Stopping and Ranges of Ions in Matter
(
Pergamon
,
New York
,
1984
).
45.
H.
Menhofer
and
H.
Heusinger
,
Int. J. Radiat. Appl. Instr.: Part C. Radiat. Phys. Chem.
29
,
243
(
1987
).
46.
A.
Kondyurin
,
P.
Naseri
,
K.
Fisher
,
D. R.
McKenzie
, and
M. M. M.
Bilek
,
Polym. Degrad. Stab.
94
,
638
(
2009
).
47.
G.
Mesyats
,
Y.
Klyachkin
,
N.
Gavrilov
, and
A.
Kondyurin
,
Vacuum
52
,
285
(
1999
).
48.
E. A.
Kosobrodova
,
A. V.
Kondyurin
,
K.
Fisher
,
W.
Moeller
,
D. R.
McKenzie
, and
M. M. M.
Bilek
,
Nucl. Instrum. Methods Phys. Res., Sect. B
280
,
26
(
2012
).
49.
E.
Kosobrodova
,
A.
Kondyurin
,
D. R.
McKenzie
, and
M. M. M.
Bilek
,
Nucl. Instrum. Methods Phys. Res., Sect. B
304
,
57
(
2013
).
50.
B.
Ranby
and
J. F.
Rabek
,
Photodegradation, Photo-Oxidation and Photostabilization of Polymers
(
Wiley
,
New York
,
1975
).
51.
Z.
Yongheng
and
Gu
Zhenan
,
J. Non-Cryst. Solids
352
,
4030
(
2006
).
52.
P. J.
Launer
and
B.
Arkles
, “
Infrared analysis of organosilicon compounds
,” in
Silicon Compounds: Silanes and Silicones
(
Gelest Inc.
,
Morrisville
,
2013
).
53.
N.
Hu
,
Y. Q.
Rao
,
S.
Sun
,
L.
Hou
,
P.
Wu
,
S.
Fan
, and
B.
Ye
,
Appl. Spectrosc.
70
,
1328
(
2016
).
54.
D. J.
Rosenberg
,
S.
Alayoglu
,
R.
Kostecki
, and
M.
Ahmed
,
Nanoscale Adv.
1
,
4878
(
2019
).
55.
P.
Innocenzi
,
J. Non-Cryst. Solids
316
,
309
(
2003
).
56.
B. J. G.
de Aragão
and
Y.
Messaddeq
,
J. Braz. Chem. Soc.
19
,
1582
(
2008
).
57.
I. F.
Husein
,
C.
Chan
,
S.
Qin
, and
P. K.
Chu
,
J. Phys. D: Appl. Phys.
33
,
2869
(
2000
).
58.
A.
Toth
,
I.
Bertoti
,
G.
Marletta
,
G. G.
Ferenczy
, and
M.
Mohai
,
Nucl. Instrum. Methods Phys. Res., Sect. B
116
,
299
(
1996
).
59.
G.
Marletta
,
A.
Toth
,
I.
Bertoti
,
T. M.
Duc
,
F.
Sommer
, and
K.
Ferencz
,
Nucl. Instrum. Methods Phys. Res., Sect. B
141
,
684
(
1998
).
60.
S.
Bhattacharya
,
Y.
Gao
,
Ve.
Korampally
,
M. T.
Othman
,
S. A.
Grant
,
K.
Gangopadhyay
,
S.
Gangopadhyay
,
Appl. Surf. Sci.
253
,
4220
(
2007
).
61.
C.
Satriano
,
E.
Conte
, and
G.
Marletta
,
Langmuir
17
,
2243
(
2001
).
62.
R.
Huszank
,
D.
Szikra
,
A.
Simon
,
S. Z.
Szilasi
, and
I. P.
Nagy
,
Langmuir
27
,
3842
(
2011
).
63.
A.
Groza
and
A.
Surmeian
,
J. Nanomater.
2015,
204296
(
2015
).
64.
G. A.
Parks
,
J. Geophys. Res.
89
,
3997
(
1984
).
65.
B. M. B.
Sansao
,
J. J.
Kellar
,
W. M.
Cross
,
K.
Schottler
, and
A.
Romkes
,
Powder Technol.
384
,
267
(
2021
).
66.
D.
Bodas
,
J.-Y.
Rauch
, and
C.
Khan-Malek
,
Euro. Poly. J.
44
,
2130
(
2008
).
67.
K.
Sondhi
,
S.
Hwangbo
,
Y.-K.
Yoon
,
T.
Nishida
, and
Z. H.
Fan
,
J. Micromech. Microeng.
28
,
125014
(
2018
).
68.
T.
Trantidou
,
Y.
Elani
,
E.
Parsons
, and
O.
Ces
,
Microsyst. Nanoeng.
3
,
16091
(
2017
).
69.
H.
Yasuda
,
Plasma Polymerization
(
Academic
,
New York
,
1985
).
70.
H.
Steinhauser
and
G.
Ellinghorst
,
Angew. Makromol. Chem.
120
,
177
(
1984
).
71.
V. I.
Povstugar
,
V. I.
Kodolov
, and
S. S.
Mikhailova
,
Composition and Properties of Surface of Polymer Materials
(
Khimiya
,
Moscow
,
1988
).
72.
N. D.
Lang
and
W.
Kohn
,
Phys. Rev. B
12
,
4555
(
1970
).
Published open access through an agreement with The University of Sydney