Biocompatibility of Antifogging SiO-doped Diamond-Like Carbon Laparoscope Coatings

Abstract

Laparoscopes can suffer from fogging and contamination difficulties, resulting in a reduced field of view during surgery. A series of diamond-like carbon films, doped with SiO, were produced by pulsed laser deposition for evaluation as biocompatible, antifogging coatings. DLC films doped with SiO demonstrated hydrophilic properties with water contact angles under 40°. Samples subjected to plasma cleaning had improved contact angle results, with values under 5°. Doping the DLC films with SiO led to an average 40% decrease in modulus and 60% decrease in hardness. Hardness of the doped films, 12.0 – 13.2 GPa, was greater than that of the uncoated fused silica substrate, 9.2 GPa. The biocompatibility was assessed through CellTiter-Glo assays, with the films demonstrating statistically similar levels of cell viability when compared to the control media. The absence of ATP released by blood platelets in contact with the DLC coatings suggests in vivo hemocompatibility. The SiO doped films displayed improved transparency levels in comparison to undoped films, achieving up to an average of 80% transmission over the visible spectrum and an attenuation coefficient of 1.1 × 10⁴ cm⁻¹ at the 450 nm wavelength. The SiO doped DLC films show promise as a method of fog prevention for laparoscopes.

1. Introduction

Since the introduction of camera guided laparoscopy, the number of laparoscopic procedures performed has increased each year. Laparoscopic procedures have been found to result in increased patient comfort and decreased mortality rates when compared to open surgical procedures [1, 2]. One drawback to these procedures is the susceptibility of the lens to fogging, which results in a reduced surgical field [3]. Intraoperative cleaning, which may require repeated removal and reinsertion of the laparoscopic lens, leads to longer surgery times and greater risk of surgical site infection [4-6].

Several methods for reducing lens fogging exist, although none are completely effective and many have substantial drawbacks. A common technique involves the use of a surfactant to reduce the surface tension of the liquids interacting with the lens [4-7]; these surfactants often need reapplication during the procedure requiring removal of the laparoscope from the body, increasing the risk of contamination. Insufflation, which uses CO₂ gas to inflate the abdomen during the procedure, and water baths have been investigated for the reduction of fogging, with limited data to support their effectiveness [7, 8]. Other methods require extensive modification to the lens, such as the addition of a heating mechanism to the end of the laparoscope, which can be costly and still leave the lens susceptible to fogging [7-9]. In recent years, there has been an interest in the development of new materials to solve the fogging issue [10, 11].

This manuscript presents a continuation of research initiated by Leonard et al., which investigated the use silicon monoxide (SiO) doping to create antifogging DLC films by pulsed laser deposition (PLD) [12, 13]. In this work, the amount of SiO dopant is increased beyond that of the original study to determine if contact angle may be further reduced. Importantly, the long-term effectiveness of antifogging DLC films, both as-made and after plasma etching is determined. In addition, the biocompatibility of the films is explored using CellTiter-Glo assays and the measurement of adenosine triphosphate (ATP) release upon exposure of the samples to platelet rich plasma (PRP). For this study, the films were deposited on fused silica, which is a durable optical material, used for laparoscope applications [14].

DLC optical components are intended for future use in laparoscopes. These tools, used during minimally invasive surgeries, are in contact with blood. Hemocompatibility is an important translational consideration for materials intended for contact with blood. Guidance for hemocompatibility testing is provided by the International Organization of Standards (ISO 10993-4). The duration of blood contact varies among the range of procedures in which laparoscopes are used, but 60 minutes is a reasonable estimate averaged over the most common procedures. Among the range of possible hemocompatibility assessments, platelet activation, activation and subsequent thrombosis arguably pose the most significant risks during the relatively short duration exposure to blood outside the vascular flow [15].

The coatings may also find use with other related medical devices, such as laryngoscopes, bronchoscopes, cystoscopes, and endoscopes, as well as consumer products such as eyewear and camera lenses.

2. Experimental Details

DLC films were deposited on UV grade Corning 7980 fused silica substrates, with thicknesses of 500 μm. Each sample deposition coated three 25 × 25 mm substrates and one 38 × 38 mm substrate. Physical and optical characterizations were performed on the 25 × 25 mm substrates while the 38 × 38 mm piece was set aside for biocompatibility testing. Before deposition, the substrates were ultrasonically cleaned in acetone and methanol for ten minutes each. They were rinsed with ultrapure water and dried with compressed nitrogen following sonication, then placed in a 1:1 (v:v) piranha solution (H₂SO₄ (99.99%) and H₂O₂ (30%)) soak for 2 minutes. The final cleaning steps were a soak in ultrapure water for 1 minute plus rinse and drying with compressed nitrogen before being affixed to the sample holder.

The thin film samples were synthesized through pulsed laser deposition (PLD) with a 193 nm ArF excimer laser. The laser pulse repetition rate was 100 Hz at a laser energy of 5.0 mJ resulting in a laser fluence of 4.1 J/cm² based on a 0.06 mm² spot size and 49% beam attenuation by the chamber window and interior shield. The target-to-substrate distance was 66 mm. For all samples, the deposition chamber pressure remained less than 4.0 × 10⁻⁴ Pa. A multicomponent PLD target was made consisting of semiconductor-grade graphite (Poco Graphite, Inc.) and SiO (Kurt J. Lesker Co.). Axial target movement is controlled by a programmable stepper motor, so that the position can be manipulated to alter the portion of the multicomponent target interacting with the laser.

Six DLC samples were synthesized with differing amounts of SiO dopant. The samples were synthesized with 400,000 total laser pulses with the amount of laser pulses on the dopant ranging from 0 – 30%. Table 1 details the number of laser pulses on each material for the respective sample. A separate sample set was made with the same process parameters as the 30% SiO doped sample on four, 25 × 25 mm substrates for additional optical characterization. Samples were stored in ambient air when not being characterized.

Table 1.

Doped and undoped DLC deposition parameters.

Sample Name	Pulses on SiO Dopant (%)	Pulses on Carbon	Pulses on SiO Dopant
ABB025	None	400000	N/A
ABB007	10	360000	40000
ABB008	15	340000	60000
ABB009	20	320000	80000
ABB010	25	300000	100000
ABB012	30	280000	120000

Tape tests were performed to assess film adhesion to the substrate following the methods dictated by ASTM D3359. These tests were performed with Adhesive ASTM D3359 Cross-Hatch Adhesion Test Tape (Elcometer 99). The samples and tape were examined visually for any signs of delamination.

The effects of differing SiO dopant amounts on the optical transmission of the samples were determined by spectrophotometry. The transmission spectra were measured from 200 nm to 900 nm at 1.0 nm increments with a GenTech Scientific TU-1901 spectrophotometer. The data was collected with a PC equipped with UVWin5.0 software. An uncoated fused silica substrate was used as a reference during each measurement. Witness marks present on the deposited substrates ensured the position of the measured area was consistent between samples.

Contact angle of the films was measured with a Krüss DSA20E Easy Drop Standard equipped with a software-controlled system of dosing. The probe liquids tested were ultrapure water, with a resistivity of 12.6 MΩ·cm or greater, and benzyl alcohol (Sigma-Aldrich) both with a drop volume of 2 μL per measurement. Five measurements were performed on each sample with both probe liquids. The water contact angle was assessed on as-made samples and samples subject to plasma cleaning. Benzyl alcohol contact angles were measured on the as-made samples and used for surface energy calculations with the Fowkes method [16]. The Fowkes method calculates surface energy, using contact angle results from two probe liquids, as a sum of the dispersive and polar components as shown below in Equation 1 [17].

$${\gamma }_{\mathrm{s}}={\gamma }_{\mathrm{s}}^{\mathrm{d}}+{\gamma }_{\mathrm{s}}^{\mathrm{p}}$$ (1)

The thicknesses of the films were determined with a J.A. Woollam M-2000VI ellipsometer. The data was collected with a Windows XP computer equipped with CompleteEASE software. The location for the reported thickness is in the center of the sample; the sample thickness was nonuniform and varied as a function of radial distance from the center of the ablation plume.

Experiments in simulated body fluid (SBF) were conducted to assess the biostability of the films. The sample deposited on the 38 × 38 mm substrate was cut into twelve 8.45 × 8.45 mm pieces to ensure they were properly sized for a close fit in a welled cell culture plate and to remove the masked portions of the sample. This allowed for multiple tests to be completed and for uniformity of the exposed film/substrate interface for all specimens. These pieces were then cleaned in accordance with their method of testing. For SBF soaks, the samples were cleaned by sonication in ultrapure water and then methanol. The SBF was made following the methods established by Cho et al., 1995 [18]. The soaks were performed and kept at body temperature of 37° C in a Lauda Ecocline E100 Immersion Thermostat water bath and assessed every 4 weeks for 48 weeks via an optical microscope. For SBF soaks, each of the doped films were scored before insertion. This allowed the assessment of any changes in the interface between the DLC and fused silica during soaking.

CellTiter-Glo assays were performed to assess the effects of DLC films on cell viability. The samples were cleaned in ultrapure water by sonication. NIH/3T3 cells (ATCC stock no: CRL-1658) were used for evaluation of cell viability. NIH/3T3 cells were cultured according to ATCC recommendations in Dulbecco’s Modified Eagle’s Medium grown with 10% iron supplemented bovine calf serum and penicillin/streptomycin. All cultures were grown in T-75 flasks (ThermoFisher Scientific) at 80% confluence or less. NIH/3T3 cells were seeded in standard 24-well plates at 25,000 cells/well in 1mL media containing DLC films and cultured for 24 hrs. The films were sterilized with 70% ethanol after fabrication, washed with 1x phosphate buffered saline (PBS) to remove residual ethanol, and positioned coated-side-up in 24-well plates before seeding with NIH/3T3 cells. Samples were maintained under sterile cell culture conditions for 24 hrs. A modified commercial CellTiter-Glo (Promega) assay was carried out to evaluate cell viability at the end of the culture period and the resulting luminescence was assessed via IVIS. At the conclusion of the 24-hour incubation period, 24-well plates were allowed to equilibrate to room temperature for 30 minutes. Before being mixed with an orbital shaker for 2 minutes, 600 μL of media was removed from each well and 400 μL of CellTiter-Glo reagent was then added directly to the media in each well. Once mixing was complete, the 24-well plates were allowed to rest for 10 minutes before 100 μL samples of the resulting solution was transferred to black-walled 96-well plates for analysis in order to eliminate any interference caused by the DLC films during IVIS imaging. The average viability of NIH/3T3 cells grown on DLC films was normalized to media controls. An assay of 1% T-x-100 treated NIH/3T3 cells served as a control to simulate complete cytotoxicity. Note that the testing of the 100% carbon, the 15% SiO doped, and 20% SiO doped samples were performed on a different day than the 10%, 25% and 30% SiO doped samples.

Blood compatibility was assessed by measuring the ATP release from platelet rich plasma (PRP) following contact with the samples. Uncoated fused silica substrates were used as control samples. Each measurement was performed using blood from a single donor; the study used three blood donations. Each sample was placed in one well of a 24 well cell culture plate (Kemtech 4422A) with the DLC coating facing up. The amount of 0.75 mL of PRP was added to each well to ensure that the sample was completely submerged. The plate was incubated for 60 min at 37 °C. ATP release was measured in a Chrono-Log Model 700 Whole Blood/Optical Lumi-Aggregometer. Positive controls were carried out by the addition of γ-thrombin, a platelet aggregating agent that also induces ATP release. Maximum available ATP was estimated from PRP exposed to ultrasound energy (Cole-Palmer EW-59989-29) for 60 s at room temperature to lyse the platelets and their ATP-containing dense granules. ATP release from each sample was estimated as per the manufacturer as (test luminescence intensity / test luminescence gain) × (ATP standard gain / ATP standard luminescence intensity) × 2 nmol.

The effects of plasma cleaning on the hydrophilicity of the films was investigated with a Harrick Plasma PDC-32G benchtop plasma cleaner with an argon processing gas. The films were subjected to 3 minutes of treatment at medium RF power before being removed for contact angle testing. The contact angles of two 30% SiO doped samples were tested, and each sample subjected to plasma cleaning was measured within 60 minutes of being treated.

A study was conducted to determine the stability of the samples’ contact angle, both as made and after plasma cleaning. The contact angle for each sample was measured in 24-hour intervals over the course of 14 days; additional measurements, at greater frequency, were taken at the early stage of testing. The films were stored in air for the duration of the time study.

Atomic force microscopy (AFM) was performed to determine the film morphology and roughness on both as-made films and a plasma cleaned film. The plasma cleaned film was characterized within 12 hours of being treated. The measurements were performed with a Bruker Dimension Icon Atomic Force Microscope connected to a computer equipped with ScanAsyst software in contact mode. A 2 μm × 2 μm sample area with a scanning speed of 10 μm/s at a resolution of 256 × 256 pixels was collected. The data were analyzed on a separate computer with NanoScope Analysis V1.2.0 software installed.

Raman Spectroscopy was performed with a Thermo Scientific DXR Raman Microscope, and the data was collected through Omnic Spectra Software. The laser wavelength was 532 nm and was focused through an X50 objective lens at a power of 10 mW. All acquisitions were for 10 seconds with 20 accumulations taken at each measured location. Gaussian deconvolution of the peaks was performed using Origin software. A Hysitron TI 980 Triboindenter with a diamond Berkovich tip was used to perform nanoindentation on the films. A second-order area function and a fused silica standard with known mechanical properties were used to calibrate for the tip area [19]. Nanoindentation was performed using a line collecting technique in which 10 indents with a spacing of 20 μm were collected on each sample. A trapezoidal loading pattern consisting of a 5 second loading period, 2 second hold period, and 5 second unloading period was utilized for each indent. A maximum force of 500 μN was applied during the loading period such that the indentation depths were at approximately 50-60% of the film thickness. The reduced elastic modulus (i.e. indentation modulus) and hardness of each indent were then calculated using the Oliver and Pharr method [20]. Because the undoped sample produced for this study, using 400,000 laser pulses, was considerably thinner than the other samples, a special undoped sample using 2.0 × 10⁶ laser pulses with a measured thickness of 80 ± 1 nm, was made specifically for nanoindentation testing to achieve more comparable results. Care was also taken to locate the indents on each sample at approximately the same distance from what would have been the center of the sample holder during film deposition to help address variability in the film thickness as a function of spatial location during deposition.

3. Results & Discussion

3.1. Visual Inspection

Representative samples of the as-made films are shown in Figure 1. The films are transparent and show no signs of delamination or other defects as seen from visual inspection. No delamination of the films was observed after performing tape tests, demonstrating the effective adhesion of the films to the substrate. The 100% carbon sample was noticeably darker than the SiO doped samples. However, the samples become increasingly darker as the amount of SiO dopant is increased.

Figure 1.

Undoped and SiO doped DLC samples. Note: Witness marks associated with the mounting hardware used during deposition can be seen in the upper left and bottom right of the samples. For each sample, identifying text can be seen through the film and substrate.

3.2. Film Thickness and Transparency

The effects of SiO dopant amount on film transparency can be seen in Figure 2. The shape of the transmission spectrum for the undoped DLC film is markedly different from the rest of the samples; it has increased transmission at the lower energy region of the spectrum and decreased transmission in the higher energy (blue and ultraviolet) regions. In general, the transparency of the doped films decreased with increasing SiO content. Through the visible spectrum, the 10% SiO doped sample had the highest measured transmission. The initial increase in the film transparency is likely due to the reduced amount of carbon in the film. Silicon has been shown to increase the sp³ hybridization of DLC films, making them more transparent, which follows the observed trend of the experimental series [21]. The hypothesis is the doped films get darker with increasing SiO content due to the incorporation of SiO particles into the matrix; SiO is a dark brown, opaque material. The effects of film thickness were determined through calculating the attenuation coefficient of the films at a wavelength of 450 nm (blue light). These results, shown in Figure 3, were calculated by following the methods by Manjunatha & Paul using the Beer-Lambert Law [22]:

$$\mathrm{I}={\mathrm{I}}_0{\mathrm{e}}^{-\alpha \mathrm{t}}$$ (2)

Figure 2.

Transmission spectra of DLC films composed of varying amounts of SiO dopant: (a) 100% carbon, (b) 10% dopant, (c) 15% dopant, (d) 20% dopant, (e) 25% dopant, and (f) 30% dopant.

Figure 3.

Attenuation coefficient of the DLC films calculated at a wavelength of 450 nm.

Where I is the transmitted intensity of light, ${\mathrm{I}}_0$ is the incident intensity, α is the attenuation coefficient, and t is the thickness of the material.

From these calculations it was determined that increasing the amount of dopant in the films increased the attenuation coefficient values, from 1.1 × 10⁴ cm⁻¹ for the 10% SiO-doped sample to 2.5 × 10⁴ cm⁻¹ for the 30% SiO-doped sample. However, the values for the doped samples are much smaller than for the undoped sample, which has an attenuation coefficient of 1.0 × 10⁵ cm⁻¹, demonstrating a marked improvement in optical transparency with the incorporation of SiO.

3.3. Water Contact Angle and Surface Energy

The water contact angle results from the as-made films are shown in Figure 4. It was observed that the addition of SiO as a dopant significantly lowered the contact angle, which is in agreement with Leonard et al [13]. The quantity of SiO was also observed to affect the contact angle with the 30% SiO doped sample having a contact angle almost 20° lower than the 10% SiO doped sample. The influence of SiO content on the contact angle decreases between 15% and 30% SiO as there was only a 4° difference between these samples. This suggests that the upper limit of SiO dopants’ ability to increase hydrophilicity is in the 15 - 30% range.

Figure 4.

The effect of silicon monoxide dopant on average water contact angle. Note: Error bars are present but difficult to observe for each sample.

Increased surface energies are known to correspond to decreased contact angles and the effects of increasing SiO dopant on surface energy are shown in Figure 5. The highest measured surface energy was just over 80 mN/m for the 30% SiO doped sample while the undoped sample had a surface energy of 41 mN/m. The difference in surface energy values of the undoped and doped samples demonstrates that the SiO dopant improved the hydrophilicity of DLC, which agrees with the results collected during contact angle measurements. Although the hydrophilicity of the films increased with increasing SiO content, the transparency of the films decreased demonstrating a trade-off between transparency and hydrophilicity. The 15% SiO doped sample could be the most effective composition for antifog coating applications as it provided a balance of decreased contact angle with limited loss in transparency.

Figure 5.

The effect of SiO dopant on average surface energy. Note: Error bars are present but difficult to observe for each sample.

The SiO doped films were significantly thicker than the undoped film, as shown in Table 2. The difference is likely due to the variation in ablation rate between the two target materials, graphite and SiO, with the ablation rate of SiO being much greater.

Table 2.

Thicknesses of the SiO Doped Films as a Function of Laser Pulses Measured at the Center of the Sample.

Sample Name	Pulses on Dopant (%)	Thickness (nm)
ABB025	None	29 ± 2
ABB007	10	112 ± 2
ABB008	15	115 ± 2
ABB009	20	101 ± 1
ABB010	25	111 ± 1
ABB012	30	108 ± 1

The difference in contact angle between an as-made sample and a plasma cleaned sample can be observed in Figure 6. Immediately following plasma cleaning, an angle of less than 5° was observed for the sample (Figure 6(a)), which is significantly less than the original 33.3° measured on the untreated films (Figure 6(b)). These results show that plasma cleaning may be an effective way to enhance the antifogging properties of the material and reverse any increase in contact angle that occurs over time due to surface contamination. Furthermore, plasma cleaning would have the additional benefit of sterilizing the instrument, as required before surgical procedures [23].

Figure 6.

Profile of water drops on 30% SiO doped samples: (a) an as-made film and (b) 60 minutes after treatment. Note: The baseline for each measurement is indicated with a yellow, dashed line.

Time study results for an as-made and a plasma cleaned film are shown in Figure 7. Conducting contact angle measurements over 14 days allowed for observation of the longevity of the plasma cleaning treatment effect on the films. The plasma cleaned films began at an average angle of 5° that was sustained for 24 hours. After two days, the contact angle measurements of the films increased to over 20° and, for the duration of the study, slowly elevated to approximately 30° by the fourteenth day of observation. This is a significant improvement on the as-made sample, which maintained an average contact angle of 33° for the duration of the time study. Two weeks after the time-studies completion, the contact angle for the treated film was measured and it had returned to approximately 33°. The decrease and subsequent increase in contact angle may be caused by the formation of hydroxyl free radicals in the film surface as detailed by Subedi et al, where they observed the same trend of increasing contact angle after a few hours of treatment [24]. Alternatively, the increase in contact angle of the plasma cleaned films could also be due to hydrocarbons and other contaminants attaching to the film surface while in ambient air [25]. Although the temporal nature of the hydrophilicity is not ideal, 24 hours of antifog protection should be sufficient for most surgeries. Furthermore, the surface can be restored through subsequent plasma treatment, which has the additional benefit of sterilizing the instrument [23].

Figure 7.

Results of average contact angle measurements of plasma cleaned vs. as-made films over 14-days: (a) as-made and (b) plasma cleaned. Note: Error bars are present but are not always visible for each measurement.

3.4. Raman Spectroscopy

The Raman spectra of the undoped and doped films can be observed in Figure 8. The sp²/sp³ ratio of DLC dictates the overall properties of the films [26, 27]. The D peak, resulting from carbon-carbon bond stretching and contractions, is at ~1360 cm⁻¹ and the G peak, caused by in plane stretching of carbon-carbon bonds, is at ~1580 cm⁻¹ [27]. The ratio of the intensity of the D and G peak (${\mathrm{I}}_{\mathrm{D}}∕{\mathrm{I}}_{\mathrm{G}}$) can give insight into the size and number of sp² clusters within the DLC films [27]. The G peak characteristics of these films are shown in Table 3. The shift in G peak in the doped samples, along with the increased ${\mathrm{I}}_{\mathrm{D}}∕{\mathrm{I}}_{\mathrm{G}}$ ratio, suggest a greater sp² content in these films. This generalization does not hold true for the 10% SiO doped sample which demonstrated a much smaller ${\mathrm{I}}_{\mathrm{D}}∕{\mathrm{I}}_{\mathrm{G}}$ ratio than the other doped samples. The decrease could indicate a reduction in aromatic cluster formation or an increase in sp³ clusters. The peaks at ~1060 cm⁻¹ and 1200 cm⁻¹ can be attributed to the fused silica substrate [28, 29].

Figure 8.

Raman spectra of DLC films with varying amounts of laser pulses on SiO dopant: (a) 100% carbon, (b) 10% dopant, (c) 15% dopant, (d) 20% dopant, (e) 25% dopant, and (f) 30% dopant. Note: This data has been normalized.

Table 3.

Gaussian fitting of Raman spectra of undoped and SiO Doped DLC films.

Sample Name	Pulses on Dopant (%)	G Peak Position	${\mathrm{I}}_{\mathrm{D}}∕{\mathrm{I}}_{\mathrm{G}}$	Full Width Half Maximum G Peak
ABB025	None	1548 ± 1	0.4 ± 0.02	266 ± 3
ABB007	10	1471 ± 6	0.3 ± 0.01	194 ± 1
ABB008	15	1473 ± 2	0.5 ± 0.02	245 ± 2
ABB009	20	1467 ± 2	0.6 ± 0.20	297 ± 2
ABB010	25	1452 ± 3	0.8 ± 0.01	303 ± 1
ABB012	30	1477 ± 5	0.5 ± 0.03	239 ± 3

3.5. Surface Roughness and Morphology

The morphology and roughness of a plasma cleaned film, as-made films, and an uncoated substrate were investigated by AFM. The AFM results and root mean square (RMS) roughness are demonstrated in Figures 9 and 10. The AFM results for the undoped sample (Figure 9(B)) demonstrated a much smoother film than the as-made SiO doped film (Figure 9(C)). This increase in roughness can be attributed to the formation of particulates of SiO during the ablation process. There are repeated bands visible in the uncoated fused silica substrate (Figure 9(A)) resulting from its fabrication; these also can be seen in the undoped and SiO doped samples. The bands likely affect the roughness of the undoped sample but not as much on the SiO doped sample because of the silica particulates. The plasma cleaned sample (Figure 9(D)) had a different topography than the as-made sample. This result could mean the plasma cleaning process not only removed contaminants, but also caused an etching effect on the surface resulting in a more textured film.

Figure 9.

Representative AFM images of DLC with varied dopant amounts and an uncoated substrate: (a) uncoated substrate; (b) 100% carbon; (c) 30% SiO dopant; (d) plasma cleaned 30% SiO dopant.

Figure 10.

RMS roughness results for an uncoated substrate, DLC with varied SiO dopant amounts, and a plasma cleaned film.

3.6. Mechanical Properties

The addition of SiO as a dopant resulted in a decrease in modulus and hardness compared to the undoped film (see Figures 11 and 12), which was consistent with the trends reported in the literature [30, 31]. The decrease was most pronounced for the transition from the undoped film to the 10% SiO film, a reduction from 113.3 ± 4.7 GPa to 75.6 ± 4.4 GPa for the modulus and 30.6 ± 2.9 GPa to 13.2 ± 0.6 GPa for the hardness, respectively. With 15% SiO, the modulus further decreased (63.2 ± 2.3 GPa) but had only a minimal additional effect on the hardness (12.4 ± 0.8 GPa). The 20-30% SiO samples had similar modulus and hardness behavior when compared to the 10-15% SiO samples, suggesting that the mechanical properties were not heavily influenced by the changing SiO doping levels. It should also be noted that even though the addition of SiO caused a decrease in hardness compared to the undoped film, the hardness values of all the doped films were still above the measured value for the uncoated fused silica substrate, 9.2 ± 0.8 GPa. This result suggests that the SiO-doped films offer an improvement in resistance to plastic deformation over that of the uncoated material. Furthermore, all sample containing SiO provided a relatively consistent decrease in the modulus, which may be useful for applications where the high stiffness of an undoped DLC film causes issues with elastic deformation.

Figure 11.

Modulus as a function of pulses on SiO dopant.

Figure 12.

Hardness as a function of pulses on SiO dopant.

3.7. Stability and Biocompatibility

3.7.1. Simulated Body Fluid Studies

A scribed 15% SiO doped sample is shown in Figure 13. In Figure 13(A) there is no delamination present before insertion. After 48 weeks of soaking, as shown in Figure 13(B), there is still no delamination or significant changes in the film. The lack of changes after soaking helps verify the stability of the film.

Figure 13.

Photographs of a 15% SiO doped DLC sample: (A) before soaking and B) after soaking in SBF for 48 weeks.

3.7.2. Assessment of Biocompatibility using CellTiter-Glo Assays

The results for acute toxicity of DLC films evaluated ex vivo 24 hours after incubation of NIH/3T3 cells with DLC films made with various SiO dopant levels are shown in Figure 14. Compared to the media, the DLC films had statistically similar levels of cell viability observed among all SiO dopant concentrations. These results help demonstrate the biocompatibility of DLC films.

Figure 14.

Average viability normalized to media controls of NIH 3T3 cells incubated for 24 hours with various SiO doped DLC films (n = 5 technical replicates). The significance of the data was evaluated via ordinary one-way ANOVA with Dunnett's Multiple Comparison Test (*p<0.05).

3.7.3. Blood Compatibility Testing

ATP released by blood platelets in contact with any of the 6 DLC coatings and the uncoated substrate was not significantly different from zero. Total ATP in the PRP samples was estimated to be 2.60 ± 1.92 nmols (n = 3, standard deviation). Using the luminescent gain settings most appropriate for this maximum ATP measurement, we estimate the limit of instrument sensitivity to be 0.01 nmols. Using this criterion, PRP exposed to any of the 6 DLC coatings or uncoated substrate yielded average ATP values that were not statistically different than the limit of instrument sensitivity that is equivalent to 0.4% of the maximum ATP release.

PRP samples treated to 100 nM γ-thrombin demonstrated moderate, bi-phasic aggregation and released 1.23 ± 0.21 nmol of ATP (Figure 15). Platelet aggregation is characterized in this system as the decrease in turbidity during the transition from many individual platelets (high turbidity) to fewer aggregates of larger size (low turbidity). Release of ATP was dependent on γ-thrombin dose with 300 nM of γ-thrombin inducing the release of 1.70 ± 0.47 nmols of ATP and associated with strong platelet aggregation.

Figure 15.

Low dose, 100 nM γ-thrombin (orange) induces biphasic, but extensive, platelet aggregation (a) and delayed, but considerable, ATP release (b). High dose, 300 nM γ-thrombin induces rapid, monophasic, extensive platelet aggregation (c) and early, substantial ATP release (d). These controls confirm the potential for platelet aggregation and ATP release that is absent in PRP samples exposed to DLC coated samples or DLC control samples.

DLC coatings failed to elicit detectable ATP release from platelets, suggesting that the use of these coatings is unlikely to trigger coagulation-associated adverse effects during 60 minutes of exposure to blood.

Despite some variability in the repeated measures that are presumed to result from biological differences among subjects, each PRP sample was capable of significant and dose-dependent aggregation and ATP release in response to γ-thrombin (see Figure 15). These aggregation and ATP release results are in agreement with similarly treated controls reported previously [32]. Thus, the γ-thrombin positive control results confirm the potential for platelet activation and ATP release due to DLC exposure. Consequently, the lack of meaningful platelet ATP release elicited by DLC sample contact strongly supports the hypothesis that DLC coatings are unlikely to trigger coagulation-related adverse events during proposed future use as components in laparoscopes. In addition, the PRP volume per DLC surface area in this study increases the contact time of any single platelet with the sample relative to the intended use in laparoscopic instruments that will be moved during use and, generally, in contact with a larger blood volume. Thus, the relatively small volume per surface area of the test increases platelet exposure to the DLC sample, amplifying the potential for platelet activation and ATP release compared to the conditions anticipated during laparoscope use. The lack of evidence for ATP release under the rigorous ex vivo conditions used in this study provides enhanced confidence in the probable platelet hemocompatibility of these DLC surfaces as laparoscopic windows in vivo.

4. Conclusion

Diamond-like carbon thin films doped with varying amounts of SiO were synthesized via pulsed laser deposition on fused silica substrates. The stability of the films was maintained while soaked in simulated body fluid stored at 37° C for 48 weeks. Plasma cleaning reduced surface roughness and decreased contact angle to 5° for over 24 hours in the doped films. Doping the DLC films led to an average 40% decrease in modulus and 60% decrease in hardness. However, the SiO dopant loading had little influence on the magnitude of the decrease, suggesting that SiO can be used across the 10-30% dopant loading range with minimal effect on the mechanical properties, and optimization of other properties that are more sensitive to the dopant loading should be possible. The SiO doped films also had average hardness values, 12.0 – 13.2 GPa, that were above that of the uncoated fused silica substrate, 9.2 GPa, indicating that the SiO doped films do provide some measure of resistance against plastic deformation. The SiO doped films also had lower average modulus values (63.2 – 75.6 GPa) than the undoped film (113.3 GPa), which may prove beneficial in applications where elasticity is a critical parameter for material performance. In vitro cell viability assays found all the films demonstrated similar biocompatibility results to the media control with no loss in cell viability. ATP released by blood platelets in contact with the DLC coatings is very small, not greater than the limit of instrument detectability, and suggests in vivo hemocompatibility. Spectrophotometry showed that the incorporation of SiO improved the transparency of the films, achieving a 1.1 × 10⁴ cm⁻¹ attenuation coefficient for the 10% SiO-doped sample at a wavelength of 450 nm, compared to an order of magnitude greater value of 1.0 × 10⁵ cm⁻¹ for the undoped sample. Because transparency decreased with increased doping level and there is little change in water contact angle beyond 15% SiO doping, a value of 15-20% is likely the optimum level of doping. The findings of this research study demonstrate the possibility of using SiO doped DLC as an antifogging coating for a laparoscopic lens.

Table 4.

Doped and undoped DLC deposition parameters.

Sample Name	Pulses on SiO Dopant (%)	Pulses on Carbon	Pulses on SiO Dopant
ABB025	None	400000	N/A
ABB007	10	360000	40000
ABB008	15	340000	60000
ABB009	20	320000	80000
ABB010	25	300000	100000
ABB012	30	280000	120000

Table 5.

Thicknesses of the SiO Doped Films as a Function of Laser Pulses.

Sample Name	Pulses on Dopant (%)	Thickness (nm)
ABB025	None	29 ± 2
ABB007	10	112 ± 2
ABB008	15	115 ± 2
ABB009	20	101 ± 1
ABB010	25	111 ± 1
ABB012	30	108 ± 1

Table 6.

Gaussian fitting of Raman spectra of undoped and SiO Doped DLC films.

Sample Name	Pulses on Dopant (%)	G Peak Position	${\mathrm{I}}_{\mathrm{D}}∕{\mathrm{I}}_{\mathrm{G}}$	Full Width Half Maximum G Peak
ABB025	None	1548 ± 1	0.4 ± 0.02	266 ± 3
ABB007	10	1471 ± 6	0.3 ± 0.01	194 ± 1
ABB008	15	1473 ± 2	0.5 ± 0.02	245 ± 2
ABB009	20	1467 ± 2	0.6 ± 0.20	297 ± 2
ABB010	25	1452 ± 3	0.8 ± 0.01	303 ± 1
ABB012	30	1477 ± 5	0.5 ± 0.03	239 ± 3

Highlights:

• The incorporation of silicon and oxygen into the carbon matrix results in a hydrophilic film.

• Argon etching decreases the water contact angle of the films to less than 5°.

• The films were assessed using CellTiter-Glo assays and demonstrated statistically similar levels of cell viability when compared to the control media.

• The absence of ATP released by blood platelets in contact with the DLC coatings suggests in vivo hemocompatibility.