Amorphous hydrogenated carbon films with diamond-like properties have attracted considerable interest. Specifically, diamond-like carbon (DLC) films have attractive properties similar to diamonds. The most popular properties of DLC films are their ability to prevent corrosion against chemicals (acids and alkaline solutions). Additionally, DLC films provide low friction, high thermal conductivity, high resistance, and extreme hardness. 1 The combination of these properties allows for many applications of DLC film coatings. For example, DLC film coatings are used where wear resistant surfaces, passivation, and corrosion protection are desirable. 2,3 Furthermore, a large number of techniques have been reported to allow the deposit of DLC films at low temperatures. 4 In particular, radiofrequency (RF) plasma chemical vapor deposition (CVD) is useful because it allows the deposit of DLC films on most substrates that are conductor or insulator substrates at room temperature. 5 Therefore, it is possible that DLC films can be deposited on polymeric materials, which are used as biomaterials.
Recent studies have indicated biomedical applications of DLC films. 6 In previous work, the authors reported the thrombosis and hemolysis properties of DLC films. 7–9 From these results, it could be ascertained that blood compatibility of the materials would be improved significantly when the substrates were coated with DLC films. The materials used to make artificial organs must have biocompatibility, blood compatibility, and dynamic compatibility, or problems will exist in relation to the duration of use. DLC film coatings, which not only have excellent blood compatibility but also meet the often criteria, are widely expected to be adapted for the purpose of improving surface technique because they add their attractive properties to material surfaces.
At the present time, an electrohydraulic total artificial heart (EHTAH) system is being developed. 10 One of there artificial hearts consists of two ellipsoidal diaphragm type blood pumps, an energy converter, and several electronic units. The diaphragm is made of polyurethane elastomer that separates blood and silicone oil in the chamber. The energy converter reciprocates and delivers hydraulic silicone oil to alternating blood pumps through a pair of flexible oil conduits. This alternating pulsation of silicone oil facilitates blood circulation. An EHTAH system and ellipsoidal diaphragm type blood pump is shown in Figure 1. In this artificial heart system, however, the penetration of silicone oil through the diaphragm could pose a problem in the long term use of this system. Penetrations might compromise the device’s hydraulic function. Moreover, it might result in the accumulation of silicone oil in the recipient’s body and exert toxic effects. It is feared that such penetration could cause serious problems.
In this study, the authors used DLC film coated on an ellipsoidal diaphragm. The purpose of such coating is to prevent the penetration of silicone oil and blood through the diaphragm. This method is expected to be adapted because DLC films are excellent as fluid and vapor transport barriers. 2 However, the diaphragm must continuously pulsate to perform its primary function, that is, blood circulation. The diaphragm would need to pulsate approximately 100,000 times per day or 50 million times annually. To apply DLC film to an artificial heart diaphragm, it is necessary that the DLC film be tested in regards to its adaptability to the diaphragm’s use. Although there are a number of reports indicating that DLC films might possibly have applications as new biomaterials, it is important to investigate the durability of DLC films for each application. The present authors estimate the functionality and stability of the DLC films to be sufficient in the dynamic transformation of the diaphragm while pulsating and in the prevention of silicone oil penetration through the diaphragm.
Materials and Methods
Diffusion Estimation by Static Condition
As a preliminary experiment, the present authors evaluated DLC film on polymeric material. Using the normal RF plasma CVD process, the DLC film was attached to polyurethane elastomer sheet substrates (model Miractran E980, Nihon Miractran, Kanagawa, Japan) that have a circular shape (φ25 mm, thickness: 0.3 mm). This material has been used to attach medical devices. The schematic system of RF plasma CVD is shown in Figure 2. This system consists of two planar electrodes, an RF generator (model SS-301AAE, Fuji Electronic Industrial Co., Ltd., Tokyo, Japan), a matching box (model HC-2000, Tokyo Hy-power Labs., Inc., Saitama, Japan), and a vacuum pump (model 1397, Sargent-Welch Scientific Co., Buffalo Grove, IL). The glow discharge plasma (13.56 MHz) of the electrical power was a constant 100 W; this separated the hydrocarbon gas molecules (CH4) and deposited DLC film on the substrate surfaces under the following conditions: methane gas pressure of 30, 50, and 100 Pa with deposition times of 15, 10, and 10 minutes, respectively. The thickness of all DLC film deposited was kept at approximately 100 nm.
To test whether fluid would penetrate the DLC film coated on the polyurethane surfaces, the following experimental system was developed (Figure 3). In the experiment, the samples were set between the two cells shown. Each cell was filled with a physiologic saline (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) and distilled water solutions (Otsuka Pharmaceutical). Static pressure of 2 kg/mm2 was applied from the physiologic saline side. After 72 hours, Na+ ion particles in the distilled water (through the polyurethane sheet with DLC film coating) were measured by microwave induced plasma mass spectroscopy (MIP-MS: model P-6000, Hitachi, Ltd., Tokyo, Japan).
Production of DLC Films on Ellipsoidal Substrate Surface
In RF plasma processing, the most common system consists of two planar electrodes (normal process). However, it is difficult to deposit DLC film uniformly on insulator material surfaces with a three-dimensional shape. To attach uniform DLC film on an ellipsoidal diaphragm surface, the authors have developed a special electrode that can fix DLC film on all of the ellipsoidal substrates (Figure 4A). This electrode is electrically conducted to the cathode electrode. In this article, the special electrode is referred to as improved electrode. Before the deposition of DLC film on the diaphragm surface by the RF plasma CVD method with the improved electrode process, it is important to investigate the usefulness of this new process. Specifically, the authors must examine the uniformity of the DLC film using the improved electrode process. The authors investigated the distribution of the DLC film thickness using the improved electrode and the film structural characterizations. To investigate the uniformity of the DLC film with the improved electrode process, the authors used silicon wafers (10 mm × 10 mm) of the crystal direction (100) type. These wafers were put on at different four positions (from the center position to the edge) of the improved electrode surface (Figure 4B). In this, the deposition of the DLC film was performed under the following conditions: the CH4 gas pressure was 50 Pa, and the deposition time was 30 minutes. At this time, the glow discharge plasma (13.56 MHz) of that electrical power was a constant 100 W.
The DLC film thickness was measured along the cross-section by scanning electron microscope (SEM: model JSM-5310LVB, JEOL Ltd., Tokyo, Japan) images. DLC films contain a mixture of sp3-bonds (tetrahedral or aliphatic), sp2-bonds (trigonal or aromatic), and sp1-bonds (linear of acetylenic) coordinated carbon atoms in a disordered network. The properties of DLC films are determined by the bonding hybridization of the carbon atoms and the relative concentrations of these bonds. 11,12 Because of the importance of the sp3/sp2 ratio hybridized carbon and hydrogen content in determining the properties of DLC films, it is of primary interest to characterize these two properties. The methods most commonly used for the study of the structure of DLC films are infrared absorption and Raman spectroscopy. 11,13 Therefore, the structure of the DLC film was investigated using infrared spectroscopy (IR: model FT/IR-620, JASCO Co., Tokyo, Japan) and Ar-laser Raman spectrophotometer (Raman: model NRS-2100, JASCO Co., Tokyo, Japan). Additionally, the chemical composition and bonding states of the DLC film were measured with x-ray photoelectron spectrometer (XPS: model JPS-9000MC, JEOL Ltd, Tokyo, Japan). The monochrometer was used with Al Kα. To consider the influence of the electrode shape, it was compared with the characteristics of conventional DLC films. In this study, the conventional DLC films were attached on silicon wafer using the normal process. These conventional DLC films are referred to as typical DLC films.
Penetration Through the Diaphragm Coated With DLC Films Under Pulsation Operation
The schematic of the pulsation and penetration test system with mock circulation is shown in Figure 5. This in vitro test system with a mock circulation consists of a blood pump, a flexible stainless steel conduit (Valqua Seiki Ltd., Shizuoka, Japan), and a hydraulic actuator. The ellipsoidal diaphragm type blood pump and the hydraulic actuator were developed by Aisin Cosmos R&D Co., Ltd., Aichi, Japan. The diaphragm of the blood pump is made of polyurethane elastomer (model Miractran E980, Nihon Miractran, Kanagawa, Japan). These units are used in the EHTAH system. 10 Using the RF plasma CVD process with the improved electrode, the DLC film was attached upon the diaphragm surface. The diaphragm with DLC film coating was put between the blood chamber of the blood pump and the hydraulic oil chamber. The surface of the diaphragm coated with DLC film was set to the side of the oil chamber at that time. The hydraulic oil chamber was filled with silicone oil (model SH-200, Toray Silicon Co., Tokyo, Japan) at a volume of 250 ml. The blood pump was connected to the receiver circulation circuit, which was filled with 2,500 ml of physiologic saline (Otsuka Pharmaceutical).
To establish that the stability of the DLC film and penetration prevention of silicone oil through the diaphragm are sufficient under the stress of dynamic transformation of the diaphragm with its accompanying pulsation, this in vitro test system was conducted for 20 days. The conditions of the pulsation operation are listed in Table 1. With regards to the penetration of silicone oil, the diffusion concentrations of silicon ions in the receiver chamber were measured by inductively coupled plasma atomic emission spectroscopy (ICP-AES: model ICAP-575 II, Nippon Jarrell-Ash Co., Ltd, Tokyo, Japan). The penetration was obtained based upon the following equations:14
Where Co is the initial concentration of silicone oil in the hydraulic oil chamber, Ct is the concentration of silicone oil in the receiver chamber at time t, P is the penetration coefficient, A is the effective area (surface area of diaphragm), and V is the volume of the receiver circulation circuit. Total flux for silicone oil into physiologic saline through the diaphragm with the DLC film coating was calculated by the equation J =PA ΔCo, where J is total flux and ΔCo is the concentration gradient.
Results and Discussion
Diffusion Estimation Under Static Condition
Table 2 shows the amount of Na+ ion in the distilled water that penetrated through the sample (polyurethane sheet). For the pure polyurethane sheet (without DLC film coating), the penetration of Na+ ion was 32.8 ppm. In contrast, the amount of solution penetration of the DLC film was approximately 1 ppb. Because Na+ ion concentration of the initial distilled water was 1.25 ppb, it could be ascertained that there was no solution penetration through the DLC film under the static pressure of 2 kg/mm2.
Production of DLC Films on Ellipsoidal Substrate Surface
The uniformity, thickness, and structure of the DLC films that were attached using the improved electrode process were investigated using SEM, IR, Raman, and XPS. The thickness of the DLC film that was attached at each position using the improved electrode surface is shown in Figure 6. Using the improved electrode process, the thickness of the DLC film at different positions was kept uniform at approximately 300 nm. On the other hand, when the normal process (the planar electrode process) was applied to ellipsoidal substrate, it was impossible to achieve uniformity in the DLC film. Thus it was demonstrated that the deposition rate was dramatically improved when using the improved electrode process.
The spectra of the IR analysis are shown in Figure 7. The IR was used to compare the structures of the DLC film attached at different positions upon the improved electrode surface. In addition, the spectrum of positions 1 ∼ 4 was compared with typical DLC films. Table 3 gives an overview on relevant peaks. 15 The main peak of all the spectra was the waveform of 2,915 cm−1 and 2,925 cm−1, and the spectra peak of sp3 and sp2 bonding was observed at 2,855 ∼ 3,000 cm−1 (Figure 7). The spectra of positions 1 ∼ 4 were similar to the typical DLC films. In this IR analysis, it was expected that the improved electrode process obtained uniformity of the DLC film equivalent to the conventional DLC films.
The Raman spectra are shown in Figure 8. As for the Raman spectra of all the positions (positions 1 ∼ 4) on the improved electrode surface, each spectrum was same in form, and two broad peaks were shown in 1,350 cm−1 and 1,530 cm−1. In the case of Ar-laser, Raman spectra of DLC films show two broad peaks:D-peak around 1,350 cm−1 and G-peak around 1,540 ∼ 1,580 cm−1. 13 Raman spectrum form of DLC films, which consists of D-peak and G-peak, strongly depends upon the ratio of sp3/sp2. 13,16 In this Raman spectra analysis (Figure 8), it is observed that all the spectra of the DLC film on the improved electrode surface were similar to the typical DLC films. The improved electrode process was shown to attach the DLC film uniformly on the ellipsoidal substrate.
Figure 9 shows C1s spectra for the DLC film, which were attached through the improved electrode process including the typical DLC films. The main feature of these spectra consisted of an asymmetric peak at 284.2 eV, and these asymmetric spectra result from π-bonding. 17 XPS spectra of DLC films are separated into two types of peaks: sp3-bonds and sp2-bonds. 18 The ratio of sp3/sp2 is important to the structural characterization of DLC films. Taki et al.13 state that examining sp3 and sp2 carbon networks in DLC films was possible by analyzing the peak positions and the full width at half maximum (FWHM) of XPS C1s spectra. All the FWHM for spectrum of positions 1 ∼ 4 were approximately 1.75 (Figure 9). Although the FWHM of the typical DLC films was 1.71, there were no clear differences in these spectra, including the main peak positions. In this analysis of XPS, it was revealed that the DLC film attached to the ellipsoidal substrate were uniform in structural characterizations.
The SEM, IR, Raman, and XPS analyses proved that the improved electrode process could attach DLC films uniformly upon the ellipsoidal substrate surface. Additionally, it was inferred that the DLC film attached using improved electrode process were equivalent to the typical DLC films.
Penetration Through the Diaphragm With DLC Films by Pulsation Operation
The in vitro test system was operated for 20 days. The DLC film on the diaphragm surface was tested under pulsation operation. Figure 10 shows the amount of the silicone oil in the receiver circulation circuit (physiologic saline side) through the diaphragm. The diffusion concentration of silicon ions were measured from ICP-AES. Twenty days after the experiment started, the amount of penetration of silicone oil had been reduced to one-third using the DLC film coating when compared with an uncoated diaphragm. It was expected that the stability of the DLC film was sufficient in the dynamic transformation of the diaphragm when pulsating because the amount of transference increased at constant rate, that is, there was a significant liner relationship in the amount of penetration. If the DLC film had been destroyed by pulsation, the amount of penetration would have increased rapidly. In the diffusion experiment under static pressure, it had been already ascertained that there was no penetration through the DLC film. Therefore it is conjectured that the cause of the penetration of silicone oil through the DLC film coated diaphragm was mainly in the composition of the blood pump.
Penetration of silicone oil through the blood pump diaphragm is a possible problem in the long term use in an EHTAH system. The total flux for silicone oil through an uncoated diaphragm was 2.1 ml/year; this is the annual amount of penetration in the EHTAH system. 10 This problem might compromise the device’s hydraulic function, and it might result in accumulation of silicone oil in the recipient’s body and exert toxic effects. It is feared that such penetration may also cause more serious problems. Hence, the application of DLC films is quite applicable to the problem of the penetration of silicone oil through a blood pump diaphragm.
Using the improved electrode process, the present authors successfully attached uniform DLC films on the ellipsoidal diaphragm surface of the EHTAH blood pump. In the in vitro test, pulsation of the diaphragm and penetration of silicone oil were examined. It was proven that the stability of the DLC film was sufficient in the dynamic transformation of the diaphragm while pulsating. Also, the degree to which silicone oil penetrates a diaphragm has been improved using DLC film coatings as compared with an uncoated diaphragm.
Accordingly, it was found that DLC films would be quite applicable to new biomaterials such as blood pump diaphragms.
This work was partially supported by the Center for Research, Tokyo Denki University. The authors thank Tsuneo Toya at Tokyo Denki University for the discussion of the SEM observations and XPS analysis.
1. Martinu L, Raveh A, Boutard D, et al: Properties and stability of diamond-like carbon films related to bonded and unbonded hydrogen. Diamond Related Mater 2: 673–677, 1993.
2. Deng J, Braun M: DLC multiplayer coatings for wear protection. Diamond Related Mater 4: 936–943, 1995.
3. Shimura K-I, Baba K: Microscale wear properties of ultrathin diamond-like carbon films. Jpn J Appl Phys 39: 626–628, 2000.
4. Tiainen V: Amorphous carbon as a bio-mechanical coating mechanical properties and biological applications. Diamond Related Mater 10: 153–160, 2001.
5. Tamor MA, Vassell WC, Carduner KR: Atomic constraint in hydrogenated “diamond-like” carbon. Appl Phys Lett 58: 592–594, 1991.
6. Butter R, Allen M, Chandra L, Lettington H, Rushton N: In vitro studies of DLC coating with silicon intermediate layer. Diamond Related Mater 4: 857–861, 1995.
7. Alanazi A, Nojiri C, Noguchi T, et al: Improved blood compatibility of DLC coated polymeric material. ASAIO J 46: 440–443, 2000.
8. Alanazi A, Nojiri C, Kido T, et al: Engineering analysis of diamond-like carbon coated polymeric materials for biomedical applications. Artif Organ 24: 624–627, 2000.
9. Ohgoe Y, Yasuda T, Hirakuri K, Funakubo A, Fukui Y: Hemolytic properties of DLC films deposited on polymeric materials. TANSO 202: 71–76, 2002.
10. Tatsumi E, Taenaka Y, Uesho K, et al: Current status of development and in vivo evaluation of the National Cardiovascular Center electrohydraulic total artificial heart system. J Artif Organs 3: 62–69, 2000.
11. Spear KE, Dismukes JP (eds): Synthetic Diamond: Emerging CVD Science and Technology. New York: John Wiley & Sons, Inc., 1994.
12. Grill A, Patel V: Characterization of diamondlike carbon by infrared spectroscopy? Appl Phys Lett 60: 2089–2091, 1992.
13. Taki Y, Takai O: XPS structural characterization of hydrogenated amorphous carbon thin films prepared by shielded arc ion plating. Thin Solid Films 316: 45–50, 1998.
14. Matsuo Y, Taenaka Y, Takano H, et al: Permeation of silicone oil through a diaphragm in an electrohydraulic ventricular assist system. ASAIO J 38: 570–573, 1992.
15. Hirakuri KK, Minorikawa T, Friedbacher G, Grasserbauer M: Thin film characterization of diamond-like carbon films prepared by r. f. plasma chemical vapor deposition. Thin Solid Films
302: 5–11, 1997.
16. Ferrari AC, Robertson J: Interpretation of Raman spectra of disordered and amorphous carbon. Am Phys Soc 61: 14095–14107, 2000.
17. Marton D, Boyd KJ, Al-Bayati AH, Todorov SS, Rabalais JW: Carbon nitride deposited using energetic species: a two-phase system. Phys Rev Lett 73: 118–121, 1994.
18. Kimura A, Azuma Y, Suzuki T, Saito T, Ikuhara Y: Microstructure of diamond-like carbon films prepared by cathodic arc deposition. Diamond Related Mater 11: 1436–1440, 2002.