Otology & Neurotology 35:1179Y1186 Ó 2014, Otology & Neurotology, Inc.
A Polymer-Based Multichannel Cochlear Electrode Array *Kyou Sik Min, †‡Seung Ha Oh, §Min-Hyun Park, *Joonsoo Jeong, and *Sung June Kim *School of Electrical Engineering and Computer Science, ÞDepartment of OtorhinolaryngologyYHead and Neck Surgery, Seoul National University College of Medicine, þSensory Organ Research Institute, Seoul National University Biomedical Research Institute, Seoul, Republic of Korea §Department of Otorhinolaryngology, Boramae Medical Center, SMG-SNU, Seoul, Korea
Objective: Compared with conventional cochlear electrode arrays, which are hand assembled and wire-based, polymer-based implants have several advantages. They are very precise, and their fabrication is inexpensive because of the use of thin-film processes. In the present study, a cochlear electrode array based on a high-performance liquid crystal polymer material is devised. Furthermore, the device is encapsulated in silicone elastomer. Methods: The fabrication steps introduced here include thin-film processes with liquid crystal polymer (LCP) films and customized self-aligning molding processes for the electrode array. To assess the feasibility of the proposed electrode array, the charge storage capacitance and impedance were measured using a potentiostat. Vertical and horizontal deflection forces were measured using a customized fixture and a force sensor. Insertion and extraction forces were also measured using a transparent human cochlear plastic model, and five cases involving human temporal insertion
trials were undertaken to assess the level of safety during the insertion process. Results: The charge storage capacity and impedance at 1 kHz were 33.26 mC/cm2 and 1.02 k6, respectively. Likewise, the vertical force and horizontal force of the electrode array were 3.15 g and 1.07 g. The insertion force into a transparent plastic cochlear model with displacement of 8 mm from a round window was 8.2 mN, and the maximum extraction force was 110.4 mN. Two cases of human temporal bone insertion showed no observable trauma, whereas 3 cases showed a rupture of the basilar membrane. Conclusion: An LCP-based intracochlear electrode array was fabricated, and its electrical and mechanical properties were found to be suitable for clinical use. Key Words: Atraumatic cochlear electrodeVCochlear implantationVIntracochlear electrode arrayV Liquid crystal polymerVpolymer-based neural interface. Otol Neurotol 35:1179Y1186, 2014.
Current commercialized cochlear implants commonly use wire-based platinum-alloy electrode arrays. The electrode array is inserted into the scala tympani to interface with the spiral ganglion cells electrochemically. The electrode array is made up of electrodes, which deliver an electrical charge to the ganglion cells, lead wires, and silicone as a carrier to encapsulating the wires and electrodes. Because most of the fabrication processes of a conventional cochlear electrode array require manual handling, resulting in a low yield and high cost, major cochlear implant companies are making
an effort to develop manufacturing processes for higher efficiency and better manufacturability. The advantages of microfabrication technology are not only better manufacturability but also a greater level of feasibility of a high-density neural interface. Although polyimide and parylene-C are widely used in biocompatible MEMS applications, their long-term reliability levels are controversial on the account of their high moisture absorption rates and because of the unstable adhesion between the laminated layers in the body environment, which contains corrosive fluids (1Y4). Because of its good mechanical and chemical stability, liquid crystal polymer (LCP) is considered as a good candidate material for manufacturable biomedical applications such as a neural electrode array and as an encapsulation material (5Y11). LCP has a much lower moisture absorption rate than those of polyimide and parylene-C, which is highly related to its long-term reliability (7,12Y19). The LCP material can also be used as a near-hermetic packaging material for MEMS applications (20Y23). There has been much effort to create high-density and manufacturable intracochlear electrode arrays using microfabrication technology (24Y27). The first LCP-based
Address correspondence and reprint requests to Sung June Kim, Ph.D., School of Electrical Engineering and Computer Science, Bldg. 301 Room 1006, Seoul National University, San 56-1, Shillim-dong, Gwanak-gu, Seoul, 151-742, Korea; E-mail: [email protected] This work was supported in part by the Ministry of Knowledge Economy of Korea under Grant 10033657; in part by the Pioneer Research Center Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (NRF-20090082961); in part by the Public Welfare and Safety research program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (NRF-2010-0020851). The authors disclose no conflicts of interest.
1179
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1180
K. S. MIN ET AL.
cochlear electrode array was reported in 2006 (10). It used 50-Km-thick LCP films with high and low melt temperatures as the substrate cover layer and as an adhesive layer, respectively. Microfabrication processes were used to pattern the electrodes and lead wires of 12 electrode contacts spaced 200 Km apart. The finished LCP-based electrode array was attached to a silicone elastomer carrier for insertion into an animal cochlea. Although the electrode array was successfully used in an acute animal implant trial (28), it was reported that the stiffness of the electrode was higher than desired for the intended cochlear implant application (10). In this study, we suggest an LCP-based cochlear electrode array. The cochlear electrode array is fabricated on a flexible LCP film using our previously reported fabrication process (8,9). Because the thickness of the LCP is closely related to the stiffness of the electrode array, the high stiffness of the first LCP-based cochlear electrode array was likely caused by the thick LCP structure. The thickness of the present LCP electrode array is only 50 Km, which is thinner and more flexible than even one trace layer of the previously reported LCP-based cochlear electrode array. To gain more flexibility, in this study, an LCP substrate among the electrodes remains to such a minimal extent
as only to maintain lead wires. The LCP-based electrode array is molded in a silicone carrier using a novel self-aligning injection molding process. Instead of aligning each electrode on the mold manually (29,30), we utilize a sawtooth-like structure for our LCP substrate to self-align and to open the electrode sites without an additional ablation process. The elastomer carrier encapsulates the sharp edges of the LCP electrode so as not to destroy the surrounding tissues of the scala tympani. To assess the feasibility of the electrode array, the electrical and the mechanical characteristics of the finished electrode array are reported in this study. Human temporal bone study results are also presented. MATERIALS AND METHODS Fabrication Figure 1A describes the structure of the LCP-based cochlear electrode array. It is fabricated using laser micromachining, a customized thermal press and injection molding. After cleaning in methanol, acetone, and isopropyl alcohol for 1 minute each, an LCP substrate (Kuraray, Vecstar FA-25, Tokyo, Japan) is dry etched using an inductive coupled plasma (ICP) etcher for 3 minutes. Subsequently, titanium and gold electroplating seed layers are deposited using an e-gun evaporator (Maestech Co., Ltd.,
FIG. 1. A, Ten micrometerYwide metal wires and sixteen 300 550 Km2 electrodes are patterned as interconnection lines and electrode sites on the LCP film (Kuraray, Vecstar FA-25, Tokyo, Japan) using photolithography and electroplating. Then, a laser-cut cover LCP layer and a patterned layer are laminated by a thermal-press bonding process. The final laser cut is followed by iridium oxide sputtering and an elastomer molding process. B, The final laser cut is performed to reduce the rigidity along the z-axis, where the basilar membrane is. The aspect ratio of the cross-section of the interconnection part of the LCP electrode is 4:1 (apex) to 10:1 (base). This property makes the electrode more flexible so that the electrode array will not destroy the basilar membrane. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A POLYMER-BASED MULTICHANNEL COCHLEAR ELECTRODE ARRAY ZZS550-2/D, Pyung Taek, South Korea) at 50 and 100 nm, respectively. Then, coating with a 5.5-Km-thick photoresist (Hoechst Celanese, AZ4330, Somerville, NJ, USA) and photolithography using a mask aligner machine (SUSS MicroTec, MA6/BA6, Garching, Germany) with 5-Km-thick gold electroplating are performed. After detaching the substrate from the elastomer-coated wafer, the electroplated seed layer is etched away using gold and titanium etchants. The patterned substrate is then thermally bonded to a laser-processed passivation layer with contact windows using a thermal press (Carver, Inc., model CH; press no. 4386, Wabash, IN, USA). Thermal press bonding is performed at 300 psi (2.1 MPa) and 285-C for 45 minutes. The electroplated seed layers of electrodes are sputter-deposited on the contact windows and the surrounding area for the further electroplating of the electrode sites. Although the LCP film is flexible along the normal direction of its surface, it is relatively rigid along the tangential direction. If a cochlear electrode is too rigid, it can cause damage to the surrounding tissues. As shown in Figure 1B, to reduce the rigidity along the z-axis (parallel to the modiolus), we reduced the width of the LCP substrate between the electrodes. Meanwhile, because the sharp edges of the thin film substrate can also destroy cochlear tissue, we molded the thin film with silicone elastomer using a customized molding process. To open the electrode sites of a silicdone-molded neural interface, oxygen plasma etching (31) or a laser ablation process are usually used (32). However, we devised a molding process, which opens the sites without an additional process. In this study, the stepwise injection molding process consisting of 2 steps is used for self-alignment and site opening. In the first injection molding, the silicone elastomer (Nusil Silicone Technology, MED 6233, Carpinteria, CA, USA) carrier part is initially molded. This molded part does not contain an electrode array but instead has a 200-Km-deep guiding groove for an outline of the sawtooth-like LCP electrode array (Fig. 2, A and E). Because the outline of the guiding groove is designed based on the laser-machined microstructure of the LCP electrode array, each electrode site is self-aligned to be fixed in
1181
its position (Fig. 2, B and F). After aligning and placing the LCP electrode array, the encapsulating silicone layer is molded using a cover mold, which has as many pillars and which is as large as the electrode sites, with precise alignment upon the electrode sites (Fig. 2C). Because the pillars on the cover mold are in contact with and pressing the electrode sites, after the release, every stimulation site is open to the outside (Fig. 2, D and G).
Characterization Cyclic voltammetry was conducted to assess the charge storage capacity (CSC). The electrolyte was a phosphate-buffered saline (PBS) solution (Invitrogen Life Technologies, Gibco 10010, Carlsbad, CA, USA) at pH 7.2. Using a potentiostat (Solartron Analytical, 1287A, Farnborough, UK) and a 3-cell electrochemical system with a platinum counter electrode and a silver/silver chloride (Ag/AgCl) reference electrode, we swept the electrochemical potential from j0.6È0.8 V versus an Ag/AgCl electrode at a sweep rate of 100 mV/s. We also measured the electrochemical impedance spectrum using an impedance analyzer (Solartron Analytical, 1286 and 1287A, Farnborough, UK). We used an identical electrochemical cell for the cyclic voltammetry measurements. The impedance spectrum was measured using a 10 mVrms excitation at frequencies ranging from 0.1 Hz to 100 kHz. The insertion characteristics and the final position of the electrode array in the scala tympani (ST) are becoming far more important in today’s cochlear implant systems because conservation of residual hearing means an expansion of cochlear implant candidates (33Y35). In this study, the mechanical characteristics are measured using a method reported in earlier work (36). The stiffness of the finished electrode array is measured after fixing the electrode array in a custom-designed fixture. The stiffness of the Nurobiosys 16-channel wire-based intracochlear electrode array used in the aforementioned study (36) is also measured using this fixture for comparison. Each electrode array is held securely, and a mechanical force gauge measures the force needed to bend
FIG. 2. Thin film molding process for self-alignment. A 200-Km-deep groove is initially molded along the outline of the LCP electrode array as an alignment pattern (A and E). The LCP electrode is positioned in the groove. Silicone rubber is mounted in the base mold used in the (A) and (B) processes (B and F ). The cover silicone layer is molded with a cover mold with rectangular pillars upon the electrode sites (CYD, G). Without laser ablation or an oxygen plasma etching process, the electrode sites are opened successfully. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1182
K. S. MIN ET AL.
FIG. 3. Automated insertion setup. The load cell (ATI, Nano17, Pinnacle Park Apex, NC, USA) is fixed to the motorized linear actuator (Newport, LTA-HS, Irvine, CA, USA). The actuator is controlled by Labview software, which also records the displacement and insertion force; 50% glycerin is used as a lubricant.
the array by 30 degrees from its normal shape. Horizontal and vertical deflection forces are measured at every 1-mm spot from the bases of the arrays. The stiffness ratio is calculated by dividing the vertical deflection force by the horizontal deflection force. The insertion force of the LCP-based electrode array is also compared with that of wire-based electrode array. The wire-based electrode array has 16 platinum-iridium electrodes and wires with a crosssectional diameter of 75 Km. The diameters of the wire-based electrode array are 600 Km at the apex and 800 Km at the base. The length of the wire-based electrode array is 22 mm. A load cell (ATI, Nano17, Pinnacle Park, Apex, NC, USA) is fixed to a motorized linear actuator (Newport, LTA-HS, Irvine, CA, USA). The actuator is controlled by Labview software, which also records the displacement and insertion force. For the insertion force measurements, 50% glycerin is used as a lubricant (Fig. 3). For an assessment of the insertion safety, 5 cases of temporal bone study results are briefly presented. The insertion trauma is standardized
using the following grading scheme: 0, no trauma; 1, elevation of the basilar membrane; 2, rupture of the basilar membrane or spiral ligament; 3, dislocation into the scala vestibuli; and 4, fracture of the osseous spiral lamina or modiolar wall (37).
RESULTS Figure 4 describes the shape of the finalized intracochlear electrode array. The photograph of the electrode array shows the iridium oxide stimulation sites molded in silicone elastomer. Although silicone elastomer mold is designed to make a straight carrier, the result shows a gently rounded shape. Through the transparent silicone elastomer, the sawtooth-like LCP leads are observed between the stimulation electrodes. Figure 4B shows a cross-sectional view and
FIG. 4. Photographs and dimensions of the LCP-based intracochlear microelectrode array. A, Photograph of the intracochlear electrode array showing the 16 gold (iridium oxide) electrode sites and silicone elastomer windows. Among the electrodes, narrow LCP leads are observed. B, Cross-sectional views: apical (left) and basal (right) dissections show its semicircular shape. The areas of cross section are 500 550 Km (apex) and 850 625 Km, respectively. The length of the electrode array is 28 mm (center). Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A POLYMER-BASED MULTICHANNEL COCHLEAR ELECTRODE ARRAY
1183
FIG. 5. Insertion force of wire-based and LCP-based electrode arrays. The insertion forces at a displacement of 8 mm from the round window are 17.2 mN (wire-based) and 8.2 mN (LCP-based). The final insertion force of the LCP-based electrode was determined to be half that of the wire-based electrode array.
gives the dimensions of the electrode array. The dimensions of the apex and base are 500 550 Km and 850 625 Km, respectively, with a semicircular cross section. The length of the electrode is 28 mm. From the impedance spectrum and cyclic voltammogram, we observed that the electrochemical impedance at 1 kHz and the charge storage capacity were 1.02 k6 and 33.26 mC/cm2, respectively. The vertical force and horizontal force of the electrode array are 3.15 and 1.07 g, respectively. The V/H force ratio is 2.77, which is comparable to that of commercial cochlear electrode arrays (36). The insertion forces at a displacement of 8 mm from the round window are 17.2 mN (wire-based) and 8.2 mN (LCP-based) (Fig. 5). The maximum extraction forces were determined to be 158.5 mN (wire-based) and 110.4 mN (LCP-based) (Fig. 6). The temporal bone study result is shown in Figures 7 and 8. Among 5 cases, 2 samples
were inserted without trauma (Grade 0), whereas a rupture of the basilar membrane (Grade 2) was observed in 3 cases. The mean insertion depth was found to be 405 degrees from the round window.
DISCUSSION The manufacturing process of the conventional cochlear electrode arrays consists of the forming of electrode sites, the welding of each electrode to a lead wire, fixation of the electrode to a mold, and injection molding. Most of the processes are manually conducted under a stereoscopic microscope to make a 12- to 22-channel electrode array. However, using the proposed manufacturing process, 208 electrode sites and lead wires on 13 electrode arrays are easily patterned at once on an LCP-coated silicon
FIG. 6. Extraction force of wire-based electrode and LCP electrode arrays. The extraction force of the wire-based straight electrode (158.5 mN) is greater than that of the LCP-based electrode array (110.4 mN). Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1184
K. S. MIN ET AL.
FIG. 7. Cross-section of the human temporal bone with the LCP-based cochlear electrode array inserted atraumatically.
wafer by photolithography. The metal-patterned LCP films are encapsulated using a thermal press bonding process followed by an automated laser machining process. In contrast to the conventional molding process, which fixes each electrode to an aligned position on a mold one by one, the proposed self-aligning injection molding process uses a micro-machined structure of an LCP-based electrode array for alignment and fixation. Because the LCPbased electrode array is inserted into a preformed silicone structure, each electrode site fixed is itself at aligned position. The proposed automated manufacturing process can lower the number of fabrication errors and increase productivity, resulting in a low-cost cochlear implant system. These aspects will contribute to an increase in the number of potential cochlear implant recipients even in developing or underdeveloped countries. The wire-based comparator electrode was tested by Rebscher et al. in 2008 (36). According to their article, the electrode showed 0% of trauma, whereas Cochlear Banded and Cochlear Contour showed rates of 37.5% and 38.9%, respectively. The insertion force data were not produced in
the study. However, the mean stiffness was 0.8 g, whereas these values of the Cochlear Banded and Cochlear Contour types were 0.6 and 1.65 g, respectively. Because the insertion trauma is related not only to the insertion force but also to the insertion depth and dimension of electrode array, whether the stiffness has a negative relationship with the insertion trauma is unclear. Nevertheless, the mechanical properties of the cochlear electrode array are still closely related to the degree of insertion safety. Several current major cochlear implant manufacturers, including Cochlear Ltd., Med-El, and Advanced Bionics, have developed electrode arrays, which have low levels of insertion force, providing evidence of atraumatic insertion (38). There are 2 types of electrode arrays, those with a straight electrode and those with a perimodiolar electrode with a stylet. The straight electrode array is inserted into the cochlear along the outer wall of the scala tympani. The initial insertion force is measured at the moment the apical electrode array tip hits the outer wall of the scala tympani and increases as the number of wires contained in the carrier increases. On the other hand, the stiffness of the LCP-based electrode array is determined by the thickness of the LCP film. In this study, 8 to 16 electrode array channels are patterned on a single layer of LCP film, which means that the difference in the stiffness between the base and the apex is not as much as that in the wire-based electrode array, as shown in Figure 5. An effort is being made to optimize the stiffness variation along the length of the LCP-based electrode array to ensure atraumatic insertion into cochlea. Results of this study will be made available in a future publication. The extraction damage should be considered because 4 to 6 reimplantations can be expected with very young recipients throughout their lifetimes (39). The plastic deformation of the straight wire-based electrode is one of the sources of the extraction force. Moreover, a perimodiolar electrode requires extra force to uncoil the preshaped carrier. Because LCP is an elastic material, the LCP-based electrode tends to maintain its original shape. Thus, the
FIG. 8. Result of the human temporal bone study. Among the 5 cases, 2 samples were inserted without trauma (Grade 0), whereas a rupture of the basilar membrane (Grade 2) was observed in 3 cases. The mean insertion depth was 405 degrees from the round window. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A POLYMER-BASED MULTICHANNEL COCHLEAR ELECTRODE ARRAY extraction force of the LCP-based electrode is less than that of the wire-based electrode array, as shown in Figure 6. The LCP-based electrode has less insertion and extraction force than conventional wire-based electrode arrays. The LCP material is a very strong and stable material when it is used for semiconductor packaging applications or in electronic components. As a film, it becomes flexible along its normal direction, but the mechanical properties of LCP are still stiff along its tangential direction. Thus, sharp edges of neural electrodes based on a rigid film substrate may be traumatic to the surrounding tissue. To protect the surrounding tissue against the edges of filmbased neural electrode arrays, encapsulation of the electrode array must be employed. There are various types of silicone rubber with mechanical properties similar to that of the surrounding tissue implanted in bodies. However, without considering the electrode site openings, it is difficult to open electrode sites, which are buried in silicone rubber. In this case, a polymer ablation process such as laser ablation or dry etching can be applied. However, the laser ablation process can burn the elastomer and electrode, and dry-etching is time consuming. In our work, a micro-molding process without an ablation process is exploited. Self-aligning during the elastomer molding process could precisely locate the electrode array on the opening structure of the upper mold and successfully opened the electrode sites. Conventional cochlear implant systems consist of a wire-based electrode array, a metal package, and a receiver coil. To fabricate an interconnection between the electronics and outer components (the electrode array and coil), a feedthrough should be used in the metal-based package. Most metal-based implant packages are manufactured by brazing and laser welding, both of which involve ultra-high temperatures and high costs. LCP is a thermoplastic polymer, which means that it can be molded, joined, and cut close to its melting temperature. The electrode array in the present study also uses the thermoelasticity to laminate the encapsulation. LCP can also be used as an RF-transparent implantable packaging material, which results in monolithic encapsulation of all components of the implantable unit. We designed a prototype of a LCP-based cochlear implant unit 20 mm (width) 20 mm (length) 2 mm (height) in size, which can be implanted with minimal incision. An accelerated soak test and animal implantation result with the LCP-based cochlear implant system will be published in the near future.
CONCLUSION An LCP-based cochlear electrode array provides suitable electrochemical properties for electrical stimulation. The vertical and horizontal deflection force ratio of the electrode array is comparable to that of a conventional wire-based electrode array. The insertion and extraction forces of the LCP-based electrode array are 47% and 69%, respectively, of the insertion force and extraction force of a wire-based electrode array, which implies that the present LCP-based
1185
electrode array can be inserted with less trauma than a conventional wire-based electrode array. Two cases of human temporal bone study results showed no observable trauma, whereas 3 cases showed a rupture of the basilar membrane. REFERENCES 1. Deiasi R, Russell J. Aqueous degradation of polyimides. J Appl Polym Sci 1971;15:2965Y&. 2. Murray S, Hillman C, Pecht M. Environmental aging and deadhesion of polyimide dielectric films. J Electron Packaging 2004;126: 390Y7. 3. Edell D. Insulating biomaterials N01-NS-2Y2347. NINDS Quarterly Progress Report 2002. 4. Schmidt E, McIntosh J, Bak M. Long-term implants of Parylene-C coated microelectrodes. Med Biol Eng Computing 1988;26: 96Y101. 5. Culbertson EC. A new laminate material for high performance PCBs: liquid crystal polymer copper clad films Electronic Components and Technology Conference, 1995. Proceedings., 45th 1995:520Y3. 6. Dean RN, Weller J, Bozack MJMJ, et al. Realization of ultra fine pitch traces on lcp substrates. Components and Packaging Technologies, IEEE Transactions on 2008;31:315Y21. 7. Lee CJ, Oh SJ, Song JK, et al. Neural signal recording using microelectrode arrays fabricated on liquid crystal polymer material. Materials Science and Engineering: C 2004;24:265Y8. 8. Lee SW, Min KS, Jeong J, et al. Monolithic encapsulation of implantable neuroprosthetic devices using liquid crystal polymers. IEEE Trans Biomed Eng 2011:58:2255Y8. 9. Lee SW, Seo JM, Ha K, et al. Development of microelectrode arrays for artificial retinal implants using liquid crystal polymers. Invest Ophth Vis Sci 2009;50:5859Y66. 10. Scott Corbett JK, Johnson T. Polymer-Based Microelectrode Arrays Materials Research Society Symposium, 2006. 11. Thompson DC, Tentzeris MM, Papapolymerou J. Packaging of MMICs in multilayer LCP substrates. Microwave Wireless Components Letters, IEEE 2006;16:410Y2. 12. Jayaraj K, Farrell B. Liquid crystal polymers and their role in electronic packaging. Adv Microelectron 1998;25:15Y8. 13. Edell DJ, Farrell B. Implantable devices having a liquid crystal polymer substrate, 2003. 14. Wang X, Engel J, Liu C. Liquid crystal polymer (LCP) for MEMS: processes and applications. J Micromech Microeng 2003;13:628. 15. Ketterl JR, Yarno JP, Corbett SS III, et al. Bio-implant and method of making the same, 2004. 16. Corbett SS III, Johnson TJ, Clopton BM, et al. Method of making high contact density electrode array, 2004. 17. Keesara VV, Durand DM, Zorman CA. Fabrication and characterization of flexible, microfabricated neural electrode arrays made from liquid crystal polymer and polynorbornene, Materials Research Society Symposium Proceedings: Cambridge Univ Press, 2006. 18. Wang K, Liu C-C, Durand DM. Flexible nerve stimulation electrode with iridium oxide sputtered on liquid crystal polymer. Biomed Eng IEEE Trans 2009;56:6Y14. 19. Dean R, Weller J, Bozack MJ, et al. Realization of ultra fine pitch traces on LCP substrates. Components Packaging Technol IEEE Trans 2008;31:315Y21. 20. Zou G, Gronqvist H, Starski JP, et al. Characterization of liquid crystal polymer for high frequency system-in-a-package applications. Adv Packaging IEEE Trans 2002;25:503Y8. 21. Kingsley N, Bhattacharya SK, Papapolymerou J. Moisture lifetime testing of RF MEMS switches packaged in liquid crystal polymer. Components Packaging Technol IEEE Trans 2008;31:345Y50. 22. Noll T, Larmouth R. A Low-Cost Near-Hermetic Multichip Module Based on Liquid Crystal Polymer Dielectrics, Proceedings-Spie The International Society for Optical Engineering: Citeseer, 1996: 385Y90. 23. Jauniskis L, Farrell B, Harvey A, et al. LCP PCB-based packaging for high-performance protection. Adv Packaging 2006;10:40Y2. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1186
K. S. MIN ET AL.
24. Bhatti PT, Wise KD. A 32-site 4-channel high-density electrode array for a cochlear prosthesis. IEEE J Solid-St Circ 2006;41:2965Y73. 25. Wise KD, Bhatti PT, Wang JB, et al. High-density cochlear implants with position sensing and control. Hear Res 2008;242: 22Y30. 26. An SK, Park SI, Jun SB, et al. Design for a simplified cochlear implant system. IEEE Trans Biomed Eng 2007;54:973Y82. 27. Bell TE, Wise KD, Anderson DJ. A flexible micromachined electrode array for a cochlear prosthesis. Sensors Actuators A: Physical 1998;66:63Y9. 28. Bonham RS B, Corbett S, Johnson T, et al. Reducing Channel Interaction through Multi-Polar Stimulation: Combining Tripolar and Monopolar Methods with High-Density Arrays Conference on Implantable Auditory Prostheses, Asilomar, CA, 2005. 29. Byers CLV, Loeb Gerald E, Merzenich Michael M, Rebscher Stephen J. Intracochlear electrode array. United States: The Regents of the University of California (Berkeley, CA), 1989. 30. Byers CLV, Loeb Gerald E, Merzenich Michael M, Rebscher Stephen J. Method for Making an Intracochlear Electrode Array. Berkeley, CA: United States: Regents of the University of California, 1987. 31. Taylor GN, Wolf TM. Oxygen plasma removal of thin polymerfilms. Polym Eng Sci 1980;20:1087Y92.
32. Schmidt EM, Bak MJ, Christensen P. Laser exposure of Parylene-C insulated microelectrodes. J Neurosci Methods 1995;62:89Y92. 33. Gantz BJ, Turner C, Gfeller KE. Acoustic plus electric speech processing: preliminary results of a multicenter clinical trial of the iowa/nucleus hybrid implant. Audiol Neurotol 2006;11:63Y8. 34. James C, Albegger K, Battmer R, et al. Preservation of residual hearing with cochlear implantation: How and why. Acta Otolaryngol 2005;125:481Y91. 35. Skarzynski H, Podskarbi-Fayette R. A new cochlear implant electrode design for preservation of residual hearing: a temporal bone study. Acta Otolaryngol 2010;130:435Y42. 36. Rebscher SJ, Hetherington A, Bonham B, et al. Considerations for design of future cochlear implant electrode arrays: electrode array stiffness, size, and depth of insertion. J Rehabil Res Dev 2008;45:731Y47. 37. Eshraghi AA, Yang NW, Balkany TJ. Comparative study of cochlear damage with three perimodiolar electrode designs. Laryngoscope 2003;113:415Y9. 38. Roland JT. A model for cochlear implant electrode insertion and force evaluation: results with a new electrode design and insertion technique. Laryngoscope 2005;115:1325Y39. 39. Jolly C, Mueller J, Helbig S, et al. New trends with cochlear implant electrodes. Otology Jpn 2010;20:239Y46.
Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A Polymer-Based Multichannel Cochlear Electrode Array *Kyou Sik Min, †‡Seung Ha Oh, §Min-Hyun Park, *Joonsoo Jeong, and *Sung June Kim *School of Electrical Engineering and Computer Science, ÞDepartment of OtorhinolaryngologyYHead and Neck Surgery, Seoul National University College of Medicine, þSensory Organ Research Institute, Seoul National University Biomedical Research Institute, Seoul, Republic of Korea §Department of Otorhinolaryngology, Boramae Medical Center, SMG-SNU, Seoul, Korea
Objective: Compared with conventional cochlear electrode arrays, which are hand assembled and wire-based, polymer-based implants have several advantages. They are very precise, and their fabrication is inexpensive because of the use of thin-film processes. In the present study, a cochlear electrode array based on a high-performance liquid crystal polymer material is devised. Furthermore, the device is encapsulated in silicone elastomer. Methods: The fabrication steps introduced here include thin-film processes with liquid crystal polymer (LCP) films and customized self-aligning molding processes for the electrode array. To assess the feasibility of the proposed electrode array, the charge storage capacitance and impedance were measured using a potentiostat. Vertical and horizontal deflection forces were measured using a customized fixture and a force sensor. Insertion and extraction forces were also measured using a transparent human cochlear plastic model, and five cases involving human temporal insertion
trials were undertaken to assess the level of safety during the insertion process. Results: The charge storage capacity and impedance at 1 kHz were 33.26 mC/cm2 and 1.02 k6, respectively. Likewise, the vertical force and horizontal force of the electrode array were 3.15 g and 1.07 g. The insertion force into a transparent plastic cochlear model with displacement of 8 mm from a round window was 8.2 mN, and the maximum extraction force was 110.4 mN. Two cases of human temporal bone insertion showed no observable trauma, whereas 3 cases showed a rupture of the basilar membrane. Conclusion: An LCP-based intracochlear electrode array was fabricated, and its electrical and mechanical properties were found to be suitable for clinical use. Key Words: Atraumatic cochlear electrodeVCochlear implantationVIntracochlear electrode arrayV Liquid crystal polymerVpolymer-based neural interface. Otol Neurotol 35:1179Y1186, 2014.
Current commercialized cochlear implants commonly use wire-based platinum-alloy electrode arrays. The electrode array is inserted into the scala tympani to interface with the spiral ganglion cells electrochemically. The electrode array is made up of electrodes, which deliver an electrical charge to the ganglion cells, lead wires, and silicone as a carrier to encapsulating the wires and electrodes. Because most of the fabrication processes of a conventional cochlear electrode array require manual handling, resulting in a low yield and high cost, major cochlear implant companies are making
an effort to develop manufacturing processes for higher efficiency and better manufacturability. The advantages of microfabrication technology are not only better manufacturability but also a greater level of feasibility of a high-density neural interface. Although polyimide and parylene-C are widely used in biocompatible MEMS applications, their long-term reliability levels are controversial on the account of their high moisture absorption rates and because of the unstable adhesion between the laminated layers in the body environment, which contains corrosive fluids (1Y4). Because of its good mechanical and chemical stability, liquid crystal polymer (LCP) is considered as a good candidate material for manufacturable biomedical applications such as a neural electrode array and as an encapsulation material (5Y11). LCP has a much lower moisture absorption rate than those of polyimide and parylene-C, which is highly related to its long-term reliability (7,12Y19). The LCP material can also be used as a near-hermetic packaging material for MEMS applications (20Y23). There has been much effort to create high-density and manufacturable intracochlear electrode arrays using microfabrication technology (24Y27). The first LCP-based
Address correspondence and reprint requests to Sung June Kim, Ph.D., School of Electrical Engineering and Computer Science, Bldg. 301 Room 1006, Seoul National University, San 56-1, Shillim-dong, Gwanak-gu, Seoul, 151-742, Korea; E-mail: [email protected] This work was supported in part by the Ministry of Knowledge Economy of Korea under Grant 10033657; in part by the Pioneer Research Center Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (NRF-20090082961); in part by the Public Welfare and Safety research program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (NRF-2010-0020851). The authors disclose no conflicts of interest.
1179
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1180
K. S. MIN ET AL.
cochlear electrode array was reported in 2006 (10). It used 50-Km-thick LCP films with high and low melt temperatures as the substrate cover layer and as an adhesive layer, respectively. Microfabrication processes were used to pattern the electrodes and lead wires of 12 electrode contacts spaced 200 Km apart. The finished LCP-based electrode array was attached to a silicone elastomer carrier for insertion into an animal cochlea. Although the electrode array was successfully used in an acute animal implant trial (28), it was reported that the stiffness of the electrode was higher than desired for the intended cochlear implant application (10). In this study, we suggest an LCP-based cochlear electrode array. The cochlear electrode array is fabricated on a flexible LCP film using our previously reported fabrication process (8,9). Because the thickness of the LCP is closely related to the stiffness of the electrode array, the high stiffness of the first LCP-based cochlear electrode array was likely caused by the thick LCP structure. The thickness of the present LCP electrode array is only 50 Km, which is thinner and more flexible than even one trace layer of the previously reported LCP-based cochlear electrode array. To gain more flexibility, in this study, an LCP substrate among the electrodes remains to such a minimal extent
as only to maintain lead wires. The LCP-based electrode array is molded in a silicone carrier using a novel self-aligning injection molding process. Instead of aligning each electrode on the mold manually (29,30), we utilize a sawtooth-like structure for our LCP substrate to self-align and to open the electrode sites without an additional ablation process. The elastomer carrier encapsulates the sharp edges of the LCP electrode so as not to destroy the surrounding tissues of the scala tympani. To assess the feasibility of the electrode array, the electrical and the mechanical characteristics of the finished electrode array are reported in this study. Human temporal bone study results are also presented. MATERIALS AND METHODS Fabrication Figure 1A describes the structure of the LCP-based cochlear electrode array. It is fabricated using laser micromachining, a customized thermal press and injection molding. After cleaning in methanol, acetone, and isopropyl alcohol for 1 minute each, an LCP substrate (Kuraray, Vecstar FA-25, Tokyo, Japan) is dry etched using an inductive coupled plasma (ICP) etcher for 3 minutes. Subsequently, titanium and gold electroplating seed layers are deposited using an e-gun evaporator (Maestech Co., Ltd.,
FIG. 1. A, Ten micrometerYwide metal wires and sixteen 300 550 Km2 electrodes are patterned as interconnection lines and electrode sites on the LCP film (Kuraray, Vecstar FA-25, Tokyo, Japan) using photolithography and electroplating. Then, a laser-cut cover LCP layer and a patterned layer are laminated by a thermal-press bonding process. The final laser cut is followed by iridium oxide sputtering and an elastomer molding process. B, The final laser cut is performed to reduce the rigidity along the z-axis, where the basilar membrane is. The aspect ratio of the cross-section of the interconnection part of the LCP electrode is 4:1 (apex) to 10:1 (base). This property makes the electrode more flexible so that the electrode array will not destroy the basilar membrane. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A POLYMER-BASED MULTICHANNEL COCHLEAR ELECTRODE ARRAY ZZS550-2/D, Pyung Taek, South Korea) at 50 and 100 nm, respectively. Then, coating with a 5.5-Km-thick photoresist (Hoechst Celanese, AZ4330, Somerville, NJ, USA) and photolithography using a mask aligner machine (SUSS MicroTec, MA6/BA6, Garching, Germany) with 5-Km-thick gold electroplating are performed. After detaching the substrate from the elastomer-coated wafer, the electroplated seed layer is etched away using gold and titanium etchants. The patterned substrate is then thermally bonded to a laser-processed passivation layer with contact windows using a thermal press (Carver, Inc., model CH; press no. 4386, Wabash, IN, USA). Thermal press bonding is performed at 300 psi (2.1 MPa) and 285-C for 45 minutes. The electroplated seed layers of electrodes are sputter-deposited on the contact windows and the surrounding area for the further electroplating of the electrode sites. Although the LCP film is flexible along the normal direction of its surface, it is relatively rigid along the tangential direction. If a cochlear electrode is too rigid, it can cause damage to the surrounding tissues. As shown in Figure 1B, to reduce the rigidity along the z-axis (parallel to the modiolus), we reduced the width of the LCP substrate between the electrodes. Meanwhile, because the sharp edges of the thin film substrate can also destroy cochlear tissue, we molded the thin film with silicone elastomer using a customized molding process. To open the electrode sites of a silicdone-molded neural interface, oxygen plasma etching (31) or a laser ablation process are usually used (32). However, we devised a molding process, which opens the sites without an additional process. In this study, the stepwise injection molding process consisting of 2 steps is used for self-alignment and site opening. In the first injection molding, the silicone elastomer (Nusil Silicone Technology, MED 6233, Carpinteria, CA, USA) carrier part is initially molded. This molded part does not contain an electrode array but instead has a 200-Km-deep guiding groove for an outline of the sawtooth-like LCP electrode array (Fig. 2, A and E). Because the outline of the guiding groove is designed based on the laser-machined microstructure of the LCP electrode array, each electrode site is self-aligned to be fixed in
1181
its position (Fig. 2, B and F). After aligning and placing the LCP electrode array, the encapsulating silicone layer is molded using a cover mold, which has as many pillars and which is as large as the electrode sites, with precise alignment upon the electrode sites (Fig. 2C). Because the pillars on the cover mold are in contact with and pressing the electrode sites, after the release, every stimulation site is open to the outside (Fig. 2, D and G).
Characterization Cyclic voltammetry was conducted to assess the charge storage capacity (CSC). The electrolyte was a phosphate-buffered saline (PBS) solution (Invitrogen Life Technologies, Gibco 10010, Carlsbad, CA, USA) at pH 7.2. Using a potentiostat (Solartron Analytical, 1287A, Farnborough, UK) and a 3-cell electrochemical system with a platinum counter electrode and a silver/silver chloride (Ag/AgCl) reference electrode, we swept the electrochemical potential from j0.6È0.8 V versus an Ag/AgCl electrode at a sweep rate of 100 mV/s. We also measured the electrochemical impedance spectrum using an impedance analyzer (Solartron Analytical, 1286 and 1287A, Farnborough, UK). We used an identical electrochemical cell for the cyclic voltammetry measurements. The impedance spectrum was measured using a 10 mVrms excitation at frequencies ranging from 0.1 Hz to 100 kHz. The insertion characteristics and the final position of the electrode array in the scala tympani (ST) are becoming far more important in today’s cochlear implant systems because conservation of residual hearing means an expansion of cochlear implant candidates (33Y35). In this study, the mechanical characteristics are measured using a method reported in earlier work (36). The stiffness of the finished electrode array is measured after fixing the electrode array in a custom-designed fixture. The stiffness of the Nurobiosys 16-channel wire-based intracochlear electrode array used in the aforementioned study (36) is also measured using this fixture for comparison. Each electrode array is held securely, and a mechanical force gauge measures the force needed to bend
FIG. 2. Thin film molding process for self-alignment. A 200-Km-deep groove is initially molded along the outline of the LCP electrode array as an alignment pattern (A and E). The LCP electrode is positioned in the groove. Silicone rubber is mounted in the base mold used in the (A) and (B) processes (B and F ). The cover silicone layer is molded with a cover mold with rectangular pillars upon the electrode sites (CYD, G). Without laser ablation or an oxygen plasma etching process, the electrode sites are opened successfully. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1182
K. S. MIN ET AL.
FIG. 3. Automated insertion setup. The load cell (ATI, Nano17, Pinnacle Park Apex, NC, USA) is fixed to the motorized linear actuator (Newport, LTA-HS, Irvine, CA, USA). The actuator is controlled by Labview software, which also records the displacement and insertion force; 50% glycerin is used as a lubricant.
the array by 30 degrees from its normal shape. Horizontal and vertical deflection forces are measured at every 1-mm spot from the bases of the arrays. The stiffness ratio is calculated by dividing the vertical deflection force by the horizontal deflection force. The insertion force of the LCP-based electrode array is also compared with that of wire-based electrode array. The wire-based electrode array has 16 platinum-iridium electrodes and wires with a crosssectional diameter of 75 Km. The diameters of the wire-based electrode array are 600 Km at the apex and 800 Km at the base. The length of the wire-based electrode array is 22 mm. A load cell (ATI, Nano17, Pinnacle Park, Apex, NC, USA) is fixed to a motorized linear actuator (Newport, LTA-HS, Irvine, CA, USA). The actuator is controlled by Labview software, which also records the displacement and insertion force. For the insertion force measurements, 50% glycerin is used as a lubricant (Fig. 3). For an assessment of the insertion safety, 5 cases of temporal bone study results are briefly presented. The insertion trauma is standardized
using the following grading scheme: 0, no trauma; 1, elevation of the basilar membrane; 2, rupture of the basilar membrane or spiral ligament; 3, dislocation into the scala vestibuli; and 4, fracture of the osseous spiral lamina or modiolar wall (37).
RESULTS Figure 4 describes the shape of the finalized intracochlear electrode array. The photograph of the electrode array shows the iridium oxide stimulation sites molded in silicone elastomer. Although silicone elastomer mold is designed to make a straight carrier, the result shows a gently rounded shape. Through the transparent silicone elastomer, the sawtooth-like LCP leads are observed between the stimulation electrodes. Figure 4B shows a cross-sectional view and
FIG. 4. Photographs and dimensions of the LCP-based intracochlear microelectrode array. A, Photograph of the intracochlear electrode array showing the 16 gold (iridium oxide) electrode sites and silicone elastomer windows. Among the electrodes, narrow LCP leads are observed. B, Cross-sectional views: apical (left) and basal (right) dissections show its semicircular shape. The areas of cross section are 500 550 Km (apex) and 850 625 Km, respectively. The length of the electrode array is 28 mm (center). Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A POLYMER-BASED MULTICHANNEL COCHLEAR ELECTRODE ARRAY
1183
FIG. 5. Insertion force of wire-based and LCP-based electrode arrays. The insertion forces at a displacement of 8 mm from the round window are 17.2 mN (wire-based) and 8.2 mN (LCP-based). The final insertion force of the LCP-based electrode was determined to be half that of the wire-based electrode array.
gives the dimensions of the electrode array. The dimensions of the apex and base are 500 550 Km and 850 625 Km, respectively, with a semicircular cross section. The length of the electrode is 28 mm. From the impedance spectrum and cyclic voltammogram, we observed that the electrochemical impedance at 1 kHz and the charge storage capacity were 1.02 k6 and 33.26 mC/cm2, respectively. The vertical force and horizontal force of the electrode array are 3.15 and 1.07 g, respectively. The V/H force ratio is 2.77, which is comparable to that of commercial cochlear electrode arrays (36). The insertion forces at a displacement of 8 mm from the round window are 17.2 mN (wire-based) and 8.2 mN (LCP-based) (Fig. 5). The maximum extraction forces were determined to be 158.5 mN (wire-based) and 110.4 mN (LCP-based) (Fig. 6). The temporal bone study result is shown in Figures 7 and 8. Among 5 cases, 2 samples
were inserted without trauma (Grade 0), whereas a rupture of the basilar membrane (Grade 2) was observed in 3 cases. The mean insertion depth was found to be 405 degrees from the round window.
DISCUSSION The manufacturing process of the conventional cochlear electrode arrays consists of the forming of electrode sites, the welding of each electrode to a lead wire, fixation of the electrode to a mold, and injection molding. Most of the processes are manually conducted under a stereoscopic microscope to make a 12- to 22-channel electrode array. However, using the proposed manufacturing process, 208 electrode sites and lead wires on 13 electrode arrays are easily patterned at once on an LCP-coated silicon
FIG. 6. Extraction force of wire-based electrode and LCP electrode arrays. The extraction force of the wire-based straight electrode (158.5 mN) is greater than that of the LCP-based electrode array (110.4 mN). Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1184
K. S. MIN ET AL.
FIG. 7. Cross-section of the human temporal bone with the LCP-based cochlear electrode array inserted atraumatically.
wafer by photolithography. The metal-patterned LCP films are encapsulated using a thermal press bonding process followed by an automated laser machining process. In contrast to the conventional molding process, which fixes each electrode to an aligned position on a mold one by one, the proposed self-aligning injection molding process uses a micro-machined structure of an LCP-based electrode array for alignment and fixation. Because the LCPbased electrode array is inserted into a preformed silicone structure, each electrode site fixed is itself at aligned position. The proposed automated manufacturing process can lower the number of fabrication errors and increase productivity, resulting in a low-cost cochlear implant system. These aspects will contribute to an increase in the number of potential cochlear implant recipients even in developing or underdeveloped countries. The wire-based comparator electrode was tested by Rebscher et al. in 2008 (36). According to their article, the electrode showed 0% of trauma, whereas Cochlear Banded and Cochlear Contour showed rates of 37.5% and 38.9%, respectively. The insertion force data were not produced in
the study. However, the mean stiffness was 0.8 g, whereas these values of the Cochlear Banded and Cochlear Contour types were 0.6 and 1.65 g, respectively. Because the insertion trauma is related not only to the insertion force but also to the insertion depth and dimension of electrode array, whether the stiffness has a negative relationship with the insertion trauma is unclear. Nevertheless, the mechanical properties of the cochlear electrode array are still closely related to the degree of insertion safety. Several current major cochlear implant manufacturers, including Cochlear Ltd., Med-El, and Advanced Bionics, have developed electrode arrays, which have low levels of insertion force, providing evidence of atraumatic insertion (38). There are 2 types of electrode arrays, those with a straight electrode and those with a perimodiolar electrode with a stylet. The straight electrode array is inserted into the cochlear along the outer wall of the scala tympani. The initial insertion force is measured at the moment the apical electrode array tip hits the outer wall of the scala tympani and increases as the number of wires contained in the carrier increases. On the other hand, the stiffness of the LCP-based electrode array is determined by the thickness of the LCP film. In this study, 8 to 16 electrode array channels are patterned on a single layer of LCP film, which means that the difference in the stiffness between the base and the apex is not as much as that in the wire-based electrode array, as shown in Figure 5. An effort is being made to optimize the stiffness variation along the length of the LCP-based electrode array to ensure atraumatic insertion into cochlea. Results of this study will be made available in a future publication. The extraction damage should be considered because 4 to 6 reimplantations can be expected with very young recipients throughout their lifetimes (39). The plastic deformation of the straight wire-based electrode is one of the sources of the extraction force. Moreover, a perimodiolar electrode requires extra force to uncoil the preshaped carrier. Because LCP is an elastic material, the LCP-based electrode tends to maintain its original shape. Thus, the
FIG. 8. Result of the human temporal bone study. Among the 5 cases, 2 samples were inserted without trauma (Grade 0), whereas a rupture of the basilar membrane (Grade 2) was observed in 3 cases. The mean insertion depth was 405 degrees from the round window. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
A POLYMER-BASED MULTICHANNEL COCHLEAR ELECTRODE ARRAY extraction force of the LCP-based electrode is less than that of the wire-based electrode array, as shown in Figure 6. The LCP-based electrode has less insertion and extraction force than conventional wire-based electrode arrays. The LCP material is a very strong and stable material when it is used for semiconductor packaging applications or in electronic components. As a film, it becomes flexible along its normal direction, but the mechanical properties of LCP are still stiff along its tangential direction. Thus, sharp edges of neural electrodes based on a rigid film substrate may be traumatic to the surrounding tissue. To protect the surrounding tissue against the edges of filmbased neural electrode arrays, encapsulation of the electrode array must be employed. There are various types of silicone rubber with mechanical properties similar to that of the surrounding tissue implanted in bodies. However, without considering the electrode site openings, it is difficult to open electrode sites, which are buried in silicone rubber. In this case, a polymer ablation process such as laser ablation or dry etching can be applied. However, the laser ablation process can burn the elastomer and electrode, and dry-etching is time consuming. In our work, a micro-molding process without an ablation process is exploited. Self-aligning during the elastomer molding process could precisely locate the electrode array on the opening structure of the upper mold and successfully opened the electrode sites. Conventional cochlear implant systems consist of a wire-based electrode array, a metal package, and a receiver coil. To fabricate an interconnection between the electronics and outer components (the electrode array and coil), a feedthrough should be used in the metal-based package. Most metal-based implant packages are manufactured by brazing and laser welding, both of which involve ultra-high temperatures and high costs. LCP is a thermoplastic polymer, which means that it can be molded, joined, and cut close to its melting temperature. The electrode array in the present study also uses the thermoelasticity to laminate the encapsulation. LCP can also be used as an RF-transparent implantable packaging material, which results in monolithic encapsulation of all components of the implantable unit. We designed a prototype of a LCP-based cochlear implant unit 20 mm (width) 20 mm (length) 2 mm (height) in size, which can be implanted with minimal incision. An accelerated soak test and animal implantation result with the LCP-based cochlear implant system will be published in the near future.
CONCLUSION An LCP-based cochlear electrode array provides suitable electrochemical properties for electrical stimulation. The vertical and horizontal deflection force ratio of the electrode array is comparable to that of a conventional wire-based electrode array. The insertion and extraction forces of the LCP-based electrode array are 47% and 69%, respectively, of the insertion force and extraction force of a wire-based electrode array, which implies that the present LCP-based
1185
electrode array can be inserted with less trauma than a conventional wire-based electrode array. Two cases of human temporal bone study results showed no observable trauma, whereas 3 cases showed a rupture of the basilar membrane. REFERENCES 1. Deiasi R, Russell J. Aqueous degradation of polyimides. J Appl Polym Sci 1971;15:2965Y&. 2. Murray S, Hillman C, Pecht M. Environmental aging and deadhesion of polyimide dielectric films. J Electron Packaging 2004;126: 390Y7. 3. Edell D. Insulating biomaterials N01-NS-2Y2347. NINDS Quarterly Progress Report 2002. 4. Schmidt E, McIntosh J, Bak M. Long-term implants of Parylene-C coated microelectrodes. Med Biol Eng Computing 1988;26: 96Y101. 5. Culbertson EC. A new laminate material for high performance PCBs: liquid crystal polymer copper clad films Electronic Components and Technology Conference, 1995. Proceedings., 45th 1995:520Y3. 6. Dean RN, Weller J, Bozack MJMJ, et al. Realization of ultra fine pitch traces on lcp substrates. Components and Packaging Technologies, IEEE Transactions on 2008;31:315Y21. 7. Lee CJ, Oh SJ, Song JK, et al. Neural signal recording using microelectrode arrays fabricated on liquid crystal polymer material. Materials Science and Engineering: C 2004;24:265Y8. 8. Lee SW, Min KS, Jeong J, et al. Monolithic encapsulation of implantable neuroprosthetic devices using liquid crystal polymers. IEEE Trans Biomed Eng 2011:58:2255Y8. 9. Lee SW, Seo JM, Ha K, et al. Development of microelectrode arrays for artificial retinal implants using liquid crystal polymers. Invest Ophth Vis Sci 2009;50:5859Y66. 10. Scott Corbett JK, Johnson T. Polymer-Based Microelectrode Arrays Materials Research Society Symposium, 2006. 11. Thompson DC, Tentzeris MM, Papapolymerou J. Packaging of MMICs in multilayer LCP substrates. Microwave Wireless Components Letters, IEEE 2006;16:410Y2. 12. Jayaraj K, Farrell B. Liquid crystal polymers and their role in electronic packaging. Adv Microelectron 1998;25:15Y8. 13. Edell DJ, Farrell B. Implantable devices having a liquid crystal polymer substrate, 2003. 14. Wang X, Engel J, Liu C. Liquid crystal polymer (LCP) for MEMS: processes and applications. J Micromech Microeng 2003;13:628. 15. Ketterl JR, Yarno JP, Corbett SS III, et al. Bio-implant and method of making the same, 2004. 16. Corbett SS III, Johnson TJ, Clopton BM, et al. Method of making high contact density electrode array, 2004. 17. Keesara VV, Durand DM, Zorman CA. Fabrication and characterization of flexible, microfabricated neural electrode arrays made from liquid crystal polymer and polynorbornene, Materials Research Society Symposium Proceedings: Cambridge Univ Press, 2006. 18. Wang K, Liu C-C, Durand DM. Flexible nerve stimulation electrode with iridium oxide sputtered on liquid crystal polymer. Biomed Eng IEEE Trans 2009;56:6Y14. 19. Dean R, Weller J, Bozack MJ, et al. Realization of ultra fine pitch traces on LCP substrates. Components Packaging Technol IEEE Trans 2008;31:315Y21. 20. Zou G, Gronqvist H, Starski JP, et al. Characterization of liquid crystal polymer for high frequency system-in-a-package applications. Adv Packaging IEEE Trans 2002;25:503Y8. 21. Kingsley N, Bhattacharya SK, Papapolymerou J. Moisture lifetime testing of RF MEMS switches packaged in liquid crystal polymer. Components Packaging Technol IEEE Trans 2008;31:345Y50. 22. Noll T, Larmouth R. A Low-Cost Near-Hermetic Multichip Module Based on Liquid Crystal Polymer Dielectrics, Proceedings-Spie The International Society for Optical Engineering: Citeseer, 1996: 385Y90. 23. Jauniskis L, Farrell B, Harvey A, et al. LCP PCB-based packaging for high-performance protection. Adv Packaging 2006;10:40Y2. Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
1186
K. S. MIN ET AL.
24. Bhatti PT, Wise KD. A 32-site 4-channel high-density electrode array for a cochlear prosthesis. IEEE J Solid-St Circ 2006;41:2965Y73. 25. Wise KD, Bhatti PT, Wang JB, et al. High-density cochlear implants with position sensing and control. Hear Res 2008;242: 22Y30. 26. An SK, Park SI, Jun SB, et al. Design for a simplified cochlear implant system. IEEE Trans Biomed Eng 2007;54:973Y82. 27. Bell TE, Wise KD, Anderson DJ. A flexible micromachined electrode array for a cochlear prosthesis. Sensors Actuators A: Physical 1998;66:63Y9. 28. Bonham RS B, Corbett S, Johnson T, et al. Reducing Channel Interaction through Multi-Polar Stimulation: Combining Tripolar and Monopolar Methods with High-Density Arrays Conference on Implantable Auditory Prostheses, Asilomar, CA, 2005. 29. Byers CLV, Loeb Gerald E, Merzenich Michael M, Rebscher Stephen J. Intracochlear electrode array. United States: The Regents of the University of California (Berkeley, CA), 1989. 30. Byers CLV, Loeb Gerald E, Merzenich Michael M, Rebscher Stephen J. Method for Making an Intracochlear Electrode Array. Berkeley, CA: United States: Regents of the University of California, 1987. 31. Taylor GN, Wolf TM. Oxygen plasma removal of thin polymerfilms. Polym Eng Sci 1980;20:1087Y92.
32. Schmidt EM, Bak MJ, Christensen P. Laser exposure of Parylene-C insulated microelectrodes. J Neurosci Methods 1995;62:89Y92. 33. Gantz BJ, Turner C, Gfeller KE. Acoustic plus electric speech processing: preliminary results of a multicenter clinical trial of the iowa/nucleus hybrid implant. Audiol Neurotol 2006;11:63Y8. 34. James C, Albegger K, Battmer R, et al. Preservation of residual hearing with cochlear implantation: How and why. Acta Otolaryngol 2005;125:481Y91. 35. Skarzynski H, Podskarbi-Fayette R. A new cochlear implant electrode design for preservation of residual hearing: a temporal bone study. Acta Otolaryngol 2010;130:435Y42. 36. Rebscher SJ, Hetherington A, Bonham B, et al. Considerations for design of future cochlear implant electrode arrays: electrode array stiffness, size, and depth of insertion. J Rehabil Res Dev 2008;45:731Y47. 37. Eshraghi AA, Yang NW, Balkany TJ. Comparative study of cochlear damage with three perimodiolar electrode designs. Laryngoscope 2003;113:415Y9. 38. Roland JT. A model for cochlear implant electrode insertion and force evaluation: results with a new electrode design and insertion technique. Laryngoscope 2005;115:1325Y39. 39. Jolly C, Mueller J, Helbig S, et al. New trends with cochlear implant electrodes. Otology Jpn 2010;20:239Y46.
Otology & Neurotology, Vol. 35, No. 7, 2014
Copyright © 2014 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
