**4.1 Dual-layered MEMS coil**

In this section, a planar coil is designed, which features: 1) dual-layer thin-film metal conductive wires sandwiched between multiple layers of Parylene C, and 2) interconnections between two adjacent layers that are formed by filling the Parylene through holes with PVD metal. Fig. 6 shows the microscope images of a fabricated coil and its interconnection via. This coil has totally 10 turns of wires made of approximately 2000 Å multiple layers of Ti/Au/Ti metallization. Titanium serves as an adhesion promoter to improve the bonding strength between gold and Parylene C. The device has overall dimensions of ~ 9.5 mm in outer diameter, ~ 5 mm in inner diameter, and ~ 11 µm in thickness, limited by the lens capsule size of the human eyes. The interconnection via occupies an area of ~ 0.06 mm2 with a contact resistance of less than 1 Ω, which can be negligible compared with the total coil ESR. The device is proven to be very flexible and foldable (Fig. 7), easing the procedure of surgical insertion and lessening physical damages in the region of implantation.

Fig. 5. Fabrication process flow of a dual-metal-layer Parylene-based MEMS structure.

(Tai et al., 2006).

**4.1 Dual-layered MEMS coil** 

in the region of implantation.

**4. Coil designs and fabrication results** 

For implantable devices, our Parylene-metal thin film technology has several unique advantages compared with conventional semiconductor-based microfabrication technologies. Using biocompatible Parylene directly as the actual substrate greatly simplifies the device integration and packaging procedures. Devices fabricated in this way are very flexible and foldable so that they can be implanted through small surgical incisions, allowing wounds to heal quickly. Moreover, the metal lines are completely padded by the Parylene material, and can therefore withstand repeated bending during surgical handling. Finally, a post-fabrication heat-molding process has been developed to modify the skins into various shapes that match the curvatures of the target implant areas

In this section, a planar coil is designed, which features: 1) dual-layer thin-film metal conductive wires sandwiched between multiple layers of Parylene C, and 2) interconnections between two adjacent layers that are formed by filling the Parylene through holes with PVD metal. Fig. 6 shows the microscope images of a fabricated coil and its interconnection via. This coil has totally 10 turns of wires made of approximately 2000 Å multiple layers of Ti/Au/Ti metallization. Titanium serves as an adhesion promoter to improve the bonding strength between gold and Parylene C. The device has overall dimensions of ~ 9.5 mm in outer diameter, ~ 5 mm in inner diameter, and ~ 11 µm in thickness, limited by the lens capsule size of the human eyes. The interconnection via occupies an area of ~ 0.06 mm2 with a contact resistance of less than 1 Ω, which can be negligible compared with the total coil ESR. The device is proven to be very flexible and foldable (Fig. 7), easing the procedure of surgical insertion and lessening physical damages

Fig. 6. (a) A fabricated dual-metal-layer coil sitting on a penny. (b) The microscope image shows the interconnection via between two metal layers. (Li et al., @ 2005 IEEE)

Fig. 7. Demonstration of the coil's flexibility and foldability. (Li et al., @ 2005 IEEE)

The electrical properties of the fabricated coil are characterized experimentally. Recall equation (7) in Section 2, by setting the derivation of the real part to zero and equating the imaginary part to zero, the self-inductance (Ls) and the parasitic capacitance (Cs) can be extracted using equations (11) and (12), where ω0 is defined as the frequency at which the real part of the impedance is maximum, and ωz is the zero-reactance frequency at which the imaginary part of the impedance is zero (Wu, 2003).

$$\mathcal{L}\_{\rm S} = \frac{\mathcal{R}\_{\rm s}}{\sqrt{2(a\_0^2 - a\_x^2)}},\tag{11}$$

$$\mathcal{C}\_{\rm S} = \frac{\sqrt{2(\alpha\_0^2 - \alpha\_\mathbf{z}^2)}}{\mathcal{R}\_\mathbf{s} (2\alpha\_0^2 - \alpha\_\mathbf{z}^2)}. \tag{12}$$

For the coil in Fig. 6, the ESR (Rs) is measured to be around 72 Ω and the resistivity of ebeam deposited gold is calculated to be around 2.25×10-6 Ω·cm. This number agrees with the resistivity of bulk gold (2.2×10-6 Ω·cm), implying that the E-beam evaporated metal is voidfree. The coil impedance is swept with an HP 4192A LF impedance analyzer over a frequency range from 5 Hz to 13 MHz. From the impedance versus frequency curves (Fig. 8), f0 and fz can be read with values of 7.5 MHz and 3.3 MHz respectively. Knowing Rs, ω0, and ωz, the coil self-inductance and capacitance are therefore calculated as Ls = 1.19 µH and

Implantable Parylene MEMS RF Coil for Epiretinal Prostheses 13

Experiments have been done at UCSC to study the feasibility of using MEMS coils as the receiver coil for the current inductive link design. A rough estimate is that, in the worst case, a minimal Q factor of 10 will be needed in order to deliver ~100 mW for chip operation and stimulation. From the theoretical analysis herein, it is known that the Q factor can be enhanced by increasing the metal thickness and/or the number of metal layers. However, ebeam evaporated metals are usually limited in film thickness due to high process cost. Electroplated and sputtered metals can be thicker alternatives but their qualities, such as density and conductivity, are typically not as good as evaporated metals. This problem becomes more serious especially when devices are implanted inside harsh biological environments. From the device design aspect, increasing metal layers is more practical for the Q factor enhancement of MEMS coils, thus a fold-and-bond technology emerges as a

In the concept of fold-and-bond technology (Fig. 10), two or more thin-film planar spiral coil segments are fabricated from the same batch so that each segment has identical selfinductance and ESR, denoted by Ls and Rs, respectively. A new coil can then be formed by stacking *n* segments together in either parallel or series connections. For parallel stacking, particularly, the new coil will have identical inductance but *n*-times larger in equivalent metal thickness, resulting in an *n*-times lower ESR. As for series stacking, the resistance remains the same while the inductance increases by *n*-times because of the mutual inductance between adjacent layers. According to the definition of the Q factor, both approaches can achieve an *n*-times Q factor enhancement. In this section, the series stacking

configuration is used to demonstrate the technology concept.

Fig. 10. Concept of the fold-and-bond technology for Q factor enhancement.

Fold-and-bond coil's fabrication involves the dual-metal-layer Parylene/metal skin realization and a post-fabrication thermal bonding process (Li et al., 2008). The Parylenemetal skin with two buried layers of metal is first fabricated in the same manner as described in Fig. 5, in which one layer of metal is used to form the conductive wires of the coil, while the other layers is used to make the interconnections between the layers. This thin film skin can be folded and stacked into multiple layers because of the flexibility of Parylene C. While hand alignment under an optical microscope is used at the current stage, special alignment jigs can be custom designed in the future to achieve precise

A coil with one fold is depicted for representation.

good candidate.

**4.2 Fold-and-bond coil** 

Cs = 201 pF. The theoretical numbers are also calculated using the abovementioned equations, and the fitting curves are plotted in Fig. 8, in comparison with the measured curves. The experimental data matches the theoretical calculations closely, with deviations of less than 19%. These errors may be attributed to the simplification of the 3-element model as well as interferences from the measurement instruments. The Q factor of the coil is obtained to be approximately 0.1 at the target frequency of 1 MHz, as expected from the design.

Fig. 8. Impedance measurement and curve fitting using the 3-element model: (a) Imaginary part; (b) Real part. (Red curves correspond to theoretical parameters of the fabricated coil: Ls = 1.0 µH, Rs = 67 Ω and Cs = 183 pF.)

The data and power transfer performances have also been verified using a custom designed data link at the University of California, Santa Cruz (UCSC). The testing waveforms are shown in Fig. 9, where the blue curve represents the data driving signal on the primary stage, the green curve represents the voltage across the primary coil, and the purple curve represents the receiving voltage across the secondary coil.

Fig. 9. Inductive coupling test waveforms: (a) received signal is 25 mV peak to peak; (b) received signal is 15 mV peak to peak.

While successful data transmission through our coils has been demonstrated, it is noted that this device has no driving capability due to its small Q factor (~ 0.1), meaning that the power cannot be delivered to the load. Therefore, enhancing coil's Q factor to achieve a higher power transfer efficiency is crucial for designing the next generation of coils. Experiments have been done at UCSC to study the feasibility of using MEMS coils as the receiver coil for the current inductive link design. A rough estimate is that, in the worst case, a minimal Q factor of 10 will be needed in order to deliver ~100 mW for chip operation and stimulation. From the theoretical analysis herein, it is known that the Q factor can be enhanced by increasing the metal thickness and/or the number of metal layers. However, ebeam evaporated metals are usually limited in film thickness due to high process cost. Electroplated and sputtered metals can be thicker alternatives but their qualities, such as density and conductivity, are typically not as good as evaporated metals. This problem becomes more serious especially when devices are implanted inside harsh biological environments. From the device design aspect, increasing metal layers is more practical for the Q factor enhancement of MEMS coils, thus a fold-and-bond technology emerges as a good candidate.
