Design considerations

One of the attractive features of the Michigan and other planar photoen-graved probes is the ability to customize design for specific experiments. The substrate can have any two-dimensional shape with single or multiple shanks, electrode sites can be of any surface area and can be placed anywhere along the shank(s) at any spacing, tips can be made very sharp or blunt, and features such as holes and barbs can be included for special applications such as sieve electrodes.74-76 The NIH NCRR-sponsored University of Michigan Center for Neural Communication Technology (CNCT) has been providing probes to investigators since 1994. At the time of this writing, over 150 designs have been fabricated through the CNCT, a subset of which comprise a catalog of basic designs. A number of scientific papers and presentations, listed at, have resulted from use of these probes. The design freedom offered by this technology makes the devices attractive to researchers in a broad range of applications; however, there are some practical and process-enforced limitations on probe features that must be considered when selecting or designing a probe. Shank length When selecting or designing a probe, a first characteristic an investigator typically considers is shank length. This is the length of the device that will be inserted into tissue or the maximum depth of the intended target. While shank length can be considered a process-limited feature due to the maximum limit imposed by the size of the wafer, it is actually more practically limited as defined by stiffness and strength. Typical 15-| m-thick, 100-| m-wide probe shanks longer than about 6 mm may bend during insertion, causing the target to be missed. As described by Najafi and Hetke,77 when choosing a shank length one must consider the buckling load, which defines the stiffness of the probe as it is pressed against the tissue, and the maximum stress at the bottom of the probe shank, which defines the strength of the probe. The buckling load is proportional to t3W/L,2 and the maximum stress is proportional to t/L.2 Therefore, to design a probe that is stiff and strong with the given thickness t = 15 |im, the substrate width (W) should be

Figure 7.6 A sharp probe tip formed using a combination of shallow and deep boron diffusions. Probes with these sharp tips have been used to penetrate tough structures such as peripheral nerve and spinal cord.

increased, and the length (L) should be the minimum necessary to reach the intended target.

Insertion properties can also be improved by increasing the sharpness of the tip.78 As described above, this can be achieved by using a shallow boron diffusion at the probe tip which results in a structure that is no more than 1 to 2 ||m in diameter. In addition to the shallow diffusion, a very sharp taper angle (<10 degrees) can be used to improve penetration characteristics. Figure 7.6 shows a probe tip that is shallow diffused and has a taper angle of 18 degrees that has successfully penetrated sciatic and auditory nerves in cat, structures which are not penetrable with standard deep diffused tips. Shank width Taking into consideration the above discussion on shank length and the relationship of shank width to the strength and stiffness of long shanks, one should still try to minimize the width of the shank to make the device as noninvasive as possible. Shank width is a process enforced limitation that is controlled by the diffusion mask and lateral diffusion. As the width of the diffusion opening decreases, lateral diffusion of boron underneath the mask eventually limits the minimum achieveable shank width for a given substrate thickness.79 Using an EDP etch alone, the minimum shank width achievable for a standard deep-diffused 15-|im-thick shank is about 15 |im because, as the mask width decreases, so does the thickness. Narrow "scaled" probes can be realized, however, using shallow-diffusion and/or an additional masking and etching step. To do this, deep RIE is used to etch into the silicon substrate and trim off the lateral diffusion, resulting in a narrower device than would be realizable using the EDP etch alone. Scaled probes have been fabricated with shanks as narrow as 5 |im. While 1.5-mm-long, 5-|im-thick scaled shanks have successfully penetrated the guinea pig pia mater and recorded from cortex, the mechanical practicality of these tiny devices has yet to be proven.

Another issue related to shank width is the interconnect lead width and spacing. Although submicron feature sizes are standard in industry, one must consider electrical crosstalk between very close leads. When more electrodes are to be accommodated on a narrower shank, the width of interconnect lines and the spacing between the lines must be reduced, resulting in an increased coupling capacitance. As described by Najafi et al.,79 for probes with line widths and spacing as small as 1 |im, the crosstalk approaches 1%. This is acceptable for recording devices, as effects will be negligible compared to background noise. Even with feature sizes as small as 0.25 | m, the crosstalk is still less than 4%. Substrate thickness Substrate thickness is a feature that is of importance when it comes to insertion into tough tissue or into deep structures. The standard Michigan probe process is capable of producing probes 5 to 15 | m in thickness based on the time and temperature of the boron diffusion and hence the depth of the etch-stop. Probes 15 |im thick and 5 mm long and tapered from 15 to 120 |im in width are capable of penetrating the pia with minimal bending in most preparations. If the device is much longer than this, however, the buckling force is lower. If a very long shank is absolutely required by the application, a stronger, stiffer device can be achieved by forming a "box-beam" substrate. This type of device uses the same fabrication process as the Michigan chemical delivery probe80 to form an open channel within the silicon substrate (Figure 7.7). This channel results in a box-beam structure that is inherently stiffer and stronger than a normal beam because the second moment of its cross-section is larger. This structure is being investigated for its efficacy in penetrating deeper and/or tougher structures, including the inferior colliculus, spinal cord, and cochlear nucleus in cats.

Figure 7.7 A probe with three channels, spaced 20 |im apart, buried within its substrate. While these probes are being developed for chemical delivery, they are also useful for applications where a stiffer device is necessary. (From Chen, J. et al., IEEE Trans. Biomed. Eng., 44, 760, 1997. With permission.)

Figure 7.7 A probe with three channels, spaced 20 |im apart, buried within its substrate. While these probes are being developed for chemical delivery, they are also useful for applications where a stiffer device is necessary. (From Chen, J. et al., IEEE Trans. Biomed. Eng., 44, 760, 1997. With permission.) Site spacing

Another characteristic one considers when choosing a probe design is the center-to-center spacing between the recording sites. This feature is basically unlimited; the spacing between sites can be as small as a few microns to as large as several hundred microns or more. The choice will depend on the location of intended use as well as the type of data to be collected. For example, for purposes of spike sorting, signal improvement, and/or cell imaging, correlated signals are desired on more than one site. A general rule of thumb for these types of recordings is that the sites should be spaced at 50 |im or less.21 Close-spaced sites can be in the form of a linear array, or in a tetrode configuration as shown in Figure 7.8. Sites that are spaced 100 |im apart will show some correlated as well as uncorrelated activity. When the sites are spaced at 200 |im or more, little or no coherency occurs between sites.21 Arrays with this larger spacing, for example, can be used to investigate the layered organizations of cortical tissue. Site area

The choice of site area for recording is an electrical consideration as it is related to thermal noise and the ability to record low-level signals and hence influences the achievable signal-to-noise ratio. There is a trade-off when choosing a site area. Although perhaps more selective in the signals it detects, a smaller site imposes a higher impedance, causing signal attenuation and added noise. In contrast, a larger site offers a lower impedance at the expense of being less selective and therefore picks up signals from a larger population of cells. While site areas on Michigan probes ranging from 70 to 4000 |im2

Figure 7.8 A two-shank tetrode array; sites are spaced at 25 |im on the diagonal, and shanks are spaced at 150 |im.

have successfully recorded neural spikes, the current typical recording site on a Michigan probe is just under 200 |im2. The average 1-kHz impedance on a 200-|im2 iridium site is around 2 Mfl and, with appropriate amplifiers, the baseline noise level is about 15 |VRMS. This site area has evolved over many years as a "best" choice for maximizing signal-to-noise ratio and, in fact, is still evolving. It is likely that a single best choice for all brain regions does not exist due to variations in cell size and density. A systematic study is currently underway to investigate the recording quality of different site areas in a variety of brain structures.

If the electrode site is to be used for stimulation, a larger site size may be required to safely deliver the charge required for the application. In addition, the charge capacity of the site can be increased by an order of magnitude by forming iridium oxide through electrochemical activation.70,72 This procedure involves cycling the electrode potential between positive and negative limits, which results in a porous, hydrous, multilayer oxide film that has a high charge capacity and is exceptionally resistant to dissolution and corrosion during stimulation.

Was this article helpful?

0 0
Peripheral Neuropathy Natural Treatment Options

Peripheral Neuropathy Natural Treatment Options

This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.

Get My Free Ebook

Post a comment