Neuromodulation: Technology at the Neural Interface Received: November 8, 2014
Revised: December 23, 2014
Accepted: January 27, 2015
(onlinelibrary.wiley.com) DOI: 10.1111/ner.12285
Improved Poly(3,4-Ethylenedioxythiophene) (PEDOT) for Neural Stimulation Himadri Shekhar Mandal, PhD*; Jemika Shrestha Kastee, BS*; Daniel Glenn McHail, BS†; Judith Faye Rubinson, PhD‡; Joseph Jewell Pancrazio, PhD*; Theodore Constantine Dumas, PhD† Objective: This study compares the stability of three variations of the conductive polymer poly(3,4-ethylenedioxythiophene) or PEDOT for neural micro-stimulation under both in vitro and in vivo conditions. We examined PEDOT films deposited with counterions tetrafluoroborate (TFB) and poly(styrenesulfonate) (PSS), and PEDOT:PSS combined with carbon nanotubes (CNTs). Methods: For the in vitro stability evaluation, implantable micro-wires were coated with the polymers, placed in a vial containing phosphate buffered saline (PBS) under accelerated aging conditions (60°C), and current pulses were applied. The resulting voltage profile was monitored over time. Following the same polymer deposition protocol, chronic neural micro-probes were modified and implanted in the motor cortex of two rats for the in vivo stability comparison. Similar stimulating current pulses were applied and the output voltage was examined. The electrochemical impedance spectroscopic (EIS) data were also recorded and fit to an equivalent circuit model that incorporates and quantifies the time-dependent polymer degradation and impedance associated with tissue surrounding each micro-electrode site. Results: Both in vitro and in vivo voltage output profiles show relatively stable behavior for the PEDOT:TFB modified microelectrodes compared to the PEDOT:PSS and CNT:PEDOT:PSS modified ones. EIS modeling demonstrates that the time-dependent increase in the polymeric resistance is roughly similar to the rise in the respective voltage output in vivo and indicates that the polymeric stability and conductivity, rather than the impedance due to the tissue response, is the primary factor determining the output voltage profile. It was also noted that the number of electrodes showing unit activity post-surgery did not decay for PEDOT:TFB as was the case for PEDOT:PSS and CNT:PEDOT:PSS. Conclusions: PEDOT:TFB may be an enabling material for achieving long lasting micro-stimulation and recording. Keywords: Conductive polymer, neural stimulation, PEDOT Conflict of Interest: The authors reported no conflict of interest.
INTRODUCTION
www.neuromodulationjournal.com
Address correspondence to: Joseph J. Pancrazio, PhD, Dept. of Bioengineering, George Mason University, 4400 University Dr., MS 1G5, Fairfax, VA 22030, USA. Email: [email protected] * Department of Bioengineering, George Mason University, Fairfax, VA, USA; † Department of Molecular Neuroscience, The Krasnow Institute for Advanced Study, George Mason University, Fairfax, VA, USA; and ‡ Department of Chemistry, Georgetown University, Washington, DC, USA For more information on author guidelines, an explanation of our peer review process, and conflict of interest informed consent policies, please go to http:// www.wiley.com/bw/submit.asp?ref=1094-7159&site=1 Sources of financial support: The Defense Advanced Research Projects Agency (DARPA) MTO sponsored this work under the auspices of Dr. Jack Judy through the Space and Naval Warfare Systems Center, Pacific Grant/Contract No. N66001-121-4026.
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
657
Electrical stimulation using implantable electrodes (IEs) is critical in neuroscience for its relevance in a wide range of brain machine interface applications (1–4). One of the major challenges for current IEs is to lower the dimensions of the IE in order to achieve microstimulation of spatially restricted regions of the nervous system, while inducing minimal tissue response (5–11). But, as the size is decreased, the charge injection capability of the electrode, which is primarily capacitive in nature and proportional to the electrochemically active surface area at the interface, becomes smaller and insufficient for effective neuronal stimulation. This scenario can be improved by increasing the surface roughness, thus expanding the effective surface area without increasing the geometric area (12–14). Unfortunately, metallic roughened surfaces are chemically reactive and show unreliable electrochemical properties over time (14–16). To address this issue, modification of micro-electrodes with conductive polymers (CPs) has been suggested to be a viable route and currently is an active area of research (17–21). CPs are, in general, porous in morphology, have orders of magnitude larger electrochemically active surface area compared to metallic counterparts of the same overall dimensions and are promising materials for developing micro-electrodes capable of large charge injection. However,
the long-term stability of CPs, especially in physiological conditions, is a serious concern for use of these materials for chronic applications. Recently, we reported that the conductivity and stability of the most commonly utilized CP poly(3,4-ethylenedioxythiophene) or PEDOT can be significantly increased by substitution of the widely used, multivalent and bulky counter-anion poly(styrenesulfonate), or PSS, by the smaller monovalent tetrafluoroborate (TFB) anion (Fig. 1)
MANDAL ET AL. The applied charge density was 0.56 mC/cm2 (assuming the geometric area of the chronic electrode site to be 177 μm2) which is below the charge-injection limit for PEDOT modified electrodes (19). Note that the temperature used here is in compliance with methods based on the ASTM F1980-07 “Standard Guide for Accelerated Aging of Sterile Barrier Systems for Medical Devices” (23). The aging factor at 60°C is at least 4 assuming the body temperature (37°C) as the baseline. Assuming a conservative Q10 for polymers, the reaction rate doubles for each 10°C increase (24). Figure 1. Chemical structures of PEDOT, PSS and TFB.
during the electro-deposition of the polymer (22). We also demonstrated that this particular variation of PEDOT, i.e. PEDOT:TFB, is even more stable than carbon nanotube (CNT)-incorporated PEDOT (CNT:PEDOT:PSS) which has been suggested to have enhanced performances due to the CNTs’ unique electromechanical and structural properties. In the present study, we examined and compared the stability of PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS in electrical stimulation based experiments under both in vitro and in vivo conditions. We also investigated their in vivo neuronal recording capability. Based on the data, PEDOT:TFB demonstrates better performances for both stimulation and recording, and could prove useful in fabricating long-lasting micro-scale brain-machine interface (BMI) devices.
MATERIALS AND METHODS Materials Gold(III) chloride hydrate, 3,4-ethylenedioxylthiophene (EDOT), PSS, acetonitrile and tetrabutylammonium tetrafluoroborate were obtained from Sigma-Aldrich (St. Louis, MO, USA). PBS was obtained from Mediatech Inc. (Manassas, VA, USA). The CNT dispersion was purchased from CheapTubes.com (Brattleboro, VT, USA). Pt/Ir microwires (PI20030.1A5) and chronic probes (A1x16-5 mm-100– 177-CM16) were purchased from MicroProbes for Life Sciences (Gaithersburg, MD, USA) and NeuroNexus Technologies (Ann Arbor, MI, USA), respectively.
Preparation of Polymer Modified Implantable Microwires and Chronic Neural Probes Details of the modification process are described elsewhere (22). In brief, Pt/Ir implantable microwires were first electroplated with a thin layer of gold. This involved cycling the wires between 0 and −1.5 V three times in a solution of HAuCl4 (5 mM) at a scan rate of 1V/s. The polymers (PEDOT:TFB, PEDOT:PSS and CNT:PEDOT:PSS) were then deposited using a CHI 660D potentiostat (CH Instruments, Inc., Austin, TX, USA). After the polymer deposition, electrodes were cycled between 0 and 0.5V (50 cycles, cyclic voltammetry) in 1x PBS. Chronic probes containing Ir electrodes (A1x16-5 mm-100–177-CM16 from NeuroNexus) for in vivo studies were modified using similar gold and polymer deposition protocols.
658
Accelerated Aging Study In Vitro The polymer modified Au coated Pt/Ir microwires (N = 3 each for PEDOT:PSS, CNT:PEDOT:PSS, and PEDOT:TFB) were placed in a vial containing PBS at 60°C, and 1 μA, 1 ms, bi-phasic, 5 pulses/train, 83.3 Hz, anodic first current pulses were applied using a potentiostat (CHI 660D, CH Instruments, Inc., Austin, TX, USA) equipped with a multiplexer (CHI 684, CH Instruments, Inc., Austin, TX, USA). www.neuromodulationjournal.com
In Vivo Stimulation and Recording Out of the 16 microelectrode sites in one NeuroNexus probe, the first set of 4 consecutive sites (from the tip) was modified with PEDOT:TFB, the second set of 4 with PEDOT:PSS, the third set of 4 was unmodified and the remaining set of 4 sites was with CNT:PEDOT:PSS. A second NeuroNexus probe was also modified where the location of the polymer was shifted by 4 sites (from the tip of the probe) to reduce the degree of bias due to position within the cortex. These probes were implanted in two female Long Evans rats (300–370 g), as described previously (22). Current pulses similar to those used for the in vitro measurements were applied. During the data (stimulation) collection, the animal was anesthetized in an induction chamber with 1 L/min oxygen and 5% isoflurane. However, the neuronal signals were recorded in freely moving rats over a recording session of 10 minutes. Data were acquired periodically during a 50-day period using a Cerebus system (Blackrock Microsystems, Salt Lake City, UT, USA). The details of data collection and processing are described in our previous publication (22). The statistical analysis for the differential expression of unit recordings as a function of coatings was performed using the Z Test Calculator for 2 population proportions (http://www.socscistatistics.com/). Two-tailed hypothesis was adopted and the significance level was set at 0.05. All animal procedures comply with the National Institute of Health guidelines and were approved by the Institutional Animal Care and Use Committee at George Mason University, Fairfax, VA.
Electrochemical Impedance Spectroscopy (EIS) and Modeling The impedance was measured at the open circuit potential from 0.1 Hz to 1 MHz, with a peak-to-peak amplitude of the sinusoidal voltage of 20 mV. EIS data was fit with ZsimpWin (Princeton Applied Research, Oak Ridge, TN, USA).
RESULTS In Vitro Stimulation Studies Figure 2b–d show representative in vitro output voltage profiles from PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS modified microwires after applying current pulses (parameters are shown in Fig. 2a) from day 1 and day 40. Anodic voltage outputs are presented. PEDOT:TFB modified electrodes show slower change in the output over time compared to the PEDOT:PSS and CNT:PEDOT:PSS modified electrodes (Fig. 2e).
In Vivo Stimulation, EIS and Recording The in vivo output voltage profiles from PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS modified sites in NeuroNexus (NNx) probes are shown in Figure 3. However, unlike the in vitro voltage output, the change over time is not gradual. For all three polymer sites, the output voltage was high on day 7, low on day 15 and relatively
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
IMPROVED PEDOT FOR NEURAL STIMULATION
Figure 2. Current pulse recordings in vitro. Parameters used for bi-phasic current stimulations (a). Output voltages elicited by current pulses delivered through PEDOT:TFB (b), PEDOT:PSS (c), and CNT:PEDOT:PSS (d) modified microwires. In each figure, black and red data are for day 1 and day 40, respectively. Anodic (e) output voltages over time: PEDOT:TFB (black rectangle), PEDOT:PSS (blue circle), and CNT:PEDOT:PSS (green triangle).
www.neuromodulationjournal.com
an electrical equivalent circuit model (Fig. 4b) based on our previously published work (22) on polymer degradation over time in vitro. Here, we included two additional parameters to model the tissue response at the electrode-solution interface which is very common for implanted probes (25,26). In the model, the charge
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
659
constant onwards (Fig. 3e). Overall, PEDOT:TFB modified sites show lower output voltage compared to the PEDOT:PSS and CNT:PEDOT:PSS modified sites. Figure 4c–e show typical in vivo EIS data for PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS modified sites. The data were fit to
MANDAL ET AL.
Figure 3. An optical image of PEDOT:TFB modified sites (first four dark regions from the tip) in a NNx probe (a). Output voltages elicited by current pulses delivered through PEDOT:TFB (b), PEDOT:PSS (c), and CNT:PEDOT:PSS (d) modified sites. In each figure, black and red data are for day 1 and 50, respectively. Anodic (e) output voltages over time: PEDOT:TFB (black rectangle), PEDOT:PSS (blue circle), and CNT:PEDOT:PSS (green triangle).
660
transfer resistance and the associated double layer capacitance across the rough polymeric interface are designated as Rpoly and Qpoly, respectively. Cd is the double layer capacitance due to the defect sites of the polymer film. Rbio and Cbio are the polarization resistance and associated capacitance of the tissue, respectively. We also recorded the pre-implant EIS data for the probes in PBS and fit them to the in vitro model (Fig. 4a) for comparison. Impedance increases for all polymer modified sites after implant (Fig. 4c– www.neuromodulationjournal.com
e). This is consistent with our previous observation and may be related, at least in part, to the inflammatory response and/or protein adsorption on the electrode sites (22,25). All the fit parameters are listed in Table 1. The polymeric resistance Rpoly and the resistance of the tissue Rbio increase over time for all the polymer modified sites. However, while Rpoly rose by factors of 1.4–2.8, the major difference between the coatings was the relatively modest change by a factor of 2 in Rbio for PEDOT:TFB versus
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
IMPROVED PEDOT FOR NEURAL STIMULATION
Figure 4. In vitro (a) and in vivo (b) EIS equivalent circuit models. Experimental impedance data and the fit (red line) for PEDOT:TFB (c), PEDOT:PSS (d), and CNT:PEDOT:PSS (e) modified sites in one of the chronic probes. In each figure, black rectangle (pre-implant EIS in 1X PBS), blue triangle (day 1 after surgery), and green circles (day 50 after surgery). Table 1. Fit Parameters for EIS Data From All Coated Microelectrodes. Pre-implant Cd (F/cm2) Rpoly (Ohm-cm2) Qpoly (S-sec1/2/cm2) n (0 < n < 1) Rbio (Ohm-cm2) Cbio (F/cm2)
4.51E-05 (1.05E-06) 5.34E-02 (9.57E-04) 1.72E-01 (3.91E-02) 9.87E-01 (1.72E-03)
PEDOT:TFB Day 1
Day 50
Pre-implant
3.82E-05 (3.31E-06) 1.19E-01 (2.26E-02) 5.47E-02 (1.44E-02) 8.54E-01 (2.85E-02) 1.21E-01 (4.51E-02) 1.05E-01 (1.91E-02)
3.24E-05 (1.75E-06) 1.71E-01 (1.93E-02) 3.12E-02 (1.18E-02) 8.11E-01 (1.99E-02) 2.49E-01 (4.92E-02) 6.95E-02 (2.89E-02)
5.61E-05 (2.61E-06) 7.79E-02 (4.51E-03) 3.72E-02 (5.71E-03) 8.66E-01 (7.12E-03)
PEDOT:PSS Day 1
Day 50
Pre-implant
3.01E-05 (7.93E-06) 4.17E-01 (1.89E-01) 1.38E-02 (4.02E-03) 8.16E-01 (2.64E-02) 2.98E-01 (7.81E-02) 2.99E-02 (6.14E-03)
2.68E-05 (4.86E-06) 1.06E+00 (7.59E-01) 9.87E-03 (2.74E-03) 5.64E-01 (5.80E-02) 1.91E+01 (1.81E+01) 1.46E-02 (7.76E-03)
3.91E-04 (3.36E-04) 9.37E-02 (2.05E-02) 2.57E-02 (2.52E-03) 8.54E-01 (2.02E-02)
CNT:PEDOT:PSS Day 1
Day 50
2.96E-05 (7.61E-06) 2.87E-01 (6.46E-02) 8.54E-03 (4.54E-03) 8.02E-01 (2.66E-02) 4.10E-01 (6.04E-02) 2.31E-02 (8.20E-03)
2.61E-05 (1.39E-06) 8.22E-01 (7.08E-02) 8.50E-03 (2.25E-03) 5.89E-01 (4.98E-02) 3.05E+01 (2.45E+01) 1.40E-02 (6.00E-03)
Values in parentheses correspond to the respective standard error of the mean (SEM).
PEDOT:PSS and CNT:PEDOT:PSS, which showed increases by factors of 65–75. We also performed in vivo neuronal recordings in awake and behaving animals with the polymer modified electrodes. Figure 5a–c show representative single units from day 2. The peakto-peak amplitude of single units for PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS coated electrodes is 87.49 ± 7.86 μV, 125.42 ± 33.80 μV, 122.75 ± 12.73 μV, respectively (mean ± SEM). The corresponding noise levels are 8.03 ± 0.70 μV (PEDOT:TFB), 9.90 ± 0.93 μV (PEDOT:PSS), and 12.28 ± 1.91 μV (CNT:PEDOT:PSS). Figure 5d shows the time-dependent activity profile for the modified sites. Although initially, the proportion of coated electrodes that show activity is not statistically significant, a clear difference (P < 0.05) is observed over time (by test of proportions, p < 0.05). PEDOT:TFB modified sites show better neuronal recording capability compared to the PEDOT:PSS and CNT:PEDOT:PSS modified sites.
DISCUSSION
www.neuromodulationjournal.com
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
661
In both in vitro and in vivo stimulation experiments, the change in the output voltage profile is reduced and remains within the “safe
water window” for the PEDOT:TFB modified electrode sites compared to the PEDOT:PSS and CNT:PEDOT:PSS coated sites. This is in line with our previous work where we reported significantly larger conductivity and thermal stability of PEDOT:TFB coated microelectrodes compared to those coated with PEDOT:PSS and CNT:PEDOT:PSS (22). Note that we continued collecting the data as long as the output was within the “safe water window” to avoid confounding variables such as the formation/adsorption of H2, O2, and free radicals, which may damage/delaminate the polymer film, promote inflammation of the surrounding tissue, and complicate the data interpretation. The increase in Rpoly over time for all the polymer modified sites may be related to the time-dependent aging of the polymers (22). To explain the origin of this Rpoly in more details, we refer to our previous work where we showed polymeric aging involves the collapse of the porous structure. When the structure is highly porous and easily accessible to the supporting electrolytes, the delocalized and already charged PEDOT backbone can be stabilized by these supporting electrolytes during the transfer of the polaronic charge carriers and the formation of Cpoly at the polymer-solution interface. When the porosity is reduced, the PEDOT backbone becomes less accessible to the electrolytes and thereby, Rpoly increases. This is
MANDAL ET AL.
Figure 5. Representative units from (a) PEDOT:TFB, (b) PEDOT:PSS, and (c) CNT:PEDOT:PSS coated electrode sites; (d) Activity profile over time: PEDOT:TFB (gray), PEDOT:PSS (blue), and CNT:PEDOT:PSS (green).
662
also consistent also with the relatively uniform decrease in the corresponding Qpoly for all the polymer modified electrodes. The smaller increase in Rpoly for PEDOT:TFB compared to PEDOT:PSS and CNT:PEDOT:PSS indicates that the former has higher stability. This could be due to a number of factors: (1) PEDOT is slightly crystalline and depending on the size of the counter anion, crystallinity can be different (27,28). Generally, the degree of crystallinity and thereby, the overall stability of PEDOT improves with smaller anions like TFB (28–30) and (2) in PEDOT:PSS and CNT:PEDOT:PSS, PEDOT nanocrystals are surrounded by a shell of excess PSS (28,31). In physiological conditions, these PSSs can be eroded away from the film leading to a faster collapse of the PEDOT structure in PEDOT:PSS and CNT:PEDOT:PSS compared to PEDOT:TFB. The degree of increase in Rpoly from day 1 to day 50, is roughly similar to the increase in the respective voltage output. This may suggest that the voltage output profile (specifically, the vertical part) is linked primarily to the resistance of the conductive polymer i.e. Rpoly and subsequent time-dependent capacitor formation (the slanted www.neuromodulationjournal.com
portion of the output) across the polymeric interface. However, compared to the in vitro condition, the in vivo change in the output voltage is higher. A number of factors could be responsible for this: the adsorbed proteins and/or cells could interact with the pi conjugated system, modify the conformation of the PEDOT backbone and change (de-dope) the electronic structure of the polymer. Rbio increased over time for all the polymers, pointing to the gradual formation of a thicker insulated protein and/or cell assembly. However, for PEDOT:TFB, the increase was moderate and occurred in parallel with improved neural recording (reduced noise, more contacts with units) compared to the other polymer coatings. While definitive analysis of tissue response requires immunohistological methods, electrical equivalent circuit models offer insight into tissue response to implanted materials (32). Overall, these findings further support our earlier demonstration of reduced neurotoxicity for this particular polymer (22) and suggest that greater biocompatibility may be responsible for its sustained recording capability.
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
IMPROVED PEDOT FOR NEURAL STIMULATION
CONCLUSIONS In this work, we compare the stability of three variations of PEDOT for electrical neural stimulations both in vitro and in vivo, and report their neural recording capability. From the EIS modeling perspective, we present a circuit model that incorporates and quantifies the time-dependent polymer degradation and biological response separately for implanted electrodes. This approach is critical for assessing the material specific performances in vivo. Based on the preliminary data presented here, PEDOT:TFB modified electrodes show relatively stable stimulation output and recording capability for a longer period of time than the PEDOT:PSS and CNT:PEDOT:PSS modified electrodes. These data provide further evidence that PEDOT:TFB may be a suitable material for chronic brain machine interfaces where both stimulation and recording are necessary.
Authorship Statements Dr. Mandal modified the neural probes with conductive polymers, initiated the study, analyzed the data and wrote the paper; Ms. Kastee and Mr. McHail collected the in vivo data; Mr. McHail performed the surgery under the supervision of Dr. Dumas; Drs. Rubinson and Pancrazio organized and led the project.
How to Cite this Article: Mandal H.S., Kastee J.S., McHail D.G., Rubinson J.F., Pancrazio J.J., Dumas T.C. 2015. Improved Poly(3,4Ethylenedioxythiophene) (PEDOT) for Neural Stimulation. Neuromodulation 2015; 18: 657–663
REFERENCES 1. Schmidt EM, Bak MJ, Hambrecht FT, Kufta CV, O’Rourke DK, Vallabhanath P. Feasibility of a visual prosthesis for the blind based on intracortical micro stimulation of the visual cortex. Brain 1996;119:507–522. doi: 10.1093/brain/119.2.507 2. Jackson A, Mavoori J, Fetz EE. Long-term motor cortex plasticity induced by an electronic neural implant. Nature 2006;444:56–60. doi: 10.1038/nature05226 3. Franke F, Jäckel D, Dragas J et al. High-density microelectrode array recordings and real-time spike sorting for closed-loop experiments: an emerging technology to study neural plasticity. Front Neural Circuits 2012;6:105. doi: 10.3389/ fncir.2012.00105 4. Kipke DR, Shain W, Buzsáki G et al. Advanced neurotechnologies for chronic neural interfaces: new horizons and clinical opportunities. J Neurosci 2008;28:11830– 11838. doi: 10.1523/JNEUROSCI.3879-08.2008 5. Buzsáki G. Large-scale recording of neuronal ensembles. Nat Neurosci 2004;7:446– 451. doi: 10.1038/nn1233 6. Kozai TDY, Langhals NB, Patel PR et al. Ultrasmall implantable composite microelectrodes with bioactive surfaces for chronic neural interfaces. Nat Mater 2012;11:1065–1073. doi: 10.1038/nmat3468 7. Sanders JE, Stiles CE, Hayes CL. Tissue response to single-polymer fibers of varying diameters: evaluation of fibrous encapsulation and macrophage density. J Biomed Mater Res 2000;52:231–237. doi: 10.1002/1097-4636(200010)52:13.0.CO;2-E 8. Seymour JP, Kipke DR.Neural probe design for reduced tissue encapsulation in CNS. Biomaterials 2007;28:3594–3607. doi: 10.1016/j.biomaterials.2007.03.024 9. Biran R, Martin DC, Tresco PA. Neuronal cell loss accompanies the brain tissue response to chronically implanted silicon microelectrode arrays. Exp Neurol 2005;195:115–126. doi: 10.1016/j.expneurol.2005.04.020 10. Bjornsson CS, Oh SJ, Al-Kofahi YA et al. Effects of insertion conditions on tissue strain and vascular damage during neuroprosthetic device insertion. J Neural Eng 2006;3:196–207. doi: 10.1088/1741-2560/3/3/002 11. Bernatchez SF, Parks PJ, Gibbons DF. Interaction of macrophages with fibrous materials in vitro. Biomaterials 1996;17:2077–2086. doi: 10.1016/0142-9612(96) 00014-2 12. Paik S-J, Park Y, Cho D. “Dan.” Roughened polysilicon for low impedance microelectrodes in neural probes. J Micromechanic Microengineer 2003;13:373–379. doi: 10.1088/0960-1317/13/3/304
13. Burke LD, Scannell RA. An investigation of hydrous oxide growth on iridium in base. J Electroanal Chem Interfacial Electrochem 1984;175:119–141. doi: 10.1016/S00220728(84)80350-2 14. Burke LD, Mulcahy JK, Whelan DP. Preparation of an oxidized iridium electrode and the variation of its potential with pH. J Electroanal Chem Interfacial Electrochem 1984;163:117–128. doi: 10.1016/S0022-0728(84)80045-5 15. Loeb GE, Peck RA, Martyniuk J. Toward the ultimate metal microelectrode. J Neurosci Methods 1995;63:175–183. doi: 10.1016/0165-0270(95)00107-7 16. Cogan SF, Guzelian AA, Agnew WF, Yuen TG, McCreery DB. Over-pulsing degrades activated iridium oxide films used for intracortical neural stimulation. J Neurosci Methods 2004;137:141–150. doi: 10.1016/j.jneumeth.2004.02.019 17. Green RA, Hassarati RT, Bouchinet L et al. Substrate dependent stability of conducting polymer coatings on medical electrodes. Biomaterials 2012;33:5875–5886. doi: 10.1016/j.biomaterials.2012.05.017 18. Gerwig R, Stett A, Stelzle M. PEDOT–CNT composite microelectrodes for recording and electrostimulation applications: fabrication, morphology, and electrical properties. Front Neuroeng 2012;5:8. doi: 10.3389/fneng.2012.00008 19. Venkatraman S, Hendricks J, King ZA et al. In vitro and in vivo evaluation of PEDOT microelectrodes for neural stimulation and recording. IEEE Trans Neural Syst Rehabil Eng 2011;19:307–316. doi: 10.1109/TNSRE.2011.2109399 20. Luo X, Weaver CL, Zhou DD, Greenberg R, Cui XT. Highly stable carbon nanotube doped poly(3,4-ethylenedioxythiophene) for chronic neural stimulation. Biomaterials 2011;32:5551–5557. doi: 10.1016/j.biomaterials.2011.04.051 21. Cogan SF. Neural stimulation and recording electrodes. Annu Rev Biomed Eng 2008;10:275–309. doi: 10.1146/annurev.bioeng.10.061807.160518 22. Mandal HS, Knaack GL, Charkhkar H et al. Improving the performance of poly(3,4ethylenedioxythiophene) for brain–machine interface applications. Acta Biomater 2014;10:2446–2454. doi: 10.1016/j.actbio.2014.02.029 23. ASTM F1980-07. Standard Guide for Accelerated Aging of Sterile Barrier Systems for Medical Devices. West Conshohocken, PA: ASTM International, 2011. www.astm.org. 24. Hukins DWL, Mahomed A, Kukureka SN. Accelerated aging for testing polymeric biomaterials and medical devices. Med Eng Phys 2008;30:1270–1274. doi: 10.1016/ j.medengphy.2008.06.001 25. Ludwig KA, Uram JD, Yang J, Martin DC, Kipke DR. Chronic neural recordings using silicon microelectrode arrays electrochemically deposited with a poly(3,4ethylenedioxythiophene) (PEDOT) film. J Neural Eng 2006;3:59–70. doi: 10.1088/ 1741-2560/3/1/007 26. Prasad A, Xue Q-S, Sankar V et al. Comprehensive characterization and failure modes of tungsten microwire arrays in chronic neural implants. J Neural Eng 2012;9:056015. doi: 10.1088/1741-2560/9/5/056015 27. Martin DC, Wu J, Shaw CM et al. The morphology of poly(3,4Ethylenedioxythiophene). Polym Rev 2010;50:340–384. doi: 10.1080/15583724 .2010.495440 28. Takano T, Masunaga H, Fujiwara A, Okuzaki H, Sasaki T. PEDOT nanocrystal in highly conductive PEDOT:PSS polymer films. Macromolecules 2012;45:3859–3865. doi: 10.1021/ma300120g 29. Crispin X, Marciniak S, Osikowicz W et al. Conductivity, morphology, interfacial chemistry, and stability of poly(3,4-ethylene dioxythiophene)–poly(styrene sulfonate): a photoelectron spectroscopy study. J Polym Sci B-Polym Phys 2003;41:2561– 2583. doi: 10.1002/polb.10659 30. Jönsson SK, Birgerson J, Crispin X et al. The effects of solvents on the morphology and sheet resistance in poly(3,4-ethylenedioxythiophene)–polystyrenesulfonic acid (PEDOT–PSS) films. Synth Met 2003;139:1–10. doi: 10.1016/S03796779(02)01259-6 31. Xing KZ, Fahlman M, Chen XW, Inganäs O, Salaneck WR. The electronic structure of poly(3,4-ethylene-dioxythiophene): studied by XPS and UPS. Synth Met 1997;89: 161–165. doi: 10.1016/S0379-6779(97)81212-X 32. McConnell GC, Butera RJ, Bellamkonda RV. Bioimpedance modeling to monitor astrocytic response to chronically implanted electrodes. J Neural Eng 2009;6:055005. doi: 10.1088/1741-2560/6/5/055005
COMMENT This paper compares three variations of conductive polymers for micro-stimulation. The authors provide promising preliminary evidence that PEDOT films deposited with tetrafluoroborate might perform better than other variations. This work will need to be confirmed in a more extensive chronic study, but these initial results indicate that modifications of the PEDOT film may result in improved electrode stability. Kevin Kilgore, PhD Cleveland, OH, USA
Comments not included in the Early View version of this paper.
663
www.neuromodulationjournal.com
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
Revised: December 23, 2014
Accepted: January 27, 2015
(onlinelibrary.wiley.com) DOI: 10.1111/ner.12285
Improved Poly(3,4-Ethylenedioxythiophene) (PEDOT) for Neural Stimulation Himadri Shekhar Mandal, PhD*; Jemika Shrestha Kastee, BS*; Daniel Glenn McHail, BS†; Judith Faye Rubinson, PhD‡; Joseph Jewell Pancrazio, PhD*; Theodore Constantine Dumas, PhD† Objective: This study compares the stability of three variations of the conductive polymer poly(3,4-ethylenedioxythiophene) or PEDOT for neural micro-stimulation under both in vitro and in vivo conditions. We examined PEDOT films deposited with counterions tetrafluoroborate (TFB) and poly(styrenesulfonate) (PSS), and PEDOT:PSS combined with carbon nanotubes (CNTs). Methods: For the in vitro stability evaluation, implantable micro-wires were coated with the polymers, placed in a vial containing phosphate buffered saline (PBS) under accelerated aging conditions (60°C), and current pulses were applied. The resulting voltage profile was monitored over time. Following the same polymer deposition protocol, chronic neural micro-probes were modified and implanted in the motor cortex of two rats for the in vivo stability comparison. Similar stimulating current pulses were applied and the output voltage was examined. The electrochemical impedance spectroscopic (EIS) data were also recorded and fit to an equivalent circuit model that incorporates and quantifies the time-dependent polymer degradation and impedance associated with tissue surrounding each micro-electrode site. Results: Both in vitro and in vivo voltage output profiles show relatively stable behavior for the PEDOT:TFB modified microelectrodes compared to the PEDOT:PSS and CNT:PEDOT:PSS modified ones. EIS modeling demonstrates that the time-dependent increase in the polymeric resistance is roughly similar to the rise in the respective voltage output in vivo and indicates that the polymeric stability and conductivity, rather than the impedance due to the tissue response, is the primary factor determining the output voltage profile. It was also noted that the number of electrodes showing unit activity post-surgery did not decay for PEDOT:TFB as was the case for PEDOT:PSS and CNT:PEDOT:PSS. Conclusions: PEDOT:TFB may be an enabling material for achieving long lasting micro-stimulation and recording. Keywords: Conductive polymer, neural stimulation, PEDOT Conflict of Interest: The authors reported no conflict of interest.
INTRODUCTION
www.neuromodulationjournal.com
Address correspondence to: Joseph J. Pancrazio, PhD, Dept. of Bioengineering, George Mason University, 4400 University Dr., MS 1G5, Fairfax, VA 22030, USA. Email: [email protected] * Department of Bioengineering, George Mason University, Fairfax, VA, USA; † Department of Molecular Neuroscience, The Krasnow Institute for Advanced Study, George Mason University, Fairfax, VA, USA; and ‡ Department of Chemistry, Georgetown University, Washington, DC, USA For more information on author guidelines, an explanation of our peer review process, and conflict of interest informed consent policies, please go to http:// www.wiley.com/bw/submit.asp?ref=1094-7159&site=1 Sources of financial support: The Defense Advanced Research Projects Agency (DARPA) MTO sponsored this work under the auspices of Dr. Jack Judy through the Space and Naval Warfare Systems Center, Pacific Grant/Contract No. N66001-121-4026.
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
657
Electrical stimulation using implantable electrodes (IEs) is critical in neuroscience for its relevance in a wide range of brain machine interface applications (1–4). One of the major challenges for current IEs is to lower the dimensions of the IE in order to achieve microstimulation of spatially restricted regions of the nervous system, while inducing minimal tissue response (5–11). But, as the size is decreased, the charge injection capability of the electrode, which is primarily capacitive in nature and proportional to the electrochemically active surface area at the interface, becomes smaller and insufficient for effective neuronal stimulation. This scenario can be improved by increasing the surface roughness, thus expanding the effective surface area without increasing the geometric area (12–14). Unfortunately, metallic roughened surfaces are chemically reactive and show unreliable electrochemical properties over time (14–16). To address this issue, modification of micro-electrodes with conductive polymers (CPs) has been suggested to be a viable route and currently is an active area of research (17–21). CPs are, in general, porous in morphology, have orders of magnitude larger electrochemically active surface area compared to metallic counterparts of the same overall dimensions and are promising materials for developing micro-electrodes capable of large charge injection. However,
the long-term stability of CPs, especially in physiological conditions, is a serious concern for use of these materials for chronic applications. Recently, we reported that the conductivity and stability of the most commonly utilized CP poly(3,4-ethylenedioxythiophene) or PEDOT can be significantly increased by substitution of the widely used, multivalent and bulky counter-anion poly(styrenesulfonate), or PSS, by the smaller monovalent tetrafluoroborate (TFB) anion (Fig. 1)
MANDAL ET AL. The applied charge density was 0.56 mC/cm2 (assuming the geometric area of the chronic electrode site to be 177 μm2) which is below the charge-injection limit for PEDOT modified electrodes (19). Note that the temperature used here is in compliance with methods based on the ASTM F1980-07 “Standard Guide for Accelerated Aging of Sterile Barrier Systems for Medical Devices” (23). The aging factor at 60°C is at least 4 assuming the body temperature (37°C) as the baseline. Assuming a conservative Q10 for polymers, the reaction rate doubles for each 10°C increase (24). Figure 1. Chemical structures of PEDOT, PSS and TFB.
during the electro-deposition of the polymer (22). We also demonstrated that this particular variation of PEDOT, i.e. PEDOT:TFB, is even more stable than carbon nanotube (CNT)-incorporated PEDOT (CNT:PEDOT:PSS) which has been suggested to have enhanced performances due to the CNTs’ unique electromechanical and structural properties. In the present study, we examined and compared the stability of PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS in electrical stimulation based experiments under both in vitro and in vivo conditions. We also investigated their in vivo neuronal recording capability. Based on the data, PEDOT:TFB demonstrates better performances for both stimulation and recording, and could prove useful in fabricating long-lasting micro-scale brain-machine interface (BMI) devices.
MATERIALS AND METHODS Materials Gold(III) chloride hydrate, 3,4-ethylenedioxylthiophene (EDOT), PSS, acetonitrile and tetrabutylammonium tetrafluoroborate were obtained from Sigma-Aldrich (St. Louis, MO, USA). PBS was obtained from Mediatech Inc. (Manassas, VA, USA). The CNT dispersion was purchased from CheapTubes.com (Brattleboro, VT, USA). Pt/Ir microwires (PI20030.1A5) and chronic probes (A1x16-5 mm-100– 177-CM16) were purchased from MicroProbes for Life Sciences (Gaithersburg, MD, USA) and NeuroNexus Technologies (Ann Arbor, MI, USA), respectively.
Preparation of Polymer Modified Implantable Microwires and Chronic Neural Probes Details of the modification process are described elsewhere (22). In brief, Pt/Ir implantable microwires were first electroplated with a thin layer of gold. This involved cycling the wires between 0 and −1.5 V three times in a solution of HAuCl4 (5 mM) at a scan rate of 1V/s. The polymers (PEDOT:TFB, PEDOT:PSS and CNT:PEDOT:PSS) were then deposited using a CHI 660D potentiostat (CH Instruments, Inc., Austin, TX, USA). After the polymer deposition, electrodes were cycled between 0 and 0.5V (50 cycles, cyclic voltammetry) in 1x PBS. Chronic probes containing Ir electrodes (A1x16-5 mm-100–177-CM16 from NeuroNexus) for in vivo studies were modified using similar gold and polymer deposition protocols.
658
Accelerated Aging Study In Vitro The polymer modified Au coated Pt/Ir microwires (N = 3 each for PEDOT:PSS, CNT:PEDOT:PSS, and PEDOT:TFB) were placed in a vial containing PBS at 60°C, and 1 μA, 1 ms, bi-phasic, 5 pulses/train, 83.3 Hz, anodic first current pulses were applied using a potentiostat (CHI 660D, CH Instruments, Inc., Austin, TX, USA) equipped with a multiplexer (CHI 684, CH Instruments, Inc., Austin, TX, USA). www.neuromodulationjournal.com
In Vivo Stimulation and Recording Out of the 16 microelectrode sites in one NeuroNexus probe, the first set of 4 consecutive sites (from the tip) was modified with PEDOT:TFB, the second set of 4 with PEDOT:PSS, the third set of 4 was unmodified and the remaining set of 4 sites was with CNT:PEDOT:PSS. A second NeuroNexus probe was also modified where the location of the polymer was shifted by 4 sites (from the tip of the probe) to reduce the degree of bias due to position within the cortex. These probes were implanted in two female Long Evans rats (300–370 g), as described previously (22). Current pulses similar to those used for the in vitro measurements were applied. During the data (stimulation) collection, the animal was anesthetized in an induction chamber with 1 L/min oxygen and 5% isoflurane. However, the neuronal signals were recorded in freely moving rats over a recording session of 10 minutes. Data were acquired periodically during a 50-day period using a Cerebus system (Blackrock Microsystems, Salt Lake City, UT, USA). The details of data collection and processing are described in our previous publication (22). The statistical analysis for the differential expression of unit recordings as a function of coatings was performed using the Z Test Calculator for 2 population proportions (http://www.socscistatistics.com/). Two-tailed hypothesis was adopted and the significance level was set at 0.05. All animal procedures comply with the National Institute of Health guidelines and were approved by the Institutional Animal Care and Use Committee at George Mason University, Fairfax, VA.
Electrochemical Impedance Spectroscopy (EIS) and Modeling The impedance was measured at the open circuit potential from 0.1 Hz to 1 MHz, with a peak-to-peak amplitude of the sinusoidal voltage of 20 mV. EIS data was fit with ZsimpWin (Princeton Applied Research, Oak Ridge, TN, USA).
RESULTS In Vitro Stimulation Studies Figure 2b–d show representative in vitro output voltage profiles from PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS modified microwires after applying current pulses (parameters are shown in Fig. 2a) from day 1 and day 40. Anodic voltage outputs are presented. PEDOT:TFB modified electrodes show slower change in the output over time compared to the PEDOT:PSS and CNT:PEDOT:PSS modified electrodes (Fig. 2e).
In Vivo Stimulation, EIS and Recording The in vivo output voltage profiles from PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS modified sites in NeuroNexus (NNx) probes are shown in Figure 3. However, unlike the in vitro voltage output, the change over time is not gradual. For all three polymer sites, the output voltage was high on day 7, low on day 15 and relatively
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
IMPROVED PEDOT FOR NEURAL STIMULATION
Figure 2. Current pulse recordings in vitro. Parameters used for bi-phasic current stimulations (a). Output voltages elicited by current pulses delivered through PEDOT:TFB (b), PEDOT:PSS (c), and CNT:PEDOT:PSS (d) modified microwires. In each figure, black and red data are for day 1 and day 40, respectively. Anodic (e) output voltages over time: PEDOT:TFB (black rectangle), PEDOT:PSS (blue circle), and CNT:PEDOT:PSS (green triangle).
www.neuromodulationjournal.com
an electrical equivalent circuit model (Fig. 4b) based on our previously published work (22) on polymer degradation over time in vitro. Here, we included two additional parameters to model the tissue response at the electrode-solution interface which is very common for implanted probes (25,26). In the model, the charge
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
659
constant onwards (Fig. 3e). Overall, PEDOT:TFB modified sites show lower output voltage compared to the PEDOT:PSS and CNT:PEDOT:PSS modified sites. Figure 4c–e show typical in vivo EIS data for PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS modified sites. The data were fit to
MANDAL ET AL.
Figure 3. An optical image of PEDOT:TFB modified sites (first four dark regions from the tip) in a NNx probe (a). Output voltages elicited by current pulses delivered through PEDOT:TFB (b), PEDOT:PSS (c), and CNT:PEDOT:PSS (d) modified sites. In each figure, black and red data are for day 1 and 50, respectively. Anodic (e) output voltages over time: PEDOT:TFB (black rectangle), PEDOT:PSS (blue circle), and CNT:PEDOT:PSS (green triangle).
660
transfer resistance and the associated double layer capacitance across the rough polymeric interface are designated as Rpoly and Qpoly, respectively. Cd is the double layer capacitance due to the defect sites of the polymer film. Rbio and Cbio are the polarization resistance and associated capacitance of the tissue, respectively. We also recorded the pre-implant EIS data for the probes in PBS and fit them to the in vitro model (Fig. 4a) for comparison. Impedance increases for all polymer modified sites after implant (Fig. 4c– www.neuromodulationjournal.com
e). This is consistent with our previous observation and may be related, at least in part, to the inflammatory response and/or protein adsorption on the electrode sites (22,25). All the fit parameters are listed in Table 1. The polymeric resistance Rpoly and the resistance of the tissue Rbio increase over time for all the polymer modified sites. However, while Rpoly rose by factors of 1.4–2.8, the major difference between the coatings was the relatively modest change by a factor of 2 in Rbio for PEDOT:TFB versus
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
IMPROVED PEDOT FOR NEURAL STIMULATION
Figure 4. In vitro (a) and in vivo (b) EIS equivalent circuit models. Experimental impedance data and the fit (red line) for PEDOT:TFB (c), PEDOT:PSS (d), and CNT:PEDOT:PSS (e) modified sites in one of the chronic probes. In each figure, black rectangle (pre-implant EIS in 1X PBS), blue triangle (day 1 after surgery), and green circles (day 50 after surgery). Table 1. Fit Parameters for EIS Data From All Coated Microelectrodes. Pre-implant Cd (F/cm2) Rpoly (Ohm-cm2) Qpoly (S-sec1/2/cm2) n (0 < n < 1) Rbio (Ohm-cm2) Cbio (F/cm2)
4.51E-05 (1.05E-06) 5.34E-02 (9.57E-04) 1.72E-01 (3.91E-02) 9.87E-01 (1.72E-03)
PEDOT:TFB Day 1
Day 50
Pre-implant
3.82E-05 (3.31E-06) 1.19E-01 (2.26E-02) 5.47E-02 (1.44E-02) 8.54E-01 (2.85E-02) 1.21E-01 (4.51E-02) 1.05E-01 (1.91E-02)
3.24E-05 (1.75E-06) 1.71E-01 (1.93E-02) 3.12E-02 (1.18E-02) 8.11E-01 (1.99E-02) 2.49E-01 (4.92E-02) 6.95E-02 (2.89E-02)
5.61E-05 (2.61E-06) 7.79E-02 (4.51E-03) 3.72E-02 (5.71E-03) 8.66E-01 (7.12E-03)
PEDOT:PSS Day 1
Day 50
Pre-implant
3.01E-05 (7.93E-06) 4.17E-01 (1.89E-01) 1.38E-02 (4.02E-03) 8.16E-01 (2.64E-02) 2.98E-01 (7.81E-02) 2.99E-02 (6.14E-03)
2.68E-05 (4.86E-06) 1.06E+00 (7.59E-01) 9.87E-03 (2.74E-03) 5.64E-01 (5.80E-02) 1.91E+01 (1.81E+01) 1.46E-02 (7.76E-03)
3.91E-04 (3.36E-04) 9.37E-02 (2.05E-02) 2.57E-02 (2.52E-03) 8.54E-01 (2.02E-02)
CNT:PEDOT:PSS Day 1
Day 50
2.96E-05 (7.61E-06) 2.87E-01 (6.46E-02) 8.54E-03 (4.54E-03) 8.02E-01 (2.66E-02) 4.10E-01 (6.04E-02) 2.31E-02 (8.20E-03)
2.61E-05 (1.39E-06) 8.22E-01 (7.08E-02) 8.50E-03 (2.25E-03) 5.89E-01 (4.98E-02) 3.05E+01 (2.45E+01) 1.40E-02 (6.00E-03)
Values in parentheses correspond to the respective standard error of the mean (SEM).
PEDOT:PSS and CNT:PEDOT:PSS, which showed increases by factors of 65–75. We also performed in vivo neuronal recordings in awake and behaving animals with the polymer modified electrodes. Figure 5a–c show representative single units from day 2. The peakto-peak amplitude of single units for PEDOT:TFB, PEDOT:PSS, and CNT:PEDOT:PSS coated electrodes is 87.49 ± 7.86 μV, 125.42 ± 33.80 μV, 122.75 ± 12.73 μV, respectively (mean ± SEM). The corresponding noise levels are 8.03 ± 0.70 μV (PEDOT:TFB), 9.90 ± 0.93 μV (PEDOT:PSS), and 12.28 ± 1.91 μV (CNT:PEDOT:PSS). Figure 5d shows the time-dependent activity profile for the modified sites. Although initially, the proportion of coated electrodes that show activity is not statistically significant, a clear difference (P < 0.05) is observed over time (by test of proportions, p < 0.05). PEDOT:TFB modified sites show better neuronal recording capability compared to the PEDOT:PSS and CNT:PEDOT:PSS modified sites.
DISCUSSION
www.neuromodulationjournal.com
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
661
In both in vitro and in vivo stimulation experiments, the change in the output voltage profile is reduced and remains within the “safe
water window” for the PEDOT:TFB modified electrode sites compared to the PEDOT:PSS and CNT:PEDOT:PSS coated sites. This is in line with our previous work where we reported significantly larger conductivity and thermal stability of PEDOT:TFB coated microelectrodes compared to those coated with PEDOT:PSS and CNT:PEDOT:PSS (22). Note that we continued collecting the data as long as the output was within the “safe water window” to avoid confounding variables such as the formation/adsorption of H2, O2, and free radicals, which may damage/delaminate the polymer film, promote inflammation of the surrounding tissue, and complicate the data interpretation. The increase in Rpoly over time for all the polymer modified sites may be related to the time-dependent aging of the polymers (22). To explain the origin of this Rpoly in more details, we refer to our previous work where we showed polymeric aging involves the collapse of the porous structure. When the structure is highly porous and easily accessible to the supporting electrolytes, the delocalized and already charged PEDOT backbone can be stabilized by these supporting electrolytes during the transfer of the polaronic charge carriers and the formation of Cpoly at the polymer-solution interface. When the porosity is reduced, the PEDOT backbone becomes less accessible to the electrolytes and thereby, Rpoly increases. This is
MANDAL ET AL.
Figure 5. Representative units from (a) PEDOT:TFB, (b) PEDOT:PSS, and (c) CNT:PEDOT:PSS coated electrode sites; (d) Activity profile over time: PEDOT:TFB (gray), PEDOT:PSS (blue), and CNT:PEDOT:PSS (green).
662
also consistent also with the relatively uniform decrease in the corresponding Qpoly for all the polymer modified electrodes. The smaller increase in Rpoly for PEDOT:TFB compared to PEDOT:PSS and CNT:PEDOT:PSS indicates that the former has higher stability. This could be due to a number of factors: (1) PEDOT is slightly crystalline and depending on the size of the counter anion, crystallinity can be different (27,28). Generally, the degree of crystallinity and thereby, the overall stability of PEDOT improves with smaller anions like TFB (28–30) and (2) in PEDOT:PSS and CNT:PEDOT:PSS, PEDOT nanocrystals are surrounded by a shell of excess PSS (28,31). In physiological conditions, these PSSs can be eroded away from the film leading to a faster collapse of the PEDOT structure in PEDOT:PSS and CNT:PEDOT:PSS compared to PEDOT:TFB. The degree of increase in Rpoly from day 1 to day 50, is roughly similar to the increase in the respective voltage output. This may suggest that the voltage output profile (specifically, the vertical part) is linked primarily to the resistance of the conductive polymer i.e. Rpoly and subsequent time-dependent capacitor formation (the slanted www.neuromodulationjournal.com
portion of the output) across the polymeric interface. However, compared to the in vitro condition, the in vivo change in the output voltage is higher. A number of factors could be responsible for this: the adsorbed proteins and/or cells could interact with the pi conjugated system, modify the conformation of the PEDOT backbone and change (de-dope) the electronic structure of the polymer. Rbio increased over time for all the polymers, pointing to the gradual formation of a thicker insulated protein and/or cell assembly. However, for PEDOT:TFB, the increase was moderate and occurred in parallel with improved neural recording (reduced noise, more contacts with units) compared to the other polymer coatings. While definitive analysis of tissue response requires immunohistological methods, electrical equivalent circuit models offer insight into tissue response to implanted materials (32). Overall, these findings further support our earlier demonstration of reduced neurotoxicity for this particular polymer (22) and suggest that greater biocompatibility may be responsible for its sustained recording capability.
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
IMPROVED PEDOT FOR NEURAL STIMULATION
CONCLUSIONS In this work, we compare the stability of three variations of PEDOT for electrical neural stimulations both in vitro and in vivo, and report their neural recording capability. From the EIS modeling perspective, we present a circuit model that incorporates and quantifies the time-dependent polymer degradation and biological response separately for implanted electrodes. This approach is critical for assessing the material specific performances in vivo. Based on the preliminary data presented here, PEDOT:TFB modified electrodes show relatively stable stimulation output and recording capability for a longer period of time than the PEDOT:PSS and CNT:PEDOT:PSS modified electrodes. These data provide further evidence that PEDOT:TFB may be a suitable material for chronic brain machine interfaces where both stimulation and recording are necessary.
Authorship Statements Dr. Mandal modified the neural probes with conductive polymers, initiated the study, analyzed the data and wrote the paper; Ms. Kastee and Mr. McHail collected the in vivo data; Mr. McHail performed the surgery under the supervision of Dr. Dumas; Drs. Rubinson and Pancrazio organized and led the project.
How to Cite this Article: Mandal H.S., Kastee J.S., McHail D.G., Rubinson J.F., Pancrazio J.J., Dumas T.C. 2015. Improved Poly(3,4Ethylenedioxythiophene) (PEDOT) for Neural Stimulation. Neuromodulation 2015; 18: 657–663
REFERENCES 1. Schmidt EM, Bak MJ, Hambrecht FT, Kufta CV, O’Rourke DK, Vallabhanath P. Feasibility of a visual prosthesis for the blind based on intracortical micro stimulation of the visual cortex. Brain 1996;119:507–522. doi: 10.1093/brain/119.2.507 2. Jackson A, Mavoori J, Fetz EE. Long-term motor cortex plasticity induced by an electronic neural implant. Nature 2006;444:56–60. doi: 10.1038/nature05226 3. Franke F, Jäckel D, Dragas J et al. High-density microelectrode array recordings and real-time spike sorting for closed-loop experiments: an emerging technology to study neural plasticity. Front Neural Circuits 2012;6:105. doi: 10.3389/ fncir.2012.00105 4. Kipke DR, Shain W, Buzsáki G et al. Advanced neurotechnologies for chronic neural interfaces: new horizons and clinical opportunities. J Neurosci 2008;28:11830– 11838. doi: 10.1523/JNEUROSCI.3879-08.2008 5. Buzsáki G. Large-scale recording of neuronal ensembles. Nat Neurosci 2004;7:446– 451. doi: 10.1038/nn1233 6. Kozai TDY, Langhals NB, Patel PR et al. Ultrasmall implantable composite microelectrodes with bioactive surfaces for chronic neural interfaces. Nat Mater 2012;11:1065–1073. doi: 10.1038/nmat3468 7. Sanders JE, Stiles CE, Hayes CL. Tissue response to single-polymer fibers of varying diameters: evaluation of fibrous encapsulation and macrophage density. J Biomed Mater Res 2000;52:231–237. doi: 10.1002/1097-4636(200010)52:13.0.CO;2-E 8. Seymour JP, Kipke DR.Neural probe design for reduced tissue encapsulation in CNS. Biomaterials 2007;28:3594–3607. doi: 10.1016/j.biomaterials.2007.03.024 9. Biran R, Martin DC, Tresco PA. Neuronal cell loss accompanies the brain tissue response to chronically implanted silicon microelectrode arrays. Exp Neurol 2005;195:115–126. doi: 10.1016/j.expneurol.2005.04.020 10. Bjornsson CS, Oh SJ, Al-Kofahi YA et al. Effects of insertion conditions on tissue strain and vascular damage during neuroprosthetic device insertion. J Neural Eng 2006;3:196–207. doi: 10.1088/1741-2560/3/3/002 11. Bernatchez SF, Parks PJ, Gibbons DF. Interaction of macrophages with fibrous materials in vitro. Biomaterials 1996;17:2077–2086. doi: 10.1016/0142-9612(96) 00014-2 12. Paik S-J, Park Y, Cho D. “Dan.” Roughened polysilicon for low impedance microelectrodes in neural probes. J Micromechanic Microengineer 2003;13:373–379. doi: 10.1088/0960-1317/13/3/304
13. Burke LD, Scannell RA. An investigation of hydrous oxide growth on iridium in base. J Electroanal Chem Interfacial Electrochem 1984;175:119–141. doi: 10.1016/S00220728(84)80350-2 14. Burke LD, Mulcahy JK, Whelan DP. Preparation of an oxidized iridium electrode and the variation of its potential with pH. J Electroanal Chem Interfacial Electrochem 1984;163:117–128. doi: 10.1016/S0022-0728(84)80045-5 15. Loeb GE, Peck RA, Martyniuk J. Toward the ultimate metal microelectrode. J Neurosci Methods 1995;63:175–183. doi: 10.1016/0165-0270(95)00107-7 16. Cogan SF, Guzelian AA, Agnew WF, Yuen TG, McCreery DB. Over-pulsing degrades activated iridium oxide films used for intracortical neural stimulation. J Neurosci Methods 2004;137:141–150. doi: 10.1016/j.jneumeth.2004.02.019 17. Green RA, Hassarati RT, Bouchinet L et al. Substrate dependent stability of conducting polymer coatings on medical electrodes. Biomaterials 2012;33:5875–5886. doi: 10.1016/j.biomaterials.2012.05.017 18. Gerwig R, Stett A, Stelzle M. PEDOT–CNT composite microelectrodes for recording and electrostimulation applications: fabrication, morphology, and electrical properties. Front Neuroeng 2012;5:8. doi: 10.3389/fneng.2012.00008 19. Venkatraman S, Hendricks J, King ZA et al. In vitro and in vivo evaluation of PEDOT microelectrodes for neural stimulation and recording. IEEE Trans Neural Syst Rehabil Eng 2011;19:307–316. doi: 10.1109/TNSRE.2011.2109399 20. Luo X, Weaver CL, Zhou DD, Greenberg R, Cui XT. Highly stable carbon nanotube doped poly(3,4-ethylenedioxythiophene) for chronic neural stimulation. Biomaterials 2011;32:5551–5557. doi: 10.1016/j.biomaterials.2011.04.051 21. Cogan SF. Neural stimulation and recording electrodes. Annu Rev Biomed Eng 2008;10:275–309. doi: 10.1146/annurev.bioeng.10.061807.160518 22. Mandal HS, Knaack GL, Charkhkar H et al. Improving the performance of poly(3,4ethylenedioxythiophene) for brain–machine interface applications. Acta Biomater 2014;10:2446–2454. doi: 10.1016/j.actbio.2014.02.029 23. ASTM F1980-07. Standard Guide for Accelerated Aging of Sterile Barrier Systems for Medical Devices. West Conshohocken, PA: ASTM International, 2011. www.astm.org. 24. Hukins DWL, Mahomed A, Kukureka SN. Accelerated aging for testing polymeric biomaterials and medical devices. Med Eng Phys 2008;30:1270–1274. doi: 10.1016/ j.medengphy.2008.06.001 25. Ludwig KA, Uram JD, Yang J, Martin DC, Kipke DR. Chronic neural recordings using silicon microelectrode arrays electrochemically deposited with a poly(3,4ethylenedioxythiophene) (PEDOT) film. J Neural Eng 2006;3:59–70. doi: 10.1088/ 1741-2560/3/1/007 26. Prasad A, Xue Q-S, Sankar V et al. Comprehensive characterization and failure modes of tungsten microwire arrays in chronic neural implants. J Neural Eng 2012;9:056015. doi: 10.1088/1741-2560/9/5/056015 27. Martin DC, Wu J, Shaw CM et al. The morphology of poly(3,4Ethylenedioxythiophene). Polym Rev 2010;50:340–384. doi: 10.1080/15583724 .2010.495440 28. Takano T, Masunaga H, Fujiwara A, Okuzaki H, Sasaki T. PEDOT nanocrystal in highly conductive PEDOT:PSS polymer films. Macromolecules 2012;45:3859–3865. doi: 10.1021/ma300120g 29. Crispin X, Marciniak S, Osikowicz W et al. Conductivity, morphology, interfacial chemistry, and stability of poly(3,4-ethylene dioxythiophene)–poly(styrene sulfonate): a photoelectron spectroscopy study. J Polym Sci B-Polym Phys 2003;41:2561– 2583. doi: 10.1002/polb.10659 30. Jönsson SK, Birgerson J, Crispin X et al. The effects of solvents on the morphology and sheet resistance in poly(3,4-ethylenedioxythiophene)–polystyrenesulfonic acid (PEDOT–PSS) films. Synth Met 2003;139:1–10. doi: 10.1016/S03796779(02)01259-6 31. Xing KZ, Fahlman M, Chen XW, Inganäs O, Salaneck WR. The electronic structure of poly(3,4-ethylene-dioxythiophene): studied by XPS and UPS. Synth Met 1997;89: 161–165. doi: 10.1016/S0379-6779(97)81212-X 32. McConnell GC, Butera RJ, Bellamkonda RV. Bioimpedance modeling to monitor astrocytic response to chronically implanted electrodes. J Neural Eng 2009;6:055005. doi: 10.1088/1741-2560/6/5/055005
COMMENT This paper compares three variations of conductive polymers for micro-stimulation. The authors provide promising preliminary evidence that PEDOT films deposited with tetrafluoroborate might perform better than other variations. This work will need to be confirmed in a more extensive chronic study, but these initial results indicate that modifications of the PEDOT film may result in improved electrode stability. Kevin Kilgore, PhD Cleveland, OH, USA
Comments not included in the Early View version of this paper.
663
www.neuromodulationjournal.com
© 2015 International Neuromodulation Society
Neuromodulation 2015; 18: 657–663
