Open Access
Add to BibSonomy
Abstract
Multifunctional optical fiber-based neural interfaces have attracted significant attention for neural stimulation, recording, and photopharmacology towards understanding the central nervous system. In this work, we demonstrate the fabrication, optoelectrical characterization, and mechanical analysis of four types of microstructured polymer optical fiber neural probes using different soft thermoplastic polymers. The developed devices have integrated metallic elements for electrophysiology and microfluidic channels for localized drug delivery, and can be used for optogenetics in the visible spectrum at wavelengths spanning from 450 nm up to 800 nm. Their impedance, measured by electrochemical impedance spectroscopy, was found to be as low as 21 kΩ and 4.7 kΩ at 1kHz when indium and tungsten wires are used as the integrated electrodes, respectively. Uniform on-demand drug delivery can be achieved by the microfluidic channels with a measured delivery rate from 10 up to 1000 nL/min. In addition, we identified the buckling failure threshold (defined as the conditions for successful implantation) as well as the bending stiffness of the fabricated fibers. Using finite element analysis, we calculated the main critical mechanical properties of the developed probes to avoid buckling during implantation and maintain high flexibility of the probe within the tissue. Our results aim to demonstrate the impact of design, fabrication, and characteristics of the materials on the development of polymer fibers as next-generation implants and neural interfaces.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
While silica glass optical fibers have been mainly used for data transfer and communications because of their low optical loss, polymer optical fibers (POFs) have been widely used for sensing applications [1]. Their unique inherent characteristics (compared to their silica counterparts) of high flexibility and biocompatibility make POFs ideal candidates for monitoring various physical parameters and biochemical sensing [2–4]. For instance, polymer fiber Bragg grating sensor fibers have been extensively used for strain, temperature, and humidity sensing with high sensitivity [4–10]. Also, the low Young’s modulus combined with the non-brittle nature makes POFs suitable for the development of biomedical devices as well as for interfaces for chronic in vivo studies [11,12].
In particular, hybrid microstructured POFs have been recently proposed not only as tunable optical components [13,14] but also as multi-functional neural interfaces for neuroscience [15]. Their low thermal processing temperature allows the use of a wide palette of materials for developing novel multi-material structures for optics [13,14,16] and neurophotonic applications [15,17]. For example, Canales et al. demonstrated the development of a multi-functional fiber device by combining polymers, metals, and microfluidic channels (MCs) for optical, electrical, and chemical interrogation of the neuronal brain circuits [17]. Jiang et al. developed expandable polymer fiber probes that can provide 3D coverage of brain tissue for multi-site neural recordings [18]. However, one critical factor that should be taken into consideration for almost every implant is its flexibility. This property is ultimately defined by the Young’s moduli of the used materials. Employing materials with lower Young’s moduli suppresses the foreign body response (FBR) that stems from the chemo-mechanical mismatch between the probes and the neural tissues [15]. This is an important factor in chronic experiments. Recently, Park et al. demonstrated that encapsulating different multimaterial fibers in a hydrogel matrix is an effective way to minimize the FBR and they showed how such an interface can provide stable stimulation and recording over 6 months [19]. However, an interface with low bending stiffness, i.e., low Young’s modulus, often results in buckling during the implantation which precludes its insertion into the neural tissue [19]. Therefore, there is a trade-off between soft “stealth” materials and effective implantation. Sui et al. interestingly demonstrated the use of a very soft Teflon-like fluorinated polymer, fluorinated ethylene propylene (FEP), for the development of functional POFs for large illumination of neuronal tissue. FEP has around 5 times lower Young’s modulus than most conventional thermoplastic polymers (e.g., poly(methyl methacrylate) (PMMA), polycarbonate (PC)) used in POF fabrication [20].
In this work, we report the fabrication and characterization of four different types of microstructured POF-based neural probes with functional elements using a thermal-drawing process. Type I: PC/FEP POFs with two MCs for drug delivery. Type II: Polysulfone (PSU)/FEP POFs. Type III: Cyclic olefin copolymers (COC) POFs with PSU jacket and tungsten electrodes for electrophysiology. Type IV: PSU/FEP POFs with indium electrodes for electrophysiology.The developed fibers were characterized in terms of their guiding properties, the impedance of their integrated electrodes, and the drug delivery ability of their MCs. Then, we numerically compared their mechanical properties using the finite element method to identify the optimum fiber characteristics to minimize buckling and deformation during implantation. We found that the microstructures’ geometry and constituting materials have a dominant role in the bending stiffness in the direction parallel to the line that connects the two microstructures’ center, and the reduction of their relative distance leads to a significant flexibility variation.
2. Materials and methods
2.1 Fiber implant fabrication
The initial preforms were developed by using the rod-in-tube method [16]. The core rod materials with a diameter of 10 mm were inserted into cladding tubes with an outer diameter of 25 mm and inner core diameter of 10.5 mm. Two microfluidic channels were introduced in the cross-section by drilling another two 5 mm diameter holes from the end of the tube preform, as shown in Fig. 1(a). During the drilling process, the starting polymer cylinders were held by a clamp and drilled by a precision drill bit (HSCO, EJOT). The assembled preforms were loaded into a furnace in the fiber draw tower and the temperature was slowly increased to fiber drawing temperature, which is between 200°C and 230 °C for PC/FEP and PSU/FEP fibers [20] and 260 °C for COC/PSU fibers [16]. The preform feeding speed was set to 0.2 mm/min and the fiber drawing speed was selected according to the desired fiber diameter. Our target fiber diameter is between 200 µm and 500 µm since this is the most widely used range for fiber-based neural probes [15,17]. The diameter of the drawn fibers was continuously monitored by a laser diameter gauge (resolution <0.1µm). Here, we fabricated four types of POFs based on the following materials. Type I: PC and FEP act as the core and cladding materials, respectively, with two embedded MCs in the cladding. Type II: PSU and FEP are the core and cladding materials, respectively. Type III: Two cyclic olefin copolymers (COC) formed a step-index waveguide structure. We integrated two tungsten electrodes by adding an extra PSU overcladding. Type IV: PSU/FEP were used as the core/cladding materials, respectively with two indium electrodes integrated after the thermal drawing process. The cross-sections of the four types of microstructured POFs can be seen in Fig. 1(b). Type III fibers were fabricated using a modified thermal drawing method (“hot/cold” method [21]) that involves direct feeding of 50 µm diameter metal wires during the drawing process. This is because of the significant thermo-mechanical incompatibility between the materials, e.g., polymers (melting point: 200°C –300°C [16]) and tungsten wires (melting point: 3395°C [22]). While the polymer preform was heated and drawn into the fibers, the two tungsten wires were fed at the top of the draw tower through the two holes in the preform at a speed equal to the one of the fiber drawing (Fig. 1(c)). Type IV fibers were developed using a different post-processing method. Firstly, we fabricated the PSU/FEP fibers with two MCs using the standard thermal drawing method. Then, another fabrication step was employed to integrate the electrodes. One end of a 5 cm long piece of with MCs was connected to a syringe needle and all the gaps between the fiber and the needle were sealed with epoxy. The other end of the fiber was immersed in molten indium metal which was infiltrated into the MCs in the fiber by applying a negative pressure (Fig. 1(d)) [23].
Fig. 1. Fabrication steps of microstructured polymer optical fiber neural probes. (a) Rod-in-tube method for preform fabrication. (b) Cross-section of the fabricated four types of POFs. Type I: PC/FEP POFs with microfluidic channels. Type II: PSU/FEP POFs. Type III: COC1/COC2 POFs with Tungsten electrodes. Type IV: PSU/FEP POFs with indium electrodes. (c) Conventional thermal drawing method (left) and modified thermal drawing method with tungsten wire feeding (right). (d) The post-processing method to integrate indium electrodes into the PSU/FEP POF.
2.2 Characterization of the fiber-based neural probes
A high-power supercontinuum fiber laser (SuperK Extreme, NKT Photonics A/S) that spans from 400 up to 2400 nm was used as the light source for the characterization of the developed fibers. As high pump power is required to achieve the broadband supercontinuum light output, leading to a total average power larger than 5 W, a variable neutral density filter at the output of the laser was employed to reduce the launch power to ∼100 mW and avoid thermal damage at the fiber end-facets. The remaining reflected light was directed to a beam trap (BT610/M, Thorlabs). An achromatic objective (x10) was employed to couple the light into the core of the POFs. The output transmitted spectrum was measured using a spectrometer (CCS200, Thorlabs).
The electrical properties of the developed probes were characterized by using electrochemical impedance spectroscopy (EIS). A chemical impedance analyzer (IM3590, Hioki) was utilized in three-terminals configuration for the EIS study: A tungsten microwire coiled around a glass rod was employed as the counter electrode; an Ag/AgCl electrode was used as the reference electrode; the integrated electrodes in the POFs-based neural probes were acting as working electrodes. This characterization was performed in phosphate-buffered saline electrolyte solution. The impedance of the fiber devices was measured from 100 Hz to 10 kHz.
For the MCs characterization, we first exposed the channels from the side of the fibers. Then, a soft Teflon tube was used to connect the side exposed channels to the needle with a fluid-filled syringe. The delivery section was fully sealed using a small amount of epoxy resin (G14250, Thorlabs). To identify the delivery rate, a trypan blue staining dye (T8154, Sigma–Aldrich) was injected and delivered from the MCs to a brain phantom (0.6% agarose) at different delivery rates of 10, 50 100, 500, and 1000 nL/min. The injecting rate can be controlled using a sensitive electrical syringe pump (PHD 2000, Harvard Apparatus) [20].
2.3 Numerical analysis of mechanical properties
As discussed in the introduction, a critical factor during the implantation of a neural probe into the tissue is its ability to successfully penetrate the neural tissue without deformations. During the probe implantation process, fracture and buckling are the main mechanical failure modes for brittle materials (e.g., silica, silicon) and ductile materials (e.g., metals and flexible polymers), respectively [24]. Since all the materials we used to fabricate our probes were either polymers or metals we mainly focused our studies on the buckling effect that appears in the shank of the probe when a uniaxial load exceeds the critical load (defined as the maximum load that leads to buckling of the neural probe). This effect will eventually lead to elastic deformation and an unstable state of the probe that directly affects its exact location in the tissue as well as its overall performance [25]. During implantation, the fiber probe is slowly inserted with high precision by a stereotaxic arm into the brain. At the moment before the probe breaks the dura layer of the brain, the load at its tip reaches its maximum value. The buckling process during the insertion of the probe inside the brain was simulated using the finite element method (COMSOL multiphysics) (Fig. 2(a)). The upper end of the developed microstructured fiber-based neural probe model was fixed (in a real surgical setting this corresponds to the stereotaxic arm) and the load was applied at the other end of the fiber along the fiber’s main axis. We used a meshed model in our calculations to obtain the implant deformation with surface displacement (Fig. 2(a)). The critical load factor was calculated to evaluate the probe’s mechanical strength against buckling failure. This factor was expressed as the ratio between the critical load that leads to the buckling failure of the probe and the maximum load during the insertion [26]:
${P_c}$ is the critical load on the probe, while ${P_i}$ is the maximum load during implantation which is estimated to be 1 mN during the insertion process [27].Fig. 2. Applied force analysis of the probes. (a) Force analysis during the insertion of the neural probe into the brain tissue and the model built in COMSOL Multiphysics for the buckling analysis. (b) Force analysis after the probe was inserted into the brain tissue and the model built in COMSOL Multiphysics for the flexibility analysis (brain and skull models were created using the BioRender software provided by BioRender.com).
After the insertion, the repetitive relative displacement between the neural probe and the tissue caused by the respiration and vascular pulsations and the mechanical mismatch between the probe and the soft tissue would lead to the disruption of glial networks and the breach of the brain-blood barrier [21,22]. Therefore, in terms of chronic neural excitation and recording, it is important to increase the flexibility of the probe for accurate interrogation of the signals, since they can be affected by glial encapsulation of the probe [28,29]. Once the implantation is completed, the fiber probe is fixed using dental cement onto the skull of the animal. Consequently, minor tissue movements would lead to the bending of the probe. To simulate this effect, one end of the probe was fixed and a load was applied to the other end of the probe in the direction perpendicular to the main axis of the fiber (Fig. 2(b)). In this case, the bending stiffness is the parameter to evaluate and define the flexibility of the probe. The bending stiffness is defined as the force (F) required to reach a certain deflection (d) [30]:
E is the composite of Young’s modulus for the device (Table 1), I is the moment of inertia, and L is the length of the probe. A Von Mises stress distribution was used here for the estimation of stress in the probe (Fig. 2(b)).During the thermal drawing process of the probes, the fiber is scaled down approximately 60 times maintaining its geometry. Therefore, the ratio between the core diameter (or optical waveguide in the case of type III fibers), outer diameter, and MCs/electrodes diameters was set to be 5:2:1 based on the original preform dimensions.
Table 1. Young’s modulus and Poisson's ratio for different materials used in the finite element analysis
3. Results and discussion
3.1 Optical characterization, impedance, and drug delivery properties
We first characterized the developed neural probes in terms of their propagation loss by using the standard cut-back method. We measured the loss of the PC (Type I), PSU (Type II, IV), and COC1 core (Type III) POFs in the wavelength range from 400 nm up to 1000 nm (Fig. 3(a)). The minimum loss value for the Type I fibers was found to be less than 0.1 dB/cm as it has been already reported for PC-based optical fibers [36] while the minimum loss of Type III (COC) was found around 0.1 dB/cm at approximately 800 nm wavelength. The loss is increasing at shorter wavelengths due to the material electronic transition [37] and scattering loss [38]. The inherent stretching vibration of the carbon-hydrogen bonds (aliphatic and aromatic) of the material leads to distinct high absorption bands e.g., 860-875 nm, 900-912 nm, 980-1025 nm, and 1090-1100 nm [24,26]. These spectral features are commonly found in POFs based on conventional thermoplastic polymers, such as PMMA, PC, and COCs [3,8,16,36]. On the other hand, PSU polymer is an amorphous high-performance thermoplastic and it is known for its high stability at high temperatures [16]. This sulfone polymer is not colorless but has an amber yellow color. The Type II and IV (PSU core) have the highest loss from all the types of fibers with a minimum loss of less than 0.2 dB/cm at around 950 nm (Fig. 3(a)). At shorter wavelengths, the loss is rapidly increasing with significantly high values below 650 nm. However, since the typical implant for in vivo applications in rodents is less than 10 mm, this polymer can be also used for optogenetics.
Fig. 3. (a) Measured transmission loss of the different POF Types. (b) Measured impedance of the tungsten and indium metal electrodes integrated into the microstructured neural probes are shown in red and blue lines, respectively. (c) The measured output rate as a function of the injection rate during fluid delivery through the MCs in a 10 mm long POF neural probe. (d) Evaluation of the MCs in the POF neural probe by injecting blue dye into a phantom brain made by 0.6% agarose gel with a speed of 100 nL/min. The three inset images were taken at 2, 4, and 8 mins during the injection, respectively.
Impedance is the main property that defines the ability of an electrode to record high-quality electrophysiology data. We measured the impedance of Type III and IV POFs integrated with tungsten and indium wires, respectively (Fig. 3(b)). Tungsten metal was selected since it is one of the most commonly used electrode materials for electrophysiology in neuroscience [39] and indium because it has a low Young’s modulus (11 GPa [34] compared with 360 GPa [35] of tungsten). Also, indium has a lower melting temperature than the polymers used here (the melting point of indium is 156°C [40] and the fiber drawing temperature of PSU/FEP fibers is between 200°C and 230°C). The red line and shaded region illustrate the measured average impedance of six tungsten electrodes and their standard deviation, respectively. Similarly, the impedance of the indium electrodes and their deviation is shown by the blue line and shaded region. The average impedance of the tungsten electrodes decreased from 26 kΩ to 1.25 kΩ as the frequency increases from 0.1 to 10 kHz. On the other hand, the average impedance of indium electrodes is relatively stable between 20 to 22 kΩ in the same frequency range. For local field potential (LFP) and action potential (AP) recordings, these electrodes can achieve a good performance as long as the impedance is less than 2MΩ [41,42]. Therefore, we anticipate that fiber probes integrated with tungsten electrodes (Type III) and indium electrodes (Type IV) can provide a high-quality neural activity recording for LFPs and APs during in vivo investigations in acute and chronic settings.
The MCs were first tested by delivering 1 µL 0.4% trypan blue dye solution into in-house produced brain phantoms (0.6% agarose gel) with an injection speed of 100 nL/min. The preparation process of the brain phantoms has been described in [20]. Figure 3(c) shows the output rate of 10 µL trypan blue solution delivered using the two 50 µm diameter MCs with distinct injection rates of 10, 50, 100, 500, and 1000 nL/min, and demonstrates that the output delivery rate was equal to the injection rate, which suggests that the MCs in the microstructure fibers can be employed for accurate drug delivery applications and photopharmacology.
The dye was injected into the brain phantom and its uniform diffusion within 8 mins is shown in Fig. 3(d). Optical images of the dye solution at different injected volumes of 0.2 µL, 0.4 µL, and 0.8 µL into the brain phantom, respectively, reveal the spread of dye (insets, Fig. 3(d)).
3.2 Mechanical properties: buckling and flexibility
The critical load factor of the POFs has been firstly calculated as a function of the effective fiber length, defined here as the distance between the hold/fixed point of the stereotaxic arm and the tip of the fiber being inserted into the tissue (Fig. 4(a) and (b)). The diameter of all POFs has been normalized to 200 µm since the most commonly used optical fiber neural probes have a diameter between 200 µm and 400 µm [17,43]. To show the effect of the MCs on the critical load factor of the probes, a solid (unstructured) POF with PC core and FEP cladding has also been included in our calculations for direct comparison. It has been reported that the insertion success rate is around 50% when the critical load factor is 1.35 and increases to 100% when the critical load factor is 3.5 [44]. The red, grey, and green highlighted areas in Fig. 4 were used to indicate the region that the likelihood of insertion is lower and higher than 50%, and up to 100%, respectively.
Fig. 4. Critical load factor as a function of the (a), (b), effective fiber length and (c), (d) fiber diameter. The red, grey, and green regions (dashed grey lines) indicate the region where the critical load factor is lower than 1.35, between 1.35 and 3.5, and higher than 3.5, respectively, which corresponds to different insertion success rates. To see how the MCs affect the critical load factor, that of an unstructured PC/FEP solid fiber was plotted in red line for comparison.
It can be seen that the longer the effective fiber length the lower is the critical load factor, and thus it is more likely to obtain buckling during implantation. Therefore, the insertion success rate can be increased by reducing the effective probe length. For instance, from Fig. 4(a), the reduction of the effective fiber length from 20 mm to 10 mm for all POFs would double the success rate from 50% to 100%. On the other hand, as shown in Fig. 4(b), the insertion success rate of Type III probe that has tungsten electrodes is always 100% due to the stiff integrated tungsten electrodes.
Based on these results, a simple experimental method to increase the successful implantation rate during the experiment is to fix the probe as near as possible to the implanted tip to reduce the effective length of the probe during the insertion process. Alternatively, a stiff sleeve with an inner diameter slightly larger than the diameter of the implanted probe can also be used to reduce the effective length of the probe. The sleeve can be positioned above the surface of the tissue when the probe is loaded and implanted into the tissue. In this way, the effective length of the probe is the distance between its tip and the end of the sleeve. Therefore, the insertion success rate can be increased without leading to any primary tissue injury.
The probe’s diameter also has a major impact on the success rate during implantation (Fig. 4(c) and (d)). Increasing the diameter of the implant is thus another way to avoid buckling. However, it should be noted that this approach will likely increase the inflammation response during chronic experiments, since it increases the bending stiffness (Fig. 5). Here, we define ‘critical diameter’ as the parameter to describe the smallest fiber diameter required to achieve a 100% insertion success rate during the implantation. Solid PC/FEP and Type II have the same critical diameter: 161 µm, with a 10 mm effective fiber length. The MCs added in the Type I fibers cause the critical diameter to increase to 168 µm, and the indium electrodes in Type IV fibers reduce the critical diameter to 159 µm. The Type III fibers have the smallest critical diameter (∼99 µm) among all the proposed probes because of the high stiffness of the integrated tungsten electrodes. These results suggest that the integration of the MCs and indium electrodes has a limited impact on the buckling failure in POFs.
Fig. 5. Bending stiffness as a function of the fiber length (a), (b), fiber diameter (c), (d), microstructure (MCs and electrodes) diameter (e), (f), and microstructure distance (g), (h). To demonstrate how the MCs affect the critical load factor, an unstructured (all solid) PC/FEP POF was plotted in red line for comparison. The bending stiffness in the x-direction and the y-direction are shown in solid and dashed lines, respectively.
The length, diameter, microstructures (MCs and electrodes) diameter, and microstructures relative distance within the cross-section of the fiber all had an effect on the bending stiffness of the neural probes (Fig. 5). For effective comparison, the effective fiber length was normalized to 3 mm (implantation in the hippocampus of adult rats [45]) and the fiber diameter was normalized to 400 µm. To show how the MCs affect the bending stiffness of the probe, the bending stiffness of an unstructured PC/FEP POF was also included in our calculations and plotted for direct comparison.
When increasing the fiber length, the probe’s flexibility also increases (Fig. 5(a)). Furthermore, it can be noticed that the difference in stiffness between an unstructured (all solid) PC/FEP POF and Type II is negligible since they share the same cladding material (FEP) and their cores (PC and PSU) have similar Young’s moduli (Table 1). It’s important to notice here that the introduction of microstructures (MCs or indium electrodes) leads not only to an overall bending stiffness variation but also to a dependence of the bending stiffness on the direction of the applied force, because of the non-centrosymmetric cross-section geometry. The bending stiffness of the Type I fiber reaches its minimum in the x-axis direction (defined as the one across the two centers of the MCs in the cross-section of the fiber as seen in the inset of Fig. 5(a)) and its maximum in the y-direction. Comparing the Type I fiber with an unstructured PC/FEP fiber, it was found that the MCs reduce the bending stiffness of the fiber by 11 N/m in the x-direction, but only 0.4 N/m in the y-direction for a length of 3 mm. On the other hand, the integration of the indium electrodes causes the bending stiffness to increase by 230 N/m in the x-direction and 4.9 N/m in the y-direction in Type IV fiber. The obtained results suggest that the introduction of microstructure in the cross-section of the fiber implant can lead to a significant bending stiffness change only along the x-axis. Moreover, the larger Young’s modulus difference between the POF and the microstructure materials, the bigger the bending stiffness difference is.
The probes become less flexible as the diameter increases (Fig. 5(c-d)). It was found that the bending stiffness exponentially increases with the increase of the fiber diameter. For instance, the bending stiffness of PC/FEP POF increases from 0.29 N/m to 4.65 N/m when the fiber diameter increases from 100 µm to 200 µm and dramatically further increases from 74.3 N/m to 181.1 N/m when the diameter increases from 400 µm to 500 µm. This indicates that the diameter of the implant has a critical role in the bending stiffness properties and consequently in the inflammation rate of the tissue.
The bending stiffness as a function of the microstructures’ diameter is shown in Fig. 5(e) and (f). Interestingly, even if the diameter of the MCs or electrodes becomes as large as 100 µm – i.e. a significant part of the fiber is constituted by the microstructures’ material - the impact on the bending stiffness of the POFs in the y-direction is very limited. In addition, the distance between the two microstructure centers also affects the fiber flexibility as shown in Fig. 5(g) and (h). Similar to the microstructure diameter impact, as the distance of the microstructures increases, this leads to a negligible effect on the probe’s flexibility in the y-direction but a pronounced impact on the x-direction for all the fabricated fibers. The bending stiffness is thus directly related (for x-direction) to the microstructure distance. As an example, for the Type IV fibers, the bending stiffness in the x-direction with the electrode distance of 320 µm is twice as the distance of 180 µm. Since the geometry and location of the microstructures play an important role in the bending stiffness of the probes, an effective way to reduce the bending stiffness is to make the MCs separation as far as possible but bring the electrodes as near as possible to the center of the cross-section of the fiber.
Despite the impact of the fiber’s asymmetry being critical for the bending stiffness, as shown in Fig. 5, it was found to be negligible in terms of the critical load factor shown in Fig. 4. The reason is that the buckling failure mode always first appears in the direction with the larger flexibility, i.e. x-direction for the fibers with MCs and y-direction for POFs with metal electrodes. As a result, the critical load factors in Fig. 4 are based on the buckling failure mode in the most flexible direction.
4. Conclusion
In summary, we have demonstrated the fabrication of four novel microstructured polymer optical fibers as multifunctional neural interfaces. We experimentally and numerically investigated their properties for deep-brain optogenetics/stimulation, electrical recording, and photopharmacology as well as the required characteristics for successful implantation with minimum inflammation. Microfluidic channels, as well as indium and tungsten electrodes, were integrated into the microstructured POF to achieve multifunctionality. The impedance of the indium and tungsten electrodes was measured and found to be 21 kΩ and 4.7 kΩ at 1 kHz, respectively. Accurate on-demand drug delivery can be achieved by the microfluidic channels in the cladding of the probes. Numerical analysis by finite element method has been performed to model the buckling failure during the probe insertion process into the tissue and inflammatory response-related probe flexibility after its insertion. Our results suggest that reducing the effective fiber length or increasing the fiber diameter can help to achieve a higher success rate during insertion. It has also been found that the geometry of microstructures can have a significant effect on the probe’s flexibility in the direction connecting the centers of the microstructure. The effect in the perpendicular direction is instead limited, even with a large difference between the Young’s moduli of the microstructure and the main POF material. In addition, the distance between the two microstructures also affects the overall bending stiffness of the probes. We believe that the presented results will provide a general guideline in the design considerations for the development of novel fiber-based neural with limited buckling failure during probe insertion and suppressed FBR in chronic experiments, towards the next-generation neural interfaces.
Funding
Villum Fonden (36063); Lundbeckfonden (Multi-BRAIN, R276-2018-869, R380-2021-1171).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. K. Peters, “Polymer optical fiber sensors—a review,” Smart Mater. Struct. 20(1), 013002 (2011). [CrossRef]
2. J. B. Jensen, P. E. Hoiby, G. Emiliyanov, O. Bang, L. H. Pedersen, and A. Bjarklev, “Selective detection of antibodies in microstructured polymer optical fibers,” Opt. Express 13(15), 5883–5889 (2005). [CrossRef]
3. C. Markos, W. Yuan, K. Vlachos, G. E. Town, and O. Bang, “Label-free biosensing with high sensitivity in dual-core microstructured polymer optical fibers,” Opt. Express 19(8), 7790–7798 (2011). [CrossRef]
4. A. G. Leal-Junior, C. A. Diaz, L. M. Avellar, M. J. Pontes, C. Marques, and A. Frizera, “Polymer optical fiber sensors in healthcare applications: A comprehensive review,” Sensors 19(14), 3156 (2019). [CrossRef]
5. C. Broadway, R. Min, A. G. Leal-Junior, C. Marques, and C. Caucheteur, “Toward commercial polymer fiber Bragg grating sensors: Review and applications,” J. Lightwave Technol. 37(11), 2605–2615 (2019). [CrossRef]
6. A. Theodosiou and K. Kalli, “Recent trends and advances of fibre Bragg grating sensors in CYTOP polymer optical fibres,” Opt. Fiber Technol. 54, 102079 (2020). [CrossRef]
7. C. Markos, A. Stefani, K. Nielsen, H. K. Rasmussen, W. Yuan, and O. Bang, “High-T g TOPAS microstructured polymer optical fiber for fiber Bragg grating strain sensing at 110 degrees,” Opt. Express 21(4), 4758–4765 (2013). [CrossRef]
8. G. Woyessa, A. Fasano, C. Markos, A. Stefani, H. K. Rasmussen, and O. Bang, “Zeonex microstructured polymer optical fiber: fabrication friendly fibers for high temperature and humidity insensitive Bragg grating sensing,” Opt. Mater. Express 7(1), 286–295 (2017). [CrossRef]
9. W. Yuan, L. Khan, D. J. Webb, K. Kalli, H. K. Rasmussen, A. Stefani, and O. Bang, “Humidity insensitive TOPAS polymer fiber Bragg grating sensor,” Opt. Express 19(20), 19731–19739 (2011). [CrossRef]
10. G. Woyessa, A. Theodosiou, C. Markos, K. Kalli, and O. Bang, “Single peak fiber Bragg grating sensors in tapered multimode polymer optical fibers,” J. Lightwave Technol. 39(21), 6934–6941 (2021). [CrossRef]
11. J. Bonefacino, H.-Y. Tam, T. S. Glen, X. Cheng, C.-F. J. Pun, J. Wang, P.-H. Lee, M.-L. V. Tse, and S. T. Boles, “Ultra-fast polymer optical fibre Bragg grating inscription for medical devices,” Light: Sci. Appl. 7(3), 17161 (2017). [CrossRef]
12. S. Farajikhah, A. F. Runge, B. B. Boumelhem, I. D. Rukhlenko, A. Stefani, S. Sayyar, P. C. Innis, S. T. Fraser, S. Fleming, and M. C. Large, “Thermally drawn biodegradable fibers with tailored topography for biomedical applications,” J. Biomed. Mater. Res., Part B 109(5), 733–743 (2021). [CrossRef]
13. C. Markos, I. Kubat, and O. Bang, “Hybrid polymer photonic crystal fiber with integrated chalcogenide glass nanofilms,” Sci. Rep. 4(1), 6057 (2014). [CrossRef]
14. C. Markos, J. C. Travers, A. Abdolvand, B. J. Eggleton, and O. Bang, “Hybrid photonic-crystal fiber,” Rev. Mod. Phys. 89(4), 045003 (2017). [CrossRef]
15. S. Park, G. Loke, Y. Fink, and P. Anikeeva, “Flexible fiber-based optoelectronics for neural interfaces,” Chem. Soc. Rev. 48(6), 1826–1852 (2019). [CrossRef]
16. P. Akrami, A. I. Adamu, G. Woyessa, H. K. Rasmussen, O. Bang, and C. Markos, “All-polymer multimaterial optical fiber fabrication for high temperature applications,” Opt. Mater. Express 11(2), 345–354 (2021). [CrossRef]
17. A. Canales, X. Jia, U. P. Froriep, R. A. Koppes, C. M. Tringides, J. Selvidge, C. Lu, C. Hou, L. Wei, Y. Fink, and P. Anikeeva, “Multifunctional fibers for simultaneous optical, electrical and chemical interrogation of neural circuits in vivo,” Nat. Biotechnol. 33(3), 277–284 (2015). [CrossRef]
18. S. Jiang, D. C. Patel, J. Kim, S. Yang, W. A. Mills III, Y. Zhang, K. Wang, Z. Feng, S. Vijayan, W. Cai, A. Wang, Y. Guo, I. F. Kimbrough, H. Sontheimer, and X. Jia, “Spatially expandable fiber-based probes as a multifunctional deep brain interface,” Nat. Commun. 11(1), 6115 (2020). [CrossRef]
19. S. Park, H. Yuk, R. Zhao, Y. S. Yim, E. W. Woldeghebriel, J. Kang, A. Canales, Y. Fink, G. B. Choi, X. Zhao, and P. Anikeeva, “Adaptive and multifunctional hydrogel hybrid probes for long-term sensing and modulation of neural activity,” Nat. Commun. 12(1), 3435 (2021). [CrossRef]
20. K. Sui, M. Meneghetti, J. Kaur, R. J. F. Sørensen, R. W. Berg, and C. Markos, “Adaptive polymer fiber neural device for drug delivery and enlarged illumination angle for neuromodulation,” J. Neural Eng. 19(1), 016035 (2022). [CrossRef]
21. M. Meneghetti, C. R. Petersen, R. E. Hansen, A. I. Adamu, O. Bang, and C. Markos, “Thermally tunable dispersion modulation in a chalcogenide-based hybrid optical fiber,” Opt. Lett. 46(10), 2533–2536 (2021). [CrossRef]
22. I. Langmuir, “The Melting-Point of Tungsten,” Phys. Rev. 6(2), 138–157 (1915). [CrossRef]
23. M. Meneghetti, J. Kaur, K. Sui, J. F. Sørensen, R. W. Berg, and C. Markos, “Soft monolithic infrared neural interface for simultaneous neurostimulation and electrophysiology,” Light: Sci. Appl. 12(1), 127 (2023). [CrossRef]
24. H. H. Draz, S. R. I. Gabran, M. Basha, H. Mostafa, M. F. Abu-Elyazeed, and A. Zaki, “Comparative mechanical analysis of deep brain stimulation electrodes,” BioMed Eng OnLine 17(1), 123 (2018). [CrossRef]
25. S. R. I. Gabran, M. T. Salam, J. Dian, Y. El-Hayek, J. L. Perez Velazquez, R. Genov, P. L. Carlen, M. M. A. Salama, and R. R. Mansour, “3-D Flexible Nano-Textured High-Density Microelectrode Arrays for High-Performance Neuro-Monitoring and Neuro-Stimulation,” IEEE Trans. Neural Syst. Rehabil. Eng. 22(5), 1072–1082 (2014). [CrossRef]
26. S. Rezaei, Y. Xu, and S. W. Pang, “Control of neural probe shank flexibility by fluidic pressure in embedded microchannel using PDMS/PI hybrid substrate,” PLoS One 14(7), e0220258 (2019). [CrossRef]
27. M. Salcman and M. J. Bak, “A new chronic recording intracortical microelectrode,” Med. Biol. Eng. 14(1), 42–50 (1976). [CrossRef]
28. W. M. Grill, S. E. Norman, and R. V. Bellamkonda, “Implanted Neural Interfaces: Biochallenges and Engineered Solutions,” Annu. Rev. Biomed. Eng. 11(1), 1–24 (2009). [CrossRef]
29. V. S. Polikov, P. A. Tresco, and W. M. Reichert, “Response of brain tissue to chronically implanted neural electrodes,” J. Neurosci. Methods 148(1), 1–18 (2005). [CrossRef]
30. Y. Dai, Y. Xue, and J. Zhang, “Vibration-Based Milling Condition Monitoring in Robot-Assisted Spine Surgery,” IEEE/ASME Trans. Mechatron. 20(6), 3028–3039 (2015). [CrossRef]
31. E. D. Steffler, J. S. Epstein, and E. G. Conley, “Energy partitioning for a crack under remote shear and compression,” Int. J. Fract. 120(4), 563–580 (2003). [CrossRef]
32. D. Olmos, S. G. Prolongo, and J. González-Benito, “Thermo-mechanical properties of polysulfone based nanocomposites with well dispersed silica nanoparticles,” Composites, Part B 61, 307–314 (2014). [CrossRef]
33. A. I. M. Greer, I. Vasiev, B. Della-Rosa, and N. Gadegaard, “Fluorinated ethylene–propylene: a complementary alternative to PDMS for nanoimprint stamps,” Nanotechnology 27(15), 155301 (2016). [CrossRef]
34. M. Calin, A. Helth, J. J. Gutierrez Moreno, M. Bönisch, V. Brackmann, L. Giebeler, T. Gemming, C. E. Lekka, A. Gebert, R. Schnettler, and J. Eckert, “Elastic softening of β-type Ti–Nb alloys by indium (In) additions,” J. Mech. Behav. Biomed. Mater. 39, 162–174 (2014). [CrossRef]
35. J. Cui, L. Ma, G. Chen, N. Jiang, P. Ke, Y. Yang, S. Wang, K. Nishimura, and J. Llorca, “Effect of twin boundaries on the strength of body-centered cubic tungsten nanowires,” Mater. Sci. Eng., A 862, 143826 (2023). [CrossRef]
36. A. Fasano, G. Woyessa, P. Stajanca, C. Markos, A. Stefani, K. Nielsen, H. K. Rasmussen, K. Krebber, and O. Bang, “Fabrication and characterization of polycarbonate microstructured polymer optical fibers for high-temperature-resistant fiber Bragg grating strain sensors,” Opt. Mater. Express 6(2), 649–659 (2016). [CrossRef]
37. T. Y. T. Yamashita and K. K. K. Kamada, “Intrinsic Transmission Loss of Polycarbonate Core Optical Fiber,” Jpn. J. Appl. Phys. 32(6R), 2681 (1993). [CrossRef]
38. A. Tanaka, H. Sawada, T. Takoshima, and N. Wakaisuki, “New Plastic Optical Fiber With Polycarbonate Core And Fluorescence-Doped Fiber For High Temperature Use,” in Fiber Optic Systems for Mobile Platforms (SPIE, 1987), 0840, pp. 19–28.
39. H. Li, J. Wang, and Y. Fang, “Recent developments in multifunctional neural probes for simultaneous neural recording and modulation,” Microsyst. Nanoeng. 9(1), 4 (2023). [CrossRef]
40. J. T. Krause, “Gold-Indium Bond for Measurement of Ultrasonic Properties in Solids at High Temperatures,” J. Appl. Phys. 39(11), 5334–5335 (1968). [CrossRef]
41. J. P. Neto, P. Baião, G. Lopes, J. Frazão, J. Nogueira, E. Fortunato, P. Barquinha, and A. R. Kampff, “Does Impedance Matter When Recording Spikes With Polytrodes?” Front. Neurosci. 12, (2018). [CrossRef]
42. J. Jiang, F. R. Willett, and D. M. Taylor, “Relationship between microelectrode array impedance and chronic recording quality of single units and local field potentials,” in 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (2014), pp. 3045–3048.
43. A. M. Aravanis, L.-P. Wang, F. Zhang, L. A. Meltzer, M. Z. Mogri, M. B. Schneider, and K. Deisseroth, “An optical neural interface:in vivocontrol of rodent motor cortex with integrated fiberoptic and optogenetic technology,” J. Neural Eng. 4(3), S143–S156 (2007). [CrossRef]
44. S. Singh, M.-C. Lo, V. B. Damodaran, H. M. Kaplan, J. Kohn, J. D. Zahn, and D. I. Shreiber, “Modeling the Insertion Mechanics of Flexible Neural Probes Coated with Sacrificial Polymers for Optimizing Probe Design,” Sensors 16(3), 330 (2016). [CrossRef]
45. S.-J. Zhang, J. Ye, C. Miao, A. Tsao, I. Cerniauskas, D. Ledergerber, M.-B. Moser, and E. I. Moser, “Optogenetic dissection of entorhinal-hippocampal functional connectivity,” Science 340(6128), 1232627 (2013). [CrossRef]
