Optoelectronic Neural Interfaces Based on Quantum Dots

Abstract

Optoelectronic modulation of neural activity is an emerging field for the investigation of neural circuits and the development of neural therapeutics. Among a wide variety of nanomaterials, colloidal quantum dots provide unique optoelectronic features for neural interfaces such as sensitive tuning of electron and hole energy levels via the quantum confinement effect, controlling the carrier localization via band alignment, and engineering the surface by shell growth and ligand engineering. Even though colloidal quantum dots have been frontier nanomaterials for solar energy harvesting and lighting, their application to optoelectronic neural interfaces has remained below their significant potential. However, this potential has recently gained attention with the rise of bioelectronic medicine. In this review, we unravel the fundamentals of quantum-dot-based optoelectronic biointerfaces and discuss their neuromodulation mechanisms starting from the quantum dot level up to electrode-electrolyte interactions and stimulation of neurons with their physiological pathways. We conclude the review by proposing new strategies and possible perspectives toward nanodevices for the optoelectronic stimulation of neural tissue by utilizing the exceptional nanoscale properties of colloidal quantum dots.

Keywords: nanocrystal; neural interface; neural stimulation; optoelectronics; quantum dot.

Figure 1

Applications of semiconductor quantum dots for neurotechnology (top). Schematics for the three main configurations that can lead to neural stimulation using quantum dots (bottom). The free-standing configuration represents the interaction between the targeted cells and the QDs in the extracellular medium without any physical, chemical, or biological attachment to the cell membrane. The second configuration (bottom middle) exhibits the interaction between the targeted cells and the QDs, which may bind to the cell membrane through QD–antibody conjugates or via conjugation with target specific ligands, such as peptides and proteins. The third configuration (bottom right) utilizes QDs in thin-film or blend form. Neuron–QD interaction depends on the chemical, physical, or ionic stimuli generated by QDs.

Figure 2

(a) Quantum dots with stepping emission from blue to red (top). Representative photoluminescence spectrum for different size quantum dots (middle). Representative conduction and valence band diagram for different sizes of semiconductor quantum dots (bottom). (b) Representative TEM images for core/shell quantum dots (scale bars: 20 and 5 nm, respectively). (c) Core/shell semiconductor nanoparticle systems with type I, quasi-type II, and type II band alignment.

Figure 3

(a) Schematic diagram of a generic photovoltaic biointerface architecture. (b) Energy band diagram of a regular (top) and inverted optoelectronic system (bottom). The layers represent the corresponding layers in panel a. The exciton generation occurs in the active layer upon illumination. Dissociated electron and the hole move toward the charge transport layers according to the energy levels between the layers.

Figure 4

Quantum mechanical simulations of type I and type II nanocrystals, showing the effect of wave function engineering on charge carrier localization. (a) Top: Blue lines show energy band alignment and black lines show minimum electron and hole discrete energy levels. R and R+H correspond to the radius of the InP core and the InP/ZnO core/shell QDs, respectively. Bottom: Simulated electron and hole wave functions for the InP core (top) and the InP/ZnO core/shell (bottom) QDs. Black and red lines show electron and hole radial probability functions, respectively. Blue line represents the electron confinement potential. Dashed black and red lines represent single electron and hole energies, respectively. Reprinted with permission from ref (69). Copyright 2018 American Chemical Society. (b) Top: Energy band alignment schematics of type I InP/ZnS and type II InP/ZnO/ZnS QDs. Bottom: Simulated electron (red lines) and hole (black lines) wave functions for InP core, InP/ZnS core/shell, and InP/ZnO/ZnS core/shell/shell nanocrystals. Blue lines represent the electron confinement potential. Reprinted with permission from ref (83). Copyright 2021 Springer Nature. In both studies, type II nanostructures exhibit electron delocalization to shells, which leads to decreased electron–hole wave function overlap, i.e., reduced exciton binding energy.

Figure 5

Primary charge injection mechanisms in QD-based biointerfaces. (a) Illustration of faradaic and capacitive charge injection mechanisms. Electrons or holes accumulate on the biointerface surface, inducing faradaic or capacitive charge injection in the electrolyte (hole accumulation was shown as a representative case). IHP, inner Helmholtz layer, OHP, outer Helmholtz layer, GCL, Gouy–Chapman diffuse-charge layer. (b) Electrical circuit model of the electrode–electrolyte interface. C_dl represents double-layer capacitance and is equivalent to the series sum of IHP, OHP, and GCL capacitances. R_CT and R_s denotes charge transfer resistance and solution resistance, respectively. W represents Warburg impedance. (c) Typical current–voltage profiles of resistive and capacitive elements.

Figure 6

Obtaining capacitive-dominant QD-based biointerfaces via donor–acceptor nanoheterojunction (a–c) and band alignment engineering (d, e). (a) Device structure of QD-fullerene donor–acceptor nanoheterojunction-based interfaces for obtaining capacitive-dominant photoresponse. (b) Schematic showing the electron transfer from QD to fullerene derivative PCBM upon photoexcitation. (c) Capacitive photocurrent generation mechanism showing each step in a consecutive manner. E_b, exciton binding energy; E_cb, E_vb, E_ss, conduction band, valence band, and surface state energy levels, respectively. Band bending at the electrolyte interface prevents electron transfer to electrolyte. Electrons will then be transferred to ZnO and ITO because of the high electron mobility of ZnO. Holes are trapped in the QD valence band because of the ZnO valence band level, which induces capacitive photocurrent. Panels a, b, and c reprinted with permission from ref (83). Copyright 2021 Springer Nature. (d) Schematic of device architecture containing photoactive layers of P3HT:PbS:PCBM, with the intermediate layer (ZnO, MoOx, or none) coated on glass/ITO substrates. Although the photoactive layer generates excitons within the device, the energy level of the intermediate layer determines the surface polarity by routing either the electrons or the holes toward the top surface layer. (e) Manipulation of the band alignment via choice of different intermediate layers. Type I and type II devices generate faradaic-dominant photocurrents due to electron transfer from the PCBM LUMO level to electrolyte. Type III architecture accumulates holes on the surface. Holes do not interact with the electrolyte because of the unfavorable energy level of P3HT HOMO and water oxidation levels, which leads to capacitive-dominant photoresponse. Panels d and e reprinted with permission from ref (71). Copyright 2018 American Physical Society.

Figure 7

Pioneering semiconductor nanoparticle-based optoelectronic neural interfaces. (a) HgTe QDs stabilized with thioglycolic acid-coated single-material device. (b) Light absorption characteristics (1, solid line) and photogenerated voltage (2, bars) of HgTe QDs and layer-by-layer films. UV–vis absorption spectrum on HgTe QD dispersion stabilized by thioglycerol used for fabrication of LBL films. (c) Action potential responses of NG108 cells grown on (PDDA/HgTe)₁₂ + (PDDA/Clay)₂ under photostimulus with and without tetrodotoxin (TTX). Panels a, b, and c reprinted with permission from ref (67). Copyright 2007 American Chemcial Society. (d) Schematic of the interaction between a QD and cell membrane. (e) UV–visible absorbance and photoluminescence (PL) characterization of CdTe QDs. (f) Current-clamped recording of cortical neurons on CdSe QD film. Fluorescence image of a micropipette coated with CdSe QDs used for single-cell stimulation. Panels d, e, and f reprinted with permission from ref (68). Copyright 2012 The Optical Society. (g) Schematic of the optoelectronic coupling between NR-conjugated CNT coated by ppAA. (h) Schematic drawing of the CdSe–GSH QDs (left), CdSe/CdS–GSH QDs (center), and CdSe/CdS–GSH NRs (right). Average photocurrents for different devices based on CdSe, CdSe/CdS, and CdSe/CdS NRs with CNTs under an excitation pulse of 30 mW cm^–2 for 100 ms with a 405 nm illumination source. (i) (Upper left) SEM image of an NR–CNT film (scale bar: 100 nm). (Upper right) CNT electrode array on a PDMS flexible support (scale bar: 1 mm). (Bottom) Extracellular voltage trace recorded from a chick retina following 100 ms light stimulation (405 nm, pulse interval of 30 ms) under different intensities (1.2, 3, 6, and 12 mW cm^–2). Panels g, h, and I reprinted with permission from ref (113). Copyright 2014 American Chemical Society

Figure 8

InP QD-based optoelectronic neural interfaces. (a) Schematic illustration of the photoelectrode fabrication steps and energy band diagram of the device architecture. (b) Photocurrent performance of TiO₂, InP core and InP/ZnO QD coated biointerfaces. (c) Photostimulation of a PC12 cell on the photoelectrode under 4 μW mm^–2 illumination (red bar, time period under illumination; blue bar, no illumination). Panels a, b and c reprinted with permission from ref (69). Copyright 2018 American Chemical Society. (d) Energy band diagram of bidirectional device architectures. (e) TEM image of the InP/ZnS QDs. (f) Transmembrane potential recordings of neurons on type I, type II, and ITO control samples (illumination: blue LED at 445 nm, 10 ms pulse width, 2 mW mm^–2 optical power density; blue bar indicates the 10 ms “light on” interval). Panels d, e, and f reprinted with permission from ref (88). Copyright 2021 Frontiers.

Figure 9

(a) Artificial antenna complexes made of rainbow InP quantum dots showing nonradiative energy transfer toward the cell interface. (b) (Upper inset) Photograph of the colloidal green-, yellow-, and red-emitting QDs under UV illumination. (Bottom) Energy band diagram of the quantum funnel biointerface (c) Photostimulation of the SH-SY5Y cell on the quantum funnel biointerface under illumination of 169 mW cm^–2 with 50 ms illumination pulses. Panels a, b, and c reprinted with permission from ref (70). Copyright 2019 American Chemical Society. (d) Energy band alignment of the QD integrated biointerface. InP/ZnS core/shell and InP/ZnO/ZnS core/shell/shell QDs were incorporated into the photoelectrode architecture. (e) Photocurrent density traces of the devices with InP/ZnO/ZnS:PCBM volume ratios of 1:1 (black), 1:3 (red), and 1:7 (orange). The inset shows the components of the photocurrent. Capacitive current is the peak photocurrent reached after the light onset, whereas resistive current is the photocurrent remained after 90% of the illumination duration has passed. (f) Ratios of the capacitive to resistive components for devices with different QD:PCBM mixing ratios. Panels d, e, and f reprinted with permission from ref (83). Copyright 2021 Springer Nature.

Figure 10

PbS- and AlSb-based neural interfaces. (a) (Top) Photocapacitive current levels of ITO/ZnO/P3HT:PCBM and ITO/ZnO/P3HT:PbS-QDs:PCBM photoelectrodes. (Bottom) Capacitive and faradaic components of type I, type II, and type III photoelectrodes under illumination of 10 ms light pulses with an intensity of 1 mW cm^–2. The architecture for different types of biointerfaces was explained in panels c and d in Figure 6. (b) Membrane potential variation of SH-SY5Y cells grown on the type III biointerface in panel d upon light illumination (10 ms, 1 mW cm^–2). Panels a and b reprinted with permission from ref (71). Copyright 2019 American Physical Society. (c) Atomic force microscopy (5 μm × 5 μm) of P3HT:PCBM surfaces with the optimized binary ratio of 2:1 on ITO/ZnO-coated glass substrates (left, 2D views; right, 3D views) with various thin film thicknesses (t) in tapping-mode. R_a shows the average surface roughness. (d) Peak photocurrent for the binary photoelectrodes as a function of various thin film thicknesses. Panels c and d reprinted with permission from ref (101). Copyright 2020 The Optical Society. (e) Structure of the AlSb integrated biointerface (left inset: cross-sectional SEM image) and energy band diagram of the proposed device. (f) Intracellular membrane potential change with respect to a distant Ag/AgCl electrode was measured after the photostimulation of primary hippocampal neurons on the glass:ITO control (red) and the biointerface (black) under illumination of 100 mW cm^–2 with 20 ms illumination pulses. Blue semitransparent area shows the 445 nm light illumination period (g) Successful spike ratio of neurons on the glass:ITO/ZnO/P3HT control (gray) and the biointerface (black) under different illumination frequencies of 20 ms, 50 mW cm^–2, and 20 pulses (n = 20, mean ± s.d.). Panels e, f and g reprinted with permission from ref (102). Copyright 2021 Springer Nature.

Figure 11

Biocompatibility of quantum dots for biomedical applications. (a) Oxidation mechanism of Cd-based nanoparticles. Reprinted with permission from ref (164). Copyright 2004 American Chemical Society. (b) Polymer encapsulation strategy for colloidal quantum dots. (A) Native nonpolar ligands remain intact and (B) amphiphilic polymer encapsulate the QD for water solubility. (C) Chemically reactive and polar group for bioconjugation. Reprinted with permission from ref (165). Copyright 2011 Elsevier. (c) Cell viability of hepatocytes as assessed by mitochondrial activity of CdSe QD-treated cultures relative to untreated controls under exposure to air and UV treatment. Reprinted with permission from ref (164). Copyright 2004 American Chemical Society. (d) Effect of ZnS coating on CdSe quantum dots on cytotoxicity and oxidation. Reprinted with permission from ref (164). Copyright 2004 American Chemical Society. (e) Cell viability of MCF-7 cells incubated with different concentrations of InP/ZnS QDs and CdSe/ZnS QDs for 24 h. Reprinted with permission from ref (167). Copyright 2017, Royal Society of Chemistry. (f) Cell viability and cytotoxicity assessment of InP/ZnO quantum dots with MTT (upper left), LDH assay (upper right), and visualized cell morphology via DAPI staining and actin immunolabeling (bottom, scale bar: 50 μm). Reprinted with permission from ref (69). Copyright 2018 American Chemical Society. (g) Immunofluorescence imaging of primary hippocampal neurons grown on AlSb NC-coated biointerfaces. PHNs costained with DAPI, Anti-NeuN (red), and anti-F-actin (green) (scale bar: 75 μm). Reprinted with permission from ref (102). Copyright 2021 Springer Nature.