Trionic optical biological voltage sensing

Abstract

Disclosed are methods, systems, and devices for biological voltage sensing using optical signals and their underlying quantum statistics. The disclosed technology enables the use of optical signals to record biological electrical activity from excitable cells using quantum transitions in a monolayer semiconductor material. In some aspects, a method for detecting biological electrical activity includes illuminating a probe material with a light source such that the probe material emits an optical signal, wherein the probe material is positioned in proximity to one or more electroactive cells and the optical signal responds to changes in a surrounding electric field; measuring the optical signal emitted by the probe material; decomposing the optical signal into an excitonic component and a trionic component; and detecting an electrical signal from the one or more electroactive cells based on a magnitude of the trionic component relative to a magnitude of the excitonic component of the optical signal.

Description

PCT Application Attorney Docket No.: 009062.8585.WO00TRIONIC OPTICAL BIOLOGICAL VOLTAGE SENSING CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This patent document claims priority to and benefits of U. S. Provisional Patent Application No. 63 / 764,337 entitled “TRIONIC ALL-OPTICAL BIOLOGICAL VOLTAGE SENSING” filed on February 27, 2025. The entire content of the aforementioned patent application is incorporated by reference as part of the disclosure of this patent document.TECHNICAL FIELD

[0002] This patent document relates to devices, systems, and processes for the optical detection of biological electrical activity.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0003] This invention was made with government support under EY033676 awarded by the National Institutes of Health and under ECCS2139416 awarded by the National Science Foundation. The government has certain rights in the invention.BACKGROUND

[0004] Recording biological electrical activity with high spatial and temporal resolution is critical for developing an understanding of how the brain and other organs work. Two existing methods used to this end are electrophysiology and optical microscopy. Conventional electrodebased technologies used in electrophysiology provide very high temporal resolution but cannot be scaled to provide cellular spatial resolution. On the other hand, optical microscopy naturally provides cellular resolution, but calcium and voltage indicators are slow, providing only limited temporal resolution.SUMMARY

[0005] Methods, devices, and systems for biological voltage sensing using optical signals and their underlying quantum statistics are described. The present technology enables the use of optical signals to record biological electrical activity from excitable cells using quantum transitions in a monolayer semiconductor material.

[0006] In some example embodiments, the present technology utilizes exciton to trionPCT Application Attorney Docket No.: 009062.8585.WO00conversion in a probe material, such as a monolayer semiconductor with picosecond time resolution, which in combination with the voltage sensitivity of fluorescence from these materials, provides a solution to the challenge of achieving high spatial and temporal resolution in detection of biological electrical activity. The materials used in these example embodiments have voltage sensitive fluorescence with a quick temporal response, providing a way to resolve fast biological electrical activity. Thus, the present technology is envisioned to improve optical microscopy based on electrophysiology.

[0007] In some embodiments in accordance with the present technology, a method for detecting biological electrical activity includes illuminating a probe material with a light source such that the probe material emits an optical signal, wherein the probe material is positioned in proximity to one or more electroactive cells and the optical signal is sensitive to changes in a surrounding electric field; measuring the optical signal emitted by the probe material; processing the optical signal into an excitonic component and a trionic component; and detecting an electrical signal from the one or more electroactive cells based on a magnitude of the trionic component relative to a magnitude of the excitonic component of the optical signal.

[0008] In some embodiments in accordance with the present technology, a system for detecting biological electrical activity includes a probe material configured to emit an optical signal when illuminated, wherein the optical signal is sensitive to changes in surrounding electric field caused by one or more electroactive cells; a light source configured to illuminate the probe material such that the probe material emits the optical signal; an optical detector configured to measure the optical signal emitted by the probe material; and one or more processors in communication with the optical detector, configured to process the optical signal into an excitonic component and a trionic component, and detect an electrical signal from the one or more electroactive cells based on a magnitude of the trionic component relative to a magnitude of the excitonic component of the optical signal.

[0009] The subject matter described in this patent document can be implemented in specific ways that provide one or more of the following features.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] FIG. 1A shows a schematic illustration of an example embodiment of a system for detecting biological electrical activity, in accordance with the present technology.PCT Application Attorney Docket No.: 009062.8585.WO00

[0011] FIG. IB shows an illustration of an example embodiment of the present technology wherein nanoparticles are used to detect biological electrical activity.

[0012] FIG. 1C shows a schematic of an example experimental test structure used to investigate the voltage dependent PL intensity from trionic monolayer materials.

[0013] FIG. ID shows a schematic showing the dependence of PL intensity on the Fermi level.

[0014] FIG. 2A shows a schematic of an example process for fabrication of encapsulated gold electrodes on a glass substrate.

[0015] FIG. 2B shows a schematic for an example transfer process for chemical vapor deposition (CVD)-grown MoS2.

[0016] FIGS. 3 A and 3B show schematic illustrations of photoluminescence modulation in MoS2via electrochemical gating.

[0017] FIG. 4 shows a data plot showing example photoluminescence spectra of monolayer MoS2in air and PBS.

[0018] FIG. 5A shows a schematic illustration of an electric double layer on the surface of a monolayer material.

[0019] FIG. 5B shows an illustration of an energy band diagram showing the Fermi level and trionic energy level for a monolayer material.

[0020] FIG. 6A shows an example plot of calculated PL intensity and calculated voltage responsivity vs Fermi level normalized to the trion energy.

[0021] FIG. 6B shows the measured external bias voltage dependence of the PL intensity of MoS2and WS2.

[0022] FIGS. 6C-6F show the measured external bias voltage dependence of the PL intensity of MoS2, WS2, WSe2, and MoSe2.

[0023] FIGS. 7 A and 7B show data plots illustrating the change in PL spectra for MoS2with applied bias voltages

[0024] FIGS. 8A-8E, 9A-9E, 10A-10E, and 11A-11E show additional details of voltage responsivity characterization in MoS2, WS2, MoSe2, and WSe2, respectively.

[0025] FIG. 12 shows a schematic illustration of an example one photon microscopy apparatus with two detection modes.

[0026] FIG. 13 shows a schematic illustration of an example two-photon microscopy apparatus.PCT Application Attorney Docket No.: 009062.8585.WO00

[0027] FIGS. 14A-14C show images of MoS2photoluminescence imaging and modulation with two-photon excitation.

[0028] FIGS. 15 A and 15B show data plots showing measured fractional PL intensity in time for different square wave amplitudes

[0029] FIGS. 16A-16C show data plots comparing signal traces of the photodetector output voltage with no input and that generated by the PL from a floating MoS2crystal.

[0030] FIG. 16D shows a data plot showing MoS2PL responsivity.

[0031] FIG. 17A shows a data plot of the measured frequency response of the voltage sensitive trionic PL for sinusoidal modulations.

[0032] FIG. 17B shows data plots showing simulated electrochemical response of the MoS2 / gold electrode in PBS using an equivalent circuit model, and the corresponding measured fraction PL change of the MoS2 / gold electrode.

[0033] FIG. 18 shows a data plot comparing the measured dependence of fractional PL of an electrically floating MoS2monolayer on the distance to a Pt / Ir electrode and a corresponding simulation.

[0034] FIG. 19A shows a schematic illustration of photoluminescence modulation via local electric field without electrical contact.

[0035] FIG. 19B shows a data plot of a COMSOL simulation of electric field and potential distributions induced by applying a bias through a microelectrode placed above MoS2.

[0036] FIG. 19C shows a data plot showing an example PL response as a function of the electrode’s distance from the substrate.

[0037] FIG. 20 shows data plots showing an example PL response to an applied emulated neuronal extracellular action potential.

[0038] FIG. 21A shows a false color micrograph of a MoS2multi-modal detection configuration.

[0039] FIG. 21B shows data plots showing the mean displacement of the cardiomyocytes extracted from microscope video recordings using optical flow analysis over 5 days.

[0040] FIG. 21C shows a schematic illustration of an example equivalent circuit corresponding to a multi-modal recording configuration.

[0041] FIG. 21D shows data plots showing simultaneously measured electrical and fractional PL signals.PCT Application Attorney Docket No.: 009062.8585.WO00

[0042] FIGS. 21 E and 2 IF show schematics of fitting the optically measured extracellular field potential using equivalent circuit models.

[0043] FIG. 22A shows a schematic illustration of MoS2trions interacting with the ionic activity from an optically porated cell.

[0044] FIG. 22B shows a micrograph of cardiomyocytes cultured on MoS2monolayers used for optical poration experiments.

[0045] FIG. 22C shows data plots showing all-optical voltage measurement from a MoS2monolayer beneath a cardiomyocyte before and after optical poration.

[0046] FIGS. 22D shows data plots showing continuous all-optical voltage recording for optically porated cells.

[0047] FIGS. 22E-22I show images and data plots showing intracellular recordings after optical poration of the cell.

[0048] FIG. 23A shows microscope images corresponding to different stages of an all-optically recorded intracellular action potential.

[0049] FIG. 23B shows data plots showing traces recorded from MoS2monolayers beneath porated and intact cardiomyocytes.

[0050] FIG. 23C shows a data plot showing a close-up view of an all-optically recorded single action potential from the monolayer beneath a porated cell.

[0051] FIG. 24 shows images and data plots showing simultaneous recording of intracellular action potentials from electrically floating MoS2using wide-field fluorescent microscopy.

[0052] FIG. 25 shows fluorescent microscope images of MoS2PL.DETAILED DESCRIPTION

[0053] Quantum confinement in nanoengineered semiconductor materials results in the material’s optical properties being intricately linked to electrons, which can be manipulated by external electric fields. These optoelectronic features make such engineered semiconductors suitable as probe materials for studying biological electrical activity. In addition to their relatively high quantum yields, picosecond level emission lifetimes make these materials particularly promising for monitoring biological voltages with high spatiotemporal resolution.

[0054] Disclosed are materials, systems, and methods created harness exciton to trion conversion in angstrom thick semiconductors to provide label-free, dual-polarity, all-opticalPCT Application Attorney Docket No.: 009062.8585.WO00detection of electrical activity, via changes in photoluminescence, in biological cells with ultrahigh temporal resolution. Described herein is a physical model in accordance with the disclosed technology devised to elucidate the conversion process utilized by the disclosed technology, which is inherently governed by the quantum statistics of the background electrons induced by biological activity.

[0055] In some embodiments in accordance with the present technology, a system for detecting biological electrical activity includes a probe material to be positioned proximate one or more electroactive cells that is configured to emit an optical signal when the probe material illuminated and be altered by surrounding electric field(s) caused by the one or more electroactive cells. The system includes a data processing device, including one or more processors and memory, in communication with the optical detector, configured to decompose (e.g., process) the optical signal into an excitonic component and a trionic component, and to detect an electrical signal from the one or more electroactive cells based on a magnitude of the trionic component relative to a magnitude of the excitonic component of the optical signal. Some example embodiments of the probe material include a transition metal dichalcogenides (TMDC) monolayer. In some embodiments, the TMDC monolayer includes a molybdenum disulfide (MoS2) monolayer.

[0056] In some implementations of the example embodiments of the present technology, we show that the exemplary monolayer MoS2enables completely bias-free tetherless operation due to its substantial trion density originating from intrinsic sulfur vacancies introduced during chemical vapor deposition. In other embodiments, other TMDCs, such as WS2, WSe2, or MoSe2may be modified (e.g., through doping) in order to obtain a suitable trion density that similarly enables bias-free, tetherless operation. In additional embodiments, TMDCs in the form of nanoparticles or nano-sized sheets may be employed to achieve similar optical sensing capabilities in a non-planar geometry. The disclosed example embodiments illustrate the use of label-free all-optical voltage sensing in angstrom thick semiconductor probe materials.

[0057] Monolayer semiconductors of angstrom thickness have emerged in recent years with their material properties arising from quantum confinement physics. These materials, belonging to the class of materials known as transition metal dichalcogenides (TMDC), have been explored for their potential in optical and electrical applications. Arguably, optoelectronic device applications have attracted the most attention, owing to room temperature stable excitons generating bright photoluminescence (PL). Excitons, quasiparticles formed due to the Coulomb interaction betweenPCT Application Attorney Docket No.: 009062.8585.WO00an electron and a hole, also intrigued scientists for their rich physics at this ultimate monolayer limit. The charged nature of excitons sets the stage for controlling optical properties of these materials, such as absorption and photoluminescence, using external electric fields, which have been investigated for optoelectronic applications. The voltage responsivity of optical properties can be of particular interest in detecting biological electrical activity. However, the use of these biocompatible monolayer semiconductors has been elusive for applications in biosensing such as dynamic imaging of biological potentials with high spatiotemporal resolution.

[0058] To this end, in some embodiments of the present technology, we harness the voltage sensitive photoluminescence in angstrom thick semiconductor MoS2arising from the exciton to trion conversion physics to all-optically detect electrical activity from human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs). Trions, i.e. charged excitons, are intricately connected to the quantum statistics of background electrons, which can be controlled by biological electrical activity in effect modulating the corresponding Fermi level through the electric field effect. The semiconductor nature of MoS2also allows it to be used simultaneously as an electrode for validating trionic all-optical voltage recording.

[0059] FIG. 1A shows a schematic of an example embodiment of a system 100 for detecting biological electrical activity, in accordance with the present technology. The system 100 includes a probe material 110, which is to be positioned proximate one or more electroactive cells 112. In some embodiments, the probe material 110 is configured on a substrate. The system 100 includes a light source 114 configured to emit a probe light 116 to illuminate the probe material 110. The system 100 includes an optical detector 120 configured to measure an optical signal 118 emitted by the probe material 110, which the optical signal 118 is sensitive (e.g., responds to or may be altered by) electrical field(s) associated with the electroactive cell(s) 112. The system includes a data processing device (not shown) in communication with the optical detector 120. In some example embodiments, detection of biological electrical activity is performed by placing the probe material 110 whose photoluminescence is sensitive to a surrounding electric field, such as a TMDC, in close proximity with (e.g., within 20 pm or less) the one or more electrically active cells 112, such as a cardiomyocyte. The probe material 110 may be optically excited by the probe light 116 emitted by the light source 114 (e.g., a laser or a light emitting diode (LED)), and the photoluminescence 118 of the probe material 110 may be detected by an optical detector 120 (e.g., a camera, a photodetector, a photomultiplier tube, etc.). The photoluminescence signal 118PCT Application Attorney Docket No.: 009062.8585.WO00measured from the probe material 110 may be analyzed to determine the relative contributions of excitons and / or trions to the signal. The photoluminescence data may be spatially and / or time resolved. Subsequently, data describing the contributions of the excitonic and trionic components may be used to detect electrical activity (e.g., extracellular potentials and / or intracellular potentials) of the electrically active cell 112. Based on the spatial and / or temporal resolution of the photoluminescence data, electrical activity at the individual cell level can be detected.

[0060] In some embodiments, the probe material 110 may be in contact with an electrode. This electrode may be used to supply an electrical bias to the probe material and / or detect an electrical signal from the probe material during sensing.

[0061] In some embodiments, the probe material 110 may be in physical contact with one or more electrically active cells 112. These cells may be selectively porated while in contact with the probe material 110, enabling the probe material 110 to detect intracellular potentials. For example, the light source 114 may be used to selectively porate the cell 112.

[0062] In some example embodiments, the cells 112 which are the target of the detection are cardiomyocytes, while in other embodiments, the target cells 112 may be other electroactive cells.

[0063] In some example embodiments, the probe material 110 for detection is configured as a monolayer TMDC, such as MoS2, WS2, MoSe2, or WSe2, wherein the monolayer TMDC has a planar configuration. In some embodiments, TMDC materials may be doped (e.g., during synthesis or through post-synthesis modification) in order to achieve desired photoluminescence characteristics. Aspects of an example embodiment of the system 100 where the probe material 110 is configured as a monolayer TMDC is illustrated later in FIG. 1C, as an example.

[0064] In some example embodiments, the probe material 110 for detection may be a TMDC configured as a nanoparticle. The shape and size of these nanoparticles may be selected such that they are suitable for injection into biological tissue, which may be advantageous in neuroscience applications or other detection applications where three-dimensional spatial resolution is desirable. For example, the nanoparticles may be configured as nanosheets or monolayer disks with a diameter in a range of 20 nm to 50 nm. The exemplary nanoparticles probe material could be selectively injected at different depths within a section of tissue to enable label-free and functionalized cell-type specific deep tissue voltage imaging. For example, functionalized MoS2quantum dots can be used as fluorescent labels for 3D optical spatial localization in targeted cancer imaging. Aspects of an example embodiment of the system 100 where the probe material 110 isPCT Application Attorney Docket No.: 009062.8585.WO00configured as a TMDC nanoparticle is illustrated in FIG. IB, as an example.

[0065] FIG. IB shows a schematic of an example embodiment where exemplary MoS2nanoparticles are used as the probe material 110, e.g., labeled 110B in FIG. IB, to detect electrical activity in a cell culture. As illustrated in this figure, the small size of these example probe material nanoparticles HOB enables them to be distributed at various vertical positions within the cell culture. This feature may be advantageous in applications where there are multiple layers of cells within a tissue or culture and where it is desirable to resolve electrical activity at specific layers or locations.

[0066] In some embodiments, injectable nanoparticles may be prepared using the liquid phase exfoliation method. Using this method, the size of nanoparticles can be engineered by controlling the centrifugation speed and time. The injectable nanoparticles may be either chemically functionalized to specifically target the cells or used as pristine. For example, MoS2nanoparticles provide the opportunity for surface functionalization via the thiolation of sulfur vacancies. In neuroscience applications, lysine-phenylalanine based peptide functionalization may be used such that the nanoparticles can target neurons in cultures. Additionally, the size, shape, and injection concentration of TMDC nanoparticles may be modified in order to enhance the biocompatibility of the nanoparticles.

[0067] Examples of quantum theory for trionic voltage responsivity

[0068] We first introduce the role played by trions in light of our experimental results and a theory devised to explain and interpret our experimental observations via quantum statistics.

[0069] FIG. 1C shows a schematic of an example experimental test structure used to investigate the voltage dependent PL intensity from trionic monolayer materials in a PBS buffer with Ag / AgCl electrodes. In the example embodiment shown in FIG. 1C, the test structure includes a glass (SiCL) substrate 102, a gold (Au) working electrode 104, and a silver / silver chloride (Ag / AgCl) reference electrode 106. A monolayer probe material 108, such as molybdenum disulfide (MoS2), molybdenum diselenide (MoSe2), tungsten disulfide, (WS2) or tungsten diselenide (WSe2), is in electrical contact with the working electrode 104.

[0070] FIG. ID shows a schematic showing the dependence of PL intensity on the Fermi level and corresponding external voltage bias in a capacitive configuration. The PL intensity, F, follows the prediction given by the Fermi-Dirac statistics of background electrons (solid red line). The arrow indicates the dynamic conversion between excitons and trions as the Fermi level changes.PCT Application Attorney Docket No.: 009062.8585.WO00Voltage sensitive PL is only achievable when both excitons and trions are present, where the slope of the PL intensity (red line) is non-zero. Here, F, mr, and EF are PL intensity, trionic contribution, and Fermi energy, respectively.

[0071] FIG. 2A shows a schematic of an example process for fabrication of SU-8 encapsulated gold (Au) electrodes on a glass coverslip substrate. In the example shown, we first patterned Au microelectrode arrays on a 22x22 mm glass coverslip substrate. The coverslip was 170 pm thick, compatible with the oil immersion objective used for the optical experiments. At step 202, we sputtered an adhesion layer of 5 nm chromium (Cr) followed by 100 nm of Au films directly onto a clean coverslip. At step 204, the sample was then patterned via photolithography using AZ1505 photoresist to define the Au leads. After patterning, at step 206, the unwanted Au / Cr areas were exposed and at step 208 were etched away using Au and Cr etchants. At step 210, the photoresist was removed, producing a clean array of Au leads on the substrate. Next, we encapsulated most of the Au leads with a 4 pm thick layer of SU-8 using another photolithography step. At step 212, SU-8 was spin-coated onto the array of Au leads and the substrate. At step 214, the array was exposed and developed. At step 216, the array was baked. This results in encapsulated Au leads with only 10x10 pm2areas exposed at the tip for contacting MoS2, and a large pad for electrical connection on the other.

[0072] FIG. 2B shows a schematic for an example transfer process for chemical vapor deposition (CVD)-grown MoS2. As illustrated in FIG. 2B, CVD-grown MoS2flakes were transferred onto the prefabricated Au leads using a polymer- assisted process. We started with commercially available CVD-grown MoS2on SiO2 / Si (2D Semiconductors). At step 218, we spun coat a thin layer of polystyrene (PS) film on top of the MoS2sample and left the film to completely dry for 5 minutes. The PS at the sample edges was gently scratched to remove PS edge bead and expose some underlying SiCL. At step 220, we then placed the sample to float in a hot 90°C IM NaOH bath for 5-10 minutes. NaOH etched away the SiO2 at the edges of the sample, opening small gaps between PS and the substrate. At step 222, once the edges of the PS started to detach, we transferred the sample to a DI water bath. The hydrophobic PS film should immediately delaminate, releasing MoS2flakes from the growth substrate. The floating MoS?-containing PS film was transferred to float in two additional DI water baths to dissolve NaOH before finally being picked up by the target substrate in step 224. The sample was then heated at 80 °C for 5 minutes to evaporate residual water, and then at 120 °C for 1 hour to promote adhesion betweenPCT Application Attorney Docket No.: 009062.8585.WO00the PS / MoS2 film and substrate. Finally, at step 226, we dissolved the PS film in a toluene bath for 40 minutes. MoS2flakes were released from the PS film and remained on the substrate.

[0073] FIGS. 3 A and 3B show schematic illustrations of photoluminescence modulation in MoS2via electrochemical gating. We employed an electrochemical gating structure (i.e. a capacitive interface) to apply an external voltage to MoS2. We used phosphate buffered saline (PBS) as an ionic liquid including Na+. K+, CF. HPO42and H2PO4 ions. Voltage was applied between the Ag / AgCl reference electrode 106 inserted into the solution, and a gold working electrode 104 in contact with an example monolayer probe material 108 (in this example, MoS2). When a positive bias is applied between the reference electrode 106 and MoS2, the positive ions in PBS are repelled away from the reference electrode and accumulated on top of MoS2surface. These charges attract more electrons into the monolayer and increase electron density. This then promotes trion formation and subsequently reduces total PL emission. And the opposite happens when a negative bias is applied. Negative ions are pushed towards MoS2surface, resulting in a lower background electron density interacting with excitons. As a result, trion formation is suppressed and the total PL intensity increases. In summary, an electric double layer (EDL) between the PBS buffer and the exemplary monolayer probe material 108 forms a capacitive interface. A voltage applied across this interface induces changes in the background electron density in the monolayer via the electric field effect (FIGS. 3 A and 3B).

[0074] FIG. 4 shows a data plot showing example photoluminescence spectra of monolayer MoS2in air (402) and PBS (404). MoS2PL was measured before and after it was submerged in phosphate-buffered saline (PBS). The monolayer was in contact with a prefabricated Au electrode. In PBS, the PL intensity significantly improved and the peak blue-shifted. This observation is consistent with electrostatic surface passivation of intrinsic MoS2defects, such as sulfur vacancies. These defects are known to have n-doping effects on MoS2. As an ionic liquid, PBS contains free ions that form an electric double layer on MoS2surface. Adsorption of ions results in reduced n-doping in MoS2, which subsequently suppresses formation of trions upon photoexcitation. The higher exciton-to-trion ratio, therefore, increases the total PL emission intensity. When there is a substantial background of electrons present due to defects (e.g. sulfur vacancies) introduced during material growth / processing, or electrostatic doping induced by the interactions with the environment, or the electric field effect in a capacitor configuration, trions can be formed via multiparticle Coulomb interactions (FIG. IB). While excitons provide a sizeable radiative quantumPCT Application Attorney Docket No.: 009062.8585.WO00efficiency and consequently bright photoluminescent emission, their charged counterparts, trions, decay primarily non-radiatively resulting in lower photoluminescence.

[0075] FIG. 5A shows a schematic illustration of an electric double layer on the surface of an example monolayer material, such as a TMDC.

[0076] FIG. 5B shows an illustration of an energy band diagram showing the Fermi level Br¬ and trionic energy level ET for an example monolayer material.

[0077] Electrons are induced in the trionic sheet due to the electric field effect as a result of an externally applied bias voltage through the electric double layer capacitor. This in turn changes the Fermi level and the probability of trion formation. An applied positive voltage causes the background electron density-and consequently the number of trions participating in the PL process-to increase.

[0078] To develop a phenomenological model that captures the role played by trions in monolayer semiconductors, we need to consider the fermionic nature of the background electrons engendering trions by introducing the Fermi-Dirac statistics into the total PL intensity. To this end, we first calculate the electronic density of states for a 2D electronic system by considering the area in phase space including the factor 2 for spin.d2p 4npdpg2DWdE = 2 -f = (Eq. 1)

[0079] Here, the density of states g2DE) has the units of J_1m‘2and h is the Planck’s constant. Since the electron energy E = p2 / 2m depends quadratically on momentum p. where m is the effective electron mass, by differentiating this expression we get pdp = mdE. By plugging this into Equation 1 and using the reduced Planck’s constant h, we see that the density of states in 2D is constant as shown in Equation 2 below.m^ ^

[0080] We can then calculate the dependence of the Fermi level EFon n the 2D electron density per area by integrating over all possible energy states with the Fermi-Dirac occupation probability f(E) as shown in Equation 3.n = g2D(E)f(E dE (Eq. 3)Jo

[0081] This integral can be calculated analytically by integration by parts resulting in the expression in Equation 4.PCT Application Attorney Docket No.: 009062.8585.WO00trk T / F \n = — 7^- log ( 1 + ekBT) (Eq.4)7THZ\ /

[0082] In Equation 4, kBis the Boltzman constant and T is the absolute temperature. Since our experiments are carried out at room temperature and near physiological temperatures, the thermal energy kBT is much smaller than EF. In this limit, the dependence of EFon the electron density simplifies to the linear relationship below in Equation 5.nh2EF= - n (Eq.5)m

[0083] We also need to find the dependence of the MoS2Fermi level on any applied external bias voltage V in our experimental configuration. To this end, we can consider the fact that electrical activity induced by the cells are coupled with the MoS2monolayer through the electric double layer capacitance CBDL= s-, where s, A, d are the effective dielectric constant, area, and effective EDL thickness, respectively (as depicted in FIG. 5A). Since the total induced charge in the MoS2layer Q = neA = CEDLV, we can solve for the voltage dependence of n, which yields An = V, where e is the single electron charge. Plugging this into Equation 5 results in the expression in Equation 6.nh2sAEF= - -V (Eq.6)med

[0084] Due to this linear relationship, we can use the Fermi level and the external bias voltage interchangeably in our analysis of the crucial role played by the trions in the desired voltage sensitive PL emission from these monolayer semiconductors.

[0085] Having worked out the voltage dependence of the background electron density induced via the electric field effect, we can develop our phenomenological model that connects the biologically induced voltages to the trion formation probability and subsequently the PL responsivity. To understand the role played by trions, we consider the fermionic nature of the background electrons engendering trions by introducing the Fermi-Dirac statistics into the total PL intensity. Instead of working with the trion density directly, in our simple model we incorporate the occupation probability for electrons being involved in the formation of trions since without the background electron density the PL will be purely due to excitons and will not be sensitive to the external biological voltages to be detected.

[0086] Here, we assume that the trion contribution nTEF) to the PL intensity is proportionalPCT Application Attorney Docket No.: 009062.8585.WO00to the electronic occupation probability given by the Fermi-Dirac statistics as shown below in Equation 7.1!+ eWCr-Eri

[0087] In Equation 7. / ? = l / kBT. The Fermi-Dirac distribution gives a zero contribution until the Fermi level approaches the trion energy when it gradually reaches unity. The PL intensity F can then be assumed to follow the equation below in Equation 8.F = l - nT(FF) (Eq.8)

[0088] Based on experimental observations, this model assumes that the PL intensity, F, is unity when only excitons are involved without the background electron density, i.e. nT= 0 due to lack of sulfur vacancies during the growth process or external voltage bias. Then F gradually decreases as the trions become more probable as the Fermi level EFapproaches the trion energy ET, i.e. nTapproaches 1, as shown in the simplified energy level diagram in FIG. 5B. This simple model captures the decreasing quantum efficiency and brightness of trionic sheets as trions decay primarily non-radiatively. We plotted the calculated PL intensity as a function of the Fermi level for a fixed trion energy ETin FIG. 6A to show the decreasing PL as the trion contribution increases.

[0089] FIG. 6A shows the calculated PL intensity (left vertical axis, 602) and calculated voltage responsivity (right vertical axis, 604) vs Fermi level normalized to the trion energy (vertical line). Since the applied external voltage changes the Fermi level linearly (Eq. 6), this plot also shows the voltage dependence of the PL intensity. The largest PL intensity change occurs when the Fermi level is near the trion energy at which point there is near equal trion contribution compared to the excitonic contribution. Using our model, next we can calculate the voltage responsivity AF / AF by simply differentiating F with respect to the Fermi level. The normalized responsivity AF / EEFis plotted in FIG. 6A showing that the maximum responsivity is achieved when the Fermi level is equal to the trion energy, where the slope of the PL intensity is maximum. Our theory clearly shows that the trionic sheets are most sensitive to external voltages when the trions are equally probable as excitons in the presence of a large background electron density. Therefore, trions are an important ingredient for the proposed all-optical voltage sensing mechanism.

[0090] FIG. 6B shows the measured external bias voltage dependence of the PL intensity of MoS2(left panel) and WS2(right panel). The circular markers correspond to the normalizedPCT Application Attorney Docket No.: 009062.8585.WO00measured PL. The error bars are the standard deviations calculated from measurements of three different samples. Both panels show calculated PL using the theory based on quantum statistics (left vertical axis, 608) and its derivative representing trionic PL voltage responsivity (right vertical axis, 610) peaking where the slope is largest. The shaded area in the MoS2panel indicates that, for the example experiments shown, the MoS2exhibits voltage responsivity near the zero-bias region for bias free operation, unlike WS2that needs to be first biased at -600 mV for voltage sensitive PL.

[0091] FIGS. 6C-6F show data plots showing the voltage bias dependence of the PL intensity (scatter plot, error bar indicates 3 different samples) for 4 different monolayer semiconductors, MoS2, WS2, MoSe2and WSe2. The data traces 608 are the fits using our theory based on Fermi-Dirac statistics showing excellent agreement with the experiments. The circular markers are the means calculated from the 3 different samples for each material. The data traces 610 correspond to responsivity calculated by differentiating the expression in Equation 8. The experimental results show excellent agreement with our theory based on Fermi-Dirac statistics. These results clearly show that the PL intensity is most sensitive to external voltages when the Fermi level aligns with the trion energy, as shown in FIG. 6A. Our experiments in agreement with our theory clearly show that without the background of electrons and consequently trions, the PL in these materials will not be sensitive to applied electric fields. Based on the data shown in FIGS. 6C-6F, MoS2shows voltage responsivity near zero bias while the other three materials require a bias voltage and, for these examples, cannot be used standalone in an all-optical voltage imaging microscopy configuration typically used for such experiments. However, if these other three materials (or other TMDCs) were suitably modified (e.g., through doping during the synthesis process), they could also be used in a configuration similar to MoS2. However, for the examples investigated here, we have identified MoS2to be a preferred candidate for this purpose due to the presence of an inherently large n-type doping as a result of the growth process owing to sulfur vacancies resulting in a Fermi level already close to the trion energy.

[0092] FIG. 7A shows data plots illustrating the change in PL spectra for MoS2with applied bias voltages due to different trionic contributions. The spectra were taken at an exposure time of 5 seconds, and the average photon counts rate for the -100 mV (702), 0 mV (704), and 100 mV (706) biases (indicated by the arrows in FIG. 6B) are 47292 counts / s, 15083 counts / s, and 7771 counts / s, respectively.PCT Application Attorney Docket No.: 009062.8585.WO00

[0093] Since voltage dependence of trionic PL intensity is imperative for all-optical biological voltage sensing, we first developed a physics-based model that connects the biologically induced voltages to the trion formation probability and subsequently the PL intensity. To understand the role played by trions, we considered the fermionic nature of the background electrons engendering trions by simply introducing the Fermi-Dirac statistics into the total PL intensity. We chose to use the Fermi level EF interchangeably with the bias voltage in our capacitive configuration, since EF is linearly proportional to induced voltages for a 2D electronic system (as discussed above). Without the background electron density, the PL will be purely due to excitons and will not be sensitive to the external biological voltages to be detected, though it will be at a much brighter intensity. To account for the role played by these electrons in the formation of trions, we assume that the trionic contribution nTEF^) to the PL intensity is proportional to the electronic occupation probability near the trion energy ET, where the distribution nTEF') is given by the Fermi-Dirac statistics as nT(EF) = ^(gr-gF)and ft is the inverse of the thermal energy. Thenormalized PL intensity E can then be assumed to follow the equation F = 1 - nTEF). Based on experimental observations, this model assumes that the PL intensity, F, is unity when only excitons are involved without the background electron density, i.e. nT= 0. Then F gradually decreases as the trion formation becomes more probable as the Fermi level EFapproaches the trion energy ET, i.e. nTapproaches 1. This model captures the voltage dependence of the PL intensity, showing excellent agreement with experimental results in FIG. 6B for MoS2and WS2(see FIGS. 6E and 6F for MoSe2and WSe2).

[0094] We can now directly calculate the voltage responsivity of the PL intensity (curves 610 in FIG. 6B), dF / dV, by differentiating the calculated PL intensity F (curves 608 in FIG. 6B). The voltage responsivity is the most important parameter for biological voltage sensing and maximized when EF is near the trion energy ET with some broadening expected due to the width of the trion peak.

[0095] FIG. 7B shows data plots showing the PL spectra at various voltages of FIG. 7A decomposed (using a Lorentzian decomposition) into exciton (left peaks, 710) and trion (right peaks, 712) components at corresponding bias voltages. Voltage sensitive PL is achieved when trion and exciton concentrations are comparable. When MoS2is illuminated with a laser, the radiative process involves a combination of excitons and trions with different contributions to the PL spectrum. Compared to the excitonic emission spectrum, trionic emission is red shifted due toPCT Application Attorney Docket No.: 009062.8585.WO00the larger trion effective mass. Such a shift in the spectrum allows us to decompose the PL spectrum into the exciton and trion contributions shown as the lower solid lines in FIG. 7B.

[0096] For example, in some embodiments, the acquired data can optionally undergo baseline corrections to remove background signal from the raw PL spectrum data, after which peak positions are identified to determine the approximate energy positions for the exciton and trion peaks. The data is fitted with multiple peak functions (e.g., using a least-squares fitting algorithm), which can leverage the distinct energetic, power-dependent or charge-dependent behaviors of the quantum confined materials. From the multi-peak fitted data set, parameters are extracted, e.g., including peak positions, linewidths, and intensities associated with exciton and trion contributions.

[0097] At zero bias, for example, the MoS2trion component is comparable to the excitonic one coinciding with the maximum voltage responsivity dF / dV. By changing the Fermi level with an applied bias, we can change the excitonic and trionic contributions, leading to different voltage sensitivities (curves 610 in FIG. 6B). Far from the condition EF~ET, (i.e. away from 0 V for MoS2corresponding to near equal trions and excitons), the voltage responsivity gradually approaches zero even when there is stronger overall PL (negative bias) due to dominance of brighter excitons (FIG. 7A).

[0098] FIGS. 8A-11E show additional details of voltage responsivity characterization in MoS2, WS2, MoSe2. and WSe2. We have conducted two sets of experiments to characterize the voltage responsivity of the TMDC materials including MoS2(FIGS. 8A-8E), WS2(FIGS. 9A-9E), MoSe2(FIGS. 10A-10E), and WSe2(FIGS. 11A-11E).

[0099] FIG. 8A shows a microscope PL image of the MoS2under characterization. The small square indicates the location of the laser excitation.

[0100] FIG. 8B shows a data plot showing PL intensity of MoS2at different DC bias. A 594 nm laser with intensity of 4 W / cm2was used in the experiments. The circular markers are the normalized averaged values from three measurements at different flakes, the error bar is the standard deviation of the three measurements. The curve 608 is the fit using the Fermi Dirac distribution shown in Equation 9. The curve 610 is the normalized derivative of the curve 608 which corresponded to the voltage responsivity of the material.

[0101] FIG. 8C shows data plots showing fractional PL traces for MoS2showing the modulation by the square wave of 1000 Hz at different amplitudes of -50 mV, -40 mV, -30 mV, -PCT Application Attorney Docket No.: 009062.8585.WO0020 mV, 20 mV, 30 mV, 40 mV, and 50 mV. All measurements were conducted at zero DC offset. A 532 nm laser with intensity of 360 W / cm2was used.

[0102] FIGS. 8D and 8E show how data plots showing averaged fraction PL changes at different square wave modulating amplitude for MoS2. The responsivity is extracted from linear fitting of the measurement results.

[0103] FIG. 9A shows a microscope PL image of the WS2under characterization. The small square indicates the location of the laser excitation.

[0104] FIG. 9B shows data plots showing PL intensity of WS2at different DC bias. A 594 nm laser with intensity of 6 W / cm2was used in the experiments. The circular markers are the normalized averaged values from three measurements at different flakes, the error bar is the standard deviation of the three measurements. The curve 608 is the fit using the Fermi Dirac distribution shown in Equation 9. The curve 610 is the normalized derivative of the curve 608, which corresponded to the voltage responsivity of the material.

[0105] FIG. 9C shows data plots showing fractional PL traces for WS2showing the modulation by the square wave of 100 Hz at different amplitudes of -50 mV, -40 mV, -30 mV, -20 mV, 20 mV, 30 mV, 40 mV, and 50 mV. All measurements were conducted with a 500 mV DC offset. A 532 nm laser with intensity of 40 W / cm2was used.

[0106] FIGS. 9D and 9E show data plots showing averaged fraction PL changes at different square wave modulating amplitude for WS2. The responsivity is extracted from linear fitting of the measurement results. We noticed that the WS? PL was not stable at a laser intensity larger than 60 W / cm2in PBS, and has the weakest voltage sensitivty among the TMDC materials characterized (i.e. MoS2, MoSe2, and WSe2).

[0107] FIG. 10A shows microscope PL image of the MoSe2under characterization. The small square indicates the location of the laser excitation.

[0108] FIG. 10B shows data plots showing PL intensity of MoSe2at different DC bias. A 594 nm laser with intensity of 6 W / cm2was used in the experiments. The circular markers are the normalized averaged values from three measurements at different flakes, the error bar is the standard deviation of the three measurements. The curve 608 is the fit using the Fermi Dirac distribution shown in Equation 9. The curve 610 is the normalized derivative of the curve 608, which corresponded to the voltage responsivity of the material.

[0109] FIG. 10C shows data plots showing fractional PL traces for MoSe2showing thePCT Application Attorney Docket No.: 009062.8585.WO00modulation by the square wave of 1000 Hz at different amplitudes of -50 mV, -40 mV, -30 mV, -20 mV, 20 mV, 30 mV, 40 mV, and 50 mV. All measurements were conducted with a 500 mV DC offset. A 532 nm laser with intensity of 40 W / cm2was used.

[0110] FIGS. 10D and 10E show data plots showing averaged fractional PL changes for MoSe2at different square wave modulating amplitude. The responsivity is extracted from linear fitting of the measurement results.

[0111] FIG. 11 A shows a microscope PL image of the WSe2under characterization. The small square indicates the location of the laser excitation.

[0112] FIG. 11B shows data plots showing PL intensity of WSe2at different DC bias. A 594 nm laser with intensity of 6 W / cm2was used in the experiments. The circular markers are the normalized averaged values from three measurements at different flakes, the error bar is the standard deviation of the three measurements. The curve 608 is the fit using the Fermi Dirac distribution shown in Equation 9. The curve 610 is the normalized derivative of the curve 608, which corresponded to the voltage responsivity of the material.

[0113] FIG. 11C shows data plots showing fractional PL traces for WSe2showing the modulation by the square wave of 1000 Hz at different amplitudes of -50 mV, -40 mV, -30 mV, -20 mV, 20 mV, 30 mV, 40 mV, and 50 mV. All measurements were conducted with a 500 mV DC offset. A 532 nm laser with intensity of 400 W / cm2was used.

[0114] FIGS. 1 ID and 1 IE show averaged fractional PL changes for WSe2at different square wave modulating amplitude. The responsivity is extracted from linear fitting of the measurement results.

[0115] In the first set of experiments, we measured PL intensity at different DC biases, as shown in FIGS. 8B, 9B, 10B, and 11B. The measurement results after normalization were then fitted using Fermi Dirac distribution of Equation 9.1= 1 -t + elWT-n(Eq.9)

[0116] In Equation 9, F is the total PL intensity, is the reciprocal of the thermal energy also including the broadening of the trion peak, VTand V are voltages that corresponded to the trion energy and Fermi-level (i.e. linearly proportional to the bias voltage in 2D), respectively. In the numerical fitting, / / and VTare used as free fitting parameters. After fitting F(V), the derivative dF(V) / dV, i.e. the voltage responsivity of PL, is calculated and plotted after normalization. BasedPCT Application Attorney Docket No.: 009062.8585.WO00on the theory discussed above, at the bias voltage corresponded to the peak in the dF(V) / dV plot, the Fermi level is close to the trion energy, and the material has the highest voltage responsivity.

[0117] In the second set of experiments, we characterized the voltage dependence of PL for different materials by adding a DC offset corresponding to the different Fermi levels in addition to a small signal square wave voltage modulation. For each example material, the DC offset value is chosen to bring the material’s Fermi level close to the trion energy, and the peak-to-peak square wave amplitude varies from -50 mV to 50 mV. It is noted that the two set of experiments were conducted at different times using different batch of TMDC materials. These materials have different intrinsic initial background electron densities and hence different Fermi levels. As a result, the DC bias voltage required to achieve the highest responsivity are different between the experiments. All the example materials used in the experiments are CVD-grown and wet-transferred onto the microelectrode arrays for characterizations.

[0118] While we tested three other trionic materials including MoSe2, WS2, and WSe2, for the samples tested, we identified MoS2as the only material among the example materials measured to possess voltage sensitive PL at zero bias (FIG. 6B and FIGS. 8-11). This is due to the large background electron density (i.e. EF~ET) in the as-grown MoS2monolayers due to sulfur vacancies and consequently large trion contribution enabling all-optical, bias-free biological voltage sensing without any electrical contacts. Therefore, without requiring any biasing voltage, MoS2can detect electrophysiology signals in the range from 10 mV to sub-mV. For comparison example WS2monolayers typically used a constant 600 mV DC bias to achieve voltage sensitive PL (FIG. 6B, left panel). This is an important feature of MoS2enabling completely tetherless, all-optical sensing and compatibility with existing optical microscopy modalities.

[0119] Examples of voltage sensing characterization

[0120] FIG. 12 shows a schematic illustration of an example one photon microscopy apparatus with two detection modes: (1) Imaging mode using the sCMOS camera 1216 with fast frame rate (bottom left shows the PL image acquired at the image mode); and (2) Single pixel mode using a photodetector 1214 (bottom right image shows the PL excited by the focused laser 1208 which will be detected by the single pixel photodetector 1214). The one-photon microscope setup used in the experiments can readout the PL signal either with a fast frame rate sCMOS camera 1216 or with a single pixel photodetector 1214 for faster response (OBJ - Objective 1204, BS - Beam splitter cube 1206, fl - Excitation filter 1212, f2 - Emission filter 1210). In the imaging mode, aPCT Application Attorney Docket No.: 009062.8585.WO00broadband LED 1208 is used as the light source sent through a 570-616 nm bandpass excitation filter 1212, and the resulting PL signal is collected through a 634 nm longpass emission filter 1210. In the single-pixel detection mode, a 594 nm laser 1208 is used as the light source. The input laser is filtered by a narrow 591 / 6 nm bandpass filter 1212, and the output PL signal is collected through a 594 nm longpass filter 1210. The microscope PL images at the bottom left and bottom right of FIG. 12 are the optical signal focused onto the sCMOS camera imaging plane and the single pixel of the photodetector, respectively. The PL spot in the bottom right image has an area of around 5 μm2responding to the laser spot- partially filling the back focal plane of the oil immersion objective. All the laser intensities listed in this paper are calculated using a laser spot size of 52lim

[0121] Some advantageous features of the present technology include the following. First, it is compatible with conventional optical imaging methods with high spatial resolution. Second, a sub-ms temporal resolution is achieved for capturing all the fine details in biological electrical activity such as fast action potentials of excitable cells. Third, a bias-free dual-polarity high voltage responsivity is attained to convert biological electrical activity into detectable changes in fractional photoluminescence ΔF / F0, where Fois the baseline photoluminescence. To demonstrate these features, we characterize the voltage responsivity and temporal response of PL changes in MoS2monolayers in the test setup shown in FIG. 12 by applying electrical pulses of varying amplitudes and frequencies (50% duty cycle) emulating biological electrical activity. This setup is based on a one-photon fluorescence microscope (60x oil immersion, NA 1.4, Nikon) that can either use a single pixel photodetector 1214 (APD440A, Thorlabs) for measuring the frequency response or a fast frame rate sCMOS camera 1216 (Zyla-4.2, Andor) for imaging. The bottom left image in FIG.12 shows the one-photon PL image of a MoS2monolayer in contact with a gold electrode. The MoS2PL was excited by a broadband LED source and imaged with the sCMOS camera.

[0122] FIG. 13 shows a schematic illustration of an example two-photon microscopy apparatus. For two-photon fluorescence microscopy, we used an ultrafast femtosecond pulsed laser 1306 (Insight X3) with a broad tunable range of 680-1300 nm. The pulse width is <120 fs with a repetition rate of 80 MHz. The input laser beam was modulated across the sample via Galvo-Resonant scanners 1308 at a maximum frame rate of 400 frame per second (Thorlabs). The sample was excited through a water-immersion objective 1302 (NA = 0.8) and emission was detected by photomultiplier tubes 1310 and 1312 (PMT) through 525 / 50 nm (channel A) and 660 / 52 nmPCT Application Attorney Docket No.: 009062.8585.WO00(channel B) bandpass filters 1314 and 1316 (Semrock).

[0123] FIGS. 14A-14C show images of MoS2photoluminescence imaging and modulation with two-photon excitation.

[0124] FIG. 14A shows a two-photon PL image excited by 1064 nm pulsed laser measured between 634-686 nm. Scale bar, 10 μm.

[0125] FIG. 14B shows an image of PL change as upon application of +20 mV (left) and -20 mV (right) bias between submerged Ag / AgCl electrode and MoS2.

[0126] For PL imaging, we chose an excitation wavelength of 1064 nm and incident power of about 10 mW. PL emission was collected in channel B filtered between 634-686 nm. FIG. 14A shows a two-photon PL image of a monolayer MoS2on an encapsulated Au electrode. Applying positive and negative biases results in a decrease and increase in PL emission intensity, respectively, as shown in FIG. 14B. Because of the lack of inversion symmetry in monolayer MoS2, the material has been shown to exhibit strong second-harmonic generation (SHG). With an excitation wavelength of λ. SHG at λ / 2 can be observed in the image channel of 500-550 nm for 1000 nm< X<1100 nm excitation.

[0127] FIG. 14C shows 2-photon microscope images of the MoS2PL under positive and negative voltage bias.

[0128] To further emphasize the compatibility of label-free, all-optical voltage sensing based on trionic PL with other widely used optical microscopies, we also characterized ΔF / F0, using a laser scanning two-photon microscope for both positive and negative voltages (FIGS. 13-14). These results demonstrate the intrinsic compatibility of trionic PL based voltage sensors with conventional microscope setups in a plug and play fashion.

[0129] FIG. 15A shows a data plot showing measured fractional PL intensity in time for different square wave amplitudes (Va) of -10 mV (1502), -4 mV (1504), 4 mV (1506), and 10 mV (1508) at 1 kHz. The inset shows the applied square waveform.

[0130] FIG. 15B shows a data plot showing measured fractional PL intensity vs the applied square wave amplitude at 1 kHz. The slope corresponds to the voltage responsivity of the PL, which is an important parameter for all-optical biological voltage sensing. Trionic PL is sensitive to both polarities, which is important for capturing dual-polarity biological information. Error bars represent fitting errors.

[0131] By switching to a photodetector and 532 nm laser excitation, we first characterized thePCT Application Attorney Docket No.: 009062.8585.WO00voltage responsivity of the MoS2trionic PL using square wave modulation at 1 kHz at different amplitudes (Va) for both polarities (FIGS. 15A and 15B). Fractional PL traces in response to square wave modulation for amplitudes of ±10 mV (1502, 1508) and ±4 mV (1504, 1506) are shown in FIG. 15 A. These fractional PL traces closely follow the applied square wave bias for both positive and negative polarities with an undistorted waveform showing sharp rising and falling edges.

[0132] By additionally varying the applied voltage amplitudes from -20 mV to 20 mV at 1 kHz, we can determine the voltage responsivity of the fractional PL changes as shown in FIG.15B. We fit the experimental results assuming a linear dependence, which allows us to calculate from the slopes the voltage responsivity of the trionic fractional PL. We estimate the voltage responsivity as 2.46 % / mV (negative bias) and 1.73 % / mV (positive bias). For these values of the responsivity and an excitation laser power of 12 μW (240 W / cm2intensity) in our current example experimental configuration, we determined the noise limited minimum detectable value of the biological potential (i.e. voltage sensitivity) to be 39 μV / √Hz (negative bias) and 59 μV / √Hz (positive bias) (See FIGS. 16A-16D). For the used laser spot size of 5 μm2, the areal sensitivity becomes 88 μVμm / √Hz (negative bias) and 132 μVμm / √Hz (positive bias). Such a sensitivity suggests that the example MoS2sensor is able to detect both intracellular and extracellular action potential signals. While conventional micro-electrode arrays (MEAs) and the more recently reported CMOS MEAs have a higher sensitivity than the MoS2sensor (< 0.1 fiV / HZ but the physical size (> 15 pm) and the areal scaling of these sensors make voltage detection with cellular and sub-cellular spatial resolution challenging. The example MoS2sensor embodiment, on the other hand, thanks to the spatially localized exciton-trion conversion, can potentially enable sub-cellular resolution electrophysiology.

[0133] FIG. 16A shows a data plot comparing temporal signal traces of the photodetector output voltage with no input (i.e. dark noise, lower data trace 1602) and that generated by the PL from a floating MoS2crystal (upper data trace 1604). Both signals were recorded at a sampling rate of 250 kHz. The focused laser spot used to excite the MoS2PL is around 5 μm2and the power is 12 μW, which corresponds to an optical intensity of 240 W / cm2. No voltage modulation was applied to the MoS2during the recording.

[0134] FIG. 16B shows a data plot showing Allan deviation calculated from the 30 minute long photodetector recordings shown in FIG. 16A. At the shorter averaging time bin range, the Allan deviation of both the dark noise (1606) and the MoS2(1608) showed a decreasing trend that has aPCT Application Attorney Docket No.: 009062.8585.WO00-1 / 2 slope on the log-log plot.

[0135] FIG. 16C shows a data plot showing power spectral density based on the photodetector output shown in FIG. 16A estimated using the Welch’s method with a window size of 250000 for the dark noise (1610) and the MoS2PL (1612).

[0136] FIG. 16D shows a data plot showing responsivity derived from FIG. 15B, which is plotted as absolute optical power vs applied bias voltage for noise calculations.

[0137] In order to estimate the minimum detectable biological electrical voltage, we first need to work out some of the conversion factors. The first step is to estimate the power detected by the photodetector. For an example laser illumination power of 12 μW focused over 5 μm2, the measured MoS2PL-generated mean photodetector voltage output is -6 mV as shown in FIG. 16A (1604). Dividing this voltage by the photodetector responsivity of 1.25 × 109V / W, we approximate the detected PL power to be ~5 pW or 1.68×107photons / s through the microscope. We can now calculate the responsivity of the MoS2PL changes to the to-be-detected biological electric potentials. The data shown in FIG. 12 gives us this responsivity factor in unit of fractional PL change per mV. Using the measured PL power of 5 pW, we replotted the responsivity in unit of absolute optical power per mV as shown in FIG. 16D. From the slope, we determined the sensor responsivity to be= 0.12 pW / mV (negative bias) and >p= 0.08 pW / mV (positive bias). In terms of photon numbers, these responsivities correspond to= 4×105photons / s / mV (negative bias) and = 2.7×105photons / s / mV (positive bias).

[0138] We also calculated the power spectral density (PSD) (FIG. 16C) corresponding to the photodetector voltage output of MoS2PL (1612) with no applied bias. We found the PSD to be constant at 5.9 μV / √Hz. Dividing this value by the photodector responsivity gives us the overall experimental optical power noise (i.e. noise equivalent power (NEP)) level of 4.72 fW / √Hz corresponding to the MoS2PL. By dividing the NEP by the MoS2sensor responsivity β, we find the minimum detectable biological voltage changes over one Hertz bandwidth, or the sensitivity S to be consistent with some of the voltage sensing literature. The experimental sensitivities for the negative and positive bias are Sn= 39 μV / √Hz and Sp= 59 μV / √Hz, respectively. Using these sensitivity values and the optical spot size of 5 μm2, we can also calculate the areal sensitivities to be = 88 μVμm / √Hz and Sp= 132 μVμm / √Hz. We also comment on the potential shot noise limited sensitivity for our current configuration. The optical noise level of 4.72 fW / √HzPCT Application Attorney Docket No.: 009062.8585.WO00used in the calculation is primarily limited by the photodetector dark noise of 3.44 fW / √Hz. For the 5 pW PL detected, the corresponding photon shot noise will be only 1.72 fW / √Hz. This will result in a shot noise limited sensitivity of Sn= 14 μV / √Hz and = 31 μVμm / √Hz if a much more sensitive photodetector such as a single photon counting device (APD, or PMT) were used.

[0139] FIG. 17A shows data plots of the measured (circular markers) frequency response of the voltage sensitive trionic PL for sinusoidal modulation and the equivalent circuit model simulation result (solid line). The inset shows representative traces at 10 Hz, 100 Hz, and 1 kHz. The fractional PL response remained constant well beyond 1 kHz and exhibited a 3 dB cutoff frequency of over 50 kHz, which is already substantially faster than needed for temporally resolving the fastest sub-ms biological electrical events. We note that while the internal exciton-trion conversion happens on the picosecond time scales, the observed response time is primarily limited by photodetector bandwidth and electrochemical interface impedance as verified by the good agreement between the PL measurement and corresponding circuit model (blue solid line in FIG. 17A). The inset of FIG. 17A shows that the fractional trionic PL response follows, with high-fidelity, the applied square wave modulations between -5 mV and 5 mV at 100 Hz, 1 kHz, and 10 kHz clearly indicating an undistorted response beyond several microseconds.

[0140] FIG. 17B shows data plots showing a simulated electrochemical response of the MoS2 / gold electrode in PBS at the frequency band of 10 to 105Hz using an equivalent circuit model (top panel, solid line), and the experimentally measured fraction PL change of the MoS2 / gold electrode at the frequency band of 10 to 105Hz (bottom panel, circular markers with error bars). The circular markers are the mean of 6 subsequent measurements on the same MoS2flake, where the error bar is the standard deviation of the measurements.

[0141] The frequency dependent electrochemical response of the MoS2interface in PBS (FIG.17B) was calculated using a simplified Randles equivalent circuit as shown in the inset. The electrical parameters in the Randles equivalent circuit were obtained via electrochemical impedance spectroscopy on an SU-8 encapsulated MoS2 / gold structure using Gamry Reference 600+ potentiostat. The obtained values are as follows: solution resistance Rs= 100 kΩ, charge transfer resistance Rct= 90 MΩ, and double layer capacitance Cdl= 50 pF, where the parallel Rctand Cdlare the circuit components representing the electrochemical double layer at the MoS2interface. In the simulation, sinusoidal input (Vin) from 10 Hz to 105Hz was applied and the output voltage Vout was read out by the MoS2 / AU electrode (solid curve in FIG. 17B). The equivalentPCT Application Attorney Docket No.: 009062.8585.WO00circuit’s response to sinusoidal input was simulated using LTSpice. The bottom panel of FIG. 17B shows the experimentally measured fractional PL response of the MoS2over the frequency band from 10 Hz to 105Hz, where the MoS2PL was modulated by the sinusoidally using a function generator (FG).

[0142] FIG. 18 shows a data plot showing the measured (circular markers with error bars) dependence of fractional PL of an electrically floating MoS2monolayer on the distance to a Pt / Ir electrode emulating electrophysiological activity and corresponding FEM simulations (solid line). Error bars calculated from the standard deviation of ten measurements.

[0143] While the MoS2monolayer is in contact with a gold electrode for the examples of voltage sensing characterization above, in some embodiments, an electrical contact is not necessary for all-optical voltage sensing. In fact, it is sometimes preferable that the monolayer be electrically floating for straightforward integration with existing optical microscopies. This ability to operate in a tetherless fashion makes the example MoS2probe material studied here stand out, as it can operate bias-free due to the intrinsically high trionic density in its as-grown state. To demonstrate tetherless bias-free, all-optical voltage sensing, we use a sharp-tipped (<2 zm diameter) platinum / iridium (Pt / Ir) electrode 1904 emulating biological activity placed near an electrically floating (i.e. no electrical contact) MoS2monolayer 1908 immersed in PBS solution and disposed on a substrate 1902. The measured fractional PL changes as a function of the distance from the electrically floating monolayer is shown in FIG. 18. Finite element method (FEM) calculations agree well with the measurements.

[0144] FIG. 19A shows a schematic illustration of photoluminescence modulation via local electric field without electrical contact. The illustration shows an example in which a positive bias is applied between a Pt / Ir microelectrode 1904 and Ag / AgCl reference electrode 1906.

[0145] FIG. 19B a data plot showing an example COMSOL simulation of electric field and potential distributions induced by applying a 40 mV bias through a 2 / zm microelectrode placed 6 μm above MoS2, with respect to a reference potential placed at r = 500 μm.

[0146] FIG. 19C shows a data plot showing an example PL response as a function of the electrode’s distance from the substrate, z, to a 40 mV electrical impulse applied between the 2 / zm Pt / Ir microelectrode 1904 and an Ag / AgCl reference electrode 1906. Error bars calculated from the standard deviation of ten measurements. The circular markers represent the mean of these measurements. The solid line shows the normalized surface charge as a function of electrodePCT Application Attorney Docket No.: 009062.8585.WO00height, which is proportional to the simulated averaged electric field at MoS2.

[0147] When MoS2is placed near an electrically active cell, its surface charge changes as the cell generates a flow of positively charged ions across the cell membrane. We studied the effect of these local electric fields on the MoS2PL by utilizing a 2 μm monopolar platinum / iridium (Pt / Ir) microelectrode 1904 (MicroProbes) to generate an electrical impulse micrometers away from the MoS2monolayer 1908. We first modeled the electric field distribution in this scenario using COMSOL Multiphysics. The simulation was set up in a 2D axisymmetric space with PBS filling the upper half space (z>0) and MoS2 / glass in the lower half (FIG. 19A). An electric potential of 40 mV was applied between the microelectrode tip (r < 2 fim) at a vertical distance z from MoS2and a reference potential placed at r = 500 μm. FIG. 19B shows a simulated 2D potential map (color scale) and electric field distribution (arrows) produced by a microelectrode placed at z = 6 μm.

[0148] On the MoS2surface (z ~ 0 μm), the electric fields are primarily out-of-plane. This suggests that the presence of the electrical impulse causes an accumulation of positive ions on the MoS2surface, which would consequently induce a change in MoS2’s PL intensity. Experimentally, we used a micromanipulator to align the microelectrode tip to the center of a single MoS2monolayer 1908, and a piezo actuator to control the distance (z) of the electrode 1904 tip from the sample surface. We applied a 100 ms long 40 mV square pulse between the microelectrode 1904 and an Ag / AgCl electrode 1906 placed several millimeters away in the solution. As seen in FIG.19C, we measured the PL change as the vertical distance z varied between 1 and 20 μm. The voltage pulse resulted in an overall increase in the PL intensity, but the magnitude of the response decayed as the microelectrode moved away from the MoS2surface. The induced surface charge on the MoS2is proportional to the electric field due to the existence of EDL at the MoS2 / PBS interface. We can then calculate the normalized surface charge by taking the surface-average of electric field values right above the MoS2surface for different values of z. The simulated result is shown as a red solid line in FIG. 19C and in good agreement with the experimental results.

[0149] FIG. 20 shows data plots showing an example PL response (bottom) to an applied emulated neuronal extracellular action potential (top).

[0150] Finally, to characterize the sensor’s capability of capturing the sub-ms temporal details of fast electrophysiological activity, we emulated the temporal dynamics of neuronal extracellular action potentials by applying an electric bias waveform with sub-ms spikes followed by aPCT Application Attorney Docket No.: 009062.8585.WO00refractory period as shown in the upper panel of FIG. 20. In the recorded PL response as shown in the lower panel of FIG. 20, details of the applied electrical sub-ms sharp spike waveform can clearly be observed.

[0151] Examples of label-free all-optical biological voltage sensing

[0152] FIG. 21A shows a false color exemplary micrograph of the MoS2multi-modal detection configuration showing MoS2monolayers and the ethanol fixed cells.

[0153] Having characterized the optical and electrical response of MoS2monolayers for trionic all-optical voltage sensing, we next assessed how well MoS2monolayers can produce in-vitro voltage images of hPSC-CM electrical activity. To demonstrate an example of label-free, all-optical voltage sensing, we plated day 15 hPSC-CMs in monolayer onto devices containing MoS2monolayers integrated with gold leads (FIG. 21A).

[0154] Prior to cell seeding, devices were sterilized via 70% ethanol soaking and UV sterilization for 30 min. Devices were washed with DPBS (IX) (Gibco, 14190-144) to remove debris and coated with Matrigel (1:100 dilution in RPMI 1640. Corning, CLS354234) at 37 °C for at least 1 hr prior to cell seeding. RUES2 human pluripotent stem cells (hPSCs) were differentiated into cardiomyocytes (hPSC-CMs) following the GiWi protocol and maintained in RPMI 1640 + B27 (Gibco, 17504044). After 15 days of cardiac differentiation, hPSC-CMs were dissociated. Briefly, hPSC-CMs were dissociated in collagenase IV (4 mg / mL, Gibco, 17104-019) for 15 min at 37 °C under constant agitation on an orbital shaker. Cells were quenched in cardiomyocyte maintenance media containing 20% fetal bovine serum (v / v) (FBS) (Omega Scientific, NC0471611) and 5 μM Y-27632 dihydrochloride (LC Laboratories, Y-5301), followed by centrifugation at 400g for 10 min. hPSC-CMs were resuspended in the quenching media and plated at 150,000 cells per device onto Matrigel-coated devices containing MoS2monolayers integrated with gold leads. Media was changed after two days of plating to the cardiomyocyte maintenance media and every other day until assays were performed.

[0155] FIG. 21B shows data plots showing the mean displacement of the cardiomyocytes extracted from microscope video recordings using optical flow analysis. We studied the biocompatibility of the MoS2sensor, including both cytotoxicity of the MoS2and the phototoxicity induced due to the laser illumination. First, cardiomyocytes derived from human induced pluripotent stem cells (hPSCs) were plated onto devices containing MoS2monolayers integrated with gold leads. All 6 dishes of healthy cells (150,000 cells per dish) remained attached to thePCT Application Attorney Docket No.: 009062.8585.WO00MoS2coated dish after 3 weeks of culturing and showed regular beating motion in microscope video recordings, which indicated that the MoS2crystals are likely not cytotoxic to cardiomyocytes. In the phototoxicity study, we exposed cells from the same region with a 60 W / cm2orange laser continuously for 5 mins every day for 5 days. Cell health is confirmed by analyzing the microscope video recordings of the cells using the optical flow method to extract the displacements of the cells during contractions, where the mean value of the displacement can be used as the indicator for cell contractability and hence cell health. The mean displacements of the cells are plotted in FIG. 21B. Day 1 of the experiment corresponded to 3 weeks after the cell was plated. The results show that the laser exposure had no effect on the cells and cells have matured over the 5-day period which can be seen from the increased displacement amplitude and the decrease in the contraction frequency. The initial increase of the cell displacement amplitude is likely due to the maturation of the cardiomyocyte cells.

[0156] FIG. 21C shows a schematic illustration of an example equivalent circuit corresponding to the multi-modal recording configuration for a cell 2102 with cell membrane 2104, where a portion of the cell membrane may be in contact with a monolayer material 2106. In the example shown, the monolayer material 2106 may be MoS2or another TMDC. The monolayer material 2106 may be disposed on a substrate 2110 and at least partially in contact with an electrically conductive lead 2108. In the equivalent circuit model, Rmand Cmare the resistance and capacitance between the free membrane (the portion of the membrane 2104 not in contact with the monolayer material 2106) and the solution, Rjmand Cjm, are the resistances and capacitances of the cell membrane 2104 in contact with the monolayer material 2106, Rsis sealing resistance between the membrane-MoS2 junction and the solution, Zinis the input impedance to the MoS22106 electrode. Vji is the modified extracellular potential that was all-optically read out by MoS22106, Vj2 is the extracellular potential electrically recorded by the MoS2electrode.

[0157] By taking advantage of the optical transparency (-85%) of the semiconducting monolayer MoS22106 acting as an electrode, we performed simultaneous all-optical voltage sensing and electrode-based electrophysiological recordings. These dual features permit us to cross-validate all-optical measurements (FIG. 21C). The impedance of MoS22106 microelectrodes connected to integrated gold leads 2108 were measured as -1 M at 1 kHz for an area of 800 μm2.

[0158] FIG. 21D shows data plots showing simultaneously measured electrical (top trace) andPCT Application Attorney Docket No.: 009062.8585.WO00fractional PL, ΔF / F (bottom trace) signals. As indicated by the arrows, the electrical and optical signals are well aligned in time.

[0159] The simultaneously recorded electrode-based and all-optical voltage sensing data are shown in FIG. 21D. For the simultaneous electrode recording, we use a commercially available data acquisition system (Intan Technologies) connected to the gold lead 2108 for reading out hPSC-CM electrical activity utilizing MoS22106 as an electrode. The electrical signal was recorded at a sampling rate of 20 kHz and was notch filtered at 60 and 120 Hz and then bandpass filtered at 1 to 1000 Hz. For all-optical recording, a 594 nm laser was focused to a ~5 μm2spot on the MoS2monolayer 2106 with a cardiomyocyte on top as shown in FIG. 21A and the PL was collected onto a photodetector through the microscope detailed in FIG. 12. The intensity was calculated based on the total optical power of 3 pW measured at the focal point of the objective lens, and a laser spot size of 5 μm2(FIG. 12). The optical signal was recorded at 250 kHz sampling rate, notch filtered at 60 and 120 Hz, lowpass filtered at 300 Hz, and then Gaussian smoothed using a Gaussian kernel of size 8001 and a standard deviation of 1000. The electrical signal was recorded at a sampling rate of 20 kHz and was notch filtered at 60 and 120 Hz and then bandpass filtered from 1 to 1000 Hz.

[0160] Comparison of the recorded electrode-based and all-optical traces show perfect alignment in time. Since the electrode-based extracellular potential recording is a well-accepted gold standard in electrophysiology, such alignment in time strongly supports the biological origin of the MoS2trionic PL signal. However, the all-optical trace appeared temporally broadened compared to the electrical trace (FIG. 21D). We hypothesize that such a delay is due to the local modification of cell membrane’s electrical properties at the excitation spot through localized laser induced partial poration (laser intensity of 60 W / cm2) while not affecting cell viability, as described above. To better understand such broadening, we fit the extracellular voltagerecorded all-optically by the MoS22106 in the corresponding equivalent circuit model shown in FIG. 21C. The all-optically recordedoriginates from the intracellular voltage (i.e. action potential) filtered through the cell membrane, which is modeled as a resistor and capacitor in parallel.

[0161] FIG. 21E shows a schematic of an example process for fitting the optically measured extracellular field potential using the equivalent circuit model. The intracellular voltage shown in the left panel was used as the input voltage VAP in the equivalent model. The time scale in the VAPPCT Application Attorney Docket No.: 009062.8585.WO00plot is 1 second, the amplitude of the waveform was scaled to match the measured optical signal and has arbitrary unit. The middle panel shows the equivalent circuit model, where the parameters are: Rm=50 MOhm, Cm=40 pF, Rjm= 3000 MOhm, Cjm= 1500 pF, and Rs= 80 MOhm. The right panel show the simulated extracellular field potential Vj (blue curve, 2112) and the scaled experimentally measured optical signal of the MoS2voltage sensor (red curve, 2114). The amplitude of the plot has arbitrary unit, the time scale in the plot is 1 second.

[0162] FIG. 21F shows a schematic of an additional equivalent circuit model simulation of the all-optically measured action potential. The equivalent circuit shown in the middle panel is identical to the optical readout equivalent circuit components shown in FIG. 21C, where the voltage input VAP is the intracellular action potential (blue curve in the upper panel). The simulated output Vout of the circuit (blue curve in the lower panel, 2116) corresponds to the extracellular voltage Vji which is indicated by the measured fractional PL. The parameters of the equivalent circuit (Rm, Cm, Rjmi, Cjmi, and Rs) are determined by fitting Voutto the scaled experimentally measured ΔF / F (red curve in the lower panel, 2118).

[0163] The extracellular field potential read out from the MoS2voltage sensor has longer duration than that read out from the gold electrode 2108 as shown in FIG. 21F. We explain this using the equivalent circuit model shown in FIG. 21E, where the membrane resistance RJm, membrane capacitance Cjm, and the sealing resistance Rsare modified by the focused laser beam. The input signal VAP is obtained by converting the experimentally measured optical intracellular action potential signal into an electrical signal using the voltage responsivity (1.73% / mV) characterized in FIG. 18. The values of Rjm, Cjm, and Rsare then found by minimizing the error between the voltage Vj in the circuit and the signal converted from the fractional PL. The simulation results showed a good agreement of the waveform between the simulated voltage that corresponds to the extracellular field potential and the measured optical signal, as shown in the rightmost panel of FIG. 21E

[0164] FIG. 21F shows that the results from the circuit simulations agree well with the all-optical voltage sensing trace in FIG. 21D. The equivalent circuit simulation results support our hypothesis that the MoS2PL reports a modified extracellular potential measured outside a laser modified portion of the cell membrane. While a lower laser intensity can always be used, the minimum laser excitation intensity used for trionic PL in this example is limited by the signal-to-noise ratio due to the relatively low quantum efficiency of the commercially sourced chemicalPCT Application Attorney Docket No.: 009062.8585.WO00vapor deposition (CVD) grown MoS2used in this example. Such limitations can be overcome by enhancing MoS2quantum efficiency via chemical or plasmonic enhancement.

[0165] FIG. 22A shows a schematic illustration of MoS2trions interacting with the ionic activity from an optically porated cell. The 594 nm laser is used for both PL excitation and membrane poration.

[0166] Besides the extracellular field potentials, it is also of great interest to all-optically measure the intracellular action potentials for which conventional penetrating or patch electrodes are used. Previous studies have shown that the cell membrane can be locally porated to expose the sensor to the cytoplasm and enable the interrogation of the intracellular action potentials through electrical means. However, these conventional on-chip intracellular recording methods require nano or microscale protruding electrodes that poke into the cell membrane to achieve a seal with the membrane, which is invasive and can result in cell apoptosis. In contrast, the naturally formed tight seal between the cell and the atomically flat MoS2monolayer is much less invasive. By briefly increasing the 594 nm excitation laser intensity to 300 W / cm2, we can locally porate cells attached to the MoS2monolayer as shown schematically in FIG. 22A without affecting cell viability (see FIG. 21B for long-term cell viability after laser illumination).

[0167] FIG. 22B shows a false-color micrograph of hPSC-CMs cultured on MoS2monolayers used for optical poration experiments.

[0168] FIG. 22C shows data plots showing all-optical voltage measurement from the same area of a MoS2monolayer beneath a cardiomyocyte before (upper panel) and after (lower panel) optical poration. The upper panel in FIG. 22C shows fractional PL change recorded all-optically with a photodetector at low laser intensities before optical poration with no sign of electrophysiological activity. Intracellular action potentials begin showing in the fractional PL signal after poration at an intensity of 300 W / cm2and then lowering the laser intensity back to 60 W / cm2as shown in the lower panel of FIG. 22C.

[0169] FIG. 22D shows data plots showing continuous all-optical voltage recording over 10 minutes showing a gradually increasing intracellular signal due to optical poration. A close-up view of fractional PL traces is shown in the lower panels, corresponding to (as indicated by arrows) 10 s windows highlighted by the rectangles in the upper panel. The degree of poration can be controlled by the laser exposure time as shown in FIG. 22D (upper panel). As the exposure time increases, the magnitude of the all-optically recorded action potentials become higher (FIG. 22D,PCT Application Attorney Docket No.: 009062.8585.WO00lower panels). Replicates of the same experiments using cells cultured on the same dish at different locations are shown in FIGS. 22E and 22F, and cells cultured on a different dish are shown in FIGS. 22G-22I.

[0170] FIGS. 22E-22I show images and data plots showing intracellular recordings after optical poration of the cell. The microscope photo on the left of each figure shows the PL image of the MoS2. indicating the region of the laser excitation. An excitation laser intensity of 300 W / cm2was used in all the recordings. The data plots in the center and right of each figure show the recorded MoS2PL modulated by the intracellular action potential, where the rightmost plot is a close-up view of the PL trace. The PL traces are first notch filtered at 60 Hz and 120 Hz, and then bandpass filtered at 0.1 Hz to 200 Hz. The scale bars are all 50 pm.

[0171] FIG. 23A shows snapshot images from a sCMOS video recording corresponding to different stages of the all-optically recorded intracellular action potential indicated by stars in FIG.23C. The dashed line 2302 indicates the gold electrode and the dashed line 2304 indicates the SU-8 encapsulation area. MoS2monolayers beneath porated and intact cells are outlined by the bowtie shape and triangle dashed lines, respectively.

[0172] FIG. 23B shows data plots showing time traces extracted from all-optical sCMOS video recording from MoS2monolayers beneath porated (top panel) and intact (bottom panel) cardiomyocytes, respectively.

[0173] FIG. 23C shows a data plot showing a close-up view of an all-optically recorded single action potential from the monolayer beneath the porated cell highlighted in FIG. 23B. The stars indicate the time points at which the snapshots in FIG. 23A were taken.

[0174] Having characterized all-optical trionic voltage sensing with a single pixel detector, we demonstrate the compatibility of trionic biological voltage sensing with wide-field single photon microscopy using a sCMOS camera and LED illumination. Using this configuration, we captured a video of all-optical recording of cellular activity. FIG. 23A shows snapshot images from this all-optical trionic video recording of hPSC-CM action potentials over a 200 pm x 200 pm area. From these video recordings, we then extracted the time traces shown in FIGS. 23B and 23C in good agreement with those observed on the single pixel detector in FIGS. 22C and 22D. Only the MoS2monolayer outlined with blue dashed line (bowtie shape) in FIG. 23A beneath an optically porated cardiomyocyte reports electrophysiological activity through modulations in bright PL while another adjacent MoS2monolayer (green dashed triangle) shows no activity. Fractional PLPCT Application Attorney Docket No.: 009062.8585.WO00calculated from the averaged pixel intensities for these MoS2monolayers under both porated and intact cells are plotted in FIG. 23B. Only the MoS2monolayer under the locally porated cardiomyocyte (FIG. 23B top panel) reports intracellular action potentials without affecting cell viability.

[0175] FIG. 24 shows images and data plots showing simultaneous recording of intracellular action potentials from electrically floating MoS2using wide-field fluorescent microscopy. The left column shows a single frame from a recorded video. The right column shows the average fractional PL corresponding to the area outlined in the image in the left column. Cells at locations indicated in the top two rows had been previously porated and obvious intracellular action potential signal can be seen in the fractional PL plots. Locations indicated in the bottom two rows have no porated cells or have no cells, and there are no intracellular action potential signals in the fractional PL plots. The scale bar is 20 pm.

[0176] When multiple cardiomyocytes were optically porated, the all-optical voltage sensing was still localized to the proximity of the porated area (FIG. 24). We also note that the monolayer underneath the porated cell appears brighter since it is connected to a grounded gold lead, which reduces the background electron density leading to slightly higher quantum efficiency. However, such contact to the grounded gold lead is not necessary for all-optical voltage sensing as shown in FIG. 24, which is crucial for the tetherless operation using MoS2monolayers. FIG. 23C shows the close-up view of a single action potential activity highlighted in the upper panel of FIG. 23B. The timing marks indicated with blue stars in FIG. 23C correspond to the camera snapshot sequence shown in FIG. 23 A marking the depolarization and repolarization of the cardiomyocytes. Thus, these results show that MoS2monolayers can all-optically report the cardiomyocyte electrical activity using a standard single photon microscope. Since this approach does not require any specialized optical set up, it may be adapted for investigation of biologically relevant questions.

[0177] Additional analysis and discussion

[0178] The results discussed above demonstrate that monolayer semiconductors can be used successfully for biocompatible label-free all-optical trionic biological voltage sensing, which can be readily adapted into conventional optical microscopies. This label-free technology compares favorably in terms of voltage sensitivity with conventional optical voltage sensing schemes such as genetically encoded voltage indicators (GEVIs) without a need for prior genetic modifications. Monolayer trionic sensing has a much faster response time compared to genetically encoded Ca2+PCT Application Attorney Docket No.: 009062.8585.WO00indicators (GECIs), exhibits no phototoxicity unlike voltage sensitive dyes (VSDs), and is biocompatible, unlike cadminum- (Cd) based core / shell quantum dots (QDots).

[0179] FIG. 25 shows fluorescent microscope images of MoS2PL taken at t=0. 1, and 2 hours. The figure demonstrates the photostability of the MoS2sensor, where under a 2-hour continuous optical excitation from a bandpass filtered LED light (570 nm to 616 nm) of 10 W / cm2the MoS2PL captured after a 634 nm long pass filter didn’t show any decrease. Such high photo stability of the MoS2can be attributed to its more stable single crystalline structure compared to other fluorescent molecules.

[0180] The intrinsic generation of trions in a protected crystalline environment yields improved photostability as confirmed by our experiments showing no photobleaching of MoS2under a 2-hour continuous optical excitation, as opposed to isolated molecular voltage indicators prone to photobleaching. In recent years, label-free optical techniques based on total internal reflection (TIR) from electrochromic materials have been proposed. Detection with electrochromic materials based on TIR and plasmonic light scattering require specialized optical setups involving dual lasers, balanced detection, or dark-field spectroscopy setups. Alternatively, nitrogen vacancy (NV) centers in diamond have been explored, but require magnetic fields in a specialized configuration limiting direct adaptation into conventional light microscopies or requires costly substrates. In contrast, trionic all-optical sensing can be directly used in existing microscopy configurations readily available in neurobiology laboratories.

[0181] In summary, we experimentally demonstrated all-optical voltage sensing and imaging of hPSC-CM electrophysiological activity both intracellularly and extracellularly using trionic MoS2PL. Trions play an important role for the voltage sensitive PL as confirmed by a theory based on the quantum statistics of background electrons. All-optical voltage measurements are validated by simultaneously using MoS2itself as an electrode interface. This work establishes monolayer semiconductors as a strong candidate for label-free all-optical voltage sensing. The compatibility of this approach with conventional optical microscopy modalities presents itself for rapid adaption by extending the voltage imaging toolbox.

[0182] Examples

[0183] In some embodiments in accordance with the present technology (example 1 ), a method for detecting biological electrical activity includes illuminating a probe material with a light source such that the probe material emits an optical signal, wherein the probe material is positioned inPCT Application Attorney Docket No.: 009062.8585.WO00proximity to one or more electroactive cells and the optical signal responds to changes in a surrounding electric field; measuring the optical signal emitted by the probe material; fitting the optical signal with a first peak function corresponding to a trionic component of the optical signal and a second peak function corresponding to an excitonic component of the optical signal; and detecting an electrical signal from the one or more electroactive cells based on an intensity of the first peak function relative to an intensity of the second peak function.

[0184] Example 2 includes the method of example 1 or any of examples 1-15, wherein the optical signal responds to changes in the surrounding electric field without applying a bias voltage to the probe material.

[0185] Example 3 includes the method of example 1 or any of examples 1-15, wherein the probe material comprises a transition metal dichalcogenide (TMDC) material.

[0186] Example 4 includes the method of example 3 or any of examples 1-15, wherein the probe material is molybdenum disulfide (MoS2).

[0187] Example 5 includes the method of example 3 or any of examples 1-15, wherein the probe material is selected from a group consisting of molybdenum disulfide (MoS2), molybdenum diselenide (MoSe2), tungsten disulfide, (WS2) and tungsten diselenide (WSe2).

[0188] Example 6 includes the method of example 3 or any of examples 1-15, wherein the probe material is configured as a monolayer comprising the TMDC material.

[0189] Example 7 includes the method of example 3 or any of examples 1-15, wherein the probe material is configured as a plurality of nanoparticles comprising the TMDC material.

[0190] Example 8 includes the method of example 7 or any of examples 1-15, wherein the nanoparticles comprising the TMDC material are configured as nanosheets or monolayer nanodisks with a diameter in a range of 20 nm to 50 nm.

[0191] Example 9 includes the method of example 7 or any of examples 1-15, further comprising depositing the plurality of nanoparticles comprising the TMDC material in a cell culture of the electroactive cells to distribute the nanoparticles along a variety of vertical positions within the cell culture.

[0192] Example 10 includes the method of example 1 or any of examples 1-15, wherein the light source comprises a laser or a light emitting diode (LED).

[0193] Example 11 includes the method of example 1 or any of examples 1-15, wherein the probe material is positioned within 10 µm or less of the one or more electroactive cells.PCT Application Attorney Docket No.: 009062.8585.WO00

[0194] Example 12 includes the method of example 1 or any of examples 1-15, wherein the detected electrical signal corresponds to an extracellular potential of the one or more electroactive cells.

[0195] Example 13 includes the method of example 1 or any of examples 1-15, wherein the probe material is in physical contact with an electrode, the method further comprising applying an electrical bias to the probe material using the electrode.

[0196] Example 14 includes the method of example 1 or any of examples 1-15, wherein the probe material is in physical contact with an electrode, the method further comprising detecting the electrical signal from the one or more electroactive cells using the electrode.

[0197] Example 15 includes the method of example 1 or any of examples 1-14, wherein the probe material is in physical contact with the one or more electroactive cells, the method further comprising selectively porating the one or more electroactive cells, such that the detected electrical signal corresponds to an intracellular potential of the one or more electroactive cells.

[0198] In some embodiments in accordance with the present technology (example 15 A), a method for detecting biological electrical activity includes illuminating a probe material with a light source such that the probe material emits an optical signal, wherein the probe material is positioned in proximity to one or more electroactive cells and the optical signal is sensitive to changes in a surrounding electric field; measuring the optical signal emitted by the probe material; processing (e.g.. decomposing) the optical signal into an excitonic component and a trionic component; and detecting an electrical signal from the one or more electroactive cells based on a magnitude of the trionic component relative to a magnitude of the excitonic component of the optical signal.

[0199] Example 15B includes the method of example 15 A, which further includes one or more features recited in any of examples 1-15.

[0200] In some embodiments in accordance with the present technology (example 16), a system for detecting biological electrical activity includes a probe material configured to emit an optical signal when illuminated, wherein the optical signal responds to changes in surrounding electric field caused by one or more electroactive cells; a light source configured to illuminate the probe material such that the probe material emits the optical signal; an optical detector configured to measure the optical signal emitted by the probe material; and a data processing unit comprising one or more processors, in data communication with the optical detector, configured to fit thePCT Application Attorney Docket No.: 009062.8585.WO00optical signal with a first peak function corresponding to a trionic component of the optical signal and a second peak function corresponding to an excitonic component of the optical signal, and detect an electrical signal from the one or more electroactive cells based on an intensity of the first peak function relative to an intensity of the second peak function.

[0201] Example 17 includes the system of example 16 or any of examples 16-31, wherein the optical signal responds to changes in the surrounding electric field without requiring application of a bias voltage to the probe material.

[0202] Example 18 includes the system of example 16 or any of examples 16-31, wherein the probe material comprises a transition metal dichalcogenide (TMDC) material.

[0203] Example 19 includes the system of example 18 or any of examples 16-31, wherein the probe material is molybdenum disulfide (MoS2).

[0204] Example 20 includes the system of example 18 or any of examples 16-31, wherein the probe material is selected from a group consisting of molybdenum disulfide (MoS2), molybdenum diselenide (MoSe2), tungsten disulfide, (WS2) and tungsten diselenide (WSe2).

[0205] Example 21 includes the system of example 18 or any of examples 16-31, wherein the probe material is configured as a monolayer comprising the TMDC material.

[0206] Example 22 includes the system of example 18 or any of examples 16-31, wherein the probe material is configured as a plurality of nanoparticles comprising the TMDC material.

[0207] Example 23 includes the system of example 22 or any of examples 16-31, wherein the nanoparticles comprising the TMDC material are configured as nanosheets or monolayer nanodisks with a diameter in a range of 20 nm to 50 nm.

[0208] Example 24 includes the system of example 22 or any of examples 16-31, wherein the plurality of nanoparticles comprising the TMDC material are distributed along a variety of vertical positions within a cell culture of the electroactive cells.

[0209] Example 25 includes the system of example 16 or any of examples 16-31, wherein the light source comprises a laser or an LED.

[0210] Example 26 includes the system of example 16 or any of examples 16-31, wherein the probe material is configured to be positioned within 10 µm or less of the one or more electroactive cells.

[0211] Example 27 includes the system of example 16 or any of examples 16-31, wherein the detected electrical signal corresponds to an extracellular potential of the one or more electroactivePCT Application Attorney Docket No.: 009062.8585.WO00cells.

[0212] Example 28 includes the system of example 16 or any of examples 16-31, further comprising an electrode in contact with the probe material, wherein the electrode is configured to apply an electrical bias to the probe material.

[0213] Example 29 includes the system of example 16 or any of examples 16-31, further comprising an electrode in contact with the probe material, wherein the electrode is configured to detect the electrical signal from the one or more electroactive cells.

[0214] Example 30 includes the system of example 16 or any of examples 16-31, further comprising a substrate to facilitate a plurality of the electroactive cells.

[0215] Example 31 includes the system of example 16 or any of examples 16-31, wherein the probe material is configured to be in contact with the one or more electroactive cells, and wherein the light source is further configured to selectively porate the one or more electroactive cells, such that the detected electrical signal corresponds to an intracellular potential of the one or more electroactive cells.

[0216] In some embodiments in accordance with the present technology (example 31 A), a system for detecting biological electrical activity includes a probe material configured to emit an optical signal when illuminated, wherein the optical signal is sensitive to changes in surrounding electric field caused by one or more electroactive cells; a light source configured to illuminate the probe material such that the probe material emits the optical signal; an optical detector configured to measure the optical signal emitted by the probe material; and one or more processors in communication with the optical detector, configured to process (e.g., decompose) the optical signal into an excitonic component and a trionic component, and detect an electrical signal from the one or more electroactive cells based on a magnitude of the trionic component relative to a magnitude of the excitonic component of the optical signal.

[0217] Example 3 IB includes the system of example 31 A, which further includes one or more features recited in any of examples 16-31.

[0218] Conclusion

[0219] Implementations of the subject matter and the functional operations described in this patent document can be implemented in various systems, digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Implementations ofPCT Application Attorney Docket No.: 009062.8585.WO00the subject matter described in this specification can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a tangible and non-transitory computer readable medium for execution by. or to control the operation of, data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter effecting a machine -readable propagated signal, or a combination of one or more of them. The term “data processing unit” or “data processing apparatus” encompasses all apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them.

[0220] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0221] The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).

[0222] Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a readPCT Application Attorney Docket No.: 009062.8585.WO00only memory or a random access memory or both. The essential elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Computer readable media suitable for storing computer program instructions and data include all forms of nonvolatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g.. EPROM, EEPROM, and flash memory devices. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

[0223] While this patent document contains many specifics, these should not be construed as limitations on the scope of any invention or of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments of particular inventions. Certain features that are described in this patent document in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.

[0224] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. Moreover, the separation of various system components in the embodiments described in this patent document should not be understood as requiring such separation in all embodiments.

[0225] Only a few implementations and examples are described and other implementations, enhancements and variations can be made based on what is described and illustrated in this patent document.

Claims

PCT Application Attorney Docket No.: 009062.8585.WO00CLAIMSWhat is claimed is:

1. A method for detecting biological electrical activity, comprising:illuminating a probe material with a light source such that the probe material emits an optical signal, wherein the probe material is positioned in proximity to one or more electroactive cells and the optical signal responds to changes in a surrounding electric field;measuring the optical signal emitted by the probe material;fitting the optical signal with a first peak function corresponding to a trionic component of the optical signal and a second peak function corresponding to an excitonic component of the optical signal; anddetecting an electrical signal from the one or more electroactive cells based on an intensity of the first peak function relative to an intensity of the second peak function.

2. The method of claim 1, wherein the optical signal responds to changes in the surrounding electric field without applying a bias voltage to the probe material.

3. The method of claim 1, wherein the probe material comprises a transition metal dichalcogenide (TMDC) material.

4. The method of claim 3, wherein the probe material is molybdenum disulfide (MoS2).

5. The method of claim 3, wherein the probe material is selected from a group consisting of molybdenum disulfide (MoS2), molybdenum diselenide (MoSe2), tungsten disulfide, (WS2) and tungsten diselenide (WSe2).

6. The method of claim 3, wherein the probe material is configured as a monolayer comprising the TMDC material.

7. The method of claim 3, wherein the probe material is configured as a plurality of nanoparticles comprising the TMDC material.

8. The method of claim 7, wherein the nanoparticles comprising the TMDC material are configured as nanosheets or monolayer nanodisks with a diameter in a range of 20 nm to 50 nm.PCT Application Attorney Docket No.: 009062.8585.WO009. The method of claim 7, further comprising:depositing the plurality of nanoparticles comprising the TMDC material in a cell culture of the electroactive cells to distribute the nanoparticles along a variety of vertical positions within the cell culture.

10. The method of claim 1, wherein the light source comprises a laser or a light emitting diode (LED).

11. The method of claim 1, wherein the probe material is positioned within 10 µm or less of the one or more electroactive cells.

12. The method of claim 1, wherein the detected electrical signal corresponds to an extracellular potential of the one or more electroactive cells.

13. The method of claim 1, wherein the probe material is in physical contact with an electrode, the method further comprising:applying an electrical bias to the probe material using the electrode.

14. The method of claim 1, wherein the probe material is in physical contact with an electrode, the method further comprising:detecting the electrical signal from the one or more electroactive cells using the electrode.

15. The method of claim 1, wherein the probe material is in physical contact with the one or more electroactive cells, the method further comprising:selectively porating the one or more electroactive cells, such that the detected electrical signal corresponds to an intracellular potential of the one or more electroactive cells.

16. A system for detecting biological electrical activity, comprising:a probe material configured to emit an optical signal when illuminated, wherein the optical signal responds to changes in surrounding electric field caused by one or more electroactive cells;a light source configured to illuminate the probe material such that the probe material emits the optical signal:an optical detector configured to measure the optical signal emitted by the probe material;PCT Application Attorney Docket No.: 009062.8585.WO00anda data processing unit comprising one or more processors, in data communication with the optical detector, configured to fit the optical signal with a first peak function corresponding to a trionic component of the optical signal and a second peak function corresponding to an excitonic component of the optical signal, and detect an electrical signal from the one or more electroactive cells based on an intensity of the first peak function relative to an intensity of the second peak function.

17. The system of claim 16, wherein the optical signal responds to changes in the surrounding electric field without requiring application of a bias voltage to the probe material.

18. The system of claim 16, wherein the probe material comprises a transition metal dichalcogenide (TMDC) material.

19. The system of claim 18, wherein the probe material is molybdenum disulfide (MoS2).

20. The system of claim 18, wherein the probe material is selected from a group consisting of molybdenum disulfide (MoS2), molybdenum diselenide (MoSe2), tungsten disulfide, (WS2) and tungsten diselenide (WSei).

21. The system of claim 18, wherein the probe material is configured as a monolayer comprising the TMDC material.

22. The system of claim 18, wherein the probe material is configured as a plurality of nanoparticles comprising the TMDC material.

23. The system of claim 22, wherein the nanoparticles comprising the TMDC material are configured as nanosheets or monolayer nanodisks with a diameter in a range of 20 nm to 50 nm.

24. The system of claim 22, wherein the plurality of nanoparticles comprising the TMDC material are distributed along a variety of vertical positions within a cell culture of the electroactive cells.

25. The system of claim 16, wherein the light source comprises a laser or an LED.PCT Application Attorney Docket No.: 009062.8585.WO0026. The system of claim 16, wherein the probe material is configured to be positioned within 10 µm or less of the one or more electroactive cells.

27. The system of claim 16, wherein the detected electrical signal corresponds to an extracellular potential of the one or more electroactive cells.

28. The system of claim 16, further comprising:an electrode in contact with the probe material, wherein the electrode is configured to apply an electrical bias to the probe material.

29. The system of claim 16. further comprising:an electrode in contact with the probe material, wherein the electrode is configured to detect the electrical signal from the one or more electroactive cells.

30. The system of claim 16, further comprising:a substrate to facilitate a plurality of the electroactive cells.

31. The system of claim 16, wherein the probe material is configured to be in contact with the one or more electroactive cells, and wherein the light source is further configured to selectively porate the one or more electroactive cells, such that the detected electrical signal corresponds to an intracellular potential of the one or more electroactive cells.

Images