Monolithically Integrated μLEDs on Silicon Neural Probes for High-Resolution Optogenetic Studies in Behaving Animals

Monolithically Integrated μLEDs on Silicon Neural Probes for High-Resolution Optogenetic Studies in Behaving Animals Wu Stark Ku Wise Buzsáki Yoon

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

Monolithically Integrated μLEDs on Silicon Neural Probes for High-Resolution Optogenetic Studies in Behaving Animals 2015 Neuron 10.1016/j.neuron.2015.10.032 Wu2015-monolithically-integrated-leds-on-silicon-ne Wu2015-monolithically-integrated-leds-on-silicon-ne.txt, the project's staged copy bc888709b35563b7e4e862aae6f0fda06d5a860f not-redistributable neuroscience

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them how to deliver light at the scale of a single neuron and record from the same place, in a freely moving animal Quotation withheld because this record is not cleared for redistribution. whether light sources small enough to match a soma can be built onto a silicon recording probe without excessive heating Quotation withheld because this record is not cleared for redistribution. a probe with twelve neuron-sized LEDs and thirty-two recording sites on one silicon shank array, driving spikes at 60 nW and moving the resolution limit from light delivery onto opsin targeting Quotation withheld because this record is not cleared for redistribution. method Quotation withheld because this record is not cleared for redistribution. optogenetic light delivery and recording in deep brain structures of freely moving animals Quotation withheld because this record is not cleared for redistribution. system-or-tool-building system-or-tool-building Quotation withheld because this record is not cleared for redistribution. mechanisms light reaching non-somatic (dendritic) ChR2 recruits the neuron even when the micro-LED is far from its soma, which smooths the gain-versus-distance relation into a graded rather than step-like profile Quotation withheld because this record is not cleared for redistribution. mechanisms current flow through the micro-LED raises the voltage of the continuous n-GaN cathode layer, which capacitively couples into the recording interconnects running above it and produces the rectified artifact on all channels Quotation withheld because this record is not cleared for redistribution. mechanisms excitatory drive arriving at stratum radiatum from CA3 recruits a faster and spatially tighter ripple-generating network than drive arriving at stratum oriens from entorhinal cortex/amygdala Quotation withheld because this record is not cleared for redistribution. coupling an LED into a waveguide is capped by etendue, which is why the light source is put at the site instead of piped to it Quotation withheld because this record is not cleared for redistribution. silicon is chosen for three separate reasons at once: it can be micromachined into needles, it conducts heat five times better than sapphire, and being opaque it confines emission to one face Quotation withheld because this record is not cleared for redistribution. heat leaves through the shank toward the proximal end, so the most distal LED is the worst case and is the one modelled Quotation withheld because this record is not cleared for redistribution. the graded fall-off of drive with distance implies light is acting on dendrites, not only somata Quotation withheld because this record is not cleared for redistribution. the artefact has two physically distinct sources - voltage coupling on the board and capacitive coupling from the continuous n-GaN layer - and each has its own fix Quotation withheld because this record is not cleared for redistribution. Driving the μLEDs induces low-frequency electrical artifacts on the recording channels. what this limits: the recorded signal during stimulation: A LIMITATION OF THE CURRENT VERSION OF uLED PROBES IS THE LOW-FREQUENCY ARTIFACTS INDUCED DURING PERIODIC ACTIVATION Quotation withheld because this record is not cleared for redistribution. Non-somatic opsin expression, not the light source, now caps the achievable spatial resolution of spiking control. what this limits: the achievable spatial resolution: THE FINAL LIMITATION TO THE CURRENT UTILITY OF THE uLED PROBES RELATES TO THE AVAILABLE OPSIN TECHNOLOGIES Quotation withheld because this record is not cleared for redistribution. The probe reports spiking only, so it complements rather than replaces intracellular and all-optical methods. what this limits: what the probe can report at all: IT ONLY PROVIDES INFORMATION ABOUT SPIKING (AND NOT ABOUT INTRACELLULAR CALCIUM, Vm, OR TRANSMEMBRANE CURRENTS) Quotation withheld because this record is not cleared for redistribution. The indirect test of heating effects on spike waveforms is weakened by concurrent high-frequency oscillations at high power. what this limits: the heating test on spike waveforms: HIGH-POWER ILLUMINATION WAS OCCASIONALLY ACCOMPANIED BY HIGH-FREQUENCY (2 kHz) OSCILLATIONS, LIMITING THE INTERPRETATIONAL POWER OF THESE OBSERVATIONS Quotation withheld because this record is not cleared for redistribution. photon extraction is poor by construction: 33% internal efficiency, half the light lost into the substrate, most of the rest trapped by total internal reflection what this limits: the light budget: APPROXIMATELY 50% OF THE PHOTONS ARE EMITTED TOWARD THE SUBSTRATE AND ARE ABSORBED AT THE GaN/SILICON INTERFACE Quotation withheld because this record is not cleared for redistribution. periodic driving injects a low-frequency artefact into the recordings, present in opsin-free animals and removed only by offline filtering what this limits: the same scope as limitations[0] - the artefact is present in opsin-free animals and removed by filtering, not absent Quotation withheld because this record is not cleared for redistribution. there is no accepted safe-temperature standard for implanted devices, so the 1 degree threshold used here is the authors' own what this limits: every safety claim about the device: AS THERE IS NO ACCEPTED THRESHOLD TEMPERATURE FOR THE SAFE OPERATION OF IMPLANTABLE NEURAL DEVICES, WE LOOSELY DEFINE THE THRESHOLD AS 1 degree - the criterion is the authors' own Quotation withheld because this record is not cleared for redistribution. whether a few milliseconds above 38 degrees harms the adjacent network is stated as unknown what this limits: operation above 3.85 V: THE PEAK TEMPERATURE CAN TRANSIENTLY REACH 38 DEGREES C; HOWEVER, WHETHER A FEW MILLISECONDS OF HEATING BEYOND 38 DEGREES C HAS ADVERSE EFFECT ON THE ADJACENT NEURAL NETWORK IS UNKNOWN Quotation withheld because this record is not cleared for redistribution. the probe reports spiking only, not calcium, membrane potential or currents, and is offered as complementary rather than superior what this limits: the same scope as limitations[2] - spiking only, offered as complementary rather than replacement Quotation withheld because this record is not cleared for redistribution. spatial resolution is now limited by where the opsin is expressed rather than by where the light goes what this limits: the same scope as limitations[1]: NON-SOMATIC OPSIN EXPRESSION FACILITATED MULTILAYERED CONTROL BUT ALSO REDUCED the achievable resolution Quotation withheld because this record is not cleared for redistribution. monolithic on-silicon LED fabrication technology is carried over into implantable neural probes for systems neuroscience Quotation withheld because this record is not cleared for redistribution. the Pennes bioheat equation, with a blood-perfusion term, is imported to simulate brain tissue heating around the probe Quotation withheld because this record is not cleared for redistribution. monolithic integration and lithographic patterning, applied so that light sources and electrodes share one substrate Quotation withheld because this record is not cleared for redistribution. etendue, used as the argument for abandoning waveguide coupling altogether Quotation withheld because this record is not cleared for redistribution. finite-element bio-heat modelling, used to set an operating envelope before any animal was implanted Quotation withheld because this record is not cleared for redistribution. recording during μLED voltage bias in saline and in wild-type animals eliminates: that the time-locked artifacts in the wide-band traces are neural responses driven by the light, rather than electrical artifacts of the μLED drive itself · outcome: held Quotation withheld because this record is not cleared for redistribution. spiking gain here = firing rate during light divided by the same neuron's rate without light, so it is a within-neuron ratio and not a comparison across cells Quotation withheld because this record is not cleared for redistribution. plug efficiency = optical power out over electrical power in; the paper gives both a measured value (0.87%) and a value derived from the loss chain (1.1%), so the measurement is checkable Quotation withheld because this record is not cleared for redistribution. iHFO is used exactly as Stark2014 defined it - an induced high-frequency oscillation - and is treated here as a proxy for somatic drive because population synchrony averages out dendritic effects Quotation withheld because this record is not cleared for redistribution. direct versus indirect activation is a CLASSIFICATION, assigned by a Gaussian mixture model over waveform features, not an observation Quotation withheld because this record is not cleared for redistribution. the safe threshold is a 1 degree rise, defined by these authors for want of a field standard and described by them as loose Quotation withheld because this record is not cleared for redistribution. claim_type: method a second reading reaching the same value. Recorded: method — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. method: monolithic fabrication of InGaN LEDs on silicon, bio-heat modelling, and in vivo recording from CA1 of anaesthetised and freely moving mice with opsin-free controls a second reading DIFFERING from the recorded value. Recorded: system-or-tool-building — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. phenomenon: optogenetic control of spiking at cellular spatial resolution in the intact brain a second reading DIFFERING from the recorded value. Recorded: optogenetic light delivery and recording in deep brain structures of freely moving animals — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. builds-on → Wu et al., 2013 Quotation withheld because this record is not cleared for redistribution. extends → Kim et al., 2013; McAlinden et al., 2013 Quotation withheld because this record is not cleared for redistribution. extends → Stark et al., 2014 Quotation withheld because this record is not cleared for redistribution. extends → Stark et al. 2014, by inducing high-frequency oscillations with focal rather than fibre illumination Quotation withheld because this record is not cleared for redistribution. critiques → fibre-coupled optogenetic probes, whose resolution is set by the bulk of the implanted fibre differs in: THE AXIS IS THE PHYSICAL BULK OF THE LIGHT DELIVERY. 'Early work involving in vivo optogenetics relied on manual assembly of commercially available recording components such as metal electrodes or passive high-density probes (STARK ET AL., 2012) with optical fibers, which are bulky and may suffer from misalignment errors. Moreover, the spatial resolution of fiber-based optogenetic devices is limited by the bulk of the [fibre].' Note the citation: this criticises the very device whose own paper names heat, a 60 um recording radius and over-precision as its limits. FOUR DIFFERENT AXES ON ONE INSTRUMENT, THREE OF THEM FROM ITS OWN AUTHORS. Quotation withheld because this record is not cleared for redistribution. critiques → sapphire-substrate microLED probes, which scatter light and cannot be machined into needles differs in: THE AXIS IS THE SUBSTRATE MATERIAL, AND IT COSTS TWICE. 'Quotation withheld.' One material choice degrades the measurement AND the tissue - so the comparison is not between two devices' performance but between what each material permits. Quotation withheld because this record is not cleared for redistribution. critiques → injectable polymer LED devices, too wide for cellular resolution and hard to mount on a microdrive Quotation withheld because this record is not cleared for redistribution. confirms → activation thresholds reported for manually assembled diode probes Quotation withheld because this record is not cleared for redistribution. distinguishes-from → two-photon holographic and intracellular methods, which cannot reach deep structures in freely moving animals differs in: THE AXIS IS WHETHER THE METHOD REACHES A FREELY MOVING ANIMAL. 'Such high spatial resolution control over spiking has been previously possible only by intra-cellular current injection or by two-photon microscopy combined with optogenetics. Intracellular methods are not yet scalable in freely moving animals.' The resolution is granted to the existing methods; what they lack is the preparation - so a claim about what is achievable needs the animal state attached. Quotation withheld because this record is not cleared for redistribution. μLED —part-of→ silicon μLED silicon strength of the relation: 1 · note: monolithic integration is the whole claim: the light source is built on the probe rather than coupled to it Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the uLEDs are fabricated monolithically on silicon; part-of is loose. Quotation withheld because this record is not cleared for redistribution. silicon —contrasted-with→ sapphire silicon sapphire strength of the relation: 1 · note: CRITERION: three at once - machinable into needles, five times the thermal conductivity, and opaque so emission stays on one face Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span lists limits of a sapphire substrate beside a silicon probe; contrasted-with is loose. Quotation withheld because this record is not cleared for redistribution. gain —operationalised-as→ spiking gain spiking strength of the relation: 1 · note: direction is measure-to-phenomenon; the ratio is within-neuron, so it cannot be compared across cells of different baseline Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but spiking 'gain' is DEFINED as a ratio of spiking rates; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. PYR —contrasted-with→ INT PYR INT strength of the relation: 1 · note: CRITERION: PYR are taken as directly driven and INT as circuit-driven, and PYR jitter is lower at every distance Quotation withheld because this record is not cleared for redistribution. stratum radiatum —contrasted-with→ stratum oriens stratum radiatum stratum oriens strength of the relation: 1 · note: illumination below or at the layer drives spiking better than above it, attributed to dendritic geometry or expression bias Quotation withheld because this record is not cleared for redistribution. ChR2 —causes→ spiking ChR2 spiking strength of the relation: 1 · note: shown by absence in wild-type controls under the same illumination Quotation withheld because this record is not cleared for redistribution. rank correlation rank correlation Quotation withheld because this record is not cleared for redistribution. permutation test permutation test Quotation withheld because this record is not cleared for redistribution. Mann-Whitney U test Mann-Whitney U test Quotation withheld because this record is not cleared for redistribution. binomial test binomial test Quotation withheld because this record is not cleared for redistribution. partial correlation partial correlation Quotation withheld because this record is not cleared for redistribution. principal component analysis principal component analysis Quotation withheld because this record is not cleared for redistribution. Gaussian mixture model Gaussian mixture model Quotation withheld because this record is not cleared for redistribution. significance level significance level Quotation withheld because this record is not cleared for redistribution. rank correlation rank correlation Quotation withheld because this record is not cleared for redistribution. permutation test permutation test Quotation withheld because this record is not cleared for redistribution. Mann-Whitney U test Mann-Whitney U test Quotation withheld because this record is not cleared for redistribution. Gaussian Mixture Model Gaussian Mixture Model Quotation withheld because this record is not cleared for redistribution. Binomial test Binomial test Quotation withheld because this record is not cleared for redistribution. Sub-microwatt light (60 nW) from a single integrated micro-LED was enough to drive spiking, and microwatt-range power drove fast population oscillations. Quotation withheld because this record is not cleared for redistribution. The probe achieved independent optical control of neurons roughly 50 micrometres apart and of separate somatic versus dendritic compartments of single neurons. Quotation withheld because this record is not cleared for redistribution. Illuminating pyramidal cell bodies and apical dendrites in stratum radiatum drove spiking more effectively than illuminating stratum oriens. Quotation withheld because this record is not cleared for redistribution. Where the light entered the layer changed the induced ripples: infra- and intra-layer illumination produced the fastest and most vertically compact oscillations, while supra-layer illumination produced slower and more widespread ones. Quotation withheld because this record is not cleared for redistribution. Spiking gain of directly driven pyramidal cells increased with light power and decreased with the distance between the micro-LED and the cell soma. Quotation withheld because this record is not cleared for redistribution. Directly light-driven pyramidal cells spiked with lower temporal jitter (median 29 ms) than indirectly driven interneurons (46 ms), at every distance bin. Quotation withheld because this record is not cleared for redistribution. Thermal simulation indicated that every drive condition below 3.4 V is safe even under ten seconds of continuous operation. Quotation withheld because this record is not cleared for redistribution. 60 to 120 nW of light drove time-locked spiking in neurons next to a single LED Quotation withheld because this record is not cleared for redistribution. two LEDs 60 microns apart controlled distinct neurons independently in freely moving mice Quotation withheld because this record is not cleared for redistribution. spiking gain rose with light power and fell with LED-to-soma distance for both directly and indirectly driven cells Quotation withheld because this record is not cleared for redistribution. directly driven pyramidal cells had lower spike jitter than indirectly driven interneurons at every distance Quotation withheld because this record is not cleared for redistribution. illumination inside or below the layer produced faster and more compact induced ripples than illumination above it Quotation withheld because this record is not cleared for redistribution. the graded gain-distance profile implies light is recruiting dendritic opsin, not only somatic Quotation withheld because this record is not cleared for redistribution. spiking was sometimes driven more strongly from stratum oriens than by direct somatic illumination Quotation withheld because this record is not cleared for redistribution. the measured peak plug efficiency was around 0.87% Quotation withheld because this record is not cleared for redistribution. CONTROL NULL: in opsin-free mice the same illumination changed no action potentials, so the effects are not thermal Quotation withheld because this record is not cleared for redistribution. DECLARED INCONCLUSIVE: the waveform-distortion test of heating was confounded by 2 kHz oscillations at high power Quotation withheld because this record is not cleared for redistribution. the low-frequency artefact appeared in wild-type animals too and was removed offline rather than at source Quotation withheld because this record is not cleared for redistribution. threshold temperature threshold temperature Quotation withheld because this record is not cleared for redistribution. spiking gain spiking gain Quotation withheld because this record is not cleared for redistribution. spike phase spike phase Quotation withheld because this record is not cleared for redistribution. temporal jitter temporal jitter Quotation withheld because this record is not cleared for redistribution. gain gain Quotation withheld because this record is not cleared for redistribution. plug efficiency plug efficiency Quotation withheld because this record is not cleared for redistribution. etendue etendue Quotation withheld because this record is not cleared for redistribution. internal quantum efficiency internal quantum efficiency Quotation withheld because this record is not cleared for redistribution. iHFOs iHFOs Quotation withheld because this record is not cleared for redistribution. voltage command voltage command Quotation withheld because this record is not cleared for redistribution. input site input site Quotation withheld because this record is not cleared for redistribution. spiking gain spiking gain Quotation withheld because this record is not cleared for redistribution. temporal jitter temporal jitter Quotation withheld because this record is not cleared for redistribution. locus of maximal iHFO power locus of maximal iHFO power Quotation withheld because this record is not cleared for redistribution. gain gain Quotation withheld because this record is not cleared for redistribution. jitter jitter Quotation withheld because this record is not cleared for redistribution. plug efficiency plug efficiency Quotation withheld because this record is not cleared for redistribution. 60 nW — light power sufficient to induce time-locked spiking in nearby neurons Quotation withheld because this record is not cleared for redistribution. 150 μm 2 — light-emitting area of a single micro-LED, comparable to a pyramidal soma cross-section Quotation withheld because this record is not cleared for redistribution. 12 μLEDs and 32 recording sites — components integrated on each implanted four-shank probe Quotation withheld because this record is not cleared for redistribution. 4,096 (2 12 ) — number of distinct on/off combinations of the twelve micro-LEDs that can be programmed at any moment Quotation withheld because this record is not cleared for redistribution. 0.87% — measured peak wall-plug (optical out over electrical in) efficiency of a micro-LED Quotation withheld because this record is not cleared for redistribution. 3.33% — photon extraction efficiency of the on-silicon micro-LED Quotation withheld because this record is not cleared for redistribution. 93 — well-isolated single units recorded from CA1 across six mice Quotation withheld because this record is not cleared for redistribution. 53 μW — saturated optical output of a micro-LED at high (~13 mA) injection current Quotation withheld because this record is not cleared for redistribution. PYR: 29 ms; INT: 46 ms — median spike-timing jitter of directly driven pyramidal cells versus indirectly driven interneurons Quotation withheld because this record is not cleared for redistribution. 60 nW — light power sufficient to induce spiking in nearby neurons Quotation withheld because this record is not cleared for redistribution. 150 μm 2 — LED emission area, matched to a pyramidal soma cross-section Quotation withheld because this record is not cleared for redistribution. 0.87% against 1.1% — measured plug efficiency against the value derived from the loss chain value source note: THE VALUE DRAWS ON THE PAPER BEYOND THIS QUOTE: part of it is printed elsewhere in the paper, not in this quote ('measured peak plug efficiency is around 0.87% ( Figure 3C )…'). Quotation withheld because this record is not cleared for redistribution. 33%, 3.33% — internal quantum efficiency and photon extraction efficiency of the on-silicon LED Quotation withheld because this record is not cleared for redistribution. 1 degree — the temperature-rise threshold the authors adopted for want of a field standard Quotation withheld because this record is not cleared for redistribution. 2°C–3°C — how much brain temperature moves on its own between rest and activity Quotation withheld because this record is not cleared for redistribution. 4,096 — programmable on-off combinations across twelve LEDs Quotation withheld because this record is not cleared for redistribution. less than 8 μA — current needed to reach the intensity that activates ChR2 Quotation withheld because this record is not cleared for redistribution. 1 MΩ at 1 kHz — impedance of the integrated recording electrodes Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.1016/j.neuron.2015.10.032.