# Fileset

[Small Sci. 2024, 4, 2400367.pdf](https://mdr.nims.go.jp/filesets/a3828be6-23a3-4609-a2cb-397541e16f2f/download)

## Creator

[Batu Ghosh](https://orcid.org/0000-0002-7339-9695), [Hiroyuki Yamada](https://orcid.org/0000-0003-0394-857X), [Kazuhiro Nemoto](https://orcid.org/0000-0001-5228-1826), [Wipakorn Jevasuwan](https://orcid.org/0000-0001-9117-2497), [Naoki Fukata](https://orcid.org/0000-0002-0986-8485), Hon‐Tao Sun, [Naoto Shirahata](https://orcid.org/0000-0002-1217-7589)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Rational Doping Strategy to Build the First Solution‐Processed p‐n Homojunction Architecture toward Silicon Quantum Dot Photodetectors](https://mdr.nims.go.jp/datasets/35863237-3cbd-400f-8bd3-dd340b63e362)

## Fulltext

Rational Doping Strategy to Build the First Solution-Processed p-n Homojunction Architecture towardSilicon Quantum Dot PhotodetectorsBatu Ghosh,* Hiroyuki Yamada, Kazuhiro Nemoto, Wipakorn Jevasuwan,Naoki Fukata,* Hon-Tao Sun, and Naoto Shirahata*1. IntroductionElectronic impurity doping in a single crystalline semiconductorhas led to dramatic advances in the control of electronic andoptoelectronic properties, enabling mostof the functionalities that enrich our sociallife today and human civilization, fromtransistors to a variety of optoelectronicdevices, including solar cells and laserdiodes. The p-n junction diode that isconstructed at homo- or heterojunctioninterface between p-type and n-type semi-conductors is the heart of device architec-ture to realize those functionalities.Colloidal quantum dots (CQDs) haveattracted increasing attention as promisingphotoactive materials for next-generationoptoelectronics due to their unique charac-teristics including size-dependent spectraltunability, multiexciton generation, andbandgap engineering.[1–6] Furthermore,they offer added benefits of scalable synthe-sis and solution processability compatiblewith printable technology, thus allowingfor large-area manufacturing, including roll-to-roll processesand low-cost fabrication without using vacuum apparatus. Aswith bulk crystals, in thin films based on CQDs, electronicimpurity doping allows controlling the concentration andmobility of charge carriers to build a p-n junction. Doping inCQDs, however, differs significantly from typical processesused in bulk crystal. A couple of methods have been reportedfor compound semiconductor CQDs, such as surface doping,substitution of impurity element at a lattice site, and introduc-tion of impurity element into an interstitial site. For surfacefunctionalization, since their energy structures can be tunedby variation of surface ligand chemistry,[7–9] the difference insurface ligand allows for having the dexterity to vary doping-type (i.e., p-type and n-type characteristics). Initial attemptshave been subjected to PbS; so far, 1,2-ethanedithiol or carbox-ylate are representative capping ligands for p-type conductiv-ity,[10,11] while n-type conductivity is given by substituting thedivalent sulfur ions for monovalent halogen ions.[12,13]Incorporating heterovalent impurities that can be substitutedfor cations introduces additional carriers in CQDs. In chalco-genide CQDs such as PbS and CdSe, the divalent cations arereplaced by Agþ ions to provide positive free carriers for p-typeconductivity.[14,15] InAs CQD is a typical n-type,[16] whereasincorporation of substitutional doping with hetero-valentZn2þ ion to replace In3þ provides free holes for p-type charac-ter.[17] In spite of those great efforts, little progress has beenB. Ghosh, H. Yamada, K. Nemoto, W. Jevasuwan, N. Fukata, H.-T. Sun,N. ShirahataResearch Center for Materials Nanoarchitectonics (MANA)National Institute for Materials Science (NIMS)1-1 Namiki, Tsukuba 305-0044, JapanE-mail: batughosh@tdbcollege.ac.in; fukata.naoki@nims.go.jp;shirahata.naoto@nims.go.jpB. GhoshDepartment of PhysicsTriveni Devi Bhalotia CollegeRaniganj, West Bengal 713347, IndiaN. ShirahataGraduate School of Chemical Sciences and EngineeringHokkaido UniversityKita 13, Nishi 8, Kita-ku, Sapporo 060-8628, JapanThe ORCID identification number(s) for the author(s) of this articlecan be found under https://doi.org/10.1002/smsc.202400367.© 2024 The Author(s). Small Science published by Wiley-VCH GmbH.This is an open access article under the terms of the CreativeCommons Attribution License, which permits use, distribution andreproduction in any medium, provided the original work is properly cited.DOI: 10.1002/smsc.202400367Semiconductor p-n homojunction is a requisite building block of operatingtransistors and diodes which make up the modern electronic circuits andoptoelectronic applications. However, it has been so far limited to bulk form of singlecrystals such as silicon (Si) or gallium arsenide. Herein, a brand-new method ofconstructing p-n homojunction architectures that breaks through the limitation ispresented. Colloidal inks of p-type and n-type Si quantum dots (QDs) are synthesizedby thermal disproportionation of (HSiO1.5)n doped with boron or phosphorus,followed by surface ligand engineering. Analysis combining UV photoelectronspectroscopy, electron spin resonance, and current–voltage characteristics confirmsthat an orthogonal solvent trick makes clean interfaces between n-type and p-typeSiQD layers without disruption on film formation. The forward and reverse current–voltage characteristics of the diode, along with various spectroscopic characteriza-tions, demonstrate the formation of the first p-n homojunction of SiQDs. Theself-powered photodiode provides a tunable response specific to the wavelength.RESEARCH ARTICLEwww.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (1 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbHmade on homogeneous p-n junctions in CQDs and their use foroptoelectronic applications with acceptable performances.Silicon (Si), a second abundant element in the earth’s crust, isa typical indirect bandgap semiconductor. A wafer of Si is a mate-rial cornerstone for microelectronics and solar cells for morethan half of century. Electronic impurity doping in a single crys-talline bulk Si is a well-established technique; phosphorus (P)and boron (B) are doped to form n-type and p-type semiconduc-tors, respectively. These crystals benefit from controlling carrierconcentration and mobility through impurity doping, as it allowsbuilding p-n junctions as a key to operating optoelectronic devi-ces in the modern industry. The electronic dopants are also beingused in other forms of Si such as nanowires to form p-n homo-junction and heterojunction for device applications includingsolar cell and phototransistor.[18,19] Considering the technologicalapplications of unique properties emerged by the quantum con-finement effects such as multiexciton generation, a more intrigu-ing challenge would be especially to dope impurity in Si QDs. Todate, many efforts have been attempted to reveal the correlationbetween optical and structural properties of QDs embedded inbulky oxide or nitride matrixes (e.g., the exact location of dopantatoms), with varying dopant concentration, most notably usingboron and phosphorus.[20–24] Such bulky materials containingdoped QDs have been used to fabricate heterojunction optoelec-tronic devices such as solar cells,[25–27] but there is little chance ofcreating a p-n homojunction as long as the QDs are embedded.Impurity doping to self-standing QDs has been achieved by lib-erating the doped QDs from the SiOx matrix with hydrofluoric(HF) acid etching.[28–32] Determination of the Fermi level, whichis subjected to hydrogen-terminated QDs doped with boron orphosphorus, was first reported in 2015 using Kelvin probemicroscopy.[33] In the following year, photoelectron yield spectro-scopic (PYS) study reported that Fermi level, highest-occupiedmolecular orbital (HOMO), and lowest-unoccupied molecularorbital (LUMO) energies of QDs codoped with boron and phos-phorus are dependent on their diameters.[34] In principle, itshould also be possible to provide extra carriers to Si CQD viaelectric impurity doping to build a p-n homojunction, butthis has not yet been realized. In this work, we report, forthe first time, emergent p-n homojunction where p-type andn-type self-standing Si QD layers have overlapped each other.The fabrication of photodiode devices using CQDs, whichefficiently detect irradiated UV light with zero bias, opens upgreat possibilities for a new paradigm in Si optoelectronics.2. Results and Discussion2.1. Synthesis of Colloidal p-Type and n-Type Si QDsB-doped and P-doped Si QDs were synthesized according toScheme 1. Triethoxysilane (TES) was hydrolyzed in the presenceof boric acid or phosphoric acid (step I), followed by filtered,washed, and dried. It has been known that the hydrolysis productof TES corresponds to amorphous hydrogen silsesquioxane, thatis, (HSiO1.5)n,[35] which can be disproportionated to Si and SiO2at temperatures above 1000 °C.[36] In this work, the hydrolysisproduct was heated at 1100 °C (step II). The resulting powderwas uniformly dark brown in color. Then, the powder was treatedwith HF acid to liberate the hydrogen-terminated Si QDs fromSiO2 matrix (step III), followed by thermal hydrosilylation of1-decene or 10-unidecenoic acid (step IV). As the reaction pro-ceeds, the outermost Si atoms were terminated by monolayersof decane or undecanoic acid. It has been known that reactionsolution turns colored transparent liquid when monomolecularcoverage exceeded ≈16%,[37,38] because those monolayers inhibitaggregation of QDs to result in complete isolation of the QDs inthe good solvent. In the present study, however, the reaction solu-tion remained slightly cloudy even after a long reaction time forhydrosilylation. This suggests that impurity ions attached to Siatoms in the outermost layer of QDs inhibited hydrosilylation,resulting in lower monolayer coverage for poor solution solubilityin some of the QDs. In step V, the QDs with low monolayercoverages were removed by centrifugation to give coloredtransparent solutions of B-doped and P-doped QDs (see digitalphotographs in Scheme 1).Scanning transmission electron microscopic (STEM) imagesshown in Figure 1 exhibit round-shaped nanoparticles with aver-age diameters of 2.1 and 2.5 nm prepared using (a) phosphoricacid and (e) boric acid. Observation of nanoparticles isolated anddispersed without aggregation indicated the termination oftheir surfaces with organic monolayers of longer chain length,[39]Scheme 1. Synthesis of colloidal inks of phosphorus-doped and boron-doped Si QDs dispersed in ethanol and toluene, respectively.www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (2 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensesuch as decane or 10-undecanoic acid. X-ray powder diffraction(XRD) patterns shown in Figure 1b,f confirmed that the drop-castnanoparticles consist of a face-centered diamond cubic crystalstructure of Si, and the diffraction angles were almost identicalto those of bulky Si. Chemical bonding states of the hydrogen-terminated samples were characterized by X-ray photoelectronmicroscopy (XPS) as shown in Figure S1 (SupportingInformation). As expected, XPS Si2p spectrum was split intotwo peaks centered at 99.3 eV for Si0 and 103.6 eV for Si4þdue to slight oxidation of Si even after HF treatment.[38] It isknown that the metallic boron and boron oxide exhibit spectralpeaks at 188 and 193 eV, respectively.[40] Observation of a broadpeak centered at 188.5 eV in the XPS B1s spectrum indicated thatthe majority is B0 while B2O3 is absent. The doped boron atomsexhibited weak Raman peaks at 620 cm�1 for 11B and 644 cm�1for 10B (see Figure S2, Supporting Information), consistent withthe previous work.[40] XPS P2p spectrum exhibited two peakscentered at 129.6 and 134.4 eV. The prominent peak was attrib-uted to the P0 bonding state and the broad tail to suboxides.These signals observed in the XPS B1s and P2p spectra disap-peared after the purification of the hydrosilylated samples viastep V shown in Scheme 1, suggesting that the signals observedas peaks have originated from the impurity atoms deposited onsample surfaces. Unlike bulk Si crystal giving TO phonon line at521 cm�1 in a symmetric vibration region, the momentum is notnecessarily conserved in a size range smaller than the Bohrradius (≈5 nm).[41] The relaxation of the momentum selectionrule allows the occurrence of Raman-active modes away fromthe Brillouin zone center, resulting in the peak broadening asym-metrically and redshifting as shown in Figure S2, SupportingInformation. Such a size-dependent characteristic could beobserved in optical properties. Optical absorbance and photolu-minescence (PL) spectra of the hydrosilylated samples are shownin Figure S3 (see Supporting Information). There was no signif-icant difference in terms of absorbance between the B-doped andthe P-doped samples. PL spectra of the P-doped and B-dopedsamples were centered at 1.77 and 1.66 eV, consistent withthe relationship between size and bandgap for the undoped10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80(a) (e)(b)(i) (j)(f)(k)10 20 30 40 50 60 70 802θ (deg)Intensity10 20 30 40 50 60 70 802θ (deg)Intensity *(111)(220)(311) (400)(331)*(111)(220)(311) (400)(331)2.04 2.02 1.98g value2.00 2.04 2.02 1.982.00g value-2-1012(ytisnetnI103 )Intensity (103 )-2-1012g =2.006g=1.998g =1.998g value2.04 2.02 1.982.00g value2.04 2.02 1.982.00-2-1012Intensity (103 )(g)-2-1012Intensity (103 )(h)g =2.005 g =2.005(c) (d)Figure 1. Summary of physical and electronic properties of P-doped and B-doped Si QDs. a,e) HR-TEM images, b,f ) XRD patterns, d,h) ESR spectra, andi,j) UPS spectra of the P-doped and B-doped Si QDs terminated with decane and undecanoic acid monolayers, respectively. c,g) ESR spectra of hydrogen-terminated Si QDs served as standards. k) Possible energy structures of the doped Si QDs terminated with surface monolayers. The XRD peaks markedwith an asterisk are attributed to organic ligands.[54]www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (3 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseSiQDs.[41,42] In addition, we measured the PL spectra by varyingexcitation wavelengths between 300 and 400 nm for each dopedQD sample. The results are shown in Figure S3, SupportingInformation. Obviously, the PL peak position remainsunchanged with varying excitation wavelengths, indicating thatthe emission is not originating from surface trap states or otherdefect-related trap states. PL quantum yields (QYs) of bothhydrogen-terminated samples were smaller than 2% butenhanced to 41% and 21% after hydrosilylation of 1-deceneand 10-undecenoic acid, respectively.Electron spin resonance (ESR) measurements wereperformed at 4.2 K using an X-band ESR spectrometer with amagnetic field modulation of 100 kHz to discuss the possibledoping sites in the crystalline structure of Si QDs terminatedwith undecanoic acid and decane monolayers (see Figure 1d,h)while ESR spectra of the hydrogen-terminated samples wereused as standard (see Figure 1c,g). In Figure 1c, the ESR signalwas clearly observed at g= 1.998, corresponding to that of con-duction electrons in P-doped crystalline Si,[28] indicating thatphosphorus atoms are doped in substitutional sites of the crys-talline Si core. As expected, this signal remained after hydrosi-lylation of 1-decene. This was evidence of successful n-typedoping although XPS P2p spectrum did not give any peak. Onthe other hand, we see the appearance of signals at g= 2.006,possibly due to the presence of dangling bonds as a defectformed in an amorphous structure.[43,44] This is consistent witha previous study that reported that the outermost layer ofhydrogen-terminated Si QD consists of an amorphous structurewith many dangling bonds as nonradiative channels.[45]Interestingly, the defect-derived signals disappeared after hydro-silylation of 1-decene, suggesting the disappearance of amorph-ized surface Si layer. This observation is consistent with ourprevious study that has reported that the passivation with decanemonolayers suppresses surface reconstruction to preserve thediamond cubic lattice (i.e., inhibits amorphization of QD surface)in a broad range from the center toward the near surface in SiQD.[46] It can be discussed that the absence of dangling bonds asnonradiative channels provoked enhanced PLQY to 41%. On theother hand, in the case of B-doped Si QDs capped with hydrogenatoms, the conduction electron signal at g= 1.998 was notobserved. The result is quite reasonable because conduction elec-trons do not exist in p-type B-doped Si QDs. The hole-related ESRsignals in Si are difficult to observe and are not observed undernormal conditions. A defect-related signal was observed ataround g= 2.005. The shape of the signal is asymmetric, suggest-ing that there are at least two types of defects with similar struc-tures. Although the g-value is slightly different from that in thecase of P-doped Si QDs, it can be discussed that the defects aredangling bond-type defects which exist in the surface amorphouslayer of the QDs as in the case of P-doped Si QDs. It is noted thatthe signal at g= 2.005 was significantly weakened by hydrosily-lation of 10-undecenoic acid, consistent with the improved PLQYdue to a decrease of nonradiative channels. Unlike P dopant thatprefers to take the sp3 configuration, B impurities tend to bemore stable even near the Si surface.[47] Therefore, the fact thata small peak remains at g= 2.005 even after the hydrosilylationsuggests that the B dopants are responsible for supplying holesas electronic charge carriers in the outermost layer of the QDs.Energy structures of the hydrosilylated QDs were exploredthrough ultraviolet photoelectron microscopy (UPS) as depictedin Figure 1i,j. This investigation delved into the intriguing effectsof doping on these nanostructures. Using the UPS spectra, thework function, representing the energy required to remove anelectron from the material’s surface, was determined to be�3.6 eV for P-doped QDs and �4.84 eV for B-doped QDs, bothrelative to the vacuum level. The rise in photoelectron intensitynearest the Fermi energy (EF) is usually assigned as the valenceband maximum energy (EVBM). To extract the EVBM with respectto EF (EF�EVBM), the rise in photoelectron intensity is fitted to aline and extrapolated to a point that intersects a linear fit to thebaseline. The UPS measurement revealed that the VBM energyfor the B-doped QDs was ≈�5.36 eV, while for P-doped QDs, itwas ≈�5.13 eV with respect to vacuum. On the other hand, thebandgap of 1.66 and 1.75 eV for the B-doped and P-doped werecalculated from the PL peak photon energy, respectively. Thus,the conduction band minima (CBM) were calculated to be �3.7and �3.37 eV respectively for the B-doped and P-doped QDs.Fermi-level positions were indicative of semiconductorbehavior. For the B-doped QDs, it lay closer to the valenceband, confirming p-type behavior, while for the P-doped QDs,it was closer to the conduction band, indicating n-type behavior.Based on the UPS study, the positions of the valence band,conduction band, and Fermi level are illustrated in Figure 1k,offering a visual insight into the intricate energy landscapeof the p-type and n-type doped Si-QD system. Rectangularblocks have been used to show schematically the bandgapwith energy scale in the vertical axis. The top line of the blocksrepresents the VBM position whereas the bottom line of theblocks represents the CBM position. The horizontal black linedrawn between band gaps represents the Fermi level, whichseparates the occupied and unoccupied states within the QD.This visual representation offers a clear understanding of theenergy structure and the positioning of key energy levels withinthe QDs.2.2. Fabrication of p-n Homojunction PhotodiodeOne of the challenges in allowing to build p-n homojunctionthrough the solution process is how to stack p-type QD layerson top of n-type QD layers, and vice versa, without mixing. Inthis work, we used two different ligands (i.e., decane and unde-canoic acid) for n-type and p-type Si QDs which make them sol-uble in two orthogonal solvents (i.e., toluene and ethanol).Figure 2a,b are cross-sectional SEM images of devices with p-typeor n-type Si QD layers sandwiched between ITO (as a cathode)and Al (as an anode) electrodes, respectively. The QD layers werespin-coated onto the surface of glass substrates covered with thinfilm of ITO, after which an Al thin film of ≈100 nm was depos-ited. The thicknesses of the QD layers were tuned to ≈95 and≈70 nm and these devices were used as a reference. To preparethe device shown in Figure 2c, p-type QDs were spin coated onthe ITO surface, followed by n-type QDs. The thickness of eachQD layer was adjusted to half the thickness of the QD layers usedfor p-type only and n-type only devices to compare the optoelec-tronic performance between the three devices. The dark I–Vcharacteristics shown in Figure 3a confirm that for the p-typewww.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (4 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseonly and the n-type only devices, I–V curves were linear in behav-ior which are obvious for any semiconductor and that theconductivities are nearly same or comparable. In contrast, theI–V characteristics of the p-type/n-type stacked device show aclear rectification behavior, suggesting that the depletion layerformed at the interface between the p-type and n-type layers givesthe diode characteristics. Therefore, current conduction wasfavorable under forward bias, whereas current conduction wasFigure 2. Schematic illustrations and cross-sectional SEM images of a) p-type, b) n-type, and c) p-n junction photodetectors. Real cross sectional SEMimages of d) p-type, e) n-type, and f ) p-n junction photodetectors.Figure 3. a) Dark IV characteristics of only p-type, only n-type, and p-n junction device. b) Ideality factor of the p-n junction diode. c) I–V characteristics ofthree devices under light of 360 nm with optical power 0.13 mW.www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (5 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenselow under reverse bias conditions. The I–V curve was fitted withthe diode equation[48]nðVÞ ¼ q=kT ½dV=dðlnIÞ� (1)where n(V ) is ideality factor, q is charge, k is boltzman constant,and T is temperature.Figure 3b shows ideality factors plotted as a function of biasvoltage. Ideality factor increases as forward bias voltage increasesin the region where the effect of the series resistance is small andthen increases slowly with increasing forward bias wherethe effect of the series resistance comes into play in the I–V char-acteristics. This indicates the appearance of p-n junctionbehavior.[48–50] Furthermore, we studied the photoresponses ofthe three devices upon UV light irradiation (λ= 360� 5 nm,0.13mW) using a xenon lamp, and the results are shown inFigure 3c. Increasing trends of photocurrent for the p-type onlyand n-type only devices are quite similar. Due to the difference inmagnitude of work function between the cathode and the anode,a little photovoltaic property has been aroused. Open-circuit volt-age (Voc) and short circuit current (Isc) of p-type only device was0.44 V and 1.7 nA, respectively while n-type only device produced0.49 V of Voc and 1.65 nA of Isc. In the p-n homojunction device,where the rectifying behavior of the diode was evident from theI–V characteristics, the values of Voc and Isc were improved to0.65 V and 15.5 nA. Such an enhanced photovoltaic performancewould be evidence of the formation of p-n homojunction at theinterface between the B-doped and P-doped Si QD layers.Next, we explicitly analyzed the dependence of photoresponsebehavior of the p-n homojunction photodiode on lightwavelength and power density. Figure 4a shows the I–V charac-teristics with different optical power varying from 0.13 to4.74mW. As the optical power increases, the photocurrent alsorises, demonstrating a direct correlation between incident lightintensity and the generated current. This trend indicates the sen-sitivity of the photocurrent to changes in optical power, reflectingthe efficiency of the device in converting light energy intoelectrical current. The variation of ISC with the optical poweris shown in Figure 4b where initially the response varies linearlywith increasing the optical power and it saturates after 2.5mW.The photocurrent saturated at higher optical powers due to thesaturation of charge carriers within the material. At a low opticalpower, there were sufficient available charge carriers to respondto the incident light, resulting in an increase in photocurrent.However, as the optical power increased, more and more chargecarriers were excited, reaching a point where the materialbecame fully populated with carriers. Beyond this point, furtherincreases in optical power did not produce carriers, and the pho-tocurrent saturated. Figure 4c shows the I–V characteristics ateach wavelength from 300 to 500 nm when the irradiation poweris fixed at 0.16mW. Photocurrent showed distinct peaks atspecific wavelengths, corresponding to the QD’s absorption spec-trum. However, at wavelengths either side of peak wavelengthphotocurrent decreased. Figure 4d shows the variation of ISCwith varying wavelengths of incident light. The peak ISC wasobtained at 360 nm and would be associated with the fact thatdirect transition at Γ point (Γ25!Γ15) occurs at 3.4 eV(λ=≈365 nm). The magnitude of ISC decreased for more than360 nm and the device did not respond to light above 500 nm.This trend is consistent with optical absorption behavior-0.4 -0.2 0.0 0.2 0.4-150-120-90-60-30030600.13mW0.16mW0.63mW1.31mW2.02mW2.71mW3.45mW4.04mW4.74mWdark)An(tne rruCVoltage (V)(a)0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.510203040 p-n junction deviceIsc (nA)Optical Power (mW)(b)-0.4 -0.2 0.0 0.2 0.4-60-40-20020)A n(tnerruCVoltage (V)400 nm 320 nm 300 nm340 nm 360 nm 380 nm 420 nm 460nm 280 nm500 nm(c)300 350 400 450 5000481216Isc (nA)Wavelength (nm)(d)0.0 0.5 1.0 1.5 2.0-14-12-10-8-6-4-20 Off)An(esnop serotohPTime (Sec)p-n Junction(e)OnFigure 4. a) I–V characteristics with different optical power. b) Variation of Isc with different optical power. c) I–V characteristics of p-n junction devicewith different wavelength. d) Isc value with different wavelengths. e) Device responses with light pulse of p-n junction.www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (6 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseshown in Figure S3a (see Supporting Information). In the UVwavelength region, the active layer absorbs light most efficiently,leading to a higher generation of charge carriers and thus anincrease in photocurrent. With increasing the wavelength, theresponsivity decreases which is obvious because of decreasinglight-absorbance of the QDs. Whereas for lower wavelengththe UV light (λ< 300 nm) is blocked by the soda-lime glass sub-strate of the device which is supported by the absorbance profileof the ITO substrate (see Figure S4, Supporting Information).Figure of merits of the three devices have been calculated asbelow. Photoresponsivity is defined as the photocurrent gener-ated per unit power of the incident light on the effective area.The photoresponsivity was described using the followingequationRðλÞ ¼ Iph=ðPop � AÞ (2)where R is the photoresponsivity; Iph, Pop, and A are thephotocurrent, power density, and illuminated area of the device(about 4mm2), respectively.Since the shot noise from the dark current is the major con-tribution to the total noise in the photodetectors or photodiodes,the specific detectivity D* was calculated by[51,52]D� ¼ RA12.ð2eIdÞ12 (3)where R is the photoresponsivity, A is the illuminated area ofdevice, and Id is the dark current. External quantum efficiency(EQE) was calculated by using the formula[53]EQEð%Þ ¼ RðλÞ � hcλ� q(4)where R(λ) is the responsivity, q is the charge, λ is the wavelengthof the irradiated light.Responsivity, detectivity, and EQE were calculated using thevalues at zero bias voltage as summarized in Table S1(Supporting Information). It was clear that the detectivity hasbeen improved 22 times and reached to value 4.4� 1010 Joneswhereas EQE also increased 22 times for the p-n homojunctiondevice when compared to the values of p-type only and n-typeonly devices. As an advantage of using photodiode as photode-tector, we detected photoresponse, that is, photocurrent at zeroprobing voltage. Photoresponse with periodic light pulse of light360 nm with light intensity 0.16mW has been shown inFigure 4e. The p-type and n-type only devices’ responses weresignificantly low, and they contain a large amount of noise whichwas not desirable for photodetection at zero bias voltage (seeFigure S5, Supporting Information), whereas p-n homojunctiondevice exhibited a good photoresponse enough to detect the lightpulse clearly even at zero bias voltage. The advantages of photo-detection in photovoltaic mode such as 1) zero bias voltage, 2) no“dark” current, 3) linear, and 4) low noise (i.e., Johnson) makethe p-n homojunction layer a suitable candidate in precisionapplications. The response speed was calculated by using a singleon/off cycle. The response speed was measured in two parts.When the 350 nm light illumination was turned on and off,the voltage was defined as the rising time (τr) and falling time(τf ), respectively, at time intervals between 10 and 90% of theoutput of the normalized voltage. The τr and τf were determinedfrom the graph for the three devices and summarized in Table S2(see Supporting Information). The rise and fall times were 4 and11.3ms, respectively.3. ConclusionWe have synthesized B-doped (p-type) and P-doped (n-type) SiQDs and made them soluble in two orthogonal solvents: ethanoland toluene by choosing proper monomolecular ligands namelyn-decane and 10-undecanoic acid, respectively. Analysis combin-ing XPS, ESR, and UPS indicated the electronic impurity dopingin the QDs and underpin the Fermi level. Using two orthogonalsolvents for p- and n-type Si QDs, we fabricated the first p-nhomojunction diodes between ITO and aluminum electrodesby a simple spin-casting method at room temperature in ambientconditions. The p-n homojunction demonstrated typical I–Vcharacteristics which clearly showed diode-like rectificationand it was used further as photodiode to detect light efficientlyeven at zero bias voltage conditions. The p-n homojunctionmight be formed at the interface between uncapped surface Siatoms of B-doped QDs and those of P-doped QDs. Based onthe quantum confinement effect, the photodiode deviceresponded selectively only to UV light. However, the presenceof long, insulating, and hydrophobic ligands is responsible forthe poor efficiency of the device. The use of p-type and n-typeSi QDs with inorganic capping ligands would dramaticallyimprove device performance. Using p-type and n-type Si QDswith inorganic short ligands through the ligand exchangemethod can dramatically improve device performance.4. Experimental SectionReagents and Materials: Triethoxysilane (97%, TES) and 10-undecenoicacid (>98.0%, UA) were purchased from Tokyo Chemical Industry (TCI)Co., Ltd. 1-Decene (≥97.0%) was purchased from Sigma–Aldrich.Hydrofluoric acid (Ultrapur, 49%) was purchased from Kanto ChemicalCo., Inc. Boric acid, phosphoric acid, toluene (HPLC grade), chloroform,ethanol (99.5), methanol, hydrochloric acid (1.0 mM), and Zn powder werepurchased from Fujifilm Wako Pure Chemical Corp. Milli-Q water(18.2Ω cm) was supplied from the Sartorius water purification system(arium 611 UV). The reagents and chemicals were used as received, withthe exception of 10-undecenoic acid.Preparation of Boron-Doped Hydrogen Silsesquioxane: The schematicdiagram of the reaction has been shown in Scheme S1, SupportingInformation. TES (10mL, 53.7 mmol) was added to a round-bottom flaskequipped with a magnetic stirring bar in an ice bath under Ar atmosphereusing standard Schlenk techniques. In another flask, 165.85mg of boricacid (2.685mmol) was added to 20mL Mill-Q water. The measured pH ofthe solution was 5.6. The pH was adjusted to 3.0 by addition of 20 μLhydrochloric acid. This acidic mixture was added drop by drop to theTES solution over 3 min. The ice bath was removed at the same timeas the addition was completed, and the flask was allowed to stand foranother 2 h to allow the solution to come to room temperature. The result-ing xerogel was filtered and washed with water several times until the pHof the product reached 7. The white solid obtained was dried in a vacuumovernight.Preparation of Phosphorus-Doped Hydrogen Silsesquioxane: TES (10mL,53.7 mmol) was added to a round-bottom flask equipped with a magneticstirring bar in an ice bath under Ar atmosphere on the same Schlenk line.The 20mL aqueous solution of phosphoric acid (2.685mmol) was addedwww.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (7 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensedrop by drop to the TES solution over 3 min. The ice bath was removed atthe same time as the addition was completed. The flask was then allowedto stand for another 2 h to allow the solution to come to room tempera-ture. The product was filtered and washed with water several times untilthe pH of the product reached 7. The white solid obtained was dried in avacuum overnight.Preparation of Hydrogen-Terminated Si QD: The dried powder wasplaced in a quartz crucible and transferred to a high-temperature tube fur-nace with vacuum flanges. First, the inside of the quarts tube where thequartz crucible was placed was evacuated by the vacuum pump until 5 Paand then refilled with 5%-H2/95%-Ar gas. This gas purge was repeatedthree times. The crucible was heated to 1100 °C in 5%-H2/95%-Aratmosphere and kept for 2 h to yield brown solids as powder. 300mgof the brown solid was ground in an agate mortar and then subjectedto HF etching by stirring for 90min in a mixture of 8 mL ethanol with16mL HF. After stirring, the solution was centrifuged at 15 000 rpm for5min at 10 °C in ethanol, acetonitrile, and dichloromethane in this order.Preparation of a Colloidal Ink of Hydrophilic p-type Si QDs: A colloidal inkof the hydrophilic p-type Si QDs was prepared by hydrosilylation of10-undecenoic acid on the H-Si QD derived from the boron-doped silses-quioxane. Prior to the reaction, the 10-undecanoic acid was degassed for2 h at 70 °C and the pressure of 30 Pa. The dichloromethane solution ofHF-etched powder was transferred to a two-neck round bottom flask con-nected to Schlenk line containing 15mL of degassed 10-undecanoic acid.Dichloromethane was completely removed by evacuation of the solution atroom temperature using a vacuum pump. After that, the solution washeated in Ar atmosphere to the temperature of 175 °C within 3minand kept for 3 h. Removing the mantle heater, the temperature of thesolution gradually dropped to room temperature. The solution was splitinto four 30 mL centrifuge tubes and the rest of the tubes were filled withethanol. Centrifugation was done at 15 000 rpm for 5 min at 10 °C.Precipitated product was dispersed in 4mL of ethanol and hexane asan antisolvent. The centrifugation for washing the QDs was repeated atleast five times. The washed QDs were dispersed in ethanol at a concen-tration of 10mgmL�1.Preparation of a Colloidal Ink of Hydrophobic n-type Si QDs: A colloidalink of the hydrophobic n-type Si QDs was prepared by hydrosilylationof 1-deceneon the H-Si QD derived from the phosphorus-doped silses-quioxane. The dichloromethane solution of HF-etched powder was trans-ferred to a two-neck round bottom flask connected to Schlenk linecontaining 15mL of degassed 1-decene. The solution was refluxed forthe reaction time as short as 10 min as the solution became transparentbrown color as soon as the solution temperature reached the boilingpoint. By removing the mantle heater, the temperature of the solutiongradually dropped to room temperature. The unreacted 1-decene wasremoved by the evaporator. Afterward, the product was purified byhigh-performance liquid chromatography (HPLC, Japan AnalyticalIndustry, Japan), and the product was redispersed in toluene at a concen-tration of 10mgmL�1.Characterization of the Si QDs: XRD pattern was measured on aMiniFlex 600 (Rigaku Corp., Japan). Samples were measured at an angularstep of 0.02° (time per step: 1 s per step) using CuKα (λ= 1.5418 Å) radi-ation. XPS was obtained using a Thermo Scientific Theta Probe utilizingmonochromatic AlKα radiation. The samples for the XPS analysis are dropcasted on the STO substrate. The XPS spectra were calibrated to the C 1sspectral peak a rising from adventitious hydrocarbons (284.8 eV). Opticalabsorption spectra were recorded using a UV-vis spectrophotometer(JASCO V-650, Japan). PL measurement was carried out using a modulardouble grating Czerny–Turner monochromator and an iHR 320 emissionmonochromator (1200 lines per mm of gratings) coupled to a photomul-tiplier tube (PMT) on a NanoLog Horiba Jovin Yvon spectrofluorometerwith a 450W xenon arc lamp. The spectral resolution of the system isaround 0.3 nm. To avoid scattered excitation lights, a cut filter for495 nm-light was placed in front of the monochromator-PMT setup.Electric energy levels such as Fermi level and ionization energy were esti-mated from the ultraviolet photoelectron spectroscopic (UPS,ThermoFisher) spectra. ESR measurement was carried out at 4.2 K usingan X-band ESR spectrometer with a magnetic field modulation of 100 kHzto investigate the state of P donors and B donors in Si QDs. The g-valueswere calibrated by means of the signals of Mn2þ in MgO as a standard.Fabrication of p-n Junction Photodiode: A 10� 20mm2 rectanglesoda-lime glass covered with 150 nm-thick indium tin oxide (ITO) witha sheet resistance of 10–14Ω sq�1 was used as the substrate. The ITOfilm was patterned by chemical etching with Zn powder and 37% HCl intoa narrow strip about 2mm wide and 20mm long. The patterned ITO-covered substrate was sonicated with acetone followed by isopropanol,ethanol, and then Milli-Q water for 15min each to remove the remainingetchant. After drying, organic contaminants on the surface were removedby exposure to VUV light (Ushio, Japan) for 30 min under a reduced pres-sure of 103 Pa and a N2 flow. In the Ar-filled glove box, the ethanol ink ofthe p-type Si QDs was spin coated over the ITO-covered substrate at1000 rpm, followed by drying at 120 °C for 30 min. Next, the tolueneink of the n-type Si QDs was spin coated over the n-type Si QD filmon the substrate, followed by drying at 120 °C for 30min. Finally, thecathode electrode of Al was vapor deposited. The stainless mask andthe substrate were tightly adhered for patterned deposition.Characterization of the Devices: All device testing was performed at roomtemperature under ambient conditions. Data for the current density–voltage (I–V ) measurements were acquired using a Keithley 2425 sourcemeter. A 300W xenon lamp was used for illumination while irradiationlight wavelength was varied between 280� 5 and 500� 5 nm using band-pass filters. In the I–V test, multiple devices on a substrate were measuredindividually and each device was characterized under light irradiation, thenunder dark conditions (no illumination). No obvious differences werefound due to light cycling or repeated measurements on the same devicewithin a few months of the first test. Data for the responsivity measure-ment was collected on a home-built setup using illumination from thesame illumination source, modulated with optical chopper (#55-783,Edmund Optics). The anode and cathode of the device were connectedto a 1 GΩ load resistor and connected to a DS-5624A oscilloscope(Iwatsu Electric Co., Ltd) to record the modulation changes of thephotovoltage under open-circuit conditions. The frequency response ofthe photocurrent was displayed as a fast Fourier transform by the oscillo-scope. The response time was calculated using the rise and fall timesbetween 0.1 and 0.9, with the background of the waveform displayedon the oscilloscope as 0 and the maximum photocurrent as 1.Supporting InformationSupporting Information is available from the Wiley Online Library or fromthe author.AcknowledgementsThe authors thank Yuka Hara at NIMS for SEM observation. This work wassupported by ARIM of MEXT (JPMXP1224NM5178). This work was par-tially supported by WPI program, JSPS KAKENHI grant nos. 24K01462,24K01278, and 24K21720. H.Y. thanks the Research Fellow of JSPS andJSPS KAKENHI Grant-in-Aid grant no. 23KJ2166.Conflict of InterestThe authors declare no conflict of interest.Author ContributionsBatu Ghosh: Conceptualization (lead); Data curation (lead); Formal anal-ysis (lead); Funding acquisition (supporting); Investigation (lead);Methodology (lead); Project administration (supporting); Resources (sup-porting); Software (equal); Supervision (supporting); Validation (equal);Visualization (equal); Writing—original draft (lead); Writing—reviewand editing (equal). Hiroyuki Yamada: Conceptualization (supporting);www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (8 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseData curation (lead); Formal analysis (lead); Funding acquisition (equal);Investigation (equal); Methodology (equal); Project administration (sup-porting); Resources (supporting); Software (equal); Supervision (support-ing); Validation (equal); Visualization (equal); Writing—original draft(supporting); Writing—review and editing (supporting). KazuhiroNemoto: Conceptualization (supporting); Data curation (equal); Formalanalysis (equal); Funding acquisition (supporting); Investigation (equal);Methodology (supporting); Project administration (supporting);Resources (supporting); Software (supporting); Supervision (supporting);Validation (equal); Visualization (supporting); Writing—original draft(supporting); Writing—review and editing (supporting). WipakornJevasuwan: Conceptualization (supporting); Data curation (equal);Formal analysis (equal); Funding acquisition (supporting); Investigation(equal); Methodology (equal); Project administration (supporting);Resources (supporting); Software (supporting); Supervision (supporting);Validation (supporting); Visualization (supporting); Writing—originaldraft (supporting); Writing—review and editing (supporting). NaokiFukata: Conceptualization (supporting); Data curation (equal); Formalanalysis (equal); Funding acquisition (supporting); Investigation (equal);Methodology (equal); Project administration (supporting); Resources(supporting); Software (supporting); Supervision (lead); Validation(equal); Visualization (equal); Writing—original draft (lead); Writing—review and editing (equal). Hon-Tao Sun: Conceptualization (supporting);Data curation (equal); Formal analysis (equal); Funding acquisition (sup-porting); Investigation (equal); Methodology (equal); Project administra-tion (supporting); Resources (supporting); Software (supporting);Supervision (lead); Validation (equal); Visualization (equal); Writing—original draft (lead); Writing—review and editing (equal). NaotoShirahata: Conceptualization (lead); Data curation (supporting); Formalanalysis (supporting); Funding acquisition (lead); Investigation (lead);Methodology (lead); Project administration (lead); Resources (lead);Software (supporting); Supervision (lead); Validation (equal);Visualization (equal); Writing—original draft (equal); Writing—reviewand editing (lead).Data Availability StatementThe data that support the findings of this study are available from thecorresponding author upon reasonable request.Keywordselectric impurity doping, p-n homojunction, photodiodes, silicon quantumdots, solution-processed optoelectronicsReceived: July 23, 2024Revised: September 17, 2024Published online:[1] N. Ahn, C. Livache, V. Pinchetti, H. Jung, H. Jin, D. Hahm, Y. S. Park,V. I. Klimov, Nature 2023, 617, 79.[2] Y. Wang, L. Peng, J. Schreier, Y. Bi, A. Black, A. Malla, S. Goossens,G. Konstantatos, Nat. Photonics 2024, 18, 236.[3] H. Jin, C. Livache, W. D. Kim, B. T. Diroll, R. D. Schaller, V. I. Klimov,Nat. Mater. 2023, 22, 1013.[4] C. J. V. Pérez, J. M. Restrepo-Florez, J. Ye, N. T. Nguyen,A. A. Tountas, R. Song, C. Mao, A. Wang, A. Gouda, S. Corapi,S. Ji, H. MacLeod, J. Wu, A. Aspuru-Guzik, C. T. Maravelias,G. A. Ozin, Energy Environ. Sci. 2023, 16, 6155.[5] F. P. G. de Arquer, D. V. Talapin, V. I. Klimov, Y. Arakawa, M. Bayer,E. H. Sargent, Science 2021, 373, eaaz8541.[6] J. Shi, D. Yoo, F. Vidal-Codina, C. W. Baik, K. S. Cho, N. C. Nguyen,H. Utzat, J. Han, A. M. Lindenberg, V. Bulović, M. G. Bawendi,J. Peraire, S. H. Oh, K. A. Nelson, Nat. Nanotechnol. 2022, 17, 1288.[7] J. H. Song, H. Choi, H. T. Pham, S. Jeong, Nat. Commun. 2018, 9,4267.[8] P. R. Brown, D. Kim, R. R. Lunt, N. Zhao, M. G. Bawendi,J. C. Grossman, V. Bulović, ACS Nano 2014, 8, 5863.[9] J. Mondala, R. Lambab, Y. Yuktac, R. Yadavd, R. Kumarce, B. Pani,B. Singh, J. Mater. Chem. C 2024, 12, 10330.[10] L. Liu, S. Z. Bisri, Y. Ishida, D. Hashizume, T. Aida, Y. Iwasa, ACSAppl. Nano Mater. 2018, 1, 5217.[11] Y. Cho, B. Hou, J. Lim, S. Lee, S. Pak, J. Hong, P. Giraud, A. Jang,Y. Lee, J. Lee, J. E. Jang, H. J. Snaith, S. M. Morris, J. I. Sohn,ACS Energy Lett. 2018, 3, 1036.[12] D. Zhitomirsky, M. Furukawa, J. Tang, P. Stadler, S. Hoogland,O. Voznyy, H. Liu, E. H. Sargent, Adv. Mater. 2012, 24, 6181.[13] A. Sahu, M. S. Kang, A. Kompch, C. Notthoff, A. W. Wills,D. Deng, M. Winterer, C. D. Frisbie, D. J. Norris, Nano Lett. 2012,12, 2587.[14] V. M. Goossens, N. V. Sukharevska, D. N. Dirin, M. K. Kovalenko,M. A. Loi, Cell Rep. Phys. Sci. 2021, 2, 100655.[15] A. Stavrinadis, A. K. Rath, F. P. G. de Arquer, S. L. Diedenhofen,C. Magén, L. Martinez, D. So, G. Konstantatos, Nat. Commun.2013, 4, 2981.[16] W. Liu, J. S. Lee, D. V. Talapin, J. Am. Chem. Soc. 2013, 135, 1349.[17] L. Asor, J. Liu, S. Xiang, N. Tessler, A. I. Frenkel, U. Banin, Adv. Mater.2023, 35, 2208332.[18] M. Dutta, L. Thirugnanam, Van P. Trinh, P. N. Fukata, ACS Nano2015, 9, 6891.[19] J. He, C. Huang, H. Wen, X. Wang, Y. He, Y. Dan, Nanotechnology2020, 31, 195201.[20] M. Fujii, S. Hayashi, K. Yamamoto, J. Appl. Phys. 1998, 83, 7953.[21] M. Frégnaux, R. Khelifi, Y. L. Gall, D. Muller, D. Mathiot, Phys. StatusSolidi C 2015, 12, 80.[22] I. Marri, E. Degoli, S. Ossicini, Prog. Surf. Sci. 2017, 92, 375.[23] A. Sarkar, R. Bar, S. Singh, R. K. Chowdhury, S. Bhattacharya,A. K. Das, S. K. Ray, Appl. Phys. Lett. 2020, 116, 231105.[24] D. Li, J. Chen, Z. Xue, T. Sun, J. Han, W. Chen, E. Talbot, R. Demoulin,W. Li, J. Xu, K. Chen, Appl. Surf. Sci. 2023, 609, 155260.[25] P. J. Wu, Y. C. Wang, I.-C. Chen, Nanoscale Res. Lett. 2013, 8, 457.[26] T. Sun, D. Li, J. Chen, J. Han, W. Li, J. Xu, K. Chen, Vacuum 2023, 207,111657.[27] P. R. Huang, Y. C. Chen, K. Y. Kuo, P. T. Lee, Sol. Energy Mater. Sol.Cells 2019, 193, 287.[28] U. R. Kortshagen, R. M. Sankaran, R. N. Pereira, S. L. Girshick,J. J. Wu, E. S. Aydil, Chem. Rev. 2016, 116, 11061.[29] A. P. Y. Wong, W. Sun, C. Qian, F. M. Ali, J. Jia, Z. Zheng, Y. Dong,G. A. Ozin, Adv. Sustainable Syst. 2017, 1, 1700118.[30] S. Chandra, Y. Masuda, N. Shirahata, F. M. Winnik, Angew. Chem.,Int. Ed. 2017, 56, 6157.[31] Z. Ni, X. Pi, S. Zhou, T. Nozaki, B. Grandidier, D. Yang, Adv. Opt.Mater. 2016, 4, 700.[32] S. Kano, Y. Tada, S. Matsuda, M. Fujii, ACS Appl. Mater. Interfaces2018, 10, 20672.[33] T. Velusamy, S. Mitra, M. Macias-Montero, V. Svrcek, D. Mariotti,ACS Appl. Mater. Interfaces 2015, 7, 28207.[34] Y. Hori, S. Kano, H. Sugimoto, K. Imakita, M. Fujii, Nano Lett. 2016,16, 2615.[35] B. Ghosh, Y. Masuda, Y. Wakayama, Y. Imanaka, J. Inoue, K. Hashi,K. Deguchi, H. Yamada, Y. Sakka, S. Ohki, T. Shimizu, N. Shirahata,Adv. Funct. Mater. 2014, 24, 7151.[36] M. Dasog, J. Kehrle, B. Rieger, J. G. C. Veinot, Angew. Chem., Int. Ed.2015, 54, 2.www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (9 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License[37] H. Yamada, N. Saitoh, B. Ghosh, Y. Masuda, N. Yoshizawa,N. Shirahata, J. Phys. Chem. C 2020, 124, 23333.[38] N. Shirahata, Sci. Rep. 2022, 12, 17211.[39] S. Chatterjee, K. Nemoto, H. T. Sun, N. Shirahata, Nanoscale Horiz.2024, 9, 817.[40] M. Fujii, H. Sugimoto, M. Hasegawa, K. Imakita, J. Appl. Phys. 2014,115, 084301.[41] L. Canham, Faraday Discuss. 2020, 222, 10.[42] N. Shirahata, J. Nakamura, J. Inoue, B. Ghosh, K. Nemoto,Y. Nemoto, M. Takeguchi, Y. Masuda, M. Tanaka, G. A. Ozin,Nano Lett. 2020, 20, 1491.[43] R. P. Wang, Appl. Phys. Lett. 2006, 88, 142104.[44] N. Fukata, J. Chen, T. Sekiguchi, S. Matsushita, T. Oshima, N. Uchida,K. Murakami, T. Tsurui, S. Ito, Appl. Phys. Lett. 2007, 90, 153117.[45] B. Ghosh, M. Takeguchi, J. Nakamura, Y. Nemoto, T. Hamaoka,S. Chandra, N. Shirahata, Sci. Rep. 2016, 6, 36951.[46] B. Ghosh, T. Hamaoka, Y. Nemoto, M. Takeguchi, N. Shirahata,J. Phys. Chem. C 2018, 122, 6422.[47] C. R. Leao, A. Fazzio, A. J. R. da Silva, Nano Lett. 2008, 8, 1866.[48] S. Agarwal, M. Seetharaman, N. K. Kumawat, A. S. Subbiah,S. K. Sarkar, D. Kabra, M. A. G. Namboothiry, P. R. Nair, J. Phys.Chem. Lett. 2014, 5, 4115.[49] G. Gomila, J. M. Rubi, J. Appl. Phys. 1997, 81, 2674.[50] H. Kim, C. Y. Jung, S. H. Kim, Y. Cho, D.-W. Kim, Curr. Appl. Phys.2016, 16, 37.[51] W. Choi, M. Y. Cho, A. Konar, J. H. Lee, G. Cha, S. C. Hong, S. Kim,J. Kim, D. Jena, J. Joo, Adv. Mater. 2012, 24, 5832.[52] H. Wu, Z. Kang, Z. Zhang, Z. Zhang, H. Si, Q. Liao, S. Zhang, J. Wu,X. Zhang, Y. Zhang, Adv. Funct. Mater. 2018, 28, 1802015.[53] J. M. Wu, W. E. Chang, ACS Appl. Mater. Interfaces 2014, 6, 14286.[54] J. J. Calvin, T. M. Kaufman, A. B. Sedlak, M. F. Crook, A. P. Alivisatos,Nat. Commun. 2021, 12, 2663.www.advancedsciencenews.com www.small-science-journal.comSmall Sci. 2024, 2400367 2400367 (10 of 10) © 2024 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202400367 by National Institute For, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License