# Fileset

[overestimation-of-operational-stability-in-polymer-based-organic-field-effect-transistors-caused-by-contact-resistance.pdf](https://mdr.nims.go.jp/filesets/ab2618e2-cc81-4cab-9035-6bc1bdf4ad40/download)

## Creator

[Kenji Sakamoto](https://orcid.org/0000-0002-1379-874X), [Takeshi Yasuda](https://orcid.org/0000-0003-4652-9105), [Takeo Minari](https://orcid.org/0000-0001-7690-221X), [Masafumi Yoshio](https://orcid.org/0000-0002-1442-4352), [Junpei Kuwabara](https://orcid.org/0000-0002-9032-5655), [Masayuki Takeuchi](https://orcid.org/0000-0002-0207-0665)

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Overestimation of Operational Stability in Polymer-Based Organic Field-Effect Transistors Caused by Contact Resistance](https://mdr.nims.go.jp/datasets/edc38086-43fb-4674-adfd-f5f9ba003c9a)

## Fulltext

Overestimation of Operational Stability in Polymer-Based Organic Field-Effect Transistors Caused by Contact ResistanceOverestimation of Operational Stability in Polymer-Based OrganicField-Effect Transistors Caused by Contact ResistanceKenji Sakamoto,* Takeshi Yasuda, Takeo Minari, Masafumi Yoshio, Junpei Kuwabara,and Masayuki TakeuchiCite This: ACS Appl. Mater. Interfaces 2024, 16, 68081−68090 Read OnlineACCESS Metrics & More Article Recommendations *sı Supporting InformationABSTRACT: The bias-stress effects of bottom-gate top-contactpolymer-based organic field-effect transistors (OFETs) with differentchannel lengths (50−500 μm) were evaluated by repeating cycles ofprolonged on-state gate-bias application and transfer characteristicsmeasurements in the linear regime. The thicknesses of poly-(didodecylquaterthiophene-alt-didodecylbithiazole) active layerswere 26 and 37 nm. All OFETs exhibited nonlinear (nonideal)transfer characteristics with a maximum transconductance within thegate-source voltage sweep range. Both a shift in threshold voltage(Vthlin) and a reduction in field-effect charge carrier mobility (μlin)were apparently observed during the bias-stress application. When μlinand Vthlin were conventionally extracted from the transfer characteristics around the maximum transconductance, the Vthlin shiftamount and μlin reduction depended on the channel length and were smaller in OFETs with short channels. After contact resistance(Rc) correction, the channel length dependence disappeared. Thus, the operational stability in OFETs with short channels: ≤50(150) μm for the 26 (37) nm-thick active layers, was found to be overestimated without Rc correction. This erroneous evaluationwould become more pronounced in short-channel, high-mobility OFETs, because the Rc becomes larger relative to the channelresistance with increasing μlin and decreasing channel length. These results suggest that one should pay attention to Rc in thefundamental research into the origin of operational instability and in evaluating the effects of active layers, gate dielectrics, and activelayer/gate dielectric interfaces on operational stability.KEYWORDS: polymer-based organic field-effect transistors, bias-stress effects, contact resistance, modified transmission line method,operational stability1. INTRODUCTIONPolymer-based organic field-effect transistors (OFETs) arepromising active devices in large-area, low-cost, lightweight,flexible, and stretchable electronics owing to good solutionprocessability and superior mechanical properties of organicsemiconducting polymers. Since the cutoff frequency and on-state current of OFETs are proportional to field-effect chargecarrier mobility, its improvement is a central issue in OFETs.As a result of enormous efforts since 1986,1 the field-effectmobility of polymer-based OFETs has been significantlyimproved,2 and is now over 10 cm2 V−1 s−1.3−5 (The field-effect mobility of highly oriented poly[4-(4,4-dihexadecyl-4H-cyclopenta[1,2-b:5,4-b’]-dithiophen-2-yl)-alt-[1,2,5]-thiadiazolo[3,4-c]pyridine] (PCDTPT) active layers wasoverestimated to be >20 cm2 V−1 s−1 in ref 4. It was re-estimated to be approximately 10 cm2 V−1 s−1 afterward.6,7)This significant improvement is due to material synthesis basedon a donor−acceptor (D−A) copolymer design concept inwhich electron-donating and electron-accepting moietiesalternate along the backbone structure and development ofalignment processes to obtain highly oriented active layers.Commercial applications of OFETs require long-termoperational stability in addition to sufficiently high field-effectmobilities to meet initial specifications. The undesirablechange in the electrical characteristics under continuousoperation: that is, bias-stress effect, must be minimized. Ingeneral, prolonged bias-stress application causes deviceperformance degradation, such as a shift in the thresholdvoltage, a reduction in the field-effect mobility, an increase inthe subthreshold swing, an increase in the off-state current,and/or increased hysteresis in the transfer characteristics.8,9 Inmany cases, the operational stability is monitored by thefollowing two methods: (I) repeating cycles of prolonged gate-bias application and transfer characteristics measurement in thelinear regime and (II) measuring drain current in the linearReceived: September 12, 2024Revised: November 13, 2024Accepted: November 18, 2024Published: December 1, 2024Research Articlewww.acsami.org© 2024 The Authors. Published byAmerican Chemical Society68081https://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−68090This article is licensed under CC-BY-NC-ND 4.0Downloaded via NATL INST FOR MATLS SCIENCE (NIMS) on December 4, 2025 at 02:02:41 (UTC).See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.https://pubs.acs.org/action/doSearch?field1=Contrib&text1="Kenji+Sakamoto"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Takeshi+Yasuda"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Takeo+Minari"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Masafumi+Yoshio"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Junpei+Kuwabara"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Masayuki+Takeuchi"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Masayuki+Takeuchi"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/showCitFormats?doi=10.1021/acsami.4c15666&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?goto=articleMetrics&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?goto=recommendations&?ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?goto=supporting-info&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=tgr1&ref=pdfhttps://pubs.acs.org/toc/aamick/16/49?ref=pdfhttps://pubs.acs.org/toc/aamick/16/49?ref=pdfhttps://pubs.acs.org/toc/aamick/16/49?ref=pdfhttps://pubs.acs.org/toc/aamick/16/49?ref=pdfwww.acsami.org?ref=pdfhttps://pubs.acs.org?ref=pdfhttps://pubs.acs.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://www.acsami.org?ref=pdfhttps://www.acsami.org?ref=pdfhttps://acsopenscience.org/researchers/open-access/https://creativecommons.org/licenses/by-nc-nd/4.0/https://creativecommons.org/licenses/by-nc-nd/4.0/https://creativecommons.org/licenses/by-nc-nd/4.0/https://creativecommons.org/licenses/by-nc-nd/4.0/https://creativecommons.org/licenses/by-nc-nd/4.0/regime as a function of time under a constant continuous on-state gate-bias application.10 Then, to quantify the observedbias-stress effects, the bias-stress time dependence of thethreshold voltage shift (ΔVthlin(t) = Vthlin(t) − Vth0lin) or theon-state drain current (Id(t)) is fitted with the followingstretched exponential functions:= ·{ [ ]}V t V V t( ) ( ) 1 exp ( )thlings th0lin(1a)and= · [ ]I t I t( ) exp ( )d d0 (1b)where Vgs is the gate-source voltage; Vth0lin and Id0 are Vthlin(0)and Id(0), respectively; τ is the trapping time constant ofcarriers to fill trap levels; and β (0 ≤ β ≤ 1) is a stretchingfactor representing the distribution of time constants.10,11 τand β are used to compare the operational stability amongdifferent OFETs. Method I has the advantage of determiningnot only the shift in threshold voltage but also the changes infield-effect mobility, subthreshold swing, off-state current, andtransfer curve hysteresis. Thus, the threshold voltage shift andmobility degradation can be discussed separately. However,parts of trapped charges are released during the transfercharacteristics measurements, so the time dependence of bias-stress effects is more or less affected. In contrast, in Method II,no release of trapped charges occurs, but the separatediscussion of the threshold voltage shift and mobilitydegradation is not possible. This is because the on-statecurrent degradation is caused by both a shift in the thresholdvoltage and a reduction in the field-effect mobility. Only if thefield-effect mobility is independent of carrier concentration(gate-source voltage Vgs) and unchanged during prolongedgate-bias application, the bias-stress time dependence of theon-state Id can be converted into that of ΔVthlin, and the τ andβ determined by Method II can be compared with thosedetermined by Method I.Most polymer-based OFETs exhibit carrier-concentration-dependent (Vgs-dependent) field-effect mobility due toconformational and energitic disorder in the polymeric activelayers.12−15 Thus, even if the field-effect mobility is unchangedduring bias-stress application, Method II cannot be used todetermine ΔVthlin as a function of bias-stress time. Although itremains a useful method for evaluating the operational stabilityof OFETs for current-driven device applications, Method I isbelieved to be suitable for fundamental research into the originof operational instability and for evaluating the effects of activelayers, gate dielectrics, and active layer/gate dielectricinterfaces on operational stability.In Method I, the field-effect mobility (μlin) and thresholdvoltage (Vthlin) are extracted from the transfer characteristicsmeasured in the linear regime using the following equation:=IWCLV V VV2idlinds gs thlin dsikjjj y{zzz (2)where Vds is the drain-source voltage; L and W are the channellength and width, respectively; and Ci is the gate dielectriccapacitance per unit area. When the μlin is carrier-concentration-dependent, the transfer curve is not a straightline (not ideal) in the on-state Vgs region. Thus, for polymer-based OFETs, great care must be taken in applying eq 2. If thecarrier-concentration-dependent μlin does not change duringbias-stress application; that is, the Vgs-dependent mobilityshifts by ΔVthlin with no change in magnitude and shape in theon-state Vgs region, ΔVthlin can be easily and accuratelydetermined by shifting the initial transfer curve so that itoverlaps with the transfer curve after bias-stressing or byapplying eq 2 to the portion of the transfer curve around acertain Id in the on-state region. In this case, the obtainedΔVthlin is not affected by whether Rc correction is performed ornot, because the relative ratio of the total contact resistance Rcat source and drain electrodes to the channel resistance Rchdoes not change during bias-stress application. In contrast,when the carrier-concentration-dependent μlin changes inshape and/or in magnitude during bias-stress application,one should pay attention to Rc. For OFETs where Rc is notnegligibly small against Rch, the operational stability may beoverestimated or underestimated without Rc correction, whichis the subject of this paper.In this study, we have investigated the effect of Rc on theevaluation of operational stability of OFETs exhibiting both ashift in Vthlin and a reduction in μlin. Bottom-gate top-contact(BG-TC) OFETs with different L’s from 50 to 500 μm werefabricated on a single device substrate using poly-(didodecylquaterthiophene-alt -didodecylbithiazole)(PQTBTz-C12) as the active layer material. PQTBTz-C12 is aD−A copolymer exhibiting a liquid crystalline (LC) phase atan elevated temperature.16 The bias-stress effects wereevaluated by Method I. The Vgs-dependent Rc at differentbias-stress times was determined using a modified trans-mission-line method (TLM), which can extract Rc with muchgreater accuracy compared to the conventional TLM.17,18 Afterthe transfer characteristics were corrected using the extractedRc, the Rc-corrected μlin and Vthlin were obtained. Since allPQTBTz OFETs exhibited nonlinear transfer characteristicswith a maximum transconductance within the Vgs sweep rangeregardless of Rc correction, here the Vthlin was conventionallyevaluated from the portion of transfer curves around Vgs atwhich μlin becomes a maximum. As a result, we found thatwithout Rc correction, the operational stability was over-estimated for OFETs with short channels. This erroneousevaluation would become more pronounced in short-channel,high-mobility OFETs, because Rc becomes larger relative toRch with increasing μlin and decreasing L. As Rc is included inthe characteristics of OFETs, Rc correction may not benecessary when evaluating the operational stability of OFETsthemselves. However, our results suggest that Rc correctionshould be performed in the fundamental research into theorigin of operational instability and in evaluating the effects ofactive layers, gate dielectrics, and active layer/gate dielectricinterfaces on operational stability. Finally, the operationalstability of PQTBTz OFETs was compared to that of poly(2,5-bis(3-hexadecylthiophen-2-yl)thieno[3,2-b]thiophene)(PBTTT-C16) and PCDTPT OFETs evaluated in ourprevious work.19,202. EXPERIMENTAL SECTION2.1. Materials. The synthesis of PQTBTz-C12 was outsourced toTCI Co. Ltd. and was carried out through the same route described inref 16. However, the details of reaction conditions were modified, andthe reaction using H2S gas and dimethylamine in ref 16 was replacedwith a reaction using NaSH and MgCl2, based on ref 21. The number-average molecular weight (polystyrene standard) and polydispersity ofPQTBTz-C12 used in this study were 24 kg mol−1 and 2.9,respectively. The differential molecular weight distribution is shown inFigure S1. It was confirmed by differential scanning calorimetryACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068082https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-as(DSC) that the polymer exhibited an LC phase between 145 and 193°C during the heating scan. The DSC curve is presented in Figure S2.In the fabrication of OFETs, anhydrous chlorobenzene (CB)purchased from Sigma-Aldrich, electronics grade acetone and 2-propanol purchased from Kishida Chemical, toluene, dehydratedtoluene, and electronics grade sulfuric acid and hydrogen peroxidepurchased from Kanto Chemical, and octadecyltrichlorosilane(ODTS) provided by Acros Organics were used as received.2.2. Fabrication of PQTBTz OFET Arrays. PQTBTz OFETarrays were fabricated on heavily doped n-type Si(100) substrates (20× 20 mm2) with a thermally grown SiO2 layer (≈100 nm thick). Thesubstrates were cleaned sequentially with acetone and a piranhasolution (a mixture of sulfuric acid and hydrogen peroxide) andimmediately treated with ODTS vapor at 120 °C for 3 h. The detailedprocedures of the cleaning and ODTS treatment were describedelsewhere.20,22 (In this study, the nanogroove formation in ref 20 andthe rubbing treatment before spin-coating in ref 22 were skipped.)The ODTS treatment was performed to minimize charge trapping atthe active layer/SiO2 gate dielectric interface. The resultant ODTS-treated surfaces showed water contact angles in excess of 110°. Spin-coating such highly hydrophobic gate dielectric surfaces withsemiconducting polymer films was difficult due to the repellentnature against organic solvents (lyophobicity). This difficulty wassolved by forming an appropriate hydrophobic−hydrophilic (lyopho-bic-lyophilic) pattern on the highly hydrophobic gate dielectricsurface, and OFETs were fabricated in the hydrophobicareas.19,20,22−24 For solution of PQTBTz-C12 in CB, a simple patternof a 15 mm-square hydrophobic region with a 2.5 mm widehydrophilic outer edge24 was sufficient. This hydrophobic−hydro-philic surface pattern was produced by exposing the ODTS-treatedsurfaces to vacuum ultraviolet (VUV) light (wavelength 172 nm)through a photomask, as described elsewhere.22 The VUV-light-exposed areas became hydrophilic (lyophilic). Then, 26 (37) nm-thick active layers were deposited on the substrates by spin-coatingwith a hot 0.40 (0.56) wt % solution of PQTBTz-C12 in CB at roomtemperature (RT) in air. The solution temperature was 60 °C. Therotation speed and duration of spin-coating were 1000 rpm and 180 s,respectively. Then, the spin-coated films were annealed at 180 °C for1 h in a nitrogen atmosphere.16 The crystallinity enhancement ofperfectly oriented crystalline lamella with a d-spacing of 2.1 nm (edge-on molecular orientation) during annealing at 180 °C was confirmedby the out-of-plane and rocking scan X-ray diffraction (XRD)measurements.16 The XRD profiles are shown in Figure S3. Themonolayer molecular steps were confirmed in the AFM images shownin Figure S4. The PQTBTz-C12 thickness was estimated with a stylustype step profiler (Kosaka ET200).The n+-Si substrate and ODTS-treated SiO2 layer served as acommon gate electrode and gate dielectrics of OFETs, respectively.To complete BG-TC OFET arrays, source and drain (S/D)electrodes were formed on the annealed PQTBTz-C12 films bysequential thermal evaporation of MoO3 (25 nm thick) and Au (40nm thick) through a shadow mask in vacuum (base pressure <3 ×10−5 Pa). The thicknesses of the S/D electrodes were the reading of athickness monitor, which was precalibrated to match the filmthickness measured by a stylus type step profiler. Two differentshadow masks were utilized. One was for producing 12 × 5 arrays ofOFETs with L/W of 50/300 μm shown in Figure 1a. The channeldirections of neighboring OFETs were orthogonal to each other. Thisarray pattern was used to evaluate the spatial uniformity of the spin-coated PQTBTz-C12 film in terms of OFET properties. The otherwas for producing sets of OFETs with different L’s from 50 to 500 μmin 50 μm intervals and a constant W of 500 μm shown in Figure 1b.Bias-stress measurements were performed on the sets of OFETs, andthe Vgs-dependent Rc at different bias-stress times were extracted fromtheir transfer characteristics. The regions enclosed by dotted lines inFigure 1 are the hydrophobic SiO2 surface areas.2.3. Electrical Characterization. The electrical characteristics ofOFETs were measured using a vacuum probe station and asemiconductor parameter analyzer system.19 Each OFET waselectrically isolated by removing the surrounding PQTBTz-C12 filmwith a tungsten needle in air. Then, an OFET substrate was set in thevacuum probe station. To remove residual oxygen and moisture in theactive layer, the array was annealed at 180 °C for 15 min in vacuum.First, the initial output and transfer characteristics of all OFETs weremeasured in the saturation regime. Then, the on-state bias-stresseffects were measured by Method I for selected sets of OFETs withdifferent L’s and a constant W of 500 μm. The electrical measurementconditions will be described with the experimental data. All the aboveelectrical measurements were performed at RT under vacuumconditions less than 10−4 Pa in the dark. The gate dielectriccapacitance was measured with an LCR meter (HIOKI 3522−50).The Ci for all OFETs reported in this paper was 31.9 nF cm−2.2.4. Other Characterization. The molecular weight and thermalbehavior of PQTBTz-C12 were examined with a gel permeationchromatography system (Shimadzu Nexera 40) and a differentialscanning calorimeter (Shimadzu DSC-60), respectively. The contactangle measurements and optical microscope observations wereperformed using a Kyowa DM 500 contact angle meter and anOlympus BX51 optical microscope, respectively. The AFM imageswere acquired with a system composed of Hitachi High-TechAFM5100N and AFM5000II. The out-of-plane and rocking scanXRD measurements were performed using an RIGAKU SmartLab X-ray diffractometer.3. RESULTS AND DISCUSSION3.1. Spatial Uniformity of PQTBTz Films from aViewpoint of OFET Characteristics. The spatial uniformityof PQTBTz-C12 films formed on ODTS-treated SiO2/Sisubstrates is crucial for extraction of reliable Vgs-dependent Rcby a modified TLM, because a set of OFETs with differentlengths of channels with nearly identical electrical propertiesshould be prepared. Thus, first the spatial uniformity wasconfirmed from the viewpoint of OFET properties. For thispurpose, a 26 nm-thick PQTBTz OFET array with the S/Delectrode pattern (L/W = 50/300 μm) shown in Figure1a wasfabricated, and the initial electrical properties of all OFETswere measured in the saturation regime. The typical outputand transfer characteristics are shown in Figure 2. The output(transfer) characteristics were acquired by negatively increas-ing Vds (Vgs) down to −30 V as a forward sweep andimmediately executing the reverse voltage sweep. In the outputFigure 1. Optical microscope images of OFET arrays with differentS/D electrode patterns. (a) 12 × 5 array of OFETs with L/W of 50/300 μm. (b) Array of OFETs with different L’s and a constant W of500 μm. The regions enclosed by the dotted lines are the hydrophobicSiO2 surface areas.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068083https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig1&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig1&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig1&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig1&ref=pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ascharacteristic measurement, the Vgs was varied from 0 to −30V in increments of −5 V, and in the transfer characteristicmeasurement, a constant Vds of −30 V was applied. All OFETsshowed good p-channel transistor behavior with almost nohysteresis in Id between the forward and reverse sweeps andmaximum current on/off ratios greater than 106. In the outputcharacteristics, the nonlinear increase of Id was observed in thelow Vds range (linear regime), which indicates that Rc is notnegligible relative to Rch for OFETs with L = 50 μm.The field-effect hole mobility μ, threshold voltage Vth, andsubthreshold swing SS were evaluated from the forward sweeptransfer curves. The μ and Vth were extracted using thefollowing equation describing Id in the saturation regime:| | =I WLC V V2( )d i gs th2(3)The SS was determined from the partial transfer curves (notshown) remeasured in intervals of 0.05 V over the turn-on Vgsrange. The histograms of μ, Vth, and SS are shown in Figures 3,S5a, and S5b, respectively. The average values (Av) andstandard deviations (σ) of μ, Vth, and SS were 0.159 ± 0.059cm2 V−1 s−1, −4.39 ± 0.23 V, and 0.265 ± 0.033 V/decade,respectively. The relative standard deviation (σ/Av) of μ wasless than 4%. In addition to μ, the device-to-device variationsof Vth and SS were also found to be very small. From these verysmall device-to-device variations, it was confirmed that theuniformity of the PQTBTz-C12 films formed in this study wasvery high. To visually indicate the very small device-to-devicevariations, that is, the high spatial uniformity, the transfercurves of all 60 OFETs are plotted in Figure S6. For ease ofviewing, only the transfer curves in the forward sweeps wereplotted. One can see their excellent overlapping.3.2. Bias-Stress Effects of OFETs with DifferentChannel Lengths. The contact resistance Rc of BG-TCOFETs consists of the metal−semiconductor interfaceresistance Ri and the access resistance Ra from the metal−semiconductor interface to the conduction channel: that is, Rc= Ri + Ra. Since Ra increases with increasing active layerthickness, OFETs with different Rc can be prepared by varyingthe active layer thickness. To clearly demonstrate the influenceof Rc on the bias-stress effect evaluation, OFET arrays with 26and 37 nm-thick PQTBTz active layers and the S/D electrodepattern shown in Figure1b were fabricated. The bias-stresseffect measurements were performed on each set of OFETswith different L’s ranging from 50 to 500 μm at 50 μmintervals. As already stated, the bias-stress effects wereevaluated by Method I. The bias-stress condition andmeasurement sequence were the same as in our previouswork,19,20 except for the Vgs sweep range and interval in thetransfer curve measurements. In the present study, the transfercurves in the linear regime were acquired by sweeping Vgsbetween +2 and −30 V with 0.1 V intervals in the forward andreverse directions sequentially under Vds = −1 V. The specificmeasurement sequence was as follows. A constant prolongedon-state bias-stress (Vgs = −30 V and Vds = −1 V) was applied1 h after completing the initial transfer curve measurement(bias-stress time of 0 s). At bias-stress times of 5 × 102, 5 ×103, and 2.5 × 104 s, the on-state bias-stress application wasinterrupted and immediately the transfer characteristics weremeasured, after which the on-state bias-stress was immediatelyreapplied, except for a bias-stress time of 2.5 × 104 s.The transfer curves of 26 nm-thick PQTBTz OFETs with L= 50 and 500 μm recorded at different bias-stress times areshown by the solid and broken curves, respectively, in Figure4a. Both OFETs exhibited nonlinear (nonideal) transfercharacteristics with a maximum transconductance (gm = dId/dVgs) within the Vgs sweep range. The Vgs-dependent μlin (=−gm × L/WCiVds) calculated from the nonlinear transfercharacteristics is shown in Figure 4b. For both OFETs,obvious bias-stress effects were observed in both Vthlin and μlin;that is, the transfer curves shifted in the negative voltagedirection with increasing bias-stress time and the maximum μlingradually decreased. Here, note that the vertical axis of Figure4a is normalized by L and W: that is, −Id × L/W. On this scale,the transfer curves in the linear regime of OFETs with differentL’s and W’s should overlap each other when Rc is negligiblysmall relative to Rch, as seen from eq 2. However, |Id| × L/W atVgs = −30 V for the OFET with L = 50 μm was reduced byroughly 45% compared to that for the OFET with L = 500 μmdue to the non-negligible Rc. This result suggests that at Vgs =−30 V, Rc is approximately equal to Rch of the OFET with L =50 μm. Since Rch increases proportionally to L, in the OFETwith L = 500 μm, Rc is one-tenth of Rch, and the effect of Rc issignificantly reduced. Thus, the transfer characteristics and Vgs-Figure 2. Typical (a) output and (b) transfer characteristics in thesaturation regime of OFETs with L/W = 50/300 μm. The solid anddotted curves show the data for forward and reverse sweeps,respectively.Figure 3. Histogram analysis of initial μ in the saturation regime ofPQTBTz OFET arrays. The bin widths are 0.005 cm2 V−1 s−1.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068084https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig3&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig3&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig3&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig3&ref=pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asdependent μlin of the OFET with L = 500 μm can beconsidered close to the intrinsic ones without the effect of Rc.Thus, it is seen from Figure 4b that the effect of Rc not onlyreduces the magnitude of μlin but also significantly changes theshape of the Vgs dependence. The Vgs at which μlin is maximumshifts in the positive direction in the OFET with L = 50 μm.For OFETs exhibiting nonideal transfer characteristics, the μlinand Vthlin are conventionally extracted from the transfercharacteristics around the maximum transconductance (mobi-lity). The shape distortion in the Vgs dependence of μlin due tothe non-negligible Rc can lead to an erroneous evaluation ofthe operational stability. Figure 4c shows the bias-stress timedependence of normalized ΔVthlin and μlin, where the Vthlin andμlin were extracted by the conventional way described above;and ΔVthlin and μlin were normalized to the initial effective bias-stress voltage (Vgs − Vth0lin) and the initial mobility μ0lin,respectively. The μ0lin and Vth0lin, respectively, of 26 nm-thickPQTBTz OFETs were 0.09 cm2 V−1 s−1 and −2.77 V for L =50 μm, and 0.16 cm2 V−1 s−1 and −4.07 V for L = 500 μm. InFigure 4c, the OFET with L = 50 μm appears to be slightlystable against the prolonged bias-stress application comparedto the OFET with L = 500 μm.The corresponding data for the 37 nm-thick PQTBTzOFETs are shown in Figure 4d−f. The μ0lin and Vth0lin,respectively, of 37 nm-thick PQTBTz OFETs were 0.08 cm2V−1 s−1 and −2.82 V for L = 50 μm, and 0.16 cm2 V−1 s−1 and−4.00 V for L = 500 μm. As expected, the influence of Rc onthe transfer characteristics and μlin can be seen morepronounced. The |Id| × L/W at Vgs = −30 V was reduced byroughly half by decreasing L from 500 to 50 μm, as shown inFigure 4d. Comparing Figure 4b,e, it is seen that the reductionof μlin in decreasing L from 500 to 50 μm is larger than that forthe 26 nm-thick PQTBTz OFETs. In Figure 4f, the OFETwith L = 50 μm appears to be significantly more stable than theOFET with L = 500 μm. In the next section, it is demonstratedthat these apparent channel length dependences in the bias-stress effects disappear after performing Rc correction on thetransfer characteristics.3.3. Contact Resistance Correction. The Vgs-dependentRc at different bias-stress times was extracted in 0.1 V intervalsby a modified TLM from a set of the forward-sweep transfercurves in the linear regime measured for the OFETs withdifferent L’s from 50 to 500 μm in 50 μm intervals. Thechannels of OFETs operating in the linear regime can beconsidered as an approximately uniform resistance, which isgiven by Rch = L/WμlinCi(Vgs − Vthlin). Thus, the totalresistance Rtotal of OFETs is expressed by Rtotal = Rch + Rc.These resistances are usually normalized by W to be universalfor OFETs with different W’s, as follows:= +R W LC V VR W( )total lini gs thlin c(4)This is the equation used in the conventional TLM. RcW canbe obtained from the intercept to the y-axis (L = 0) in a plot ofRtotalW versus L. Xu et al.17 proposed a modified TLM as amethod to determine RcW with higher accuracy. This methodis based on the equation that is obtained by dividing both sidesof eq 4 by L:= +R WL C V VR WL1( )( )1totallini gs thlin c(5)RcW can be obtained from the linear regression slope in a plotof RtotalW/L versus 1/L. For a detail discussion on the accuracyin the determination of RcW, see ref 17.The modified TLM plot of the 26 and 37 nm-thick PQTBTzOFETs at a bias-stress time of 0 s is shown in Figure 5a, whereonly data for Vgs = −10, −20, and −30 V are plotted for ease ofviewing. The corresponding plots at bias-stress times of 5 ×102, 5 × 103, and 2.5 × 104 s, are presented in Figure S7. TheFigure 4. Bias-stress effects of PQTBTz OFETs with different active layer thicknesses of 26 (a−c) and 37 nm (d−f) and different L’s of 50 (solidcurves) and 500 μm (broken curves). (a,d) Transfer curves measured in the linear regime (Vds = −1 V) at bias-stress times of 0, 5 × 102, 5 × 103,and 2.5 × 104 s. (b,e) Vgs-dependent μlin extracted from the transfer curves in (a,d). (c,f) Bias-stress time dependence of normalized μlin (opensymbols) and ΔVthlin (filled symbols) for PQTBTz OFETs with L = 50 (squares) and 500 mm (diamonds). The solid and broken curves are thefitting results with eq 1a.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068085https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig4&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig4&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig4&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig4&ref=pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asfilled and open symbols (solid and dotted straight lines) arethe data points (regression lines) for the 26 and 37 nm-thickPQTBTz OFETs, respectively. The coefficients of determi-nation R2 for the 26 (37) nm-thick OFETs were 0.922 (0.976),0.977 (0.990), and 0.985 (0.995) at Vgs of −10, −20, and −30V, respectively. Since the linear regression slopes correspond toRcW, it can be seen at a glance that the RcW for the 37 nm-thick PQTBTz OFETs is larger than that for the 26 nm-thickPQTBTz OFETs. Figure 5b shows the Vgs-dependent RcWextracted by the modified TLM. The very thick, thick, thin,and very thin curves show the results for bias-stress times of 0,5 × 102, 5 × 103, and 2.5 × 104 s, respectively. The solid andbroken curves are RcW for the 26 and 37 nm-thick PQTBTzOFETs, respectively. As seen from Figure S8, when Vgsapproached Vthlin from the negative side, the fitting accuracydegraded rapidly. This is because in such a Vgs region, thecontribution of RchW to RtotalW becomes much larger than thatof RcW; that is, the first term on the right-hand side of eq 5becomes dominant, making it difficult to determine the linearregression slope in the modified TLM plots with sufficientaccuracy. Thus, only RcW extracted with R2 ≥ 0.8 are plottedin Figure 5b and used in Rc correction. Although the unrealisticsagging of RcW is still seen in the Vgs-dependence at bias-stresstimes of 5 × 102, 5 × 103, and 2.5 × 104 s for the 37 nm-thickPQTBTz OFETs, it does not significantly affect the next Rccorrection.Now, we can perform Rc correction on the transfercharacteristics obtained in the bias-stress effect measurements.The Rc-corrected Id was calculated by multiplying Id by a Rc-correction factor F, which is given by F = (Rch + Rc)/Rch =Rtotal/(Rtotal − Rc), where Rtotal = Vds/Id. The Rc-correctedtransfer curves and μlin are shown in Figure 6a,d and b,e,respectively. The bias-stress time dependence of ΔVthlin/(Vgs −Vth0lin) and μlin/μ0lin evaluated from the Rc-corrected transferFigure 5. (a) Modified TLM plot of 26 and 37 nm-thick PQTBTzOFETs with different L’s from 50 to 500 μm at a bias-stress time of 0s; the squares, circles, and diamonds are the data points for Vgs = −10,−20, and −30 V, respectively. The straight lines are the regressionlines. The filled and open symbols (solid and broken lines) are thedata points (fitting results) for 26 and 37 nm-thick PQTBTz OFETs,respectively. (b) RcW extracted by modified TLM at different bias-stress times. The very thick, thick, thin, and very thin curves show theresults for bias-stress times of 0, 5 × 102, 5 × 103, and 2.5 × 104 s,respectively. The solid and broken curves are the results for 26 and 37nm-thick PQTBTz OFETs, respectively.Figure 6. Bias-stress effects of PQTBTz OFETs with different active layer thicknesses of 26 (a−c) and 37 nm (d−f) and different L’s of 50 (solidcurves) and 500 μm (broken curves) after Rc correction. (a,d) Rc-corrected transfer curves in the linear regime (Vds = −1 V) at bias-stress times of0, 5 × 102, 5 × 103, and 2.5 × 104 s. (b,e) Rc-corrected Vgs-dependent μlin extracted from the transfer curves in (a,d). (c,f) Rc-corrected bias-stresstime dependence of normalized μlin (open symbols) and ΔVthlin (filled symbols) for PQTBTz OFETs with L = 50 (squares) and 500 μm(diamonds). The solid and broken curves are the fitting results with eq 1a.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068086https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig5&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig5&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig5&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig5&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig6&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig6&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig6&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig6&ref=pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ascurves are shown in Figure 6c,f, where the Rc-corrected μ0linand Vth0lin, respectively, of 26 (37) nm-thick PQTBTz OFETswere 0.17 (0.18) cm2 V−1 s−1 and −4.48 (−4.39) V for L = 50μm and 0.18 (0.18) cm2 V−1 s−1 and −4.45 (−4.43) V for L =500 μm. From these figures, it was found that the channellength dependence seen in Figure 4 disappeared after Rccorrection. Therefore, the apparent higher operational stabilityof the OFETs with L = 50 μm compared to that of the OFETswith L = 500 μm in Figure 4c,f was due to the effect of a non-negligible Rc.Figure 7a,b shows the channel length dependence of ΔVthlin/(Vgs − Vth0lin) and μlin/μ0lin at a bias-stress time of 2.5 × 104 sbefore (squares) and after (circles) Rc correction for the 26and 37 nm-thick PQTBTz OFETs, respectively. The solid linesin these figures are guides to the eye, and the dotted lines showthe average values for the OFETs with L’s from 50 to 500 μm.Both ΔVthlin/(Vgs − Vth0lin) and μlin/μ0lin are seen to beconstant independent of L after Rc correction. The influence ofRc can be seen more clearly in Figure 7b. Before Rc correction,the reduction in μlin caused by bias-stress application increaseswith increasing L but it is still underestimated even at L = 500μm compared to that after Rc correction. On the other hand,ΔVthlin/(Vgs − Vth0lin) increases with increasing L up to 200μm, then being constant and almost equal to the average valueof the Rc-corrected ΔVthlin/(Vgs − Vth0lin). The similar trendcan be seen in Figure 7a. From these figures, we found thatwithout Rc correction, the reduction of μlin was underestimatedeven for the OFETs with L = 500 μm, and ΔVthlin wasunderestimated in the OFETs with short channels: ≤50 (150)μm for the 26 (37) nm-thick active layers. That is, theoperational stability was found to be overestimated without Rccorrection.The fitting of the bias-stress time dependence of ΔVthlin/(Vgs− Vth0lin) before and after Rc correction with eq 1a wasperformed for both sets of the OFETs with different L’s. Theobtained τ and β were plotted as a function of L in Figure S9and as τ−β mapping in Figure 7c. These figures explicitly showthat the channel length dependence disappears after Rccorrection, which indicates the reliability of Rc correctionperformed in this study. Therefore, we successfully evaluatedthe bias-stress effect in BG-TC PQTBTz OFETs with anODTS-treated SiO2 gate dielectric excluding the influence ofRc. The average values of Rc-corrected τ and β for the 26 (37)nm-thick PQTBTz OFETs were 1.2 (0.6) × 106 s and 0.29(0.30), respectively, and the Rc-corrected μlin/μ0lin was onaverage 0.85 (0.86) at 2.5 × 104 s. Figures 7 and S9 suggest asimple and effective way to avoid erroneous estimation in τ, β,and ΔVthlin, when a set of OFETs with different L's cannot beprepared in sufficient intervals to determine RcW by themodified TLM; it is to evaluate the bias-stress effect on twoOFETs whose channel lengths differ by more than a factor of 2and ensure that they are equal within the experimentaluncertainty. Before closing this subsection, we would like tonote that the erroneous evaluation of the operational stabilitywould become more pronounced in short-channel, high-mobility OFETs, because Rc becomes larger relative to Rch withincreasing μlin and decreasing L.3.4. Comparison of Operational Stability amongPBTTT, PCDTPT, and PQTBTz OFETs. In our previouswork,19,20 the operational stability of BG-TC OFETs with 24nm-thick PBTTT-C16 and 16 nm-thick PCDTPT active layersformed on ODTS-treated SiO2 (≈100 nm thick)/Si substrateswas evaluated under vacuum environment. Since extrinsicfactors such as the atmospheric oxygen and moisture affect theoperational stability of OFETs, burying the intrinsic relation-ship between the operational stability and the properties ofconstituent materials, the evaluation under vacuum environ-ment is essential to investigate the origin of intrinsicoperational instability. Interestingly, these two OFETs showedonly a shift in Vthlin with almost no change in the carrier-concentration-dependent μlin, as shown in Figure S10. Asexplained in Introduction, Rc correction is no need for theseOFETs.In the present study, the operational stability of PQTBTzOFETs in a vacuum environment was evaluated under thesame bias-stress condition and measurement sequence as in theprevious work.19,20 The device structure of the 26 nm-thickPQTBTz OFETs with L/W = 50/500 μm was almost the sameas those of the PBTTT and PCDTPT OFETs. Thus, theoperational stability among the three OFETs can be comparedwith minimum uncertainty, allowing us to discuss the origincausing the difference in operational stability. This comparisonis also interesting from the perspective of the difference inthermal behavior of semiconducting polymers; PBTTT-C16and PQTBTz-C12 exhibit LC phases at elevated temperatures,whereas PCDTPT reveals no LC phase up to 300 °C.25 ForPBTTT26 and PQTBTz, the active layers consisting ofperfectly oriented crystalline lamellae throughout the entirefilm thickness are formed by annealing at the LC temperature(150 and 180 °C, respectively). For PCDTPT, the formationof perfectly oriented crystalline lamellae is confined to less than10 nm near the ODTS-treated SiO2 gate insulator surfaces,even after annealing at 200 °C.19 The difference in thermalbehavior appeared as that in the surface morphology; wideFigure 7. (a,b) μlin/μ0lin (open symbols) and ΔVthlin/(Vgs − Vth0lin) (filled symbols) without (squares) and with (circles) Rc correction at a bias-stress time of 2.5 × 104 s as a function of L: (a) 26 nm-thick PQTBTz OFETs and (b) 37 nm-thick PQTBTz OFETs. (c) τ and β determined for26 nm- (squares and circles) and 37 nm- (diamonds and triangles) PQTBTz OFETs with different L’s. Circles and triangles (squares anddiamonds) are data with (without) Rc correction.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068087https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig7&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig7&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig7&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig7&ref=pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asterraces with the single monolayer step height could be seenonly for the PBTTT and PQTBTz active layers as shown inFigure S11.To compare the operational stability of the PQTBTz,PBTTT and PCDTPT OFETs, their ΔVthlin/(Vgs − Vth0lin) at abias-stress time of 2.5 × 104 s, τ and β are summarized in Table1. A smaller ΔVthlin/(Vgs − Vth0lin) and a longer τ (if β is thesame) indicate higher stability in Vthlin. In addition to ΔVthlin,the reduction in μlin should be considered in the evaluation ofoperational stability. The PBTTT and PCDTPT OFETsexhibited no reduction in μlin in contrast to the PQTBTzOFETs. Therefore, the operational stability was found to behigher in the following order: PBTTT ≫ PCDTPT >PQTBTz.The Vthlin is considered the sum of the flat-band voltage(VFB) of the active layer and the Vthlin component (Vdeep)originating in the charge carriers trapped in deep interfacetrapping sites.19 The Vdeep is given by Vdeep = −Qdeep/Ci, whereQdeep is the immobilized charge carrier density. Assuming thatthe VFB is equal to the turn-on voltage (Von) in the transfercharacteristics in the linear regime,27,28 the Vdeep can beevaluated by Vthlin − Von, as shown in Figure S12. Tounderstand the origin of the difference in operational stabilityamong the PBTTT, PCDTPT, and PQTBTz OFETs, theirΔVthlin, ΔVFB, and ΔVdeep at a bias-stress time of 2.5 × 104 s areplotted in Figure 8. Interestingly, there was almost nodifference in ΔVdeep among the three OFETs. The differencein the operational stability was found to come mainly fromΔVFB. The |ΔVFB| of the PQTBTz and PCDTPT OFETs wasabout 6 and 4 times larger than that of the PBTTT OFETs,respectively.The negative ΔVFB means an increase in the downward bandbending near the gate dielectric/active layer interface. Themost likely mechanism causing ΔVFB in a vacuum environmentis the charge carrier transfer from the channel to the gatedielectric under on-state gate-bias application.29,30 For p-channel OFETs, the HOMO level mismatch between theactive layer and gate dielectric (surface) acts as an energybarrier to limit the charge carrier (hole) transfer from thechannel to the localized HOMO states of the gate dielectric(surface). Since the gate dielectrics of the PBTTT, PCDTPT,and PQTBTz OFETs are the same: ODTS-treated 100 nm-thick SiO2 layers, the HOMO level mismatch decreases as theionization potential (IP) of active layer increases. That is, thetransfer rate of charge carriers increases with increasing IP ofthe active layer, reducing the operational stability. The IPs ofPBTTT-C14, PCDTPT, and PQTBTz-C12 were reported tobe 4.7−4.8,31−33 5.16,25 and 5.19 eV,16 respectively. Althoughthe comparison of IPs reported by different research groupsinvolves uncertainty, we believe that the trend in themagnitude of IP is as follows: PBTTT < PCDTPT ≤PQTBTz. This is consistent with the trend in the |ΔVFB|shown in Figure 8. Therefore, the difference in the operationalstability among the three OFETs can be explained by thecharge carrier transfer from the channel to the gate dielectric.This conclusion is in line with that in our previous study,19 inwhich the charge carrier transfer was discussed in detail as themost likely mechanism causing ΔVFB. To avoid repetition, wewill not discuss it further.Finally, we would like to note that the formation of activelayers through LC phases does not necessarily lead to highoperational stability. Forming the active layers through the LCphase is expected to improve the molecular packing in the π-stacking direction within the perfectly oriented crystallinelamellas, increasing the resistivity of confirmational changeagainst the prolong gate-bias stress application. This shouldsuppress the generation of both shallow and deep interfacetrapping sites. Unexpectedly, no significant difference in ΔVdeepwas observed among the three OFETs. Moreover, a decreasein μlin, indicating the increase in shallow trap density,34 wasobserved only in the PQTBTz OFETs. Therefore, thesuperiority of active layer formation through the LC phase inimproving the operational stability was not confirmed in thisstudy.4. CONCLUSIONSThe operational stability of polymer-based OFETs exhibitingboth a shift in Vthlin and a reduction in μlin has beeninvestigated. BG-TC PQTBTz OFETs with different L’s from50 to 500 μm in 50 μm intervals were fabricated on ODTS-treated SiO2 (100 nm thick)/Si substrates and their bias-stresseffects were evaluated. The ΔVthlin and μlin/μ0lin caused by aprolonged on-state gate-bias apprication depended on L; the|ΔVthlin| and the reduction in μlin increased with increasing L.Apparently, this result indicates that the OFET with a shorterL is more stable against the on-state gate-bias application. AfterRc correction was performed on the transfer characteristics, theapparent channel length dependence disappeared. Thereduction in μlin before Rc correction was smaller than thatafter, regardress of L. Without Rc correction, the |ΔVthlin| wasunderestimated in the OFETs with short channels, in which Rcwas non-negligible relative to Rch. That is, the operationalTable 1. Summary of ΔVthlin/(Vgs − Vth0lin) at 2.5 × 104 s, τ, and β of OFETs with PQTBTz, PCDTPT, and PBTTT ActiveLayersasemiconductor (thickness) gate-dielectric (L/W of OFET) VV V( )thlings th0lin at 25k s τ [s] β atmosphere refPQTBTz (26 nm) ODTS-SiO2 (50/500 μm) 0.27 1 × 106 0.29 in vac. this workPCDTPT (16 nm) ODTS-SiO2 (50/300 μm) 0.24 1 × 106 0.32 in vac. 19pBTTT-C16 (24 nm) ODTS-SiO2 (50/300 μm) 0.08 3 × 108 0.26 in vac. 19,20aThickness of SiO2 is approximately 100 nm for all OFETs.Figure 8. ΔVthlin, ΔVFB, and ΔVdeep at a bias-stress time of 2.5 × 104 sfor the PQTBTz, PCDTPT, and PBTTT OFETs.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068088https://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig8&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig8&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig8&ref=pdfhttps://pubs.acs.org/doi/10.1021/acsami.4c15666?fig=fig8&ref=pdfwww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asstability was found to be overestimated for short-channelOFETs without Rc correction. As Rc becomes larger relative toRch with increasing μlin and decreasing L, the erroneousevaluation of the operational stability would become morepronounced in short-channel, high-mobility OFETs. There-fore, one should pay attention to Rc in the fundamentalresearch into the origin of operational instability and inevaluating the effects of active layers, gate dielectrics, andactive layer/gate dielectric interfaces on operational stability.Finally, the operational stability of the PQTBTz OFETs in avacuum environment was compared to that of PCDTPT andPBTTT OFETs reported in our previous work. The opera-tional stability was found to be higher in the following order:PBTTT ≫ PCDTPT > PQTBTz. The difference in theoperational stability among the three OFETs came from thedifference in ΔVFB. This can be explained by the charge carriertransfer from the channel to the gate dielectric during on-stategate-bias application.■ ASSOCIATED CONTENT*sı Supporting InformationThe Supporting Information is available free of charge athttps://pubs.acs.org/doi/10.1021/acsami.4c15666.Additional figures as mentioned in the text, includingdifferential molecular weight distribution and a DSCcurve of PQTBTz-C12; XRD profiles and AFM imagesof PQTBTz-C12 films; histogram analysis of Vth and SS;overlaid transfer curves of 60 OFETs; modified TLMplots; extracted RcW with R2; channel length depend-ence of τ and β; bias-stress effects of PBTTT andPCDTPT OFETs; AFM images of PBTTT, PCDTPT,and PQTBTz active layers; and relationship among VFB(Von), Vdeep, and Vthlin (PDF)■ AUTHOR INFORMATIONCorresponding AuthorKenji Sakamoto − Research Center for Macromolecules andBiomaterials, National Institute for Materials Science(NIMS), Tsukuba, Ibaraki 305-0044, Japan; orcid.org/0000-0002-1379-874X; Email: SAKAMOTO.Kenji@nims.go.jpAuthorsTakeshi Yasuda − Research Center for Macromolecules andBiomaterials, National Institute for Materials Science(NIMS), Tsukuba, Ibaraki 305-0047, Japan; orcid.org/0000-0003-4652-9105Takeo Minari − Research Center for Macromolecules andBiomaterials, National Institute for Materials Science(NIMS), Tsukuba, Ibaraki 305-0044, Japan; orcid.org/0000-0001-7690-221XMasafumi Yoshio − Research Center for Macromolecules andBiomaterials, National Institute for Materials Science(NIMS), Tsukuba, Ibaraki 305-0047, Japan; orcid.org/0000-0002-1442-4352Junpei Kuwabara − Tsukuba Research Center for EnergyMaterials Science (TREMS), Institute of Pure and AppliedSciences, University of Tsukuba, Tsukuba, Ibaraki 305-8573,Japan; orcid.org/0000-0002-9032-5655Masayuki Takeuchi − Research Center for Macromoleculesand Biomaterials, National Institute for Materials Science(NIMS), Tsukuba, Ibaraki 305-0047, Japan; orcid.org/0000-0002-0207-0665Complete contact information is available at:https://pubs.acs.org/10.1021/acsami.4c15666Author ContributionsThe manuscript was written through contributions of allauthors. All authors have given approval to the final version ofthe manuscript.NotesThe authors declare no competing financial interest.■ ACKNOWLEDGMENTSA part of this work was supported by JSPS KAKENHI GrantNo. 20K05310 and by “Advanced Research Infrastructure forMaterials and Nanotechnology in Japan (ARIM)” of theMinistry of Education, Culture, Sports, Science and Technol-ogy (MEXT), Proposal Number JPMXP1223NM5185.■ REFERENCES(1) Tsumura, A.; Koezuka, H.; Ando, T. Macromolecular electronicdevice: Field-effect transistor with a polythiophene thin film. Appl.Phys. Lett. 1986, 49, 1210−1212.(2) Dong, H.; Fu, X.; Liu, J.; Wang, Z.; Hu, W. 25th AnniversaryArticle: Key Points for High-Mobility Organic Field-Effect Tran-sistors. Adv. Mater. 2013, 25, 6158−6183.(3) Yamashita, Y.; Hinkel, F.; Marszalek, T.; Zajaczkowski, W.;Pisula, W.; Baumgarten, M.; Matsui, H.; Müllen, K.; Takeya, J.Mobility Exceeding 10 cm2 /(V·s) in Donor-Acceptor PolymerTransistors with Band-like Charge Transport. Chem. Mater. 2016, 28,420−424.(4) Luo, C.; Kyaw, A. K. K.; Perez, L. A.; Patel, S.; Wang, M.;Grimm, B.; Bazan, G. C.; Kramer, E. J.; Heeger, A. J. General Strategyfor Self-Assembly of Highly Oriented Nanocrystalline Semiconduct-ing Polymers with High Mobility. Nano Lett. 2014, 14, 2764−2771.(5) Park, Y.; Jung, J. W.; Kang, H.; Seth, J.; Kang, Y.; Sung, M. M.Single-Crystal Poly[4-(4,4-dihexadecyl-4H-cyclopenta[1,2-b:5,4-b′]-dithiophen-2-yl)-alt-[1,2,5]thiadiazolo[3,4-c]pyridine] Nanowireswith Ultrahigh Mobility. Nano Lett. 2019, 19, 1028−1032.(6) Bittle, E. G.; Basham, J. I.; Jackson, T. N.; Jurchescu, O. D.;Gundlach, D. J. Mobility overestimation due to gated contacts inorganic field-effect transistors. Nat. Commun. 2016, 7, No. 10908.(7) Okachi, T. Mobility overestimation due to minority carrierinjection and trapping in organic field-effect transistors. Org. Electron.2018, 57, 34−44.(8) Sirringhaus, H. Reliability of Organic Field-Effect Transistors.Adv. Mater. 2009, 21, 3859−3873.(9) Park, S.; Kim, S. H.; Choi, H. H.; Kang, B.; Cho, K. RecentAdvances in the Bias Stress Stability of Organic Transistors. Adv.Funct. Mater. 2020, 30, No. 1904590.(10) Lee, W. H.; Choi, H. H.; Kim, D. H.; Cho, K. 25th AnniversaryArticle: Microstructure Dependent Bias Stability of OrganicTransistors. Adv. Mater. 2014, 26, 1660−1680.(11) Kunii, M.; Iino, H.; Hanna, J. Bias-Stress Characterization ofSolution-Processed Organic Field-Effect Transistor Based on HighlyOrdered Liquid Crystals. Appl. Phys. Lett. 2017, 110, No. 243301.(12) Noriega, R.; Rivnay, J.; Vandewal, K.; Koch, F. P. V.; Stingelin,N.; Smith, P.; Toney, M. F.; Salleo, A. A general relationship betweendisorder, aggregation and charge transport in conjugated polymers.Nat. Mater. 2013, 12, 1038−1044.(13) Mathijssen, S. G. J.; Cölle, M.; Gomes, H.; Smits, E. C. P.; deBoer, B.; McCulloch, I.; Bobbert, P. A.; de Leeuw, D. M. Dynamics ofthreshold voltage shifts in organic and amorphous silicon field-effecttransistors. Adv. Mater. 2007, 19, 2785−2789.(14) Venkateshvaran, D.; Nikolka, M.; Sadhanala, A.; Lemaur, V.;Zelazny, M.; Kepa, M.; Hurhangee, M.; Kronemeijer, A. J.; Pecunia,ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068089https://pubs.acs.org/doi/10.1021/acsami.4c15666?goto=supporting-infohttps://pubs.acs.org/doi/suppl/10.1021/acsami.4c15666/suppl_file/am4c15666_si_001.pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Kenji+Sakamoto"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-1379-874Xhttps://orcid.org/0000-0002-1379-874Xmailto:SAKAMOTO.Kenji@nims.go.jpmailto:SAKAMOTO.Kenji@nims.go.jphttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Takeshi+Yasuda"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0003-4652-9105https://orcid.org/0000-0003-4652-9105https://pubs.acs.org/action/doSearch?field1=Contrib&text1="Takeo+Minari"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0001-7690-221Xhttps://orcid.org/0000-0001-7690-221Xhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Masafumi+Yoshio"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-1442-4352https://orcid.org/0000-0002-1442-4352https://pubs.acs.org/action/doSearch?field1=Contrib&text1="Junpei+Kuwabara"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-9032-5655https://pubs.acs.org/action/doSearch?field1=Contrib&text1="Masayuki+Takeuchi"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-0207-0665https://orcid.org/0000-0002-0207-0665https://pubs.acs.org/doi/10.1021/acsami.4c15666?ref=pdfhttps://doi.org/10.1063/1.97417https://doi.org/10.1063/1.97417https://doi.org/10.1002/adma.201302514https://doi.org/10.1002/adma.201302514https://doi.org/10.1002/adma.201302514https://doi.org/10.1021/acs.chemmater.5b04567?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.chemmater.5b04567?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/nl500758w?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/nl500758w?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/nl500758w?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.8b04302?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.8b04302?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.8b04302?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/ncomms10908https://doi.org/10.1038/ncomms10908https://doi.org/10.1016/j.orgel.2018.02.034https://doi.org/10.1016/j.orgel.2018.02.034https://doi.org/10.1002/adma.200901136https://doi.org/10.1002/adfm.201904590https://doi.org/10.1002/adfm.201904590https://doi.org/10.1002/adma.201304665https://doi.org/10.1002/adma.201304665https://doi.org/10.1002/adma.201304665https://doi.org/10.1063/1.4985628https://doi.org/10.1063/1.4985628https://doi.org/10.1063/1.4985628https://doi.org/10.1038/nmat3722https://doi.org/10.1038/nmat3722https://doi.org/10.1002/adma.200602798https://doi.org/10.1002/adma.200602798https://doi.org/10.1002/adma.200602798www.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asV.; Nasrallah, I.; Romanov, I.; Broch, K.; McCulloch, I.; Emin, D.;Olivier, Y.; Cornil, J.; Beljonne, D.; Sirringhaus, H. Approachingdisorder-free transport in high-mobility conjugated polymers. Nature2014, 515, 384−388.(15) Liu, C.; Huang, K.; Park, W.-T.; Li, M.; Yang, T.; Liu, X.; Liang,L.; Minari, T.; Noh, Y.-Y. A unified understanding of charge transportin organic semiconductors: the importance of attenuated delocaliza-tion for the carriers. Mater. Horiz. 2017, 4, 608−618.(16) Kim, D. H.; Lee, B.-L.; Moon, H.; Kang, H. M.; Jeong, E. J.;Park, J.-I.; Han, K.-M.; Lee, S.; Yoo, B. W.; Koo, B. W.; Kim, J. Y.;Lee, W. H.; Cho, K.; Becerril, H. A.; Bao, Z. Liquid-CrystallineSemiconducting Copolymers with Intramolecular Donor−AcceptorBuilding Blocks for High-Stability Polymer Transistors. J. Am. Chem.Soc. 2009, 131, 6124−6132.(17) Xu, Y.; Gwoziecki, R.; Chartier, I.; Coppard, R.; Balestra, F.;Ghibaudo, G. Modified transmission-line method for contactresistance extraction in organic field-effect transistors. Appl. Phys.Lett. 2010, 97, No. 063302.(18) Liu, C.; Xu, Y.; Li, Y.; Scheideler, W.; Minari, T. Critical Impactof Gate Dielectric Interfaces on the Contact Resistance of High-Performance Organic Field-Effect Transistors. J. Phys. Chem. C 2013,117, 12337−12345.(19) Sakamoto, K.; Bulgarevich, K.; Yasuda, T.; Minari, T.;Takeuchi, M. Origin of intrinsic operational instability in organicfield-effect transistors with aligned high-mobility donor-acceptorcopolymer active layers. Adv. Mater. Technol. 2024, 9, No. 2301503.(20) Bulgarevich, K.; Sakamoto, K.; Minari, T.; Yasuda, T.; Miki, K.;Takeuchi, M. Polymer-Based Organic Field-Effect Transistors withActive Layers Aligned by Highly Hydrophobic Nanogrooved Surfaces.Adv. Funct. Mater. 2019, 29, No. 1905365.(21) Bronstein, H.; Hurhangee, M.; Fregoso, E. C.; Beatrup, D.;Soon, Y. W.; Huang, Z.; Hadipour, A.; Tuladhar, P. S.; Rossbauer, S.;Sohn, E.-H.; Shoaee, S.; Dimitrov, S. D.; Frost, J. M.; Ashraf, R. S.;Kirchartz, T.; Watkins, S. E.; Song, K.; Anthopoulos, T.; Nelson, J.;Rand, B. P.; Durrant, J. R.; McCulloch, I. Isostructural, DeeperHighest Occupied Molecular Orbital Analogues of Poly(3-hexylth-iophene) for High-Open Circuit Voltage Organic Solar Cells. Chem.Mater. 2013, 25, 4239−4249.(22) Bulgarevich, K.; Sakamoto, K.; Minari, T.; Yasuda, T.; Miki, K.Spatially Uniform Thin-Film Formation of Polymeric OrganicSemiconductors on Lyophobic Gate Insulator Surfaces by Self-Assisted Flow-Coating. ACS Appl. Mater. Interfaces 2017, 9, 6237−6245.(23) Bulgarevich, K.; Sakamoto, K.; Yasuda, T.; Minari, T.;Takeuchi, M. Operational Stability Enhancement of PolymericOrganic Field-Effect Transistors by Amorphous PerfluoropolymersChemically Anchored to Gate Dielectric Surfaces. Adv. Electron.Mater. 2020, 6, No. 2000161.(24) Horii, Y.; Sakaguchi, K.; Chikamatsu, M.; Azumi, R.; Yase, K.;Kitagawa, M.; Konishi, H. High-Performance Solution-Processed n-Channel Organic Thin-Film Transistors Based on a Long ChainAlkyl-Substituted C60 Derivative. Appl. Phys. Express 2010, 3,No. 101601.(25) Ying, L.; Hsu, B. B. Y.; Zhan, H.; Welch, G. C.; Zalar, P.; Perez,L. A.; Kramer, E. J.; Nguyen, T.-Q.; Heeger, A. J.; Wong, W.-Y.;Bazan, G. C. Regioregular Pyridal[2,1,3]thiadiazole π-ConjugatedCopolymers. J. Am. Chem. Soc. 2011, 133, 18538−18541.(26) Himmelberger, S.; Dacuña, J.; Rivnay, J.; Jimison, L. H.;McCarthy-Ward, T.; Heeney, M.; McCulloch, I.; Toney, M. F.; Salleo,A. Effects of Confinement on Microstructure and Charge Transport inHigh Performance Semicrystalline Polymer Semiconductors. Adv.Funct. Mater. 2013, 23, 2091−2098.(27) Haneef, H. F.; Zeidell, A. M.; Jurchescu, O. D. Charge carriertraps in organic semiconductors: a review on the underlying physicsand impact on electronic devices. J. Mater. Chem. C 2020, 8, 759−787.(28) Iqbal, H. F.; Ai, Q.; Thorley, K. J.; Chen, H.; McCulloch, I.;Risko, C.; Anthony, J. E.; Jurchescu, O. D. Suppressing bias stressdegradation in high performance solution processed organictransistors operating in air. Nat. Commun. 2021, 12, No. 2352.(29) Kim, J.; Jang, J.; Kim, K.; Kim, H.; Kim, S. H.; Park, C. E. TheOrigin of Excellent Gate-Bias Stress Stability in Organic Field-EffectTransistors Employing Fluorinated-Polymer Gate Dielectrics. Adv.Mater. 2014, 26, 7241−7246.(30) Lee, B.; Wan, A.; Mastrogiovanni, D.; Anthony, J. E.; Garfunkel,E.; Podzorov, V. Origin of the bias stress instability in single-crystalorganic field-effect transistors. Phys. Rev. B 2010, 82, No. 085302.(31) Zhuo, J.-M.; Zhao, L.-H.; Png, R.-Q.; Wong, L.-Y.; Chia, P.-J.;Tang, J.-C.; Sivaramakrishnan, S.; Zhou, M.; Ou, E. C. -W.; Chua, S.-J.; Sim, W.-S.; Chua, L.-L.; Ho, P. K. -H. Direct SpectroscopicEvidence for a Photodoping Mechanism in Polythiophene andPoly(bithiophene-alt-thienothiophene) Organic Semiconductor ThinFilms Involving Oxygen and Sorbed Moisture. Adv. Mater. 2009, 21,4747−4752.(32) Zhao, L.-H.; Png, R.-Q.; Chiam, C. C. H.; Guo, H.; Zhuo, J.-M.; Chua, L.-L.; Wee, A. T. S.; Ho, P. K. H. Polarization effects onenergy-level alignment at the interfaces of polymer organic semi-conductor films. Appl. Phys. Lett. 2012, 101, No. 053304.(33) Yamashita, Y.; Tsurumi, J.; Ohno, M.; Fujimoto, R.; Kumagai,S.; Kurosawa, T.; Okamoto, T.; Takeya, J.; Watanabe, S. Efficientmolecular doping of polymeric semiconductors driven by anionexchange. Nature 2019, 572, 634−638.(34) Li, C.; Duan, L.; Li, H.; Qiu, Y. Universal Trap Effect in CarrierTransport of Disordered Organic Semiconductors: Transition fromShallow Trapping to Deep Trapping. J. Phys. Chem. C 2014, 118,10651−10660.ACS Applied Materials & Interfaces www.acsami.org Research Articlehttps://doi.org/10.1021/acsami.4c15666ACS Appl. Mater. Interfaces 2024, 16, 68081−6809068090https://doi.org/10.1038/nature13854https://doi.org/10.1038/nature13854https://doi.org/10.1039/C7MH00091Jhttps://doi.org/10.1039/C7MH00091Jhttps://doi.org/10.1039/C7MH00091Jhttps://doi.org/10.1021/ja8095569?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja8095569?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja8095569?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1063/1.3479476https://doi.org/10.1063/1.3479476https://doi.org/10.1021/jp4023844?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/jp4023844?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/jp4023844?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1002/admt.202301503https://doi.org/10.1002/admt.202301503https://doi.org/10.1002/admt.202301503https://doi.org/10.1002/adfm.201905365https://doi.org/10.1002/adfm.201905365https://doi.org/10.1021/cm4022563?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/cm4022563?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/cm4022563?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsami.6b15398?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsami.6b15398?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsami.6b15398?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1002/aelm.202000161https://doi.org/10.1002/aelm.202000161https://doi.org/10.1002/aelm.202000161https://doi.org/10.1143/APEX.3.101601https://doi.org/10.1143/APEX.3.101601https://doi.org/10.1143/APEX.3.101601https://doi.org/10.1021/ja207543g?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja207543g?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1002/adfm.201202408https://doi.org/10.1002/adfm.201202408https://doi.org/10.1039/C9TC05695Ehttps://doi.org/10.1039/C9TC05695Ehttps://doi.org/10.1039/C9TC05695Ehttps://doi.org/10.1038/s41467-021-22683-2https://doi.org/10.1038/s41467-021-22683-2https://doi.org/10.1038/s41467-021-22683-2https://doi.org/10.1002/adma.201402363https://doi.org/10.1002/adma.201402363https://doi.org/10.1002/adma.201402363https://doi.org/10.1103/PhysRevB.82.085302https://doi.org/10.1103/PhysRevB.82.085302https://doi.org/10.1002/adma.200901120https://doi.org/10.1002/adma.200901120https://doi.org/10.1002/adma.200901120https://doi.org/10.1002/adma.200901120https://doi.org/10.1063/1.4734502https://doi.org/10.1063/1.4734502https://doi.org/10.1063/1.4734502https://doi.org/10.1038/s41586-019-1504-9https://doi.org/10.1038/s41586-019-1504-9https://doi.org/10.1038/s41586-019-1504-9https://doi.org/10.1021/jp5022906?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/jp5022906?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/jp5022906?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-aswww.acsami.org?ref=pdfhttps://doi.org/10.1021/acsami.4c15666?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://www.cas.org/solutions/biofinder-discovery-platform?utm_campaign=GLO_ACD_STH_BDP_AWS&utm_medium=DSP_CAS_PAD&utm_source=Publication_ACSPubs