# Fileset

[large-magnetoresistance-in-a-si-based-double-tunnel-junction-with-purely-organic-radical-molecules.pdf](https://mdr.nims.go.jp/filesets/2fd2755e-66df-4ff6-a1a7-2da2eb623592/download)

## Creator

Jayanta Bera, Tuhin Shuvra Basu, Jannic Wolf, Haitao Zhang, [Kazuhiro Marumoto](https://orcid.org/0000-0001-9792-0775), [Yutaka Wakayama](https://orcid.org/0000-0002-0801-8884), [Carmen Herrmann](https://orcid.org/0000-0002-9496-0664), [Thomas Huhn](https://orcid.org/0000-0001-6292-4215), [Ryoma Hayakawa](https://orcid.org/0000-0002-1442-8230)

## Rights

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

## Other metadata

[Large Magnetoresistance in a Si-Based Double-Tunnel Junction with Purely Organic Radical Molecules](https://mdr.nims.go.jp/datasets/1388d5df-bd0e-40a9-8b7d-d3fdb961db55)

## Fulltext

Large Magnetoresistance in a Si-Based Double-Tunnel Junction with Purely Organic Radical MoleculesLarge Magnetoresistance in a Si-Based Double-Tunnel Junction withPurely Organic Radical MoleculesJayanta Bera, Tuhin Shuvra Basu, Jannic Wolf, Haitao Zhang, Kazuhiro Marumoto, Yutaka Wakayama,Carmen Herrmann,* Thomas Huhn,* and Ryoma Hayakawa*Cite This: Nano Lett. 2026, 26, 8257−8264 Read OnlineACCESS Metrics & More Article Recommendations *sı Supporting InformationABSTRACT: Organic radicals have shown promise for tunableand low-cost spintronic devices. However, integrating the radicalswith a Si metal−oxide−semiconductor (MOS) structure remains achallenge. Here, we incorporate stable (4-(((2,5-bis(2-(phenyl)-ethynyl)phenyl)carbonyl)(methyl)amino)-2,2,6,6-tetramethylpi-peridin-1-yl)oxidanyl (TEMPO-OPE) radicals in a Si-MOS-baseddouble-tunnel junction and demonstrate a huge positive magneto-resistance of up to 400% at a magnetic field of 7 T and atemperature of 3 K. This goes along with a significant reduction ofthe differential conductance peak corresponding to the highestoccupied molecular orbital (HOMO) of TEMPO-OPE underexternal magnetic fields. First-principles calculations suggest thatthe singly occupied molecular orbital can mix with the HOMO of TEMPO-OPE. This could lead to suppression of the HOMOconductance peak under magnetic fields and, thus, provide a possible origin of the large magnetoresistance. These findings suggest apath toward incorporating magnetic molecular functionalities into conventional Si devices, leading to large-scale integration ofmolecular spintronic devices.KEYWORDS: organic radicals, magnetoresistance, unpaired electrons, resonant tunneling, molecular orbitals,Si-based double-tunnel junctionsMolecular spintronics, which utilizes both the degree offreedom of spin and charge, is an emerging field basedon spin-dependent carrier transport through individualmolecules or their assemblies.1 This feature is expected tooffer significant applications, including low-power memory,spin-based logic, and quantum computing.2−10 Stable organicradicals possess a paramagnetic nature owing to their open-shell system with unpaired electrons. Due to their light-element compositions with carbon, hydrogen, nitrogen, andoxygen, they exhibit low spin−orbit coupling and weakhyperfine interactions.11 Therefore, organic radicals have alonger spin coherence time (∼7 μs) compared to inorganiccounterparts, like transition metal complexes and single-molecule magnets, even at room temperature.12−14 Thisfeature contributes to the protection of the information storedin the electronic spin.15,16 These attractive features of purelyorganic radicals make them promising candidates for molecularspintronic devices.The spin-dependent carrier transport through individualmolecules and their assemblies was generally investigated usingbreak-junction techniques, conductive probe atomic forcemicroscopy, and scanning tunneling microscopy (STM)techniques.17−32 For example, Liu et al. observed Kondoresonance in 1,3,5-triphenyl-6-oxoverdazyl (TOV) moleculesusing scanning tunneling spectroscopy (STS), which arises dueto the interaction between a localized spin of TOV andconduction electrons of metal electrodes.17 Müllegger et al.demonstrated Kondo resonance in α,γ-bisdiphenylene-β-phenylallyl (BDPA) molecules on Au(111) surfaces.18Frisenda et al. detected the presence of an unpaired electronspin of a polychlorotriphenylmethyl (PTM) radical usingmechanically controlled break junction (MCBJ) and electro-migrated break junction techniques.24 Additionally, largemagnetoresistance (MR) was observed in molecular junctionswith non-magnetic molecules, such as fullerene (C60) andbenzene, using magnetic electrodes.33,34 Similar MR effectswere visualized in transition metal complexes in single-molecular junctions formed by STM.22,35 Furthermore, Mitraet al. observed Kondo resonance and MR in the molecularjunctions with PTM radicals25 or Blatter radicals26 usingMCBJ techniques. However, the above-mentioned techniquesReceived: March 29, 2026Revised: May 28, 2026Accepted: June 11, 2026Published: June 19, 2026Letterpubs.acs.org/NanoLett© 2026 The Authors. Published byAmerican Chemical Society8257https://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−8264This article is licensed under CC-BY 4.0Downloaded via NATL INST FOR MATLS SCIENCE (NIMS) on July 3, 2026 at 04:12:41 (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="Jayanta+Bera"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Tuhin+Shuvra+Basu"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Jannic+Wolf"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Haitao+Zhang"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Kazuhiro+Marumoto"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Yutaka+Wakayama"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Carmen+Herrmann"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Carmen+Herrmann"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Thomas+Huhn"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Ryoma+Hayakawa"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/showCitFormats?doi=10.1021/acs.nanolett.6c01526&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?goto=articleMetrics&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?goto=recommendations&?ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?goto=supporting-info&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=tgr1&ref=pdfhttps://pubs.acs.org/toc/nalefd/26/25?ref=pdfhttps://pubs.acs.org/toc/nalefd/26/25?ref=pdfhttps://pubs.acs.org/toc/nalefd/26/25?ref=pdfhttps://pubs.acs.org/toc/nalefd/26/25?ref=pdfpubs.acs.org/NanoLett?ref=pdfhttps://pubs.acs.org?ref=pdfhttps://pubs.acs.org?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://pubs.acs.org/NanoLett?ref=pdfhttps://pubs.acs.org/NanoLett?ref=pdfhttps://creativecommons.org/licenses/by/4.0/are not adopted for large-scale integration of molecularspintronics devices. To overcome this problem, we demon-strated Si-based double-tunnel junctions, where isolatedmolecules are sandwiched as quantum dots (QDs) betweentwo oxide layers of a metal−oxide−semiconductor (MOS)structure.36−38 In our previous work, we demonstratedresonant tunneling via the singly occupied molecular orbital(SOMO) of adamantyl nitronyl nitroxide p-terphenyl (NN-TP) radicals in the Si-based double-tunnel junction.39 Thisresult clarified the survival of unpaired electrons of the organicradicals in the device structure. However, we had not observedobvious magnetoresistance (MR) in the samples with NN-TP.In this work, we investigated magnetic-field-dependentcarrier transport via TEMPO-OPE radicals incorporated intothe Si-based double-tunnel junction. TEMPO-OPE possessesan unpaired electron on the TEMPO radical, which is notconjugated with the π orbitals of the OPE backbone molecule(Figure 1a). Rather, the radical part of TEMPO-OPE iselectrically isolated from the main transport channel, incontrast to the case of NN-TP. This property is expected topreserve the localization of the unpaired electron on theTEMPO radical and to enable the weak coupling to the πorbitals of the OPE backbone due to the spatial proximitybetween the TEMPO radical and the OPE backbone. Actually,we have achieved large MR values of up to 287% at a magneticfield of 4 T in the single-molecule junction formed by a MCBJtechnique in our previous study.27 This finding inspired us toexplore the carrier transport via TEMO-OPE in the Si-baseddouble-tunnel junction under magnetic fields. Although similarexamples, e.g., grafting of magnetic molecules onto Sisubstrates, have been reported, no MR effects have appearedin Si-based molecular junctions.40−42 Here, we observed a hugepositive MR of up to 400% in the TEMPO-OPE sample underan applied magnetic field of 7 T and a temperature of 3 K. Thiscan be attributed to a significant reduction in the differentialconductance (dI/dV) peak corresponding to the highestoccupied molecular orbital (HOMO) of TEMPO-OPEunder magnetic fields. In contrast, no significant MR wasobserved in non-radical OPE (closed-shell type) samples. Ourfindings thus have the potential to integrate magnetic functionsof organic radicals into large-scale-integrated Si devices in thefuture.A Si-based double-tunnel junction with TEMPO-OPEmolecules embedded as quantum dots was formed on a highlydoped Si (p+) substrate (Figure 1a). Here, individualTEMPO-OPE molecules were embedded in an Al2O3 layer,in which the number of molecules was estimated to be in theorder of 1012−1013 cm−2 based on our previous study.36 Theinsulating layers (SiO2 and Al2O3) function as the tunnelbarriers, while the p+ Si substrate serves as the bottomelectrode. The detailed formation processes are shown insection 1.2 of the Supporting Information. It is noteworthythat the TEMPO-OPE molecule exhibits two isomericconfigurations, namely, (Z)- and (E)-TEMPO-OPE, depend-ing on the spatial configuration of the TEMPO group withrespect to rotation around the amide bond (C−N) that linksthe TEMPO substituent to the backbone. Therefore, weconsider both isomeric configurations in the Si-based double-tunnel junction when modeling their properties by first-principles methods.Figure 1b and c shows the optimized molecular structureand energy diagram of an isolated (Z)-TEMPO-OPE moleculeFigure 1. (a) Schematic illustration of the Si-based double-tunnel junction with TEMPO-OPE molecules. (b) Optimized structure of (Z)-TEMPO-OPE in vacuum. (c) Molecular orbital energy diagram and corresponding isosurfaces for (Z)-TEMPO-OPE in vacuum. (d) I−Vcharacteristic and (e) corresponding dI/dV curve of the Si-based double-tunnel junction with TEMPO-OPE molecules measured at a temperatureof 20 K.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648258https://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig1&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig1&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig1&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig1&ref=pdfpubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asevaluated by Kohn−Sham density functional theory (KS-DFT).43 We focus here on (Z)-TEMPO-OPE as it tends to bemore energetically preferable than (E)-TEMPO-OPE (TableS3) when adsorbed on an oxide surface due to a larger contactarea (it should be kept in mind that, in their isolated form, thetwo isomers are very similar in terms of their total energy,molecular orbital energies, and vibrational modes (sections 5and 6 of the Supporting Information) and that the isomers’exact atomic configuration when embedded in an oxide isunknown; thus, both could be present in the experiment). Theoptimized molecular structure, the molecular orbital energydiagram of (E)-TEMPO-OPE, and the detailed computationalmethods are described in section 5 of the SupportingInformation. The SOMO is mainly located on the TEMPOradical part in both the isomeric configurations, correspondingto a π*N−O orbital.Figure 1d and e depicts a typical current (I)−voltage (V)characteristic and the corresponding dI/dV curve of theTEMPO-OPE sample measured at a temperature of 20 K,where the p+ Si substrate was grounded and a voltage wasapplied to the top gold electrode. Three distinct “staircase”-likesignals were observed at the negative voltage range, and one“staircase” was visible in the positive voltage range in the I−Vcurve (Figure 1d). The corresponding dI/dV peaks wereobserved at −0.63, −0.99, −1.31, and 1.12 V in Figure 1e.Similar dI/dV peaks were observed in 25 out of a total 119devices, corresponding to a yield of approximately 21%, whichis consistent with that in the devices with NN-TP molecules.39Conversely, no dI/dV peaks appeared in the reference sampleswithout any molecules (Figure S11b). Thus, the prominent dI/dV peaks in Figure 1e indicate that the TEMPO-OPEmolecules remain intact in the double-tunnel junction.Our previous studies revealed that the tunneling currentreflecting the occupied MOs arises in the negative voltagerange due to the resonant tunneling of holes from theunderlying Si substrate to the embedded molecules.36,38,39Likewise, unoccupied MOs arise in the positive voltage rangeowing to resonant tunneling of electrons from the underlyingSi substrate to the embedded molecules. Given the peakassignment in the non-radical OPE device (Figures S9b andS10d), the dI/dV peaks visible at −0.63, −0.99, and −1.31 V,which are indicated by red arrows in Figure 1d and e, can beascribed to the HOMO, HOMO−1, and HOMO−2 ofTEMPO-OPE molecules, respectively. The dI/dV peak at1.12 V, which is indicated by a green arrow, correspondsaccordingly to the lowest unoccupied molecular orbital(LUMO) of TEMPO-OPE molecules. The mean values ofpeak positions were estimated to be −0.68 ± 0.12 V for theHOMO and 1.08 ± 0.09 V for the LUMO, respectively, fromthe statistical dI/dV measurements in the TEMPO-OPEdevices (Figure S10c). Accordingly, the HOMO−LUMO gapwas estimated at 1.8 eV, which is comparable to that of non-radical OPE (1.4 eV) (Figure S10d).The carrier transport of Si-based double tunnel junctionswith TEMPO-OPE molecules was examined under magneticfields to clarify the role of the unpaired electron spin. Figure 2ashows the I−V characteristics of the sample under magneticfields ranging from 0 to 7 T, where magnetic fields wereapplied in the perpendicular direction to the sample plane. Themeasurement temperature was fixed at 3 K. It is noted that 10-point-data smoothing with the Savitzky−Golay method wasimplemented in Figure 2a and b to eliminate large noise. Thecomparison of raw data and the smoothed curves is given inFigure S12. Significantly, the tunneling current via the HOMOlevel was strongly reduced by magnetic fields, as indicated bythe red arrow in Figure 2a. The current level was returned tothe original one by the reduction of the magnetic field from 7to 0 T (Figure S13). In contrast, no changes in the tunnelingcurrent through the LUMO level were observed under thesame magnetic fields.This intriguing occurrence was more clearly visible in thedI/dV curves. As shown in Figure 2b, the dI/dV peakcorresponding to the HOMO level was closely suppressedunder magnetic fields (Figure 2c). Consequently, the dI/dVpeak disappeared in magnetic fields of above 3 T. Thisreduction in the HOMO dI/dV peak is also confirmed in two-dimensional (2D) color maps of dI/dV curves as a function ofmagnetic fields (Figure S14). On the other hand, no significantchanges were observed in the dI/dV peak attributed to theLUMO level (Figure 2d). Moreover, the same behavior,Figure 2. (a) I−V characteristics and (b) corresponding dI/dV curves of TEMPO-OPE samples under magnetic fields ranging from 0 to 7 T. ThedI/dV peaks corresponding to the HOMO and LUMO levels of TEMPO-OPE molecules are shown in enlarged views with (c) black and (d)orange dotted rectangles, respectively. The magnetic field was applied perpendicular to the sample plane.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648259https://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig2&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig2&ref=pdfpubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asnamely, the reduction in the HOMO dI/dV peak, wasobserved in a different device (Figure S15). Such variationsin tunneling currents under magnetic fields did not appear inboth samples with non-radical OPE molecules (Figure S9c andd) and without any molecules (Figure S11a and b).In order to probe the exact variations in tunneling currentvia the HOMO and LUMO levels, we performed magneto-resistance (MR) measurements on our samples at fixedvoltages. MR values were calculated as ×( ) 100 (%)R RRM 00,where RM and R0 are the resistances obtained at ±7 and 0 T,respectively. The MR was measured with the followingsequence of magnetic fields: 0 → 7 → 0 → −7 → 0 T.Figure 3a and b shows the variation in MR curves obtained at−0.75 and 1.2 V, respectively, as a function of applied magneticfields, where the measurement temperature was fixed at 3 K. Alarge positive MR value of up to 400% was observed inTEMPO-OPE samples at −0.75 V, which corresponds to theHOMO level of the TEMPO-OPE molecules. The observedMR value in the TEMPO-OPE sample is higher than thepreviously reported experimental values obtained using organicradicals and metal phthalocyanines in single-moleculejunctions.22,23,25−27 In contrast, no significant MR wasobserved at 1.2 V, which agrees with the LUMO level of theTEMPO-OPE molecules. For comparison, we carried out MRmeasurements on the non-radical OPE samples (Figure S9eand f) as well as the sample without molecules (Figure S11cand d). In both cases, we did not observe any MR effect. Thesefindings clarify that the MR effect in the TEMPO-OPE sampleis caused by the TEMPO radical group.Figure 3c and d depicts the I−V and MR curves of TEMPO-OPE samples measured at a varied temperature ranging from 3to 20 K. No significant changes in I−V curves were observedwith the temperature changed. However, a drastic reduction inthe MR value was observed when the temperature increased to5 K, eventually disappearing at temperatures higher than 5 K.Next, we discuss the possible origins of the large positivemagnetoresistance (MR) in the double-tunnel junction withTEMPO-OPE molecules. Such a large MR of up to 400% inthe double-tunnel junction with radical molecules has not yetbeen reported. A negative MR was observed by Sugawara et al.in Au nanoparticles connected via a nitronyl nitroxide radicalmolecule. The negative MR was attributed to a decrease inspin-flip scattering because an applied magnetic field alignslocalized spins, thereby limiting the spin-flip scattering ofconduction electrons with an increasing magnetic field.44,45 Inour case, we observed only positive MR, and thus, this scenariois excluded. Mitra et al. reported a positive MR of 140% at amagnetic field of 6 T in single-molecule junction ofperchlorotriphenylmethyl radicals, where asymmetricallycoupled junctions exhibited Kondo resonance, while sym-metrically coupled junctions showed high MR with bothpositive and negative signs. The origin of MR was attributed tothe spin polarization of the SOMO in combination with spin-dependent scattering at metal−molecule interfaces.25 InTEMPO-OPE molecules, the position of the radicalcomponent is far from the OPE backbone, resulting in equaltransmission probabilities for both up and down spins fortransport pathways through the backbone.27 Therefore, spinpolarization is less likely in TEMPO-OPE molecules. Previousexperiments suggested that a large MR value of up to 278% at amagnetic field of 4 T in single-molecule junctions of TEMPO-OPE could originate from a reduced coupling between themolecular orbitals and the metal electrodes.27 However, in thispresent study, the TEMPO-OPE molecules are not directlycoupled to the metal electrodes, and therefore, this scenario isalso ruled out. Warner et al. reported MR (both positive andnegative) in an asymmetrically coupled single-molecularjunction, where an iron phthalocyanine (FePc) molecule onFigure 3. Variation of MR in TEMPO-OPE samples with an applied magnetic field at (a) −0.75 V, which corresponds to the HOMO peak, and(b) 1.2 V, which agrees with the LUMO peak. The magnetic field varied within the range of ±7 T. The magnetic field sweep sequence was asfollows: 0 to +7 T (black line) → +7 to 0 T (red line) → 0 to −7 T (blue line) → −7 to 0 T (green line). The measurement temperature was fixedat 3 K in panels a and b. (c) I−V and (d) MR curves obtained at −0.75 V (HOMO) of TEMPO-OPE samples measured at a varied temperatureranging from 3 to 20 K.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648260https://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig3&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig3&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig3&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig3&ref=pdfpubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asa thin insulator layer (copper nitride) on a Cu(001) surfacewas probed by a STM tip (PtIr). The system configuration(metal/vacuum/molecule/insulator/metal) is similar to our Si-based double-tunnel junction embedding TEMPO-OPE. TheMR was assumed to arise from the shift of negative differentialresistance (NDR) in dI/dV spectra under a varied magneticfield. The shift of the NDR peak was attributed to the spin-polarized and non-degenerate resonant levels caused by theexchange splitting between the up and down spins of Fe dstates.22 However, we did not observe any such shift in the dI/dV peaks; rather, the HOMO peak was suppressed andeventually disappeared under the magnetic field. Themagnetic-field-dependent carrier transport in our sample israther attributed to the fading away of the dI/dV peakcorresponding to the HOMO level.To validate our results, we have performed DFT-basedcalculations of the effective Kohn−Sham single-particle energylevels (MOs) and density of states (DOS) of the TEMPO-OPE molecule embedded in a double-layer system composedof SiO2 and Al2O3. The simulations for the molecule in thetunnel junction were performed employing the PBE functional.The calculation details are described in section 7 of theSupporting Information. Figure 4a and Figure S7a show theoptimized structure of (Z)- and (E)-TEMPO-OPE, respec-tively, where the molecules are sandwiched between hydroxyl(OH−)-terminated SiO2 and Al2O3 layers. Termination of SiO2and Al2O3 surfaces by the OH− group is confirmed from X-rayphotoelectron spectroscopy measurements (Figures S3 andS4). The spin-density isosurfaces of (Z)- and (E)-TEMPO-OPE embedded between the hydro-α-SiO2 and the hydro-α-Al2O3 layers are shown in Figure 4b and Figure S7b. The spindensity is concentrated on the nitroxyl part of the TEMPOgroup, which means that the open-shell nature of TEMPO-OPE is well-preserved, even in insulating layers. The total DOSof the hydro-α-Al2O3/(Z)-TEMPO-OPE/hydro-α-SiO2 struc-ture and partial DOS (PDOS) of (Z)-TEMPO-OPE in thesandwiched structure are shown in Figure 4c and d,respectively. The DOS and PDOS curves indicate that energylevels of (Z)-TEMPO-OPE molecules are located in the energygaps of the two insulating layers and that they are well-separated from the continuum DOS of SiO2 and Al2O3. Theestimated HOMO−LUMO gap obtained from the PDOScurve (Figure 4d) is approximately 2.5 eV, which slightlydiffers from the value of 1.8 eV, as estimated from theexperimental dI/dV curve (Figure S10c). This differencewould be caused by the assumption of an idealized junctionconfiguration. In this calculation, a single TEMPO-OPE islocated at the interface of the two oxide layers. However, in thereal device, individual molecules are completely surrounded bythe Al2O3 layer, and further, each molecule can interact withthe other molecules. The detailed discussion is provided insection 9 of the Supporting Information.The most important observation from the PDOS ofTEMPO-OPE is that the SOMO is energetically split into anoccupied spin-up part and an unoccupied spin-down part, withthe occupied spin-up part located close to the Fermi level ofthe junction (Figure 4d and Figure S7d). It is noted thatrelated aspects of close spatial proximity, with a perpendicularmolecular configuration and potentially different mechanisticimplications, were discussed in our previous work on TEMPO-OPE in gold break junctions.27,46 Although we likely did notobserve the SOMO level in the experimental dI/dV spectra inour TEMPO-OPE samples due to its localization on TEMPO’sNO part, the SOMO may interact with the HOMO ofTEMPO-OPE, e.g., through the following mechanisms.Figure 4. (a) Optimized structure, (b) spin-density isosurfaces, (c) total DOS, and (d) partial DOS of (Z)-TEMPO-OPE embedded between thehydro-α-SiO2 and hydro-α-Al2O3 layers. The Fermi level is set to 0 eV, as shown by the green dashed line in DOS plots. The states below the Fermilevel are labeled as the occupied molecular orbital, and the states above Fermi level are labeled as the unoccupied molecular orbital of TEMPO-OPE.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648261https://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig4&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig4&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig4&ref=pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?fig=fig4&ref=pdfpubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-asA first possible scenario is derived from the Zeeman effect.When a magnetic field is applied, the SOMO energy is shiftedby the Zeeman effect. This change in the SOMO would betranslated to the HOMO but not to the LUMO. In the DFTcalculations of TEMPO-OPE in the tunnel junction, theHOMO is extended onto the radical part in (Z)-TEMPO-OPE[and to a lesser extent in (E)-TEMPO-OPE] and could thusinteract with the localized SOMO (Tables S4 and S6).Conversely, such mixing is much less pronounced for LUMO.Given the apparent sensitivity of the mixing betweenbackbone-centered and radical-centered contributions in theHOMO, it may be that magnetic-field-induced shifts in theSOMO can also change the shapes of the orbitals, in particularthe HOMO. This could then lead to the observed suppressionof the HOMO dI/dV peak with an increased magnetic field.Another scenario is the conformation or orientation changeof TEMPO-OPE molecules under magnetic fields. Althoughthe molecules are embedded in the oxide layers, a substantialportion of the molecules could be able to rearrange this way.Indeed, we confirmed a similar conformation change withphotochromic molecules even in the oxide layers in ourprevious work,37 and magnetic-field-induced structural changeshave also been reported in the literature.47−49 Such changes inboth the geometrical and electronic structures could bepossible in a magnetic field, affecting the subtle couplingbetween the SOMO and HOMO.Both mechanisms would be expected to exhibit a temper-ature-dependent transport. For the first scenario, a Zeemansplitting of a single unpaired electron would not be expected tobe measurable at more than 10 K. For the second one, theconformational or orientational effect under magnetic fieldswould likely be overridden by thermal fluctuations. However, itshould be noted that the suggested mechanisms are notdefinitive, and validating them would require furtherinvestigation.In summary, we evaluated magnetic-field-dependent carriertransport via discrete molecular levels of TEMPO-OPEmolecules incorporated into a Si-based double-tunnel junction.A significant reduction of the dI/dV peak corresponding to theHOMO level of TEMPO-OPE molecules was observed underexternal magnetic fields. A large positive MR of 400% at themaximum was achieved in TEMPO-OPE samples at amagnetic field of 7 T and a temperature of 3 K. However,no significant MR was observed in reference samples with non-radical OPE and without any molecules. DFT analysis suggeststhat magnetic-field-induced changes in the SOMO or in theconformation of the molecules can translate to the HOMOand the mixing between backbone-centered and radical-centered contributions in the HOMO. These effects maylead to the observed suppression of the HOMO conductancepeak under a magnetic field. Thus, our approach provides aneffective way to integrate magnetic functionality into Sidevices, offering great potential for large-scale integration ofmolecular spintronic devices.■ ASSOCIATED CONTENT*sı Supporting InformationThe Supporting Information is available free of charge athttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526.Experimental methods, FTIR spectra and ESR measure-ments of TEMPO-OPE films, XPS measurements of Si-based double-tunnel junctions, first-principles calcula-tions for molecular orbitals of TEMPO-OPE in vacuumas well as in the double-tunnel junction, vibrationalspectra and structural stabilities of optimized (Z)- and(E)-TEMPO-OPE, I−V and dI/dV curves of non-radicalOPE samples without the TEMPO group and referencedouble-tunnel junction without any molecules underexternal magnetic fields, histograms of the tunnelingcurrent and dI/dV peak positions corresponding to theHOMO and LUMO in TEMPO-OPE and non-radicalOPE samples, raw and 10-point smoothed data of I−Vand dI/dV of TEMPO-OPE samples, and I−V and dI/dV of TEMPO-OPE samples under magnetic fields(PDF)■ AUTHOR INFORMATIONCorresponding AuthorsCarmen Herrmann − Institute for Inorganic and AppliedChemistry, University of Hamburg, 20146 Hamburg,Germany; orcid.org/0000-0002-9496-0664;Email: carmen.herrmann@uni-hamburg.deThomas Huhn − Department of Chemistry, University ofKonstanz, 78457 Konstanz, Germany; orcid.org/0000-0001-6292-4215; Email: thomas.huhn@uni-konstanz.deRyoma Hayakawa − Semiconductor Functional Device Group,Research Center for Materials Nanoarchitectonics (MANA),National Institute for Materials Science (NIMS), Tsukuba,Ibaraki 305-0044, Japan; orcid.org/0000-0002-1442-8230; Email: hayakawa.ryoma@nims.go.jpAuthorsJayanta Bera − Semiconductor Functional Device Group,Research Center for Materials Nanoarchitectonics (MANA),National Institute for Materials Science (NIMS), Tsukuba,Ibaraki 305-0044, JapanTuhin Shuvra Basu − Semiconductor Functional DeviceGroup, Research Center for Materials Nanoarchitectonics(MANA), National Institute for Materials Science (NIMS),Tsukuba, Ibaraki 305-0044, JapanJannic Wolf − Department of Chemistry, University ofKonstanz, 78457 Konstanz, GermanyHaitao Zhang − Institute for Inorganic and AppliedChemistry, University of Hamburg, 20146 Hamburg,GermanyKazuhiro Marumoto − Department of Materials Science,Institute of Pure and Applied Sciences, University of Tsukuba,Tsukuba, Ibaraki 305-8573, Japan; orcid.org/0000-0001-9792-0775Yutaka Wakayama − Semiconductor Functional Device Group,Research Center for Materials Nanoarchitectonics (MANA),National Institute for Materials Science (NIMS), Tsukuba,Ibaraki 305-0044, Japan; orcid.org/0000-0002-0801-8884Complete contact information is available at:https://pubs.acs.org/10.1021/acs.nanolett.6c01526Author 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.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648262https://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?goto=supporting-infohttps://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.6c01526/suppl_file/nl6c01526_si_001.pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Carmen+Herrmann"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-9496-0664mailto:carmen.herrmann@uni-hamburg.dehttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Thomas+Huhn"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0001-6292-4215https://orcid.org/0000-0001-6292-4215mailto:thomas.huhn@uni-konstanz.dehttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Ryoma+Hayakawa"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-1442-8230https://orcid.org/0000-0002-1442-8230mailto:hayakawa.ryoma@nims.go.jphttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Jayanta+Bera"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Tuhin+Shuvra+Basu"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Jannic+Wolf"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Haitao+Zhang"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://pubs.acs.org/action/doSearch?field1=Contrib&text1="Kazuhiro+Marumoto"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0001-9792-0775https://orcid.org/0000-0001-9792-0775https://pubs.acs.org/action/doSearch?field1=Contrib&text1="Yutaka+Wakayama"&field2=AllField&text2=&publication=&accessType=allContent&Earliest=&ref=pdfhttps://orcid.org/0000-0002-0801-8884https://orcid.org/0000-0002-0801-8884https://pubs.acs.org/doi/10.1021/acs.nanolett.6c01526?ref=pdfpubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-as■ ACKNOWLEDGMENTSThis research was supported by the Research Center forMaterials Nanoarchitectonics (MANA) of the NationalInstitute for Materials Science (NIMS), Tsukuba, Japan, theJSPS KAKENHI (Grants 23K22802 and 24KF0270), and theAdvanced Research Infrastructure for Materials and Nano-technology in Japan (ARIM) of the Ministry of Education,Culture, Sports, Science and Technology (MEXT) (GrantJPMXP1223NM5170). Additionally, H.Z. and C.H. thank theGerman Research Foundation (DFG) for support via projectsHE-5675/6-1 and GRK 2536 NANOHYBRID, as well as theHigh-Performance Computing Center at University ofHamburg for computational resources.■ REFERENCES(1) Aggarwal, A.; Kaliginedi, V.; Maiti, P. K. Quantum Circuit Rulesfor Molecular Electronic Systems: Where Are We Headed Based onthe Current Understanding of Quantum Interference, Thermoelectric,and Molecular Spintronics Phenomena? Nano Lett. 2021, 21 (20),8532−8544.(2) Rocha, A. R.; García-suárez, V. M.; Bailey, S. W.; Lambert, C. J.;Ferrer, J.; Sanvito, S. Towards Molecular Spintronics. Nat. Mater.2005, 4 (4), 335−339.(3) Vincent, R.; Klyatskaya, S.; Ruben, M.; Wernsdorfer, W.;Balestro, F. Electronic Read-out of a Single Nuclear Spin Using aMolecular Spin Transistor. Nature 2012, 488 (7411), 357−360.(4) Miura, T.; Wasielewski, M. R. Manipulating PhotogeneratedRadical Ion Pair Lifetimes in Wirelike Molecules Using MicrowavePulses: Molecular Spintronic Gates. J. Am. Chem. Soc. 2011, 133 (9),2844−2847.(5) Nelson, J. N.; Krzyaniak, M. D.; Horwitz, N. E.; Rugg, B. K.;Phelan, B. T.; Wasielewski, M. R. Zero Quantum Coherence in aSeries of Covalent Spin-Correlated Radical Pairs. J. Phys. Chem. A2017, 121 (11), 2241−2252.(6) Rugg, B. K.; Krzyaniak, M. D.; Phelan, B. T.; Ratner, M. A.;Young, R. M.; Wasielewski, M. R. Photodriven Quantum Tele-portation of an Electron Spin State in a Covalent Donor−Acceptor−Radical System. Nat. Chem. 2019, 11 (11), 981−986.(7) Schmaus, S.; Bagrets, A.; Nahas, Y.; Yamada, T. K.; Bork, A.;Bowen, M.; Beaurepaire, E.; Evers, F.; Wulfhekel, W. GiantMagnetoresistance through a Single Molecule. Nat. Nanotechnol.2011, 6 (3), 185−189.(8) Aradhya, S. V.; Venkataraman, L. Single-Molecule Junctionsbeyond Electronic Transport. Nat. Nanotechnol. 2013, 8 (6), 399−410.(9) Puebla, J.; Kim, J.; Kondou, K.; Otani, Y. Spintronic Devices forEnergy-Efficient Data Storage and Energy Harvesting. Commun.Mater. 2020, 1 (1), 24.(10) Bazarnik, M.; Bugenhagen, B.; Elsebach, M.; Sierda, E.; Frank,A.; Prosenc, M. H.; Wiesendanger, R. Toward Tailored All-SpinMolecular Devices. Nano Lett. 2016, 16 (1), 577−582.(11) Ding, S.; Tian, Y.; Li, Y.; Zhang, H.; Zhou, K.; Liu, J.; Qin, L.;Zhang, X.; Qiu, X.; Dong, H.; Zhu, D.; Hu, W. Organic Single-CrystalSpintronics: Magnetoresistance Devices with High Magnetic-FieldSensitivity. ACS Nano 2019, 13 (8), 9491−9497.(12) Kuzhelev, A. A.; Trukhin, D. V.; Krumkacheva, O. A.;Strizhakov, R. K.; Rogozhnikova, O. Yu.; Troitskaya, T. I.; Fedin,M. V.; Tormyshev, V. M.; Bagryanskaya, E. G. Room-TemperatureElectron Spin Relaxation of Triarylmethyl Radicals at the X- and Q-Bands. J. Phys. Chem. B 2015, 119 (43), 13630−13640.(13) Sanvito, S. Spintronics Goes Plastic. Nat. Mater. 2007, 6 (11),803−804.(14) Pramanik, S.; Stefanita, C.-G.; Patibandla, S.; Bandyopadhyay,S.; Garre, K.; Harth, N.; Cahay, M. Observation of Extremely LongSpin Relaxation Times in an Organic Nanowire Spin Valve. Nat.Nanotechnol. 2007, 2 (4), 216−219.(15) Herrmann, C.; Solomon, G. C.; Ratner, M. A. Organic RadicalsAs Spin Filters. J. Am. Chem. Soc. 2010, 132 (11), 3682−3684.(16) Liu, J.; Zhao, X.; Al-Galiby, Q.; Huang, X.; Zheng, J.; Li, R.;Huang, C.; Yang, Y.; Shi, J.; Manrique, D. Z.; Lambert, C. J.; Bryce,M. R.; Hong, W. Radical-Enhanced Charge Transport in Single-Molecule Phenothiazine Electrical Junctions. Angew. Chem., Int. Ed.2017, 56 (42), 13061−13065.(17) Liu, J.; Isshiki, H.; Katoh, K.; Morita, T.; Breedlove, B. K.;Yamashita, M.; Komeda, T. First Observation of a Kondo Resonancefor a Stable Neutral Pure Organic Radical, 1,3,5-Triphenyl-6-oxoverdazyl, Adsorbed on the Au(111) Surface. J. Am. Chem. Soc.2013, 135 (2), 651−658.(18) Müllegger, S.; Rashidi, M.; Fattinger, M.; Koch, R. Surface-Supported Hydrocarbon π Radicals Show Kondo Behavior. J. Phys.Chem. C 2013, 117 (11), 5718−5721.(19) Komeda, T.; Isshiki, H.; Liu, J.; Zhang, Y.-F.; Lorente, N.;Katoh, K.; Breedlove, B. K.; Yamashita, M. Observation and ElectricCurrent Control of a Local Spin in a Single-Molecule Magnet. Nat.Commun. 2011, 2 (1), 217.(20) Zhang, Y.; Kahle, S.; Herden, T.; Stroh, C.; Mayor, M.;Schlickum, U.; Ternes, M.; Wahl, P.; Kern, K. Temperature andMagnetic Field Dependence of a Kondo System in the WeakCoupling Regime. Nat. Commun. 2013, 4 (1), 2110.(21) Patera, L. L.; Sokolov, S.; Low, J. Z.; Campos, L. M.;Venkataraman, L.; Repp, J. Resolving the Unpaired-Electron OrbitalDistribution in a Stable Organic Radical by Kondo ResonanceMapping. Angew. Chem., Int. Ed. 2019, 58 (32), 11063−11067.(22) Warner, B.; El Hallak, F.; Prüser, H.; Sharp, J.; Persson, M.;Fisher, A. J.; Hirjibehedin, C. F. Tunable Magnetoresistance in anAsymmetrically Coupled Single-Molecule Junction. Nat. Nanotechnol.2015, 10 (3), 259−263.(23) Yang, K.; Chen, H.; Pope, T.; Hu, Y.; Liu, L.; Wang, D.; Tao,L.; Xiao, W.; Fei, X.; Zhang, Y.-Y.; Luo, H.-G.; Du, S.; Xiang, T.;Hofer, W. A.; Gao, H.-J. Tunable Giant Magnetoresistance in aSingle-Molecule Junction. Nat. Commun. 2019, 10 (1), 3599.(24) Frisenda, R.; Gaudenzi, R.; Franco, C.; Mas-Torrent, M.;Rovira, C.; Veciana, J.; Alcon, I.; Bromley, S. T.; Burzurí, E.; van derZant, H. S. J. Kondo Effect in a Neutral and Stable All OrganicRadical Single Molecule Break Junction. Nano Lett. 2015, 15 (5),3109−3114.(25) Mitra, G.; Low, J. Z.; Wei, S.; Francisco, K. R.; Deffner, M.;Herrmann, C.; Campos, L. M.; Scheer, E. Interplay betweenMagnetoresistance and Kondo Resonance in Radical Single-MoleculeJunctions. Nano Lett. 2022, 22 (14), 5773−5779.(26) Mitra, G.; Zheng, J.; Schaefer, K.; Deffner, M.; Low, J. Z.;Campos, L. M.; Herrmann, C.; Costi, T. A.; Scheer, E. Conventionalversus Singlet-Triplet Kondo Effect in Blatter Radical MolecularJunctions: Zero-Bias Anomalies and Magnetoresistance. Chem. 2025,11 (9), 102500.(27) Hayakawa, R.; Karimi, M. A.; Wolf, J.; Huhn, T.; Zöllner, M. S.;Herrmann, C.; Scheer, E. Large Magnetoresistance in Single-RadicalMolecular Junctions. Nano Lett. 2016, 16 (8), 4960−4967.(28) Xie, Z.; Shi, S.; Liu, F.; Smith, D. L.; Ruden, P. P.; Frisbie, C. D.Large Magnetoresistance at Room Temperature in Organic MolecularTunnel Junctions with Nonmagnetic Electrodes. ACS Nano 2016, 10(9), 8571−8577.(29) Low, J. Z.; Kladnik, G.; Patera, L. L.; Sokolov, S.; Lovat, G.;Kumarasamy, E.; Repp, J.; Campos, L. M.; Cvetko, D.; Morgante, A.;Venkataraman, L. The Environment-Dependent Behavior of theBlatter Radical at the Metal−Molecule Interface. Nano Lett. 2019, 19(4), 2543−2548.(30) Tan, Y.; Li, J.; Li, S.; Yang, H.; Chi, T.; Shiring, S. B.; Liu, K.;Savoie, B. M.; Boudouris, B. W.; Schroeder, C. M. Enhanced ElectronTransport in Nonconjugated Radical Oligomers Occurs by Tunneling.Nano Lett. 2023, 23 (13), 5951−5958.(31) Chelli, Y.; Sandhu, S.; Daaoub, A. H. S.; Sangtarash, S.;Sadeghi, H. Controlling Spin Interference in Single Radical Molecules.Nano Lett. 2023, 23 (9), 3748−3753.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648263https://doi.org/10.1021/acs.nanolett.1c02390?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.1c02390?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.1c02390?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.1c02390?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/nmat1349https://doi.org/10.1038/nature11341https://doi.org/10.1038/nature11341https://doi.org/10.1021/ja110789q?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja110789q?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja110789q?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.jpca.7b00587?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.jpca.7b00587?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/s41557-019-0332-8https://doi.org/10.1038/s41557-019-0332-8https://doi.org/10.1038/s41557-019-0332-8https://doi.org/10.1038/nnano.2011.11https://doi.org/10.1038/nnano.2011.11https://doi.org/10.1038/nnano.2013.91https://doi.org/10.1038/nnano.2013.91https://doi.org/10.1038/s43246-020-0022-5https://doi.org/10.1038/s43246-020-0022-5https://doi.org/10.1021/acs.nanolett.5b04266?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.5b04266?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsnano.9b04449?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsnano.9b04449?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsnano.9b04449?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.jpcb.5b03027?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.jpcb.5b03027?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.jpcb.5b03027?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/nmat2050https://doi.org/10.1038/nnano.2007.64https://doi.org/10.1038/nnano.2007.64https://doi.org/10.1021/ja910483b?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja910483b?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1002/anie.201707710https://doi.org/10.1002/anie.201707710https://doi.org/10.1021/ja303510g?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja303510g?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/ja303510g?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/jp310316b?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/jp310316b?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/ncomms1210https://doi.org/10.1038/ncomms1210https://doi.org/10.1038/ncomms3110https://doi.org/10.1038/ncomms3110https://doi.org/10.1038/ncomms3110https://doi.org/10.1002/anie.201904851https://doi.org/10.1002/anie.201904851https://doi.org/10.1002/anie.201904851https://doi.org/10.1038/nnano.2014.326https://doi.org/10.1038/nnano.2014.326https://doi.org/10.1038/s41467-019-11587-xhttps://doi.org/10.1038/s41467-019-11587-xhttps://doi.org/10.1021/acs.nanolett.5b00155?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.5b00155?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.2c01199?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.2c01199?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.2c01199?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1016/j.chempr.2025.102500https://doi.org/10.1016/j.chempr.2025.102500https://doi.org/10.1016/j.chempr.2025.102500https://doi.org/10.1021/acs.nanolett.6b01595?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.6b01595?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsnano.6b03853?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsnano.6b03853?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.9b00275?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.9b00275?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.3c00978?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.3c00978?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.2c05068?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-aspubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-as(32) Baum, T. Y.; Fernández, S.; Peña, D.; van der Zant, H. S. J.Magnetic Fingerprints in an All-Organic Radical Molecular BreakJunction. Nano Lett. 2022, 22 (20), 8086−8092.(33) Yoshida, K.; Hamada, I.; Sakata, S.; Umeno, A.; Tsukada, M.;Hirakawa, K. Gate-Tunable Large Negative Tunnel Magneto-resistance in Ni−C60−Ni Single Molecule Transistors. Nano Lett.2013, 13 (2), 481−485.(34) Rakhmilevitch, D.; Sarkar, S.; Bitton, O.; Kronik, L.; Tal, O.Enhanced Magnetoresistance in Molecular Junctions by GeometricalOptimization of Spin-Selective Orbital Hybridization. Nano Lett.2016, 16 (3), 1741−1745.(35) Yang, K.; Chen, H.; Pope, T.; Hu, Y.; Liu, L.; Wang, D.; Tao,L.; Xiao, W.; Fei, X.; Zhang, Y.-Y.; Luo, H.-G.; Du, S.; Xiang, T.;Hofer, W. A.; Gao, H.-J. Tunable Giant Magnetoresistance in aSingle-Molecule Junction. Nat. Commun. 2019, 10 (1), 3599.(36) Hayakawa, R.; Hiroshiba, N.; Chikyow, T.; Wakayama, Y.Single-Electron Tunneling through Molecular Quantum Dots in aMetal-Insulator-Semiconductor Structure. Adv. Funct. Mater. 2011, 21(15), 2933−2937.(37) Hayakawa, R.; Higashiguchi, K.; Matsuda, K.; Chikyow, T.;Wakayama, Y. Photoisomerization-Induced Manipulation of Single-Electron Tunneling for Novel Si-Based Optical Memory. ACS Appl.Mater. Interfaces 2013, 5 (21), 11371−11376.(38) Hayakawa, R.; Chikyow, T.; Wakayama, Y. Vertical ResonantTunneling Transistors with Molecular Quantum Dots for Large-ScaleIntegration. Nanoscale 2017, 9 (31), 11297−11302.(39) Bera, J.; Kabdulov, M.; Wakayama, Y.; Huhn, T.; Hayakawa, R.Multilevel Resonant Tunneling through Purely Organic RadicalMolecules in a Si-Based Double-Tunnel Junction. ACS Appl. Mater.Interfaces 2025, 17 (15), 23018−23024.(40) Mannini, M.; Bertani, F.; Tudisco, C.; Malavolti, L.; Poggini, L.;Misztal, K.; Menozzi, D.; Motta, A.; Otero, E.; Ohresser, P.;Sainctavit, P.; Condorelli, G. G.; Dalcanale, E.; Sessoli, R. MagneticBehaviour of TbPc2 Single-Molecule Magnets Chemically Grafted onSilicon Surface. Nat. Commun. 2014, 5 (1), 4582.(41) Pellegrino, G.; Motta, A.; Cornia, A.; Spitaleri, I.; Fragala,̀ I. L.;Condorelli, G. G. One Pot Grafting of Tetrairon(III) Single MoleculeMagnets on Silicon. Polyhedron 2009, 28 (9), 1758−1763.(42) Chang, C.-C.; Sun, K. W.; Lee, S.-F.; Kan, L.-S. Self-AssembledMolecular Magnets on Patterned Silicon Substrates: Bridging Bio-Molecules with Nanoelectronics. Biomaterials 2007, 28 (11), 1941−1947.(43) Kohn, W.; Sham, L. J. Self-Consistent Equations IncludingExchange and Correlation Effects. Phys. Rev. 1965, 140 (4A), A1133−A1138.(44) Sugawara, T.; Minamoto, M.; Matsushita, M. M.; Nickels, P.;Komiyama, S. Cotunneling Current Affected by Spin-Polarized WireMolecules in Networked Gold Nanoparticles. Phys. Rev. B 2008, 77(23), 235316.(45) Nickels, P.; Matsushita, M. M.; Minamoto, M.; Komiyama, S.;Sugawara, T. Controlling Co-Tunneling Currents in NanoparticleNetworks Using Spin-Polarized Wire Molecules. Small 2008, 4 (4),471−475.(46) Zhang, H.; Herrmann, C. Interaction between TEMPORadicals and Gold Surfaces. J. Phys. Chem. C 2023, 127 (38),19202−19212.(47) Kolotovska, V.; Friedrich, M.; Zahn, D. R. T.; Salvan, G.Magnetic Field Influence on the Molecular Alignment of VanadylPhthalocyanine Thin Films. J. Cryst. Growth 2006, 291 (1), 166−174.(48) Takami, S.; Furumi, S.; Shirai, Y.; Sakka, Y.; Wakayama, Y.Impact of Magnetic Field on Molecular Alignment and ElectricalConductivity in Phthalocyanine Nanowires. J. Mater. Chem. 2012, 22(17), 8629−8633.(49) Uchida, Y.; Tamura, R.; Ikuma, N.; Shimono, S.; Yamauchi, J.;Shimbo, Y.; Takezoe, H.; Aoki, Y.; Nohira, H. Magnetic-Field-Induced Molecular Alignment in an Achiral Liquid Crystal Spin-Labeled by a Nitroxyl Group in the Mesogen Core. J. Mater. Chem.2009, 19 (3), 415−418.Nano Letters pubs.acs.org/NanoLett Letterhttps://doi.org/10.1021/acs.nanolett.6c01526Nano Lett. 2026, 26, 8257−82648264https://doi.org/10.1021/acs.nanolett.2c02326?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.2c02326?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/nl303871x?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/nl303871x?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.5b04674?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.nanolett.5b04674?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/s41467-019-11587-xhttps://doi.org/10.1038/s41467-019-11587-xhttps://doi.org/10.1002/adfm.201100220https://doi.org/10.1002/adfm.201100220https://doi.org/10.1021/am403616m?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/am403616m?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1039/C7NR02463Khttps://doi.org/10.1039/C7NR02463Khttps://doi.org/10.1039/C7NR02463Khttps://doi.org/10.1021/acsami.5c00839?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acsami.5c00839?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1038/ncomms5582https://doi.org/10.1038/ncomms5582https://doi.org/10.1038/ncomms5582https://doi.org/10.1016/j.poly.2008.11.049https://doi.org/10.1016/j.poly.2008.11.049https://doi.org/10.1016/j.biomaterials.2006.11.048https://doi.org/10.1016/j.biomaterials.2006.11.048https://doi.org/10.1016/j.biomaterials.2006.11.048https://doi.org/10.1103/PhysRev.140.A1133https://doi.org/10.1103/PhysRev.140.A1133https://doi.org/10.1103/PhysRevB.77.235316https://doi.org/10.1103/PhysRevB.77.235316https://doi.org/10.1002/smll.200700461https://doi.org/10.1002/smll.200700461https://doi.org/10.1021/acs.jpcc.3c04401?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1021/acs.jpcc.3c04401?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-ashttps://doi.org/10.1016/j.jcrysgro.2006.02.016https://doi.org/10.1016/j.jcrysgro.2006.02.016https://doi.org/10.1039/c2jm30179bhttps://doi.org/10.1039/c2jm30179bhttps://doi.org/10.1039/B809502Ghttps://doi.org/10.1039/B809502Ghttps://doi.org/10.1039/B809502Gpubs.acs.org/NanoLett?ref=pdfhttps://doi.org/10.1021/acs.nanolett.6c01526?urlappend=%3Fref%3DPDF&jav=VoR&rel=cite-as