Realization of a double-slit SQUID geometry by Fermi arc surface states in a WTe 2 Weyl semimetal
O.O. Shvetsov, A. Kononov, A.V. Timonina, N.N. Kolesnikov, E.V. Deviatov
aa r X i v : . [ c ond - m a t . m e s - h a ll ] J a n Realization of a double-slit SQUID geometry by Fermi arc surface states in a WTe Weyl semimetal.
O.O. Shvetsov, A. Kononov, A.V. Timonina, N.N. Kolesnikov, and E.V. Deviatov Institute of Solid State Physics of the Russian Academy of Sciences,Chernogolovka, Moscow District, 2 Academician Ossipyan str., 142432 Russia (Dated: January 30, 2018)We experimentally study electron transport between two superconducting indium leads, coupledto the WTe crystal surface. WTe is characterized by presence of Fermi arc surface states, as apredicted type-II Weyl semimetal candidate. We demonstrate Josephson current in unprecedentedlylong 5 µ m In-WTe -In junctions, which is confirmed by I − V curves evolution with temperatureand magnetic field. The Josephson current is mostly carried by the topological surface states, whichwe demonstrate in a double-slit SQUID geometry, realized by coupling the opposite WTe crystalsurfaces. PACS numbers: 73.40.Qv 71.30.+h
I. INTRODUCTION
Recent renewal of interest to semimetals is mostly con-nected with topological effects. Weyl semimetals areconductors whose low-energy bulk excitations are Weylfermions . Like other topological materials , Weylsemimetals are characterized by topologically protectedmetallic surface states, which are known as Fermi arc sur-face states. This concept of the Fermi arc surface stateshas now been extended to type II materials , like MoTe and WTe , which contain electron and hole pockets .The non-trivial properties of these materials have beendemonstrated in magnetotransport experiments .Topological materials exhibit non-trivial physics inproximity with a superconductor . For the topolog-ical insulators , it is expected to allow topologicalsuperconductivity regime , which stimulates a searchfor Majorana fermions . In the case of Weyl semimetals,the proximity is predicted to produce specular Andreevreflection , or even to superconducting correlationswithin a semimetal . Moreover, topological trans-port is responsible for Josephson current in 1-2 µ m longsuperconductor-normal-superconductor (SNS) junctionsin graphene .The edge current contribution can be retrieved evenfor systems with conducting bulk by analyzing Joseph-son current suppression in low magnetic fields . Themaximum supercurrent is periodically modulated, withperiod which is defined by the magnetic flux quantumΦ = π ¯ hc/e . It is well known, that the homogeneoussupercurrent density in the conductor corresponds to asingle-slit Fraunhofer pattern . As the edge currentsemerge in a two-dimensional topological system, the sinu-soidal oscillation pattern appears , which is a finger-print of a superconducting quantum interference device(SQUID) . It is therefore reasonable to study Josephsoncurrent suppression in a long SNS junction on a three-dimensional Weyl semimetal surface.Here, we experimentally study electron transport be-tween two superconducting indium leads, coupled to the WTe crystal surface. WTe is characterized by pres-ence of Fermi arc surface states, as a predicted type-IIWeyl semimetal candidate. We demonstrate Josephsoncurrent in unprecedentedly long 5 µ m In-WTe -In junc-tions, which is confirmed by I − V curves evolution withtemperature and magnetic field. The Josephson currentis mostly carried by the topological surface states, whichwe demonstrate in a double-slit SQUID geometry, real-ized by coupling the opposite WTe crystal surfaces. II. SAMPLES AND TECHNIQUE
WTe compound was synthesized from elements by re-action of metal with tellurium vapor in the sealed silicaampule. The WTe crystals were grown by the two-stageiodine transport , that previously was successfully ap-plied for growth of other metal chalcogenides likeNbS and CrNb S . The WTe composition is veri-fied by energy-dispersive X-ray spectroscopy. The X-raydiffraction (Oxford diffraction Gemini-A, MoK α ) con-firms P mn I orthorhombic single crystal WTe withlattice parameters a = 3 . b = 6 . c = 14 . crystals demon-strate large, non-saturating positive magnetoresistanceup to 14 T field, as it has been shown for WTe Weylsemimetal .A sample sketch is presented in Fig. 1. Superconduct-ing leads are formed by lift-off technique after thermalevaporation of 100 nm indium on the insulating SiO sub-strate. A WTe single crystal ( ≈ . × µ m × . µ mdimensions) is weakly pressed to the indium leads pat-tern, so that planar In-WTe junctions (10 × ( ≈ µ m )are formed at the bottom surface of the crystal WTe inFig. 1.Charge transport is investigated between two super-conducting indium leads in a four-point technique. Anexample of electrical connections is presented in Fig. 1 :the S1 electrode is grounded; a current I is fed throughthe S2; a voltage drop V is measured between these S1 FIG. 1. (Color online) Sketch of the sample with indium con-tacts to the bottom surface of a WTe crystal (not to scale).Right inset demonstrates top-view images of the indium leadsand WTe crystal. 10 µ m wide indium superconducting leads(S1-S4) are formed on the insulating SiO substrate. 5 µ mlong In-WTe -In junctions are fabricated by weak pressing aWTe crystal ( ≈ . × µ m × . µ m) to the indiumleads pattern. Charge transport is investigated between twosuperconducting electrodes in a four-point technique: the S1electrode is grounded; a current I is fed through the S2; a volt-age drop V is measured between these S1 and S2 electrodesby independent wires because of low normal In-WTe -In re-sistance. and S2 electrodes by independent wires. In this connec-tion scheme, all the wire resistances are excluded, whichis necessary for low-impedance In-WTe -In junctions (be-low 0.5 Ohm normal resistance in the present experi-ment). The measurements are performed in standardHe cryostat in the temperature range 1.4 K – 4.2 K.The indium leads are superconducting below the criticaltemperature T c ≈ . K . III. EXPERIMENTAL RESULTS
To obtain I − V characteristics, we sweep the dc current I and measure the voltage drop V . Fig. 2 presents I − V examples in two different experimental configurations.In zero magnetic field, at low temperature 1.4 K < T c ,transport between two 5 µ m spaced contacts S1 and S2 isof clear Josephson-like behavior , as shown by the bluecurve in Fig. 2: (i) by the four-point connection schemewe directly demonstrate zero resistance region at low cur-rents; (ii) the non-zero resistance appears as sharp jumpsat current values ± I c ≈ I − V branches inFig. 2. Because of similar preparation technique, dif-ferent samples demonstrate even quantitatively similarbehavior: the obtained I c values differ within 10% ofmagnitude for different samples and in different coolings.In contrast, the resistance is always finite between 80 µ mseparated S1 and S3 indium leads, see the red curve in -15 -10 -5 0 5 10 I (mA) -0.8-0.400.40.8 V ( m V ) S1-S3 junctionS1-S2 junction-30 -15 0 15 30 I (mA) -15015 V ( m V ) T=4.2 KT=1.4 K -4 0 4 I (mA) d V / d I ( m Ω ) B=31 mTB=0S1-S2, B=0T=1.4 K, B=0 S1-S2, T=1.4 K I c FIG. 2. (Color online) Examples of I − V characteristics in twodifferent experimental configurations in zero magnetic field at1.4 K < T c . The blue curve is obtained for 5 µ m long In-WTe -In junction between the superconducting leads S1 andS2, as depicted in Fig. 1. A clear Josephson behavior canbe seen: there is no resistance at low currents, it appearsabove ± I c ≈ µ m separated S1 and S3 indium leads, see the redcurve. Left inset: superconductivity suppression in indiumleads S1 and S2 by current ≈ ±
30 mA (the solid curve, T =1 . T =4.2 K > T c ) inzero magnetic field. Right inset: dV /dI ( I ) characteristics forthe S1-WTe -S2 junction at minimal T = 1 . B c = 31 mT(the green one). Fig. 2.Even for the closely-spaced contacts S1 and S2, I − V curve can be switched to standard Ohmic behavior, ifthe indium superconductivity is suppressed by temper-ature or high current (above ≈
30 mA for the presentdimensions), as depicted in the left inset to Fig. 2. Thezero-resistance state can also be suppressed by magneticfield, as it is demonstrated in the right inset to Fig. 2.Thus, we demonstrate in Fig. 2, that two supercon-ducting contacts induce Josephson current in an unprece-dentedly long 5 µ m >> ξ In In-WTe -In junction, where ξ In ≈
300 nm is the indium correlation length .As usual for SNS junctions, an important informationcan be obtained from the maximum supercurrent I c sup-pression by temperature and magnetic field. To analyze I c ( B, T ) behavior, we use dV /dI ( I ) characteristics likeones presented in the right inset to Fig. 2: the dc cur-rent is additionally modulated by a low ac component(100 nA, 10 kHz), an ac part of V ( ∼ dV /dI ) is detectedby a lock-in amplifier. We have checked, that the lock-insignal is independent of the modulation frequency in the6 kHz – 30 kHz range, which is defined by applied acfilters. To obtain I c values with high accuracy for given( B, T ) values, we sweep current I ten times from zero(superconducting state) to above I c (resistive state), andthen determine I c as the average value of dV /dI jump B (mT) I c / I c ( B = ) T (K) I c / I c ( . K ) B || planeB ⊥ plane 0 0.2 0.40.9511.05 I c / I c ( B = ) a) B=0 B ⊥ planetwo different coolings b) T=1.4 K FIG. 3. (Color online) Suppression of the maximum super-current I c by temperature (a) or magnetic field (b). (a) I c ( T )monotonously falls to zero at 3.5 K, which well correspondsto the indium critical temperature (different symbols refer todifferent sample coolings). The curves are obtained in zeromagnetic field. (b) I c ( B ) suppression pattern crucially de-pends on the magnetic field orientation to the In-WTe -Injunction plane: it is extremely strong for the perpendicularfield, while it is very slow (within 10% until the critical field)for the parallel orientation. For both orientations, there areoscillations in I c ( B ), the period is much higher for the parallelmagnetic field (2 mT and 0.1 mT, respectively). The curvesare obtained at minimal 1.4 K temperature. positions in different sweeps.The results are presented in Fig. 3. I c ( T )monotonously falls to zero at 3.5 K, which well corre-sponds to the indium critical temperature , see Fig. 3(a). However, I c ( T ) does not demonstrate the exponen-tial decay, which is expected for long L >> ξ In SNSjunctions. Instead, the experimental I c ( T ) dependenceis even slower than the linear function of T in Fig. 3 (a),as it is usually realized for the short L < ξ In junctionregime .To our surprize, I c ( B ) suppression pattern cruciallydepends on the magnetic field orientation to the In-WTe -In junction plane, see Fig. 3 (b). If the magneticfield is perpendicular to the plane, strong suppression of I c ( B ) is observed, which is usual for standard Josephsoneffect . In contrast, I c ( B ) is diminishing very slowly(within 10% until the indium critical field) for the par-allel magnetic field. For both orientations, we observeequidistant I c ( B ) oscillations within 5% of I c magnitude,see also inset to Fig. 3 (b). The oscillations are charac-terized by high ∆ B = 2 mT period for the parallel fieldand by low ∆ B = 0 . I c ( B ) supressionin parallel magnetic fields resembles double-slit SQUIDbehavior , so the surface transport is important inWTe . FIG. 4. (Color online) Schematic diagram of a double-slitSQUID geometry, realized by Weyl surface states in WTe semimetal. Superconductivity is proximity induced near theIn leads (blue regions) between the opposite sample surfaces,since d ∼ ξ . Josephson current (denoted by arrows) is trans-ferred by top and bottom surface states simultaneously. Thus,there are two parallel SNS junctions, which form a double-slitSQUID geometry. The effective SQUID area is denoted byyellow (see the main text for details). IV. DISCUSSION
The experimental I c ( T, B ) dependencies allows us tounambiguously identify the topological effects.Slow I c ( T ) decay has been reported in long 1.5-2 µ m >> ξ SNS junctions on graphene, and has been con-nected with topological transport . WTe is regardedas type-II Weyl semimetal , which contains topologicalFermi arc surface states. These surface states are usuallydecoupled from the bulk . On the other hand, Weylsurface states inherit the chiral property of the Chern in-sulator edge states . Because of topological protection,they can efficiently transfer the Josephson current. Thismight be a reason to have slow I c ( T ) dependence in our nominally long >> ξ In ≈
300 nm devices.We should connect I c ( B ) behavior with the distribu-tion of the Josephson current within the WTe crystal,see Fig. 4. The sample thickness is comparable with in-dium coherence length ∼ ξ In ≈
300 nm, so the regionsof proximity-induced superconductivity couples two op-posite sample surfaces near the In leads (blue regions inFig. 4). The Josephson current is transferred by topo-logical surface states, so there are two parallel weak linksbetween the superconducting leads. In other words, anon-symmetric double-slit SQUID geometry is real-ized, see Fig. 4.The experimental I c ( B ) suppression pattern well cor-responds to the double-slit SQUID with two non-equivalent weak links. Parallel magnetic field induces aphase shift between the opposite WTe surfaces, so itcontrols the magnetic flux through the effective SQUIDarea, see Fig. 4. The latter can be estimated ( S ∆ B ∼ Φ )from ∆ B = 2 mT as S ≈ − cm , which gives 0.3 µ msample thickness for our 5 µ m long junctions. This esti-mation is in good correspondence with the known WTe crystal thickness.If the magnetic field is perpendicular to the WTe crys-tal plane, there is no phase shift between the the oppositesample surfaces. Instead, I c ( B ) reflects homogeneous su-percurrent distribution within the surface state in twoequivalent SNS junctions. Thus, we observe strong I c ( B )suppression in Fig. 3 (b) with oscillations in low fields ,which reflects the effective junction area S . The experi-mentally observed period ∆ B = 0 . S ≈ × − cm , i.e. to the ≈ µ m × µ m SNS junctions, which well correspond tothe sample dimensions. V. CONCLUSION
As a conclusion, we experimentally study electrontransport between two superconducting indium leads,coupled to the WTe crystal surface. WTe is charac- terized by presence of Fermi arc surface states, as a pre-dicted type-II Weyl semimetal candidate. We demon-strate Josephson current in unprecedentedly long 5 µ mIn-WTe -In junctions, which is confirmed by I − V curvesevolution with temperature and magnetic field. TheJosephson current is mostly carried by the topologi-cal surface states, which we demonstrate in a double-slit SQUID geometry, realized by coupling the oppositeWTe crystal surfaces. ACKNOWLEDGMENTS
We wish to thank Ya. Fominov, V.T. Dolgopolov,V.A. Volkov for fruitful discussions, and S.S Khasanov forX-ray sample characterization. We gratefully acknowl-edge financial support by the RFBR (project No. 16-02-00405) and RAS. As a recent review see N. P. Armitage, E. J. Mele, andAshvin Vishwanath, Reviews of Modern Physics (2017),arXiv:1705.01111 M. Z. Hasan and C. L. Kane, Rev. Mod. Phys. 82, 3045(2010). X.-L. Qi and S.-C. Zhang, Rev. Mod. Phys. 83, 1057(2011). A. Bansil, H. Lin, and T. Das, Rev. Mod. Phys. 88, 021004(2016). C.-K. Chiu, J. C. Teo, A. P. Schnyder, and S. Ryu, Rev.Mod. Phys. 88, 035005 (2016). Deng K., Wan G., Deng P., Zhang K., Ding S., Wang E.,Yan M., Huang H., Zhang H., Xu Z., Denlinger J., FedorovA., Yang H., Duan W., Yao H., Wu Y., Fan S., Zhang H.,Chen X., Zhou S., Nat. Phys. 12, 11051110 (2016). Chenlu Wang et al., Phys. Rev. B 94, 241119(R) Mazhar N. Ali, Jun Xiong, Steven Flynn, Jing Tao, QuinnD. Gibson, Leslie M. Schoop, Tian Liang, Neel Hal-dolaarachchige, Max Hirschberger, N. P. Ong and R. J.Cava Nature 514, 205 (2014). doi:10.1038/nature13763 Wang, Y. et al., Nat. Commun. 7, 13142 (2016). doi:10.1038/ncomms13142 A. Kononov, V.A. Kostarev, B.R. Semyagin, V.V. Pre-obrazhenskii, M.A. Putyato, E.A. Emelyanov, and E.V.Deviatov, Physical Review B 96, 245304 (2017). DOI:10.1103/PhysRevB.96.245304 A. Kononov, S. V. Egorov, Z. D. Kvon, N. N. Mikhailov,S. A. Dvoretsky, and E. V. Deviatov, Phys. Rev. B 93,041303(R) (2016) A. Kononov, S. V. Egorov, N. Titova, Z. D. Kvon, N. N.Mikhailov, S. A. Dvoretsky, E. V. Deviatov, JETP Lett.,101 , 41 (2015). S. Murakami, N. Nagaosa, S.-C. Zhang, Phys. Rev. Lett.93, 156804 (2004). C. L. Kane, E. J. Mele, Phys. Rev. Lett. 95, 146802 (2005). B. A. Bernevig, S.-C. Zhang, Phys. Rev. Lett. 96, 106802(2006). L. Fu and C. L. Kane, Phys. Rev. Lett. 100, 96407 (2008). S. Hart, H. Ren, T. Wagner, Ph. Leubner, M. M¨uhlbauer,C. Br¨une, H. Buhmann, L. W. Molenkamp and A. Yacoby, Nature Physics 10, 638643 (2014). For recent reviews, see C. W. J. Beenakker, Annu. Rev.Con. Mat. Phys. 4, 113 (2013) and J. Alicea, Rep. Prog.Phys. 75, 076501 (2012). Wei Chen,Liang Jiang,R. Shen,L. Sheng,B. G. Wang,D. Y.Xing EPL 103, 27006 (2013) C. W. J. Beenakker, Physical Review Letters 97 (2006). C. W. J. Beenakker, Reviews of Modern Physics 80, 1337(2008). T. Meng and L. Balents, Phys. Rev. B 86, 054504 (2012). G. Y. Cho, J. H. Bardarson, Y.-M. Lu, and J. E. Moore,Phys. Rev. B 86, 214514 (2012). H. Wei, S. P. Chao, and V. Aji, Phys. Rev. B 89, 014506(2014). V. E. Calado, S. Goswami, G. Nanda, M. Diez, A. R.Akhmerov, K. Watanabe, T. Taniguchi, T. M. Klapwijk &L. M. K. Vandersypen, Nature Nanotechnology 10, 761764(2015). doi:10.1038/nnano.2015.156 I.V. Borzenets, F. Amet, C.T. Ke, A.W. Drae-los, M.T. Wei, A. Seredinski, K. Watanabe, T.Taniguchi, Y. Bomze, M. Yamamoto, S. Tarucha, andG. Finkelstein, Phys. Rev. Lett. 117, 237002 (2016),http://dx.doi.org/10.1103/PhysRevLett.117.237002 Sean Hart, Hechen Ren, Timo Wagner, Philipp Leubner,Mathias Mhlbauer, Christoph Brne, Hartmut Buhmann,Laurens W. Molenkamp & Amir Yacoby, Nature Physics10, 638643 (2014), doi:10.1038/nphys3036 Vlad S. Pribiag, Arjan J. A. Beukman, Fanming Qu, MajaC. Cassidy, Christophe Charpentier, Werner Wegscheider& Leo P. Kouwenhoven, Nature Nanotechnology 10, 593(2015) M. Tinkham, Introduction to Superconductivity (2d ed.,McGrawHill, New York, 1996). E. B. Borisenko, V. A. Berezin, N. N. Kolesnikov, V. K.Gartman, D. V. Matveev, O. F. Shakhlevich, Physics ofthe Solid State, 59, 1310, (2017). A. Sidorov, A.E. Petrova, A.N. Pinyagin, N.N. Kolesnikov,S.S. Khasanov, S.M. Stishov, JETP, 122, 1047, (2016). A. M. Toxen Phys. Rev. 123, 442 (1961). I.O. Kulik and A.N. Omelyanchuk, Fiz. Nizk. Temp. 3, 945(1977) [Sov. J. Low Temp. Phys. 3, 459 (1977)]. I. O. Kulik, Sov. Phys. JETP 30, 944 (1970). Marco Bianchi, Richard C. Hatch, Jianli Mi, Bo Brummer-stedt Iversen, and Philip Hofmann, Phys. Rev. Lett. 107,086802 (2011)36