Few-hole double quantum dot in an undoped GaAs/AlGaAs heterostructure
aa r X i v : . [ c ond - m a t . m e s - h a ll ] D ec Few-hole double quantum dot in an undoped GaAs/AlGaAs heterostructure
L. A. Tracy, a) T. W. Hargett, and J. L. Reno Sandia National Laboratories, Albuquerque, New Mexico 87185, USA Center for Integrated Nanotechnologies, Sandia National Laboratories, Albuquerque, New Mexico 87185,USA (Dated: 27 March 2018)
We demonstrate a hole double quantum dot in an undoped GaAs/AlGaAs heterostructure. The interdotcoupling can be tuned over a wide range, from formation of a large single dot to two well-isolated quantumdots. Using charge sensing, we show the ability to completely empty the dot of holes and control the chargeoccupation in the few-hole regime. The device should allow for control of individual hole spins in single anddouble quantum dots in GaAs.One of the leading candidates for a solid-state quan-tum bit is the spin of a single electron confinedin a semiconductor . The pioneering initial experi-ments demonstrating coherent control of individual elec-tron spins in quantum dots utilized high-mobility two-dimensional (2D) electron systems in GaAs/AlGaAsheterostructures . The major source of decoherence insuch experiments is coupling between electron spins andnuclear spins in the host GaAs semiconductor . It hasbeen proposed that hole spins in GaAs would be bettersuited for such experiments due to a lesser coupling be-tween hole and nuclear spins . The stronger spin-orbitinteraction for holes, as compared to electrons, may alsoprovide a means for electrical spin manipulation .To date, experiments on single spins in semiconductorquantum dots have primarily focused on electron spins.Confinement of single hole spins in nanowire devices and self-assembled quantum dots has been demon-strated, but fabrication of quatum dots using conven-tional 2D heterostructures has potential advantages interms of flexibility of tuning the confinement potentialand for fabrication of mulitple dot devices . One ofthe main reasons for the lack of experiments on holequantum dots in GaAs is the difficulty of fabricatingelectrically stable nanostructures (such as quantum dots)in p-doped GaAs/AlGaAs heterostructures . How-ever, undoped, enhancement-mode devices provide analternate route to fabricating p-type nanostructures inGaAs. High-mobility two-dimensional hole systems havealready been demonstrated in undoped devices andrecent results have shown that a stable many-hole quan-tum dot can be formed in an enhancement-mode devicein a (311)A oriented heterostructure with a p-doped caplayer .In this article, we report fabrication and measurementof a hole double quantum dot (DQD) in an undoped (100)oriented GaAs/AlGaAs heterostructure. The mean freepath in similarly processed bulk 2D devices at T = 4K, at density of p = 2 × cm − is ∼ µ m, whichis larger than typical nanostructure dimensions, aidingin the formation of low-disorder few-hole nanostructures. a) Electronic mail: [email protected] (a) (b)
Fig. 1
FIG. 1. (a) Scanning electron micrograph of partially pro-cessed device, showing Ti/Au gates on GaAs/AlGaAs het-erostructure surface used to form a quantum dot and QPCcharge sensor. (b) Sketch of cross section of left half of de-vice.
The gate design provides a high degree of tunability,allowing for independent control over individual dot oc-cupation and tunnel barriers, as well as the ability touse a nearby quantum point contacts (QPCs) to sensedot charge occupation. We show the ability to controlthe coupling between dots, tuning the device across thetransition from one large dot to two well-isolated quan-tum dots. Using charge sensing, we determine the chargeoccupation of the DQD and demonstrate operation of thedevice in the few-hole regime.Figure 1(a) shows a scanning electron micrograph ofTi/Au gates on the surface of a GaAs/AlGaAs het-erostructure (VA0582) for a partially processed device.Figure 1(b) shows a schematic cross section of the left halfof the final device. The upper Al gate is used to accumu-late holes at the GaAs/Al x Ga − x As (x = 0.5) interface,100 nm below the heterostructure surface, as sketched inFig. 1(b). The lower, patterened Ti/Au (10 nm Ti, 40 nmAu) gates are used to locally deplete to define the QPCand DQD. Ohmic contacts to the hole layer are formedvia AuBe evaporation and anneal. A 110 nm thick layerof Al O grown via atomic layer deposition electricallyisolates the upper gate from AuBe Ohmic contacts andlower Ti/Au gates . The device conductance is exper-imentally determined via standard low-frequency lock-inmeasurements with an rms ac source-drain bias of 50 -100 µ V. For all measurements shown, the Al upper gatevoltage is held constant at -6 V. The device was measuredin a He refrigerator with a temperature of T = 380 mK.Figure 2(a) shows quantum dot conductance versus left (a) (b) Fig. 2
FIG. 2. Dot conductance for a large single dot for fixed gatevoltages V CP = 0 V, V LQPC = 0.3 V, V L = 0.2 V, and V CP = 0.275 V. (a) Dot conductance G dot vs. V LP and V RP (b) dI dot /dV sd versus V sd and V LP showing Coulomb diamonds. and right plunger gate voltages V LP and V RP with V CP = 0 V. In this regime, the device behaves like a largesingle dot, with roughly equal capacitance between thedot and gates LP and RP, where C dot − LP = 3 . C dot − RP = 3 . . In Fig. 2(b) we show a stabilitydiagram with Coulomb diamonds for this dot. The lastvisible diamond indicates a dot charging energy of ∼ . E orb ∼ . E orb ∼ π ~ /m ∗ l , we obtain arough estimate of the dot size l ∼
100 nm. Althoughthe effective hole mass will depend on the details of thedot confinement potential , here we use the effectivemass for heavy holes in 2D systems in GaAs/AlGaAssingle-interface heterostructures, m ∗ HH ∼ . m e . Forfuture devices, it would be of interest to utilize het-erostructures with shallow 2D hole layers ( <
100 nmdepth) . This should help to decrease the size of theelectrostatic confinement potential, which may be re-quired in order to achieve similar orbital level spacings tothose obtained in electron quantum dots, since the heavyhole effective mass is larger than the electron effectivemass in GaAs ( m ∗ HH ∼ . m e − . m e > m ∗ e ≈ . m e ).Figure 3(a) shows quantum dot conductance versus V LP and V RP after increasing the center plunger voltageto V CP = 0.2 V. Transport gradually evolves from singledot to double dot-like as the confinement is increased. At (a) (b) (c) FIG. 3. (a) Dot conductance G dot vs. V LP and V RP for fixedgate voltages V CP = 0.2 V, V LQPC = V L = 0 V, and V CP =0.5 V. The yellow box outlines the gate voltage region spannedin (c). (b) Left QPC conductance vs. V LQPC for V L = 0 V.(c) QPC transconductance dG qpc /dV T vs. V LP and V LP with V CP = 0.2 V. V LQPC is varied from 0.43 to 0.4 V in order tomaintain constant sensitivity. the upper right corner of Fig. 3(a), the dot tunnel bar-riers become too opaque to measure conduction directlythrough the dot. Figure 3(b) shows conductance throughthe left QPC versus V LQP C . As expected for a QPC, thedata show plateaux in the conductance, with the second-to-last plateau occuring near 2 e /h . The last plateauoccuring below 2 e /h may be the so-called ”0.7 struc-ture”, which has been previously observed in electron and hole QPCs . The precise conductance values, es-pecially for the higher conductance plateaux, are likelyaffected by lead resistance since we use a two-terminalmeasurement. In Fig. 3(c) we show the QPC transcon-ductance dG qpc /dV T versus V LP and V RP in the regionof gate voltage indicated by the yellow box in Fig. 3(a).In order to use the QPC to charge sense the occupationof the dot, we tune V LQP C in order to sit on the steepportion of the G qpc versus V LQP C curve below the lastconductance plateau. The data clearly show sensing ofboth left and right dot charge occupation, where singlehole changes in the left dot occupation produce a largerchange in dG qpc /dV T than for the right dot due to thecloser physical proximity between the left dot and QPC.In Fig. 4(a)-(c) we show QPC transconductance ver-sus left and right plunger gate voltages V LP and V RP atthree different center plunger gate voltages V CP = 0, 0.2,and 0.4 V. The data demonstrate the ability to use theCP gate to tune the DQD from a highly-coupled regime,where the transport is reminiscent of that expected for alarge, single dot, to a weakly-coupled regime where thecharge sensing signal shows two well-isolated dots. Fig. 4 (a) (b) (c) (d)
FIG. 4. Left QPC transconductance dG qpc /dV T vs. V LP and V LP for (a) V CP = 0 V, (b) V CP = 0.2 V, (c) V CP = 0.4 V, and(d) V CP = 0.7 V. Figure 4(d) shows a continuation of the charge sensingdata to larger V LP and V LP voltages, for V CP = 0.7 V.The absence of transitions in the charge sensing signal inthe upper right region of the plot, for both the left andright dot, over a wide voltage range, indicate that theDQD is empty. This allows us to label the various regionsbetween charge transitions with DQD hole occupation( N , M ), as shown in Fig. 4(d), where N ( M ) indicatesthe number of holes in the left (right) dot, respectively.In conclusion, we have demonstrated a few-hole DQDin an undoped (100) oriented GaAs/AlGaAs heterostruc-ture. The device shows good charge stability and negli-gible hysteresis with respect to gate voltage. The inter-dot coupling can be tuned over a wide range, controlingthe transition from a large single dot to two well-isolatedquantum dots. Using charge sensing we show that thedot can be completely emptied of holes and operated inthe few-hole regime. The device may provide a meansfor future experiments focusing on manipulation of sin-gle hole spins in GaAs quantum dots. ACKNOWLEDGMENTS
We would like to thank J. Dominguez for assistancewith sample fabrication. This work was performed atthe Center for Integrated Nanotechnologies, a U.S. DOE,Office of Basic Energy Sciences user facility, and SandiaNational Laboratories, a multi-program laboratory oper-ated by Sandia Corporation, a wholly owned subsidiaryLockheed-Martin Company, for the U. S. Department ofEnergy’s National Nuclear Security Administration un-der Contract No. DE-AC04-94AL85000. D. Loss and D. P. DiVincenzo, Phys. Rev. A , 120 (1998). J. R. Petta, A. C. Johnson, J. M. Taylor, E. A. Laird, A. Ya-coby, M. D. Lukin, C. M. Marcus, M. P. Hanson, A. C. Gossard,Science , 2180 (2005). F. H. L. Koppens, C. Buizert, K. J. Tielrooij, I. T. Vink, K. C.Nowack, T. Meunier, L. P. Kouwenhoven, and L. M. K. Vander-sypen, Nature , 766 (2006). R. Hanson, L. P. Kouwenhoven, J. R. Petta, S. Tarucha, and L.M. K. Vandersypen, Rev. Mod. Phys. , 1217 (2007). J. Fischer, W. A. Coish, D. V. Bulaev, and D. Loss, Phys. Rev.B , 155329 (2008). D. Brunner, B. D. Geradot, P. A. Dalgarno, G. W¨ust, K. Karrai,N. G. Stoltz, P. M. Petroff, R. J. Warburton, Science , 70(2009). K. De Greve, P. L. McMahon, D. Press, T. D. Ladd, D. Bisping,C. Schneider, M. Kamp, L. Worshech, S. H¨ofling, A. Forchel, andY. Yamamoto, Nat. Phys , 872 (2011). A. Greilich, S. G. Carter, D. Kim, A. S. Bracker, and D. Gam-mon, Nat. Photon. , 702 (2011). D. V. Bulaev, and D. Loss, Phys. Rev. Lett. , 097202 (2007). V. S. Pribiag, S Nadj-Perge, S. M. Frolov, J. W. G. van der Berg,I. van Weperen, S. R. Plissard, E. P. A. M. Bakkers, and L. P.Kouwenhoven, Nat. Nanotech. , 170 (2013). Y. Hu, F. Kuemmeth, C. M. Lieber, and C. M. Marcus, Nat.Nanotech. , 47 (2012). I. van Weperen, B. D. Armstrong, E. A. Laird, J. Medford, C.M. Marcus, M. P. Hanson, A. C. Gossard, Phys. Rev. Lett. ,030506 (2011). L. Gaudreau, G. Granger, A. Kam, G. C. Aers, S. A. Studenikin,P. Zawadzki, M. Pioro-Ladrie`ere, Z. R. Wasilewski, and A. S.Sachrajda, Nat. Phys. , 54 (2012). B. Grbi´c, R. Leturcq, K. Ensslin, D. Reuter, and A. D. Wieck,Appl. Phys. Lett , 232108 (2005). B. Grbi´c et. al , AIP Conf. Proc. , 777 (2007). A. M. Burke, D. Waddington, D. Carrad, R. Lyttleton, H. H.Tan, P. J. Reece, O. Klochan, A. R. Hamilton, A. Rai, D. Reuter,A. D. Wieck, and A. P. Micolich, Phys. Rev. B , 165309 (2012). B. E Kane, L. N. Pfeiffer, K. W. West, and C. K. Harnett, Appl.Phys. Lett. , 2132 (1993). H. Noh, M. P. Lilly, D. C. Tsui, J. A. Simmons, E. H. Hwang,S. Das Sarma, L. N. Pfeiffer, and K. W. West, Phys. Rev. B ,165308 (2003); H. Noh, M. P. Lilly, D. C. Tsui, J. A. Simmons, L.N. Pfeiffer, and K. W. West, Phys. Rev. B , 241308(R) (2003). T. M. Lu, D. R. Luhman, K. Lai, D. C. Tsui, L. N. Pfeiffer, andK. W. West, Appl. Phys. Lett. , 112113 (2007). O. Klochan, J. C. H. Chen, A. P. Micolich, A. R. Hamilton, K.Muraki, and Y. Hirayama, Appl. Phys. Lett. , 092103 (2010). M. Ciorga, A. S. Sachrajda, P. Hawrylak, C. Gould, P. Zwadzki,S. Jullian, Y. Feng, and Z. Wasilewski, Phys. Rev. B , R16315 (2000). R. L. Willett, M. J. Manfra, L. N. Pfeiffer, and K. W. West,Appl. Phys. Lett. , 033510 (2007). L. P. Kouwenhoven, C. M. Marcus, P. L. McEuen, S. Tarucha,R. M. Westervelt, and N. S. Wingreen, in
Mesoscopic ElectronTransport , Proceedings of the Advanced Study Institute, editedby L. L. Sohn, L. P. Kouwenhoven, and G. Sch´”on (Kluwer,Dordercht, 1997). R. Winkler, 2003,
Spin-orbit Coupling Effects in Two-Dimensional Electron and Hole Systems (Springer-Verlag,Berlin). T. M. Lu, Z. F. Li, D. C. Tsui, M. J. Manfra, L. N. Pfeiffer, andK. W. West, Appl. Phys. Lett. , 012109 (2008). W. Y. Mak, F. Sfigakis, K. Das Gupta, O. Klochan, H. E. Beere,I. Farrer, J. P. Griffiths, G. A. C. Jones, A. R. Hamilton, and D.A. Ritchie, Appl. Phys. Lett. , 103507 (2013). K. J. Thomas, J. T. Nicholls, M. Y. Simmons, M. Pepper, D. R.Mace, and D. A. Ritchie, Phys. Rev. Lett. , 135 (1996). A. R. Hamilton, R. Danneau, O. Klochan, W. R. Clarke, A. P.Micholich, L. H. Ho, M. Y. Simmons, D. A. Ritchie, M. Pep-per, K. Muraki, and Y. Hirayama, J. Phys. Condens. Matter ,164205 (2008). Y. Komijani, M. Csontos, I. Shorubalko, T. Ihn, K. Ensslin, Y.Meir, D. Reuter, and A. D. Wieck, Europhys. Lett.91