Detection of defect-induced magnetism in low-dimensional ZnO structures by Magnetophotocurrent
Israel Lorite, Yogesh Kumar, Pablo Esquinazi, Carlos Zandalazini, Silvia Perez de Heluani
aa r X i v : . [ c ond - m a t . m t r l - s c i ] A p r Detection of defect-induced magnetism in low-dimensional ZnO structures byMagnetophotocurrent
Israel Lorite, Yogesh Kumar, and Pablo Esquinazi
Division of Superconductivity and Magnetism, Institut f¨ur Experimentelle Physik II,Fakult¨at f¨ur Physik und Geowissenschaften, Universit¨at Leipzig, Linn´estrasse 5, 04103 Leipzig, Germany
Carlos Zandalazini and Silvia Perez de Heluani
Laboratorio de F´ısica del S´olido, Dpto. de F´ısicaFacultad de Ciencias Exactas y Tecnolog´ıa, Universidad Nacional de Tucum´an, Argentina (Dated: July 20, 2018)The detection of defect-induced magnetic order in single low-dimensional oxide structures is ingeneral difficult because of the relatively small yield of magnetically ordered regions. In this workwe have studied the effect of an external magnetic field on the transient photocurrent measured afterlight irradiation on different ZnO samples at room temperature. We found that a magnetic fieldproduces a change in the relaxation rate of the transient photocurrent only in magnetically orderedZnO samples. This rate can decrease or increase with field depending whether the magnetic orderregion is in the bulk or only at the surface of the ZnO sample. The phenomenon reported here isof importance for the development of magneto-optical low-dimensional oxides devices and providea new guideline for the detection of magnetic order in low-dimensional magnetic semiconductors.
The detection of low levels of magnetic moment is ofimportance for the study of new materials and new phe-nomena in solid state magnetism. New technologies suchas the simultaneous use of a superconducting quantuminterferometer device (SQUID) and the Moke effect, arebeing developed to detect low levels of magnetic signals .This development is of particular interest for the studyof materials that show magnetic order at surprisinglyhigh temperatures due to a relatively high concentrationof defects, like vacancies . Defect-induced magnetism(DIM) in semiconducting materials, especially oxides, isof interest also because of the expected advantages onehas combining optical and semiconducting properties forspintronics applications. However, at present there areseveral difficulties, partially in the reproducibility of theDIM due to the unknown systematic preparation of thenecessary defects density, and also because of the low lev-els of magnetic signal due to the small amount of mag-netically ordered regions in a sample. It has recently been shown that the photocurrent issensitive to the minority carriers at the interphase be-tween two semiconductors and it can be changed by anexternal magnetic field, opening new possibilities to de-sign new devices based on photocurrent processes andinterphases properties . On the other hand, it has beenobserved that the photocurrent can be affected by rel-atively low applied fields of the order of 0.5 T or less,in diluted magnetic semiconductors at room tempera-ture thanks to the exchange interaction . However, lit-tle attention has been drawn to persistent photocurrentphenomena and their sensitivity to the variation of theoxygen absorption after photoexcitation . Such pho-tocurrent phenomena are of interest to study the effect ofexternal magnetic fields on photo-transport mechanismsat the surface and/or in bulk of defect-induced magneticoxides.There is consent nowadays that the photocurrent in ZnO is mainly a surface effect . The model generallyused to understand photocurrent effects is based on thedesorption and absorption of oxygen at the ZnO sur-face. Oxygen from the environment are chemisorbedat the surface and are negatively charged by capturinga free carrier, producing at the same time a depletionzone or surface band bending . During UV-illuminationelectron-holes are created. The electrons are promotedto the conduction band and the holes from valence bandand/or intraband donor defects such as oxygen vacan-cies (V O ) and O interstitials (O i ) , can migrate to thesurface to discharge the oxygen at the surface. This pro-duces its photodesorption and decreases the band bend-ing, allowing the free carrier to diffuse in both directions,i.e. from bulk to surface and vice versa. After turn-ing off the light excitons are recombined and the oxygenchemisorbed at the surface captures a free carrier, re-ducing the free carrier density and reestablish finally theinitial dark state.The oxygen reabsorption at the surface provides themain contribution that affects the transient photocur-rent. It depends on the environmental oxygen concentra-tion and inversely on the rate of recombination of elec-trons at traps centres . In this work we studied in detailthe effect of a magnetic field on the transient photocur-rent produced after turning off the light. The presentedresults indicate that the influence of an external magneticfield on the transient photocurrent is observed only insamples with defect-induced magnetic order. The effectsare detectable thanks to the sensitivity of the phototrans-port processes. They help to detect the existence of mag-netic order at high temperatures in low-dimensional ZnOstructures, as single nano/microwires or microstructuredthin films, which would remain undetectable by standardtechniques. I. METHOD
For the photocurrent measurements we used four dif-ferent ZnO samples. Three samples were 300 nm thinfilms grown on 6 × a -plane Al O substrates byPulsed Lased Deposition (PLD). Two different films weregrown, one in oxygen atmosphere (ZO) and another in ni-trogen (ZN) enviroment with partial pressures of 1 . × − mbar from a base vacuum of 10 − mbar . TheZO film was treated afterwards with hydrogen plasma(ZOH) ( ≈ H + /cm at an energy of 300 eV at roomtemperature) . All samples were patterned to a size of ∼ × µ m by e-beam lithography and wet-etchingtechniques .The fourth sample was a single ZnO microwire witha diameter of 10 µ m and 200 µ m length. The wire wasdoped with Li (5%) and treated with hydrogen plasma(ZLH) in a similar way. Previous studies revealed thatthe room temperature hydrogen-plasma treatment in Li-doped ZnO samples (Li-concentration ≥ Zn ) in 10 nm depth from the wire surface.The used procedure produces a V Zn concentration similarto the Li one and magnetic order can be observed atroom temperature, as revealed by XMCD and SQUIDtechniques .The magneto-photocurrent measurements were per-formed in a home made close-cycle cryostat with the pos-sibility to apply a magnetic field of 0.4 T. The field wasalways applied parallel to the input current direction,also parallel to the main film area or the main axis of themicrowire. During the measurements the samples werekept always at 5x10 − mbar to ensure a constant oxygenpartial pressure at 305 K. To prove the importance of theoxygen partial pressure in the persistent photocurrent re-laxation, we performed similar experiments after purgingthe chamber with He to reduce the partial pressure ofoxygen. The reduction of the oxygen partial pressure inthe chamber increased the photocurrent relaxation timeby several hours, in agreement with previous reports .It points out the importance of the oxygen absorption forthe persistent photocurrent process. The light irradiationwas done using a Xenon lamp coupled to a monochroma-tor from which we selected a wavelength of λ = 370 nmfor all experiments. II. RESULTS
The methodology used to obtain the change of thetransient photocurrent on time and under a magneticfield consists of measuring consecutively several light-on/off cycles of the photocurrent. In such an experimentthe samples are illuminated until the photocurrent in-creases to a previously selected value i ON different fromsaturation. After reaching it, the light is turned off andthe transient photocurrent decreases. When it reachesa previously selected value i OFF , see Fig. 1, the light isturned on till the photocurrent reaches i ON . The time ∆ t t Light on
B= 0T P ho t o c u rr en t ( - A ) t (min) B=0.4 T
Light off i FIG. 1. Experimental method to measure the effect of anexternal magnetic field on the transient photocurrent. Thetwo curves represent the cycles measured at B = 0 T and B = 0 . for every transient photocurrent cycle was measured for apreviously fixed ∆ i = i ON − i OFF . The same procedure isrepeated several times to obtain the time dependence ofthe transient photocurrent rate (TPR) defined as ∆ i ∆ t ( t ).The first cycle performed in every measurement is doneat zero magnetic field ( B ) to ensure an equal startingstate for every measurement of the same sample. Notethat the selected ∆ i depends on the sample because thephotocurrent varies for every sample.To check the equipment performance and the repro-ducibility of previously published variation of the TPRwith magnetic field , we have measured the ZN film,which shows magnetic order at room temperature dueto the existence of V Zn produced during preparation (seeinset in Fig. 2(a)) . The cycles measured for the ZNsample at zero and 0.4 T field are shown in Fig. 1. Forboth fields we observe that there is an increase of ∆ t af-ter every cycle. This is due to an increase of the numberof traps(holes) created during photoexcitation . Sincenot all the holes are recombined during the light is off anadditional number of photo-generated holes are producedevery cycle. The larger the number of holes the larger theexciton recombination rate, which implies that the num-ber of carriers moving from bulk to surface, necessary toinitiate the oxygen reabsorption and reestablish the ini-tial dark conditions, reduces. Thus, the TPR decreaseswith time, see Fig. 2(a).When an external magnetic field is applied, we observequalitatively a similar variation for the TPR to the oneobserved at zero field, see Fig. 2(a), however, the vari-ation of ∆ t is larger. It means there is an increase ofthe transient photocurrent time due to the effect of theexternal magnetic field, which should be related to a re-duction of the oxygen absorption rate in comparison tothe zero field state. The observed behavior is in good -7 -7 -7 -7 -7 -7 -11 -10 -7 -7 -7 ZLHZ0 (c)(b) -6 -4 -2 0 2 4 6-8x10 -2 -4x10 -2 -2 -2 M agne t i z a t i on ( e m u / g ) Magnetic Field (T)300K i /t ( A / m i n ) (a)ZN i /t ( A / m i n )
0T 0.2T 0.4T i /t ( A / m i n ) t (min) FIG. 2. Variation of the transient photocurrent rate at dif-ferent applied magnetic fields, as a function of the total timefor the thin film samples (a)ZN and (b) ZO and the singlemicrowire (c) ZLH. All measurements were done at a fixedtemperature of 305 K. The inset in (a) shows the field loop ofthe magnetization measured for the ZN film at 300 K. agreement with that reported previously .After observing the magnetic field effect on the TPR ofa magnetic sample, a ZO thin film was measured, whichdoes not show any sign of magnetic order at room tem-perature. The TPR(t) results shown in Fig. 2(b) indi-cates that there is no dependence with the external mag-netic field.To gain a deeper insight into the field dependence ofthe TPR we compare the obtained results with thosefrom a sample that shows magnetic order only within asurface shell, instead of a bulk magnetic sample such as in -8 -8 i /t ( A / m i n ) time (min) 0T 0.2T 0.4T FIG. 3. Time dependence of the transient photocurrent ratefor ZOH film, at different applied magnetic fields at 305 K.
ZN. The implantation of H + in the ZLH sample producesa significant amount of defects within a 10 nm surfaceregion . Different experimental methods includingXMCD as well as numerical calculations indicate that theproduced concentration of V Zn ( ∼ .The results for ZLH are shown in Fig. 2(c). In contrastto ZN, the TPR increases with magnetic field.Because the results indicate that the TPR can dependon the applied magnetic field only for magnetically or-dered samples, we implanted 10 H + /cm in the non-magnetic ZO thin film shown in Fig. 2(b), labeled nowZOH. Such implantation of H + can produce a near sur-face magnetic layer within ≃
10 nm, similarly to the oneobserved in ZnO single crystals . The results of theTPR for this sample are shown in Fig. 3. The observedvariation with field is similar to the one observed for theZLH microwire suggesting that its origin should be re-lated to the formation of a magnetically surface shell. Itis worth to note that the amount of magnetically orderedmass in this sample is too small to be measurable witha commercial SQUID, i.e. its ferromagnetic moment atsaturation is smaller than 2 × − emu. Nevertheless,the photocurrent is influenced by the thin magnetic layeremphasizing the sensitivity of this property to magneticorder produced by defects in this case.To understand qualitatively the observed phenomenonwe take into account the role of the induced magneticorder in the bulk and at the near surface region. Inboth cases we have the Zeeman splitting produced bythe finite magnetic field, as consequence of the spin-orbit( L − S ) coupling with the magnetic defects, V Zn , whichis large for wide band gap magnetic semiconductors .Due to optical selection rules, which accompany the Zee-man splitting and where the spin orientation of the mag-netic defects plays a role , the trapping probability ofconduction electrons by the defects increases with field. FIG. 4. Sketch of the magnetically ordered near surface re-gion of a low-energy H + implanted ZnO sample, at B = 0 T(a) and at B =0 T (b). The magnetic field is applied parallelto the input current I . The blue surface region of thickness ∼
10 nm, denotes the defect-induced magnetically order re-gion and the pink part is the non-magnetic bulk region. Undera magnetic field the magnetic region shows a band splittingdue to the Zeeman effect, changing the probability of carrieraccumulation at the surface and influencing the transient pho-tocurrent rate. The Zeeman splitting is not in scale with thedrawn energy gap.
This effect reduces the amount of electrons that can reachthe surface and capture oxygen from the environment toreestablished the initial dark conditions. In this case theTPR decreases with applied magnetic field, see Fig. 2(a).In the second case, observed for samples ZLH andZOH, we must consider the heterostructure with onlythe near surface region magnetically ordered, see Fig. 4.When an external magnetic field is applied, similarly tothe first case, a Zeeman splitting occurs in the magneticregion. This energy shift leads to a redistribution of car-riers with larger probability to be at the deeper potential.In this case and just after removal of the light excitation,a larger accumulation of electrons occurs at the magneticregion due to the carriers coming from the non-magneticbulk region. Therefore, more electrons reach the surfaceincreasing the probability for oxygen trapping at the sur-face and as consequence a faster TPR is observed.The presented results clearly indicate a dependence ofthe TPR with relatively low magnetic fields at room tem-perature. This magnetic field dependence is qualitativelydifferent for ZnO samples with magnetic order in the bulkor at the near surface region. The observed change inthe TPR is related to the variation of the rate of oxygenabsorption to reach the initial dark state. This oxygenabsorption rate depends on the number of free electronsreaching the surface, and their amount depends on mag-netic field and the magnetically ordered region in thesample. Our results indicate that low fields can be usedto tune the transient photocurrent in magnetic samples,which can be useful to detect sensitively room tempera-ture magnetic order in low-dimensional systems, a mag-netic order that may remain undetectable using routinetechniques such as SQUID. ACKNOWLEDGMENTS
We thank W. Hergert for fruitful discussions. Thiswork was funded by the DFG through the CollaborativeResearch Center SFB 762 “Functionality of Oxide Inter-faces” in Germany, and by the Secyt-UNCor, CIUNT andConicet in Argentina. W. D. Rice, P. Ambwani, M. Bombeck, J. D. Thompson,G. Haugstad, C. Leighton, and S. A. Crooker, NatureMaterials , 481 (2014). G. Z. Xing, Y. H. Lu, Y. F. Tian, J. B. Yi, C. C. Lim,Y. F. Li, G. P. Li, D. D. Wang, B. Yao, J. Ding, and Y. P.Feng, AIP ADVANCES , 022152 (2011). D. W. Abraham, M. M. Frank, and S. Guha, AppliedPhysics Letters , 252502 (2005). C. Jagadish and S. J. Pearton,
Zinc Oxide bulk, thin filmsand nanostructures (Elsevier Press, UK, 2006). Z. Sheng, M. Nakamura, W. Koshibae, T. Makino,Y. Tokura, , and M. Kawasaki, Nature communications , 5584 (2014). L.-C. Chen, C.-H. Tien, and Y.-Y. Hsu, Jpn. J. Appl.Phys , 063002 (2010). C. Zapata, M. Khalid, G. Simonelli, M. Villafuerte, S. P.Heluani, and P. Esquinazi, Appl. Phys. Lett. , 112503(2011). F. Cai, J. Wang, Z. Yuan, and Y. Duan, J. of PowerSources , 269 (2012). W. Park, G. Jo, W.-K. Hong, J. Yoon, M. Choe, S. Lee,Y. Ji, G. Kim, Y. H. Kahn, K. Lee, D. Wang, and T. Lee,Nanotechlogy , 205204 (2011). J. C. Moore and C. V. Thompson, Sensors , 9921 (2013). B. F. Spencer, D. M. Graham, S. J. O. Hardman, E. A.Seddon, M. J. Cliffe, K. L. Syres, A. G. Thomas, S. K.Stubbs, F. Sirotti, M. G. Silly, P. F. Kirkham, A. R. Ku-marasingh, G. J. Hirst, A. J. Moss, S. F. Hill, D. A. Shaw,S. Chattopadhyay, and W. R. Flavell, Phys. Rev. B ,195301 (2013). M. Khalid, M. Ziese, A. Setzer, P. Esquinazi, M. Lorenz,H. Hochmuth, M. Grundmann, D. Spemann, T. Butz,G. Brauer, W. Anwand, G. Fischer, W. A. Adeagbo,W. Hergert, and A. Erns, Phys. Rev. B , 035331 (2009). M. Khalid, P. Esquinazi, D. Spemann, W. Anwand, andG. Brauer, New Journal of Physics , 063017 (2011). G. Bridoux, J. Barzola-Quiquia, F. Bern, W. B¨ohlmann,I. Vrejoiu, P. Esquinazi, and M. Ziese, Nanotechnology , 085302 (2012). I. Lorite, B. Straube, H. Ohldag, P. Kumar, M. Villafuerte,P. Esquinazi, R. Torres, S. P. de Heluani, V. N. Antonov,L. V. Bekenov, A. Ernst, M. Hoffmann, S. K. Nayak, W. A.Adeagbo, G. Fischer, and W. Hergert, Appl. Phys. Lett. , 082406 (2015). D. C. Hou, A. Dev, K. Frank, A. Rosenauer, and T. Voss,J. Phys. Chem. C , 19604 (2012). P. I. Reyes, C.-J. Ku, Z. Duan, Y. Xu, E. Garfunkel, andY. Lu, Appl. Phys. Lett. , 031118 (2011). C. F. Klingshirn, A. Waag, A. Hoffmann, and J. Geurts,
Zinc Oxide: From Fundamental Properties Towards NovelApplications , Springer Series in Material Science (Springer,Heidelberg, Dordrecht, London, New York, 2010). M. D. Vece, B. Kolaric, K. Baert, G. Schweitzer,M. Obradovic, R. A. L. Valle, P. Lievens, and K. Clays,Nanotechlogy , 135203 (2011). Y. Takahashi, M. Kanamori, A. Kondoh, H. Minoura, andY. Ohya, Jnp. J. Appl. Phys.33