Molecular beam epitaxy of n-type ZnS: A wide band gap emitter for heterojunction PV devices

Jeffrey Bosco, Faisal Tajdar, and Harry Atwater

California Institute of Technology, Department of Chemical Engineering and Department of Applied Physics, Pasadena, California, 91125, USA.

Abstract — Low-temperature epitaxy of zinc-blende ZnS

films on GaAs(001) substrates was demonstrated by compound-source molecular beam epitaxy. The epitaxial relationship between the film and substrate was determined by reflection high-energy electron diffraction, high-resolution X-ray diffraction, and selected area electron diffraction measurements to be ZnS(001)||GaAs(001). Strain-relaxation of the ZnS lattice occurred within the first 300 nm of film growth. The resistivity of the films could be tuned through the incorporation of an Al impurity dopant. The lowest thin-film resistivity achieved was 0.003 Ω-cm, with corresponding electron carrier concentration and mobility of roughly 4.5×1019 cm-3 and 46 cm2 V-1 s-1, respectively. Al, Ag, and In metals were found to make good ohmic contact to heavily doped ZnS films, whereas ITO and AZO transparent conductive oxides did not. Applications to novel PV devices incorporating low electron affinity absorbers are discussed. Index Terms — Conductive films, epitaxial layers, photovoltaic

cells, X-ray diffraction, zinc compounds.

I. INTRODUCTION

The growing interest in scalable, thin-film photovoltaics (PV) has led to a demand for low-cost, non-toxic, earth-abundant emitter materials for heterojunction solar cells. We have found n-type, zinc-blende zinc sulfide (ZnS) to be an ideal candidate for such applications since the material has a wide band gap of 3.68 eV, it has good antireflection properties, and it exhibits excellent surface passivation on a number of common photovoltaic absorber materials such as Si [1], Cu(In,Ga)Se2 [2], CdTe [3], and GaAs [4]. ZnS has already found applications as a buffer layer material in modern Cu(In,Ga)Se2 devices [5] as well as an alternative material for transparent conductive films [6]. Finally, ZnS is composed of abundant and inexpensive elements and thin films can be fabricated through a number of facile deposition techniques.

Incorporation of ZnS into an active device has been inhibited by the difficulty of achieving sufficiently low electrical resistivity. Fabrication of reliable ohmic contacts has also posed a challenge, particularly on bulk, single-crystal substrates [7]. However, it has been reported that low-resistivity ZnS can be achieved with Al [8, 9], In [10], Cl [11], or I [12] extrinsic dopant incorporation through non-equilibrium growth techniques such as molecular beam epitaxy (MBE) and metal-organic chemical vapor deposition

(MOCVD). Furthermore, ohmic contact can be facilitated by implementing a heavily doped region at the surface of the film.

Here we report on the growth of low-resistivity, epitaxial ZnS films grown by compound source MBE. Previous studies on epitaxial growth of ZnS have been performed on a variety of different substrates, including ZnS, Si, GaP, and GaAs. In this work we have chosen GaAs(001) as an epitaxial substrate since it favors the growth of zinc-blende ZnS and also demonstrates the ease at which ZnS can be implemented in III-V devices. Tunable n-type doping was achieved through incorporation of Al-metal impurities using a second evaporation source. The morphology and electrical properties of as-grown layers were characterized in detail using both in situ and ex situ techniques. Finally, methods for achieving suitable ohmic contact to doped films were addressed.

II. EXPERIMENTAL

ZnS films were grown in an ultra-high vacuum MBE chamber with an ultimate pressure of <2×10-10 Torr. A standard Knudsen effusion cell loaded with pure ZnS (99.9999%) was employed as a compound sublimation source.

Fig 1. ZnS growth rate dependence on substrate temperature and beam pressure as determined by both profilometry and ellipsometry film thickness measurements.

978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE 002513

Epi-ready, semi-insulating GaAs(001) single crystal wafers (AXT) were used as an epitaxial substrate. The GaAs native oxide was removed prior to film growth by exposure to a flux of RF generated atomic hydrogen for roughly 5 min. The removal of the native oxide was verified in situ using reflection high-energy electron diffraction (RHEED)evidenced by a streaky 1×1 surface reconstructionof an oxide-free, disordered surface [13]. pressure (BEP) was maintained between roughly 2.5×10-6 Torr (source temperatures of 850respectively) throughout the entire study. The growth temperature was varied between 50 oC and 350 was used extensively to characterize the surface morphology of the films during and after growthtype doping of the ZnS films during epitaxial growth was performed using electron-beam evaporation of pure (99.999%, Kurt J. Lesker). The dopant incorporattuned by varying the electron-beam power.

Ex situ characterization of the ZnS films included profilometry, ellipsometry, atomic force microscopy (AFM), high-resolution X-ray diffraction (HRXRD), electron microscopy (TEM), and selecteddiffraction (SAED). For doped films, the electrical propertiesincluding the ability to make ohmic contact, thinresistivity, electron concentration, and electron mobilitymeasured by four-point probe, Van der Pauwmethods. The actual Al impurity concentration was measured by secondary-ion mass spectrometry (SIMS)determine the extent of dopant activation.

III. RESULTS AND ANALYSIS

ZnS epilayers with specular surfaces were achieved at growth temperatures ranging from 350 oC down to 50

Fig 2. RHEED images recorded along the [0ZnS films grown at substrate temperatures ranging from 350 to 50oC

GaAs(001) single crystal wafers (AXT) were used as an epitaxial substrate. The GaAs native

by exposure to a flux of RF generated atomic hydrogen for roughly 5 min. The removal of the native oxide was verified in situ using

energy electron diffraction (RHEED) as streaky 1×1 surface reconstruction, indicative

. The ZnS beam roughly 1.5×10-6 and

850 and 875 oC, study. The growth

C and 350 oC. RHEED orientation and

during and after growth. N-type doping of the ZnS films during epitaxial growth was

beam evaporation of pure Al metal ). The dopant incorporation was

characterization of the ZnS films included atomic force microscopy (AFM),

HRXRD), transmission , and selected area electron

the electrical properties, ability to make ohmic contact, thin-film

electron mobility were Van der Pauw, and Hall effect

concentration was measured ion mass spectrometry (SIMS) in order to

with specular surfaces were achieved at C down to 50 oC

. RHEED images recorded along the [011] zone axis of substrate temperatures ranging from 350

Fig 3. (a) The zinc-blende ZnS unit cell, (b) inarrangement of the (001) surface. (c) RHEED patterns of the (001) surface with along different zone axes with diffraction simulations for the ZnS crystal structure. despite a >4% lattice mismatch with GaAs. average RMS surface roughness of greater than 200 nm in total thickness. The growth rateas a function of substrate temperature wasellipsometry and profilometry measurementsFigure 1., there was excellent agreement between the two thickness measurement techniques. Tfound to decrease rapidly for substrate temperatures oC. This trend can be attributed to a decrease in the coefficient with increasing temperatureobserved for ZnS growth on the Si(001) surface growth rate was also found to increase linearly with the ZnS BEP for a given growth temperature.

RHEED images collected along the [0as a function the substrate temperthickness are displayed in Figure 2found to change from spots to streaks, indicating a transition from 3D to 2D growth modes. patterns obtained for a given film thickness grown at two different temperatures (not shown) transition is primarily due to thickness rather than temperature. Figure 3 also displays RHEED images collected at various zone axes for a 13 nm thick film. Comparison with simulated electron diffraction patterns verified that the films exhibited the zinc-blende crystal aligned with the unit cell of the GaAs substrateto the growth axis.

Figure 4 displays HRXRD of the GaAsreflections for films of varying thicknesses. No other ZnS orientations were observed. The ZnSfound to shift lower in 2θ with increasing film thickness, suggesting that there was a relaxation of tensile strain.

ZnS unit cell, (b) in-plane atomic arrangement of the (001) surface. (c) RHEED patterns of the (001) surface with along different zone axes with electron

simulations for the ZnS crystal structure.

despite a >4% lattice mismatch with GaAs. AFM revealed an average RMS surface roughness of less than ±1 nm for films greater than 200 nm in total thickness. The growth rate of ZnS as a function of substrate temperature was determined by both ellipsometry and profilometry measurements. As displayed in Figure 1., there was excellent agreement between the two

. The ZnS growth rate was to decrease rapidly for substrate temperatures above 250

C. This trend can be attributed to a decrease in the Zn sticking with increasing temperature as was previously

observed for ZnS growth on the Si(001) surface [14]. The growth rate was also found to increase linearly with the ZnS BEP for a given growth temperature.

RHEED images collected along the [011] zone axis of ZnS as a function the substrate temperature and hence film

2. The RHEED pattern was found to change from spots to streaks, indicating a transition from 3D to 2D growth modes. The likeness of RHEED patterns obtained for a given film thickness grown at two

suggests that the observed transition is primarily due to thickness rather than

also displays RHEED images collected thick film. Comparison with

patterns verified that the films blende crystal structure and were well

aligned with the unit cell of the GaAs substrate perpendicular

displays HRXRD of the GaAs(004) and ZnS (004) reflections for films of varying thicknesses. No other ZnS

ZnS(004) peak position was with increasing film thickness,

suggesting that there was a relaxation of tensile strain. From

978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE 002514

Fig 4. HRXRD of the ZnS (004) reflection for films grown at various substrate temperatures. The inset shows relaxation of the ZnS lattice parameter with increasing film thickness. the 2θ position of the ZnS(004) peak, the outlattice parameter was calculated and plotted as a fuepilayer thickness (inset). The lattice parameter was observed to increase from an initial value of 0.5360thickness towards its relaxed value of 0.5420 indication that the partial film relaxation occurs immediately upon initial film growth with some slower relaoccurring afterwards that can be considered complete after roughly 250 nm.

A high-resolution transmission electron micrographZnS-GaAs interface is displayed in Figure 5. The imageinterface shows the clear presence of interfacwell as extra half-planes. Stacking faults and crystal twinwere also observed within the bulk of the ZnS film. However, these defects/dislocations are expected considering therelatively large in-plane lattice mismatch between GaAs and ZnS. The inset of Figure 5 shows SAED patterns collected at the epitaxial interface. The SAED shows a roughly the ZnS(001) direction with respect to the GaAs(001direction. The tilt is likely a growth mechanism for compensating the large lattice mismatch mentioned earlier. This tilt mechanism appears to be common for lattice mismatched growth of II-VI compounds and was previously observed for growth of ZnSe epilayers on Ge(001) substrates by MBE [15].

Figure 6 compares the resistivities and dopant concentrations obtained from Van der Pauw and Hall effect measurements in this work with those from previous doping studies involving MBE, MOCVD, and MOPVE growth techniques [8-10, 12]. The highest electron

. HRXRD of the ZnS (004) reflection for films grown at various substrate temperatures. The inset shows relaxation of the ZnS lattice parameter with increasing film thickness.

(004) peak, the out-of-plane ZnS parameter was calculated and plotted as a function of

epilayer thickness (inset). The lattice parameter was observed 0 nm with film

420 nm. This is an occurs immediately

upon initial film growth with some slower relaxation that can be considered complete after

ransmission electron micrograph of the igure 5. The image of the

the clear presence of interfacial defects as planes. Stacking faults and crystal twins

were also observed within the bulk of the ZnS film. However, xpected considering the

plane lattice mismatch between GaAs and shows SAED patterns collected at

the epitaxial interface. The SAED shows a roughly 1o tilt in with respect to the GaAs(001) crystal

direction. The tilt is likely a growth mechanism for compensating the large lattice mismatch mentioned earlier.

appears to be common for lattice VI compounds and was previously Se epilayers on Ge(001) substrates

compares the resistivities and dopant from Van der Pauw and Hall effect

in this work with those from previous ZnS doping studies involving MBE, MOCVD, and MOPVE

The highest electron

Fig 5. Transmission electron micrograph of the interface between the ZnS epilayers and GaAs substrate. The inset shows an SAED pattern collected at the interface. concentration we achieved was 4.5×10to a film resistivity of 0.003 Ω-cm andcm2 V-1 s-1. Slightly higher mobilities (>55 cmachieved at lower dopant concentrations. Our results compare well with the previous work, with the highest previously reported electron concentration being 3.9×10Ω-cm).

The actual Al concentration incorporated in the ZnS films was determined by SIMS analysis measured electron concentration in Figactivation was observed to range from appears that the low growth temperatures accessiblecompound source MBE help to avoid the issue of selfcompensation commonly encountered for II-VI compounds [16]. However, Hall measurements were not possible for films with nominal Al concentrations of less than 1×1018 cm-3 due to difficulty in making good electrical contact. This is a common issue for very widesemiconductors.

In order to further explore the problem of obtaining ohmic contact to sublimation-grown ZnS filmscontact materials, including In, Al, indium-tin-oxide (ITO), and Al-doped zinc oxide (transparent conductive oxides (TCOs)four-point probe measurements. deposited using e-beam evaporation whereas the TCO contacts were sputter deposited. It was found that for moderate to highly doped ZnS films (ne = 0.1~1×10Ag all made excellent electrical contact. On the other ohmic contact was not achieved for ITO or AZO for any ZnS dopant concentration. For moderately doped and undoped ZnS

Transmission electron micrograph of the interface the ZnS epilayers and GaAs substrate. The inset

shows an SAED pattern collected at the interface.

we achieved was 4.5×1019 cm-3, corresponding cm and electron mobility of 46

lities (>55 cm2 V-1 s-1) were achieved at lower dopant concentrations. Our results compare well with the previous work, with the highest previously reported electron concentration being 3.9×1019 cm-3 (ρ = 0.002

incorporated in the ZnS films by SIMS analysis and is compared to the

measured electron concentration in Figure 5. The Al dopant was observed to range from roughly 25 to 100%. It

appears that the low growth temperatures accessible using compound source MBE help to avoid the issue of self-compensation commonly encountered for extrinsic doping of

Hall measurements were not nominal Al concentrations of less than

due to difficulty in making good electrical contact. This is a common issue for very wide-band gap

e problem of obtaining ohmic films, a number of different

In, Al, and Ag metals as well as doped zinc oxide (AZO)

oxides (TCOs), were studied using The metal contacts were

beam evaporation whereas the TCO It was found that for moderate

= 0.1~1×1019 cm-3), In, Al, and Ag all made excellent electrical contact. On the other hand, ohmic contact was not achieved for ITO or AZO for any ZnS

For moderately doped and undoped ZnS

978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE 002515

Fig 6. A comparison of Al-doped ZnS resistivity vs. electron concentration with previous studies. The lines represent ideal resistivity for a given carrier concentration and mobilities of 1000 (˗˗˗˗˗˗), 100 (-----), and 10 cm2 V-1 s-1 (-•-•-). films, ohmic contact with metals was only possible after annealing the contacts under vacuum at >300 oC. However, at this temperature the contact metal was found to diffuse into the ZnS film and cause substantial bulk doping as confirmed by hall effect measurements.

IV. DISCUSSION

Our work clearly demonstrates the possibility of depositing high quality, low-resistivity, n-type ZnS emitter layers using a compound-source sublimation process. Congruent sublimation provides a simple and economical route for thin film deposition of high quality ZnS films. The very low growth temperatures demonstrated in this work are possible due to the much higher source temperature required to sublimate the ZnS compound as compared to individual Zn and S elemental sources. A low growth temperature not only expands the choice of substrate materials, but also reduces the negative effects of the fabrication process on electronic device performance commonly caused by diffusion and interfacial reactions as well as differences in thermal expansivities of the device materials.

Previous issues involving the poor electrical conductivity of ZnS appear to be limited to bulk crystal growth methods and not to non-equilibrium thin-film growth techniques. With the latter, extrinsic dopants can be incorporated at relatively low temperatures in order to avoid dopant self-compensation mechanisms which are common for wide band gap II-VI materials.

The problem of providing a good ohmic contact to n-type ZnS is also simplified if adequate dopant concentrations can be obtained (n>1×1018 cm-3). The fact that conventional TCOs like ITO and AZO do not make ohmic contact to ZnS does not pose a problem since the as-measured film resistivities for

Fig 7. Electron concentration determined by Hall Effect vs. SIMS Al impurity concentration for Al-doped ZnS films. heavily doped ZnS were sufficiently low such that a TCO is not required for fabrication of a PV device top contact. Instead, only correctly spaced contact fingers are needed since sheet resistances of <100 Ω/ can be achieved with a 100~200 nm thick heavily doped ZnS film.

Finally, ZnS makes an excellent choice of n-type heterojunction partner choice for PV devices requiring a low electron affinity emitter material. Heterojunction devices incorporating a low electron affinity absorber material necessitate an emitter layer of similar electron affinity in order to avoid substantial losses in open circuit voltage and thus overall photovoltaic conversion efficiency. ZnS has a relatively low reported electron affinity of 3.9 eV [17] and is of particular interest in novel PV devices incorporating earth-abundant absorber materials like zinc phosphide (Zn3P2) and cuprous oxide (Cu2O), with reported electron affinities of 3.6 eV [18] and 3.2 eV [19], respectively.

V. SUMMARY

We have successfully demonstrated low-temperature epitaxy of n-type ZnS films on single crystal GaAs(001) substrates by compound-source MBE. All ZnS films were found to have zinc-blende crystal structure with (001) orientation. HRXRD showed that strain-relaxation of the ZnS out-of-plane lattice parameter occurred within the first 300 nm of film growth. The incorporation of an Al impurity dopant by electron-beam evaporation was used to tune the ZnS resistivity between 1 and 0.003 Ω-cm. The highest electron carrier concentration achieved was 4.5×1019 cm-3 with corresponding electron mobility of 46 cm2 V-1 s-1. It was determined that Al, Ag, and In metals made good ohmic contact to doped ZnS films, whereas ITO and AZO conductive oxides did not. Future studies will focus on the fabrication of heterostructure PV devices incorporating ZnS as an emitter for low electron affinity absorber materials.

978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE 002516

ACKNOWLEDGEMENT

This work was supported by The Dow Chemical Company and by the Dept. of Energy, Office of Basic Energy Sciences under Grant No. DE-FG02-03ER15483. The authors would like to thank Steven Roosevelt for his assistance with the transmission electron microscopy measurements. J.P.B. acknowledges support under an NSF graduate research fellowship.

REFERENCES

[1] G. A. Landis, J. J. Loferski, R. Beaulieu, P. A. Sekulamoise, S. M. Vernon, M. B. Spitzer, and C. J. Keavney, "Wide-bandgap epitaxial heterojunction windows for silicon solar-cells," IEEE Transactions on Electronic Devices, vol. 37, pp. 372-381, 1990.

[2] T. Nakada, M. Mizutani, Y. Hagiwara, and A. Kunioka, "High-efficiency Cu(In,Ga)Se2 thin-film solar cells with a CBD-ZnS buffer layer," Solar Energy Materials and Solar Cells, vol. 67, pp. 255-260, 2001.

[3] Y. H. Kim, S. Y. An, J. Y. Lee, I. Kim, K. N. Oh, S. U. Kim, M. J. Park, and T. S. Lee, "Photoluminescence study on the effects of the surface of CdTe by surface passivation," Journal of Applied Physics, vol. 85, pp. 7370-7373, 1999.

[4] J. M. Woodall, G. D. Pettit, T. Chappell, and H. J. Hovel, "Photoluminescent properties of GaAs-GaAlAs, GaAs-oxide, and GaAs-ZnS heterojunctions," Journal of Vacuum Science and Technology, vol. 16, pp. 1389-1393, 1979.

[5] A. Pudov, J. Sites, and T. Nakada, "Performance and loss analyses of high-efficiency chemical bath deposition (CBD)-ZnS/Cu(In1-xGax)Se2 thin-film solar cells," Japanese Journal of Applied Physics Part 2-Letters, vol. 41, pp. L672-L674, 2002.

[6] X. J. Liu, X. Cai, J. S. Qiao, H. F. Mao, and N. Jiang, "The design of ZnS/Ag/ZnS transparent conductive multilayer films," Thin Solid Films, vol. 441, pp. 200-206, 2003.

[7] R. G. Kaufman and P. Dowbor, "Mechanism of formation of Ohmic contacts to ZnSe, ZnS, and mixed crystals ZnSxSe1-x," Journal of Applied Physics, vol. 45, pp. 4487-4490, 1974.

[8] T. Yasuda, K. Hara, and H. Kukimoto, "Low resistivity Al-doped ZnS grown by MOVPE," Journal of Crystal Growth, vol. 77, pp. 485-489, 1986.

[9] M. Kitagawa, Y. Tomomura, A. Suzuki, and S. Nakajima, "Homoepitaxial growth of low-resistivity Al-doped ZnS single-crystal films by molecular-beam epitaxy," Journal of Crystal Growth, vol. 95, pp. 509-511, 1989.

[10] L. C. Olsen, R. C. Bohara, and D. L. Barton, "Vacuum-evaporated conducting ZnS films," Applied Physics Letters, vol. 34, pp. 528-529, 1979.

[11] M. Grün, A. Storzum, M. Hetterich, A. Kamilli, W. Send, T. Walter, and C. Klingshirn, "Chlorine doping of cubic CdS and ZnS grown by compound source molecular-beam epitaxy," Journal of Crystal Growth, vol. 201–202, pp. 457-460, 1999.

[12] S. Yamaga, A. Yoshikawa, and H. Kasai, "Growth and properties of Iodine-doped ZnS films grown by low-pressure MOCVD using Ethyliodide as a dopant source," Journal of Crystal Growth, vol. 106, pp. 683-689, 1990.

[13] A. Khatiri, J. Ripalda, T. Krzyzewski, G. Bell, C. McConville, and T. Jones, "Atomic hydrogen cleaning of GaAs (0 0 1): a scanning tunnelling microscopy study," Surface science, vol. 548, pp. L1-L6, 2004.

[14] M. Yokoyama, K. Kashiro, and S. Ohta, "Molecular-beam epitaxial-growth of ZnS on a (100)-oriented Si substrate," Journal of Crystal Growth, vol. 81, pp. 73-78, 1987.

[15] J. Kleiman, R. M. Park, and H. A. Mar, "On epilayer tilt in ZnSe/Ge heterostructures prepared by molecular-beam epitaxy," Journal of Applied Physics, vol. 64, pp. 1201-1205, 1988.

[16] I. P. McClean and C. B. Thomas, "Conductive ZnS:Zn thin-films grown by molecular-beam deposition," Journal of Applied Physics, vol. 72, pp. 4749-4752, 1992.

[17] R. K. Swank, "Surface properties of II-VI compounds," Physical Review, vol. 153, p. 844, 1967.

[18] A. J. Nelson, L. L. Kazmerski, M. Engelhardt, and H. Hochst, "Valence-band electronic-structure of Zn3P2 as a function of annealing as studied by synchrotron radiation photoemission," Journal of Applied Physics, vol. 67, pp. 1393-1396, 1990.

[19] L. C. Olsen, R. C. Bohara, and M. W. Urie, "Explanation for low-efficiency Cu2O Schottky-barrier solar-cells," Applied Physics Letters, vol. 34, pp. 47-49, 1979.

978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE978-1-4673-0066-7/12/$26.00 ©2011 IEEE 002517