Transparent p/n diode device from a single zinc nitride sputtering target

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Transparent p/n diode device from a single zinc nitride sputtering target

V. Kambilafka a, A. Kostopoulos a, M. Androulidaki a, K. Tsagaraki a, M. Modreanu b, E. Aperathitis a,⁎a Microelectronics Research Group, Institute of Electronic Structure and Laser, Foundation for Research and Technology—FORTH-Hellas, P.O. Box 1385, Heraklion 71110, Crete, Greeceb Photonics Group, Tyndall National Institute, Cork, Ireland

a b s t r a c ta r t i c l e i n f o

Available online 30 June 2011

Keywords:SputteringZinc nitrideZinc oxideTransparent diodep-Type ZnO

Zinc oxide (ZnO) thin films showing bipolar conductivity were fabricated by sputtering of zinc nitride targetin plasma containing mixture of Ar–O2 gasses. Sputtering in pure Ar plasma produced conductive and opaquezinc nitride (ZnN) films while upon introduction of oxygen up to 30% into the plasma highly transparentsingle phase polycrystalline n-type ZnO films have been grown. ZnN sputtering in Ar plasma containing morethan 30% oxygen produced p-type ZnO films. Hall-effect and photoluminescence measurements revealed thepresence of zinc vacancies and nitrogenwhich are acting as acceptor dopants in p-type ZnO. A heterostructurewas fabricated in a single deposition run consisting of n-ZnN and p-ZnO which exhibited rectifying behaviorwith 2–2.5 V turn-on voltage. Improvements on the formed p/n heterostructure as well as the potential ofusing single sputtering target in fabrication of Zn-based homo- and hetero-junctions are discussed.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

The compound Zn-N, zinc nitride, is known since 1906 [1] but itwas in 1993 when the first publication was made on the properties ofthis compound as thin film (fabricated by ammonization of Zn [2]).Since then, zinc nitride thin films have beenmade by other depositiontechniques like radio frequency molecular beam epitaxy (RF-MBE)and metal-organic chemical vapor deposition (MOCVD) [3], sputter-ing of Zn [4–6] or ZnN [7] or electrochemically [8] and recently zincnitride was used as channel layer for the fabrication of transparentthin film transistor [9]. Zinc nitride (Zn3N2 in stoichiometric form) isan n-type semiconductor and, depending on themethod of depositionand time-exposure to ambient, its optical band gap can be between1 eV and 3.2 eV [10]. Due to the technological importance infabricating reliable p-type ZnO and consequently the realization oftransparent optoelectronic devices, zinc nitride has been used for thefabrication of p-type ZnO:N. This has been achieved either by in situ orex situ (post-deposition) thermal oxidation of zinc nitride attemperatures higher than 400 °C [6,11,12].

In a previous investigation we had examined the properties of zincnitride films which had been deposited by sputtering from zinc nitridetarget in plasma containing Ar and N2 gasses [7]. Here we report on thefabrication of both n-type and p-type ZnO:N by sputtering from zincnitride target in plasma containing Ar and O2 gasses as well as therealization of a heterostructure diode, namely n-ZnN/p-ZnO, fabricatedin a single deposition run without breaking vacuum. Improvements on

the properties of the layers ormodifications of the structure of the diodefor enhancing its output characteristics are also addressed.

2. Experimental details

RF magnetron sputtering (Nordiko, RFG-2500) was employed forthe deposition of films using commercially available zinc nitridetarget (Testbourne Ltd., Zn:N=1:1, purity 99.9%, 6 in diameter×0.25 in thick). Plasma was formed by introducing Ar and O2 gasses(having purity of 5 N and 3.5 N, respectively) in the chamber and theirflow rates were controlled by mass flow controllers. Films depositedin pure Ar plasma were grown on Corning 7059 glass substratewhereas all other films were grown on fused silica and pieces ofcrystalline Si wafer (orientation (001), resistivity 1–10 Ω cm)substrates. All substrates were cleaned ultrasonically for 15 min inpropanol and acetone then rinsed in de-ionized water and dried inflowing nitrogen. Prior to the deposition, the target was pre-sputteredfor 15 min (Ar plasma, 0.65 Pa, 100 W rf power) to remove anycontaminants from the target surface and to enable growthequilibrium conditions to be reached. All films were deposited at100 W rf power and 0.65 Pa total pressure of mixed Ar and O2 gasseson unintentionally heated substrates and the distance between thetarget and the substrates was 11 cm. The plasma species duringdeposition were monitored by an Optical Emission Spectroscopy(OES) system (DGTWIN 350, JY-Horiba) whose details can be foundelsewhere [13].

Hall-effect, X-ray diffraction (XRD) and optical measurements wereperformed on films deposited on a single 1 in×1 in glass substratehaving the appropriate patterns, for the characterization methods

Thin Solid Films 520 (2011) 1202–1206

⁎ Corresponding author. Tel.: +30 2810 394123; fax: +30 2810 394106.E-mail address: [email protected] (E. Aperathitis).

0040-6090/$ – see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.tsf.2011.06.072

Contents lists available at ScienceDirect

Thin Solid Films

j ourna l homepage: www.e lsev ie r.com/ locate / ts f

Author's personal copy

employed, which were made by using standard photolithographictechnique. XRD was employed for identifying the structural phases inthe films (Siemens D5000, Cu Ka line 1.54 Å, film size 1.5 cm×1.5 cm).The electrical properties of the films were determined by Hall-effectmeasurements using the four-probe Van der Pauw technique (film size3.5 mm×3.5 mm) and the optical properties were examined by normalincidence transmittance in the UV–Visible–NIR range using PerkinElmer Lambda 950 dual beam spectrophotometer (film size1.5 cm×1.5 cm). Room temperature photoluminescence (RT-PL) emis-sions fromfilmsdeposited on Si substratewere recordedusinganHe-Cdlaser (CW, 325 nm) and the spectra were dispersed by a UV blazed600 g/mm grating monochromator and detected by a calibrated cooledCCD detector. The film thickness varied from 120 to 170 nm asdetermined by using a Veeco Dektak 150 stylus profilometer.

For the fabrication of the n-ZnN/p-ZnO diode, first a bottomcontact, Ti/Au, was deposited on the glass substrate and, usingstandard photolithographic technique, dots with diameter 700 μmwere formed. Then the patterned substrate was transferred in thesputtering system where the ZnN layer was deposited in 100% Arplasma followed by the ZnO layer which was deposited in %Ar–O2

plasma. The deposition has been done without breaking the vacuumof sputtering chamber. The formation of the diode was completed bylift-off process and by depositing Al as the top ohmic contact for theformed diode. The current–voltage (I–V) characteristics of the diodewere recorded using a programmable curve tracer (Sony, Tektronix370).

3. Results and discussion

3.1. Properties of ZnN and ZnO thin films

Thin films deposited in plasma containing pure Ar had significanthigher deposition rate than those deposited in plasma containing amixture of Ar and O2 gasses, as seen in Fig. 1. The deposition rate wasalmost constant for the Ar plasma containing 10–75% O2 (0.5–0.7 nm/min) but was further reduced for the plasma containing pureO2 gas (0.25 nm/min). This decrease in deposition rate is due to theatom of oxygen being lighter than that of argon. However, thepotential impact of the target surface stoichiometry changes into Zn-O-N during deposition in oxygen containing plasma (target poison-ing) cannot be completely excluded. Due to the very low depositionrates of O2-containing plasma, sputtering growth experiments withmore than 75% O2 in the plasma were not performed in this work.

The species in plasma during deposition were identified using OES,an in situ plasma-monitoring technique. Apart from the emission linesarising from the gasses forming the plasma (Ar and/or O2), zinc and

nitrogen species arising from the zinc nitride target could be identifiedregardless of the amount of oxygen in plasma. In Table 1 we report theidentified emission lines from zinc and nitrogen species in 0%, 50% and100%O2 in plasma. Only emission lines fromneutral and excited Zn andmolecular nitrogen atoms could be identified. The presence of atomicnitrogen species were not detected in plasma. From Table 1 we noticethat, as the amount of oxygen was increasing in the plasma, less Zn andN2 species arising from the target could be detected and also new onesappeared in pure oxygen plasma. The low deposition rate resulted inless detectable species in plasma which can explain the formerobservation but the appearance of new species in pure O2 plasma isnot yet understood. Taking into account the high chemical activity ofoxygen, it is anticipated that films formed in oxygen-rich plasmaconditions should be oxygen-rich, having zinc vacancies and lessnitrogen than the films formed in oxygen deficient plasma.

It has been reported that films sputtered from zinc nitride target ina mixture of Ar–O2–N2 gasses do have nitrogen incorporated in theirstructure as revealed from the energy dispersive X-ray analysis (EDX)experiments [14]. However, the films of the present work, whichweredeposited from zinc nitride target in Ar–O2 plasma, could notunambiguously give nitrogen peak when examined by EDX since itssignal was almost at the detection limit of the system. This indicatedthat the nitrogen content in the structure of the ZnO films should beless than 1%. These results will be later correlated with the electricaland photoluminescence (PL) emissions of the films. It should bepointed out, however, that all films deposited from the zinc nitridetarget in this work are expected to contain nitrogen in their structure.

XRD measurements revealed that all films in this investigationwere polycrystalline. The XRD patterns of the films deposited in Ar–O2

plasma as well as in pure Ar plasma for comparison are shown inFig. 2. The broad peak seen between 20 and 30° is the only one arisingfrom the glass substrates used. Films deposited in pure Ar plasmawere Zn-rich Zn3N2 films with small amount of ZnO, according to theJoint Committee of Powder Diffraction System (JCPDS) cards 04-0831,75-0762 and 36-1451 of Zn, Zn3N2 and ZnO, respectively [7]. Thestoichiometric Zn3N2 has the cubic antibixbyte structure. The oxidephase in this film must have formed due to adsorbed oxygen duringhandling of the films. By introducing oxygen in Ar plasma, XRDanalysis of the formed films revealed only the (002) peak of ZnOindicating that these films were single-phase ZnO with hexagonalwurtzite structure having high c-axis preferential orientation. The(002) ZnO peak appeared at slightly lower angle, by ~0.15°, than that

0 20 40 60 80 1000.0

0.5

1.0

7

7.05

7.1

7.15

O2% in Ar plasma

Dep

osi

tio

n r

ate

(nm

/min

)

Fig. 1. Deposition rate of films as a function of oxygen percentage in Ar–O2 plasma.

Table 1Zinc and nitrogen species (neutral and excited) along with their emission lines(wavelength in nanometers) in plasma containing % Ar–O2 gasses.

Species Wavelength(nm)

% O2 in Ar plasma

0% O2 50% O2 100% O2

Zn 467.7 √Zn 471.9 √Zn 480.6 √Zn+ 610.3 √ √Zn 636.3 √Zn+ 773.2 √ √

N2 333.4 √N2+ 354.4 √ √

N2+ 356.6 √

N2 367.1 √N2+ 388.7 √

N2 394.1 √N2 414.7 √ √N2+ 426.6 √ √

N2 591.2 √ √N2 639.6 √N2 687.5 √N2 727.5 √

1203V. Kambilafka et al. / Thin Solid Films 520 (2011) 1202–1206

Author's personal copy

of the JCPDS card, at 34.42°, must be attributed to the stress in thestructure due to existence of nitrogen in the Zn-O structure (theatomic radius of N is greater than O and smaller than Zn). It shouldalso be reminded that all films in this investigation were deposited onunintentionally heated substrates with no post-deposition annealingtreatment. However, since there are no diffraction data (JCPDS cards)available on zinc oxynitride phases, the existence of such phases in thefilms cannot be excluded. Using Scherer formula for the (002) peak ofZnO films, the grain size of the crystallites was found to be 11–14 nm.No apparent dependence of the shift of (002) peak to lower angles andthe grain size on the amount of oxygen in plasma during depositionwas observed. A typical Atomic Force Microscope (AFM) image of theZnO films is seen in Fig. 2 (b).

The presence of Zn in the films deposited in pure Ar plasma, asrevealed from the XRD features of Fig. 2, leads to opaque filmswhereas those films deposited in O2–Ar plasma, being single phaseZnO, were transparent. The transmittance of films deposited in % Ar–O2 mixtures is seen in Fig. 3. It is seen that all ZnO films (deposited inplasma containing O2) have a transparency of ~80–90% in the visibleregion of the spectrum. The increased transmittance below theabsorption edge (softening of the absorption edge) seen for the filmsdeposited in Ar–O2 plasma has also been observed for nitrogen doped[15] or undoped [16] ZnO films. The energy band gap of the films,extracted from Tauc plot, varied between 3.2 and 3.35 eV and thelargest energy gap was observed for films deposited in 50% O2 plasma,whereasmore than 50% oxygen in plasma seems to give filmswith thelower band gap. Taking into account the 1.5% error that can beintroduced in the calculation of the energy band gap, there was no

apparent relation between the band gap of the films and the % O2 inplasma. These changes in the energy gap, as will be seen in the nextparagraph, are not related with respective changes in the carrierconcentration of the films so to be ascribed to the Moss-Bursteineffect.

The electrical properties of the films deposited on glass substratesare depicted in Fig. 4. Films deposited in plasma containing up to 10%O2 were n-type materials, with the carrier concentration decreasingfrom ND~2×1019 cm−3 to around ND~1018 cm−3 as the amount ofoxygen in plasma was increasing (Fig. 4(a)). Increasing further thecontent of oxygen in plasma, Hall measurements could not giveunambiguous results on the type of carriers (electrons or holes)contributing to conduction. However, more than 30% O2 in plasmaresulted in p-type films with NA~1018 cm−3. Thus, increasing theamount of O2 in plasma the carrier concentration decreased by almosttwo orders of magnitude and the films turned from n-type into p-typematerial. The resistivity of films increased by around 2.5 orders ofmagnitude and their mobility decreased from ~4×100 cm2/Vs to~5×10−2 cm2/Vs with increasing oxygen in Ar plasma (Fig. 4(b)).

The Hall measurements confirmed that the ZnN sputtering usingplasma containing less than 20% O2 resulted in n-type ZnO films. Then-type conduction is attributed to interstitial Zn (Zni) and oxygenvacancies (VO) since Zni and VO are known to act as donors in ZnO,with the former being shallower than the latter [17]. The existence ofnitrogen in the structure of the films, though undetectable in ourfilms, is believed to be associated with the decrease in carriers'concentration and the small increase in carriers' mobility as theoxygen increases in plasma up to 20%. As plasma is becoming oxygen-rich, p-type conduction is taking over n-type conduction which isclearly observed for films deposited in more than 30% of O2 in Arplasma. Even though zinc vacancies (VZn), oxygen interstitial (Oi) andoxygen antisites (OZn) have been reported as acceptor states in ZnO[18,19], N-on-O substitution (NO) is considered the main acceptorcontributing to p-type conduction in ZnO:N films [20,21] since it hasthe shallower energy level in ZnO. In all published works on rf-sputtered p-type ZnO:N, the starting material was mainly ZnO [22–26] target and the dopant-nitrogen was incorporated in the structureof the film by introducing N2[22,24–26] or N2O [23] gasses in plasma(along with O2 when needed). In our work, the dopant-nitrogen wascoming straight from the target (zinc nitride) and the amount ofnitrogen incorporated in the ZnO film was controlled only by thepercentage of oxygen in plasma. It was found that, even if no atomic Ncould be identified in plasma as shown by OES results presented inTable 1, ZnN sputtering using Ar plasma containing 40–50% O2

20 30 40 50 60 70 80

* Zn o Zn N x ZnO

x

x

o

**

Inte

nsi

ty (

arb

. un

its)

2 (degree)

100% Ar - 0% O

70% Ar - 30% O

50% Ar - 50% O

25% Ar - 75% O

Fig. 2. XRD patterns of films deposited in plasma containing O2–Ar mixtures and typicalAFM image of ZnO films.

300 700 1100 1500 19000

20

40

60

80

100

Tran

smit

tace

(%

)

Wavelength (nm)

100% Ar - 0% O2

95% Ar - 5% O2

90% Ar - 10% O2

70% Ar - 30% O2

25% Ar - 75% O2

Fig. 3. Transmittance of films deposited in plasma containing O2–Ar mixtures.

1204 V. Kambilafka et al. / Thin Solid Films 520 (2011) 1202–1206

Author's personal copy

allowed us to produce p-type ZnO:N films. We can conclude that,under these specific sputtering conditions, the nitrogen in ZnO wasactivated and therefore giving rise to p-type conduction in the films.Even though the mechanism is not clearly understood yet, it isanticipated that since less zinc species were detected in oxygen-richplasma conditions than those deposited in oxygen-deficient plasmaconditions which formed n-type ZnO, the zinc vacancies must not beexcluded as contributors to p-type conduction. As seen in Fig. 4, p-ZnOfilms showed lower carrier concentration and mobility but higherresistivity than those of n-ZnO films. More work is in progress tounderstand the introduced defects and the scattering mechanisms inoxygen-rich p-ZnO films by examining films grown on heatedsubstrate and different sputtering conditions.

The room temperature photoluminescence (RT-PL) spectrum offilms deposited in plasma containing 5% O2, 30% O2 and 50% O2 on Sisubstrates (thereafter called n-ZnO, n+p-ZnO and p-ZnO films,respectively) is seen in Fig. 5. A general observation of RT-PL spectrareveals that all films showed a near-band-edge emission in theultraviolet region (UV, around 3.2–3.3 eV) and a broad one in thevisible region (yellow, around 2 eV), whereas the n+p-ZnO filmshowed an additional violet emission (at around 2.9 eV). The yellowemission (~2 eV) has also been observed for ZnO:N films made fromZnN and it has been associated with deep intrinsic defects of ZnO(Zn/O interstitials, VO) which interact with the valence and conduc-

tion bands [12,27]. Even though the energy of the UV emission is ingood agreement with the optical band gap of the films (Fig. 3), theratio however of UV-to-yellow intensities was much higher for the n-ZnO film than that of the other two films (2.43, 0.68 and 0.63,respectively). These observations indicate that oxygen-deficientplasma conditions produced n-ZnO films with suppressed radiativerecombination at mid-gap defects which gave rise to visible emissionsfor films deposited in oxygen-rich conditions. It is known that visibleemissions are related to VO and Zni states [17] and it is reasonable toanticipate, in agreement with the electrical properties of the filmsdiscussed in the previous section, that films deposited in oxygendeficient plasma (n-ZnO films) contained Zni and VO donors [28].However, as plasma was becoming richer in oxygen, NO as well as toVZn acceptor states [21] start playing a major role in the RT-LTemission. Violet emission was observed from n+p-ZnO films due toelectron-hole radiative recombination at acceptor states (oxygenrelated and VZn) and donor states (Zni and VO) [21,28]. In oxygen-richplasma, the violet emission was suppressed and emissions due todeep acceptor–donor states were dominating over shallow near-band-edge states, the latter however prevailing over the donor onesgiving rise to p-type ZnO [29]. More work is needed in understandingthe effect of dopants and defects on the PL emission spectra.

0 10 20 30 40 501017

1018

1019

1020

NA

NA

ND,A

NDND

ND

% O2 in Ar plasma

% O2 in Ar plasma

Car

rier

s co

nce

ntr

atio

n (

cm-3

)

(a)

0 10 20 30 40 5010-2

10-1

100

101

102

10-2

10-1

100

101

102 resistivity, (b)

<-- n-type -->

(cm

)

<-- p-type -->

mobility,

µ (

cm2/V

s)

Fig. 4. Electrical properties, as deduced fromHall-effectmeasurements, of films depositedin plasma containing % O2–Ar mixtures: (a) carrier concentration and (b) resistivity andmobility.

1.0 1.5 2.0 2.5 3.0 3.5 4.0

(b)

(a)

(a) 5% O - 95% Ar

(b) 30% O - 70% Ar

(c) 50% O - 50% Ar

RT-

PL

(ar

b. u

nit

s)

E (eV)

(c)

Fig. 5. Room temperature photoluminescence (RT-PL) of ZnO films deposited in plasmacontaining 3 different % O2–Armixtures: (a) 5% O2–95% Ar (b) 30% O2–70% Ar and (c) 50%O2–50% Ar. Near-band-edge emission: 3.2–3.3 eV, violet emission: 2.9 eV, yellowemission: 2 eV.

-5 -4 -3 -2 -1 0 1 2 3 4 5-200.0µ

0.0

200.0µ

400.0µ

600.0µ

800.0µ

1.0m

Dar

k C

urr

ent

(A

)

Voltage (V)

Fig. 6. Current–Voltage characteristics of the fabricated p-ZnO/n-ZnN diode from asingle sputtering target. The inset shows cross-section of the p/n diode.

1205V. Kambilafka et al. / Thin Solid Films 520 (2011) 1202–1206

Author's personal copy

3.2. Properties of n-ZnN/p-ZnO diode

As seen in the previous sections, by using zinc nitride as sputteringtarget it was possible to fabricate n-type material (n-ZnN) in Ar plasmaand p-type material (p-ZnO) by substituting 40–50% of Ar with O2 inplasma during deposition. Thus, the development of both p-type and n-type layers froma single sputtering targetwithoutbreakingvacuumandthe realization of a p-ZnO/n-ZnN diode could be achieved. The result isseen in Fig. 6. The inset of Fig. 6 shows a schematic structure of thefabricated p/n diode. The ZnN layerwas deposited in 100%Ar plasma onTi/Au-coated glass substrate which was followed by the deposition ofZnO layer in 50% Ar–50% O2 plasma. The thickness of ZnN and ZnO stacklayers was 230 nm. The p/n diodewas completed by the formation of Altop contact. All contacts were formed by electron-gun evaporation.Ti/Au was used as bottom contact since linear Transmission LineMeasurements (TLM) of Ti/Au on ZnN films gave contact resistance1.9×10−4Ω cm2. As top metal contact, various metal configurationswere examined, likeNi/Au, Pt/Au, Au andAl since it is known that ohmiccontacts formation on p-ZnO is an open technological issue [30].However, only Al, having the lowest work function (4.06 eV) than theothermetals examined [31] gave diodewith the characteristics of Fig. 6.It should bementioned that Al has been reported as ohmic contact on p-ZnO:Nfilms [31]. It is seen that the current–voltage (I–V) characteristicsof the diode showed rectifying behavior with turn on voltage at around2–2.5 V. Rectifying diodes fabricated from a single source have beenreported in the literature but these diodes were homojunctions sincetheyweremade fromZnO (p/n ZnO) byMBE [32] and sputtering [33] orp/n-CuInO2 by Pulsed LaserDeposition (PLD) [34]. The diode of Fig. 6 is aheterojunction since it ismade of two different zinc compounds namelyZnN and ZnO. Detailed analysis of diode's characteristics is in progressand it will be published elsewhere.

Furthermore, strictly speaking, the fabricated p-ZnO:N/n-ZnN diodein this work is not a transparent one. The ZnN layer was fabricated atroom temperature with no in situ (heated substrate) or ex situ (post-deposition) annealing [9] resulting in an opaque film as seen in Fig. 3.However, the above presented results are very promising towards thefabrication of transparent diodes, both heterostructure p-ZnO/n-ZnN aswell as homostructure p/n ZnO, in a rather economical way, since itrequires only one sputtering target. Further improvements of the diodeconfiguration and optimizing the deposition parameters as well as theprocessing procedure for the fabrication of the diode, enhancement ofits output characteristics can be achieved. These improvements indiode's performance have to be examined in conjunctionwith themaintechnological unresolved issue of p-ZnO namely reproducibility andlong term stability.

4. Conclusions

We reported here of an approach where the use of a single zincnitride rf sputtering target led to the growth of n-type ZnN andnitrogen-containing ZnO as well as p-type ZnO:N films by changingthe composition of sputtering plasma containing Ar–O2 gassesmixture. The films grown under plasma containing pure Ar gas wereopaque n-type ZnN films, whereas plasma containing up to 30%oxygen produced n-type ZnO films. A further increase of the oxygencontent in the plasma allowed us to grow p-type ZnO films. All ZnOfilms were transparent, conductive and the carrier concentration wasdecreasing as oxygen contain was increasing in plasma.

A heterostructure, based on p-ZnO and n-ZnN fabricated in a singledeposition run, exhibited rectifying behavior with turn-on voltage 2–2.5 V is reported here. Although more work is needed for improvingthe p-ZnO/n-ZnN diode output characteristics, we have demonstratedhere a technological approach, based on the room temperature rfsputtering of zinc nitride in plasma environment, for the fabrication ofboth Zn-based p/n diodes heterostructures as well as homostructures.

Acknowledgments

This work was partially supported by the European Community'sSeventh Framework Programme (FP7/2007–2013) under grantagreement number 246334.

References

[1] A.H. White, L. Kirschbraun, J. Amer. Chem. Soc. 28 (1906) 1343.[2] K. Kuriyama, Y. Takahashi, F. Sunohara, Phys. Rev. B 48 (1993) 2781.[3] T. Suda, K. Kakishita, J. Appl. Phys. 99 (2006) 076101.[4] M. Futsuhara, K. Yoshioka, O. Takai, Thin Solid Films 322 (1998) 274.[5] F. Zong, H. Ma, W. Du, J. Ma, X. Zhang, H. Xiao, F. Ji, C. Xue, Appl. Surf. Sci. 252

(2006) 7983.[6] Z. Jun, X. Shuwen, S. Lexi, J. Semicond. 31 (2010) 043001.[7] P. Voulgaropoulou, S. Dounis, V. Kambilafka, M. Androulidaki, M. Ružinský, V. Šály,

P. Prokein, Z. Viskadourakis, K. Tsagaraki, E. Aperathitis, Thin Solid Films 516(2008) 8170.

[8] K. Toyoura, H. Tsujimura, T. Goto, K. Hachiya, R. Hagiwara, Y. Ito, Thin Solid Films492 (2005) 88.

[9] E. Aperathitis, V. Kambilafka, M. Mondreanu, Thin Solid Films 518 (2009) 1036.[10] M. Bar, K.S. Ahn, S. Shet, Y. Yan, L. Weinhardt, O. Fuchs, M. Blum, S. Pookpanratana,

K. George, W. Yang, J.D. Denlinger, M. Al-Jassim, C. Heske, Appl. Phys. Lett. 94(2009) 012110.

[11] V. Kambilafka, P. Voulgaropoulou, S. Dounis, E. Iliopoulos, M. androulidaki, K.Tsagaraki, V. Saly, M. Ruzinsky, P. Prokein, E. Aperathitis, Thin Solid Films 515(2007) 8573.

[12] C.W. Zou, R.Q. Chen, W. Gao, Solid State Commun. 149 (2009) 2085.[13] M. Koufaki, M. Sifakis, E. Iliopoulos, N. Pelekanos, M. Modreanu, V. Cimalla, G.

Ecke, E. Aperathitis, Appl. Surf. Sci. 253 (2006) 405.[14] M. Nicolescu, M. Anastasescu, S. Preda, J.M. Calderon-Moreno, P. Osiceanu, M.

Gartner, V.S. Teodorescu, A.V. Maraloiu, V. Kambilafka, E. Aperathitis, M.Modreanu, J. Optoelectron. Adv. Mater. 12 (2010) 1045.

[15] T.M. Barnes, J. Leaf, S. Hand, C. Fry, C.A. Wolden, J. Appl. Phys. 96 (2004) 7026.[16] J.F. Muth, R.M. Kolbas, A.K. Sharma, S. Oktyabrsky, J. Narayan, J. Appl. Phys. 85

(1999) 7884.[17] H.W. Liang, Y.M. Lu, D.Z. Shen, J.F. Yan, B.H. Li, J.Y. Zhang, Y.C. Liu, X.W. Fan, J. Cryst.

Growth 278 (2005) 305.[18] S.B. Zhang, S.H. Wie, A. Zunger, Phys. Rev. B 63 (2001) 075205.[19] Y.J. Zeng, Z.Z. Ye, W.Z. Xu, J.G. Lu, H.P. He, L.P. Zhu, B.H. Zhao, Y. Che, S.B. Zhang,

Appl. Phys. Lett. 88 (2006) 262103.[20] Y. Yan, S.B. Zang, S.T. Pantelides, Phys. Rev. Lett. 86 (2001) 5723.[21] J. Lu, Q. Liang, Y. Zhang, Z. Ye, S. Fujita, J. Phys. D: Appl. Phys. 40 (2007) 3177.[22] M.-L. Tu, Y.K. Su, C.Y. Ma, J. Appl. Phys. 100 (2006) 053705.[23] J. Lim, K. Shin, C. Lee, J. Mater. Sci. 39 (2004) 3195.[24] J. Wang, V. Sallet, F. Jomard, A.M. Botelho do Rego, E. Elamurugu, R. Martins, E.

Fortunato, Thin Solid Films 515 (2007) 8780.[25] J. Karamdela, C.F. Dee, B.Y. Majlis, Appl. Surf. Sci. 256 (2010) 6164.[26] T. Yen, M. DiNezza, A. Haungs, S.J. Kim, W.A. Anderson, A.N. Cartwright, J. Vac. Sci.

Technol. B 27 (2009) 1943.[27] Y. Nakano, T. Morikawa, T. Ohwaki, Y. Taga, Appl. Phys. Lett. 88 (2006) 172103.[28] Y.W. Heo, D.P. Norton, S.J. Pearton, J. Appl. Phys. 98 (2005) 073502.[29] N. Yuan, L. Fan, J. Li, X. Wang, Appl. Surf. Sci. 253 (2007) 4990.[30] K. Ip, G.T. Thaler, H. Yang, S.Y. Han, Y. Li, D.P. Norton, S.J. Pearton, S. Jang, F. Ren, J.

Cryst. Growth 287 (2006) 149.[31] S. Majumdar, P. Banerji, J. Appl. Phys. 107 (2010) 063702;

Sens. Actuators B: Chem. 140 (2009) 134.[32] Z.P. Wei, Y.M. Lu, D.Z. Shen, Z.Z. Zhang, B. Yao, B.H. Li, J.Y. Zhang, D.X. Zhao, X.W.

Fan, Z.K. Tang, Appl. Phys. Lett. 90 (2007) 042113.[33] G. Xiong, J. Wilkinson, B. Mischuck, S. Tuzemen, K.B. Ucer, R.T. Williams, Appl.

Phys. Lett. 80 (2002) 1195.[34] H. Yanagi, K. Ueda, H. Ohta, M. Orita, M. Hirano, H. Hosono, Solid State Commun.

121 (2002) 15.

1206 V. Kambilafka et al. / Thin Solid Films 520 (2011) 1202–1206