Materials

Download as pdf or txt
Download as pdf or txt
You are on page 1of 18
 
Citation:
 Rodríguez-Tapiador, M.I.;Merino, J.; Jawhari, T.; Muñoz-Rosas,A.L.; Bertomeu, J.; Fernández, S.Impact of the RF Power on theCopper Nitride Films Deposited in aPure Nitrogen Environment forApplications as Eco-Friendly SolarAbsorber.
 Materials
 2023
,
 16
, 1508.https://doi.org/10.3390/ma16041508Academic Editor: Daniel GeorgievReceived: 5 December 2022Revised: 3 February 2023Accepted: 9 February 2023Published: 10 February 2023
Copyright:
 © 2023 by the authors.Licensee MDPI, Basel, Switzerland.This article is an open access articledistributed under the terms andconditions of the Creative CommonsAttribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
materials
 Article
Impact of the RF Power on the Copper Nitride Films Depositedin a Pure Nitrogen Environment for Applications asEco-Friendly Solar Absorber
M. I. Rodr
í
guez-Tapiador
 1,
*, J. Merino
 2
 , T. Jawhari
 3
 , A. L. Muñoz-Rosas
 4
 , J. Bertomeu
 4
and S. Fern
á
ndez
 1,
*
1
Energy Department, CIEMAT, Av. Complutense 40, 28040 Madrid, Spain
2
Technology Support Center CAT, University Rey Juan Carlos, Tulip
á
n, s/n, 28039 M
ó
stoles, Spain
3
Unitat d’Espectrosc
ò
pia Raman, Centres Cient
í
fics i Tecnol
ò
gics de la Universitat de Barcelona—CCiTUB,Llu
í
s Sol
é
 i Sabar
í
s, 1-3, 08028 Barcelona, Spain
4
Departament de F
í
sica Aplicada, Universitat de Barcelona, 08028 Barcelona, Spain
*
 Correspondence: mariaisabel.rodriguez@ciemat.es (M.I.R.-T.); susanamaria.fernandez@ciemat.es (S.F.)
Abstract:
 Thismaterialcanbeconsideredtobeaninterestingeco-friendlychoicetobeusedinthepho-tovoltaic field. In this work, we present the fabrication of Cu
3
N thin films by reactive radio-frequency
(RF) magnetron sputtering at room temperature, using nitrogen as the process gas. Different RFpower values ranged from 25 to 200 W and gas pressures of 3.5 and 5 Pa were tested to determinetheir impact on the film properties. The morphology and structure were exhaustively examined by Atomic Force Microscopy (AFM), Fourier Transform Infrared (FTIR) and Raman Spectroscopiesand X-ray Diffraction (XRD), respectively. The AFM micrographs revealed different morphologiesdepending on the total pressure used, and rougher surfaces when the films were deposited at thelowest pressure; whereas FTIR and Raman spectra exhibited the characteristics bands related to the
Cu-N bonds of Cu
3
N. Such bands became narrower as the RF power increased. XRD patterns showed
the (100) plane as the preferred orientation, that changed to (111) with the RF power, revealing a
worsening in structural quality. Finally, the band gap energy was estimated from transmission spectracarried out with a Perkin Elmer 1050 spectrophotometer to evaluate the suitability of Cu
3
N as a light
absorber. The values obtained demonstrated the capability of Cu
3
N for solar energy conversionapplications, indicating a better film performance under the sputtering conditions 5.0 Pa and RF
power values ranged from 50 to 100 W.
Keywords:
 copper nitride; reactive magnetron sputtering; novel absorbers; photovoltaic devices
1. Introduction
Transition metal nitride materials such as copper nitride (Cu
3
N) show interestingoutstanding properties, such as optical, electrical and energy storage properties, which
have allowed this material to be used in several application fields. Specifically, Cu
3
N has
attractedgreatinterestasanewsolarabsorbermaterialforflexibleandlightweightthinfilm
solar cells [
1
,
2
]. This metastable semiconductor is non-toxic, composed of earth-abundant
elements and its band gap energy can be easily tunable depending on both the manufactur-
ing conditions and the deposition methods. Among the application fields, it can be found
in integrated circuits, photodetectors, optoelectronics, energy conversion applications andother technologies [
3
5
]. Highlighting the use of Cu
3
N, as a new solar absorber material for
flexible and lightweight thin film solar cell technology [
6
]. Its development has attractedconsiderable attention with the purpose of being incorporated into novel designs withina future generation of cost-effective photovoltaic devices. This increased interest as anabsorber in photovoltaic (PV) applications has mainly focused on its non-toxicity and its
earth abundance, which makes it an eco-friendly material. In addition, the band gap energy
value for the band gap of the Cu
3
N is ~0.9 eV [
7
,
8
], but the experimental results of the
 Materials
 2023
,
 16
 
 Materials
 2023
,
 16
, 1508 2 of 18
indirect and direct band gap show that they range from 1.17 to 1.69 eV and from 1.72 to
2.38 eV, respectively [
8
,
9
]. These obtained values mean that the Cu
3
N can be considered a
potential next-generation solar absorber, a possible candidate to replace the silicon in solar
cell applications. Such variations in the band gap can be attributed to a change in the stoi-
chiometry of the Cu-N system and to the presence of unintentional oxygen impurities into
the lattice. In addition, this material shows a very high absorption coefficient of 10
5
cm
1
above
 ≈
2.0 eV [
], which reinforces the idea of being used as an absorber for thin-film PVtechnologies [
1
]. Furthermore, this material can also show a p-type character, which can beobtained by doping it with interstitial fluorine [
]. The possibility to easily achieve both nand p characters by doping for the same material is useful. In this sense, the development
of both p-type and n-type sandwiched thin layers, establishing an n-p junction to obtainpower from sunlight in an economical way, entails an important saving in raw materials,
among other advantages.
At room temperature (RT), Cu
3
N is a metastable semiconductor with an anti-ReO
3
cubic crystal structure (see Figure 1). Its structure exhibits the nitrogen (N) atoms atthe corners of the unit cell and the copper (Cu) atoms, located at the center of the cubicedges. Because of that, this crystal structure is extremely favorable to insert metal atomsat the interstitial body center site (
12
,
12
,
12
) [
], a fact that could lead to different chemicalinteractions between Cu and N atoms, affecting the electronic structure of the material.
Among the most common elements that can be used to be inserted in that position, we can
find alkali metals (Li) [
13
], transition metals (Ti, Pb, Ni, Zn, Cr, Fe, Mn, Al and Sc) [
14
]
and/or non-metals (H and O) [17,18].
 
Figure 1.
 Picture of the anti-ReO
3
 crystal structure of the Cu
3
N compound. The copper atoms are
represented by the yellow balls, and the nitrogen by the grey balls.
Depending on its chemical composition, the Cu-N compound family can offer a wide
variety of optoelectronic properties, i.e., the Cu-rich Cu
4
N films that shows a metallic
 behavior, the Cu-rich, N-rich and stoichiometric Cu
3
N films that present a semiconductingperformance [
]. Hence, it is clear that there is a strong relationship, between its chemical bonds and its electronic properties. This fact is mainly attributed to the hybridization effect
 between the Cu 3d-N 2p bands and the Cu (4s, 4p) conduction bands, which gives rise to
the desired covalent bonding effect [7].
On the other hand, the different natures observed in this material can be reached by modifying its deposition conditions and/or depending on the fabrication technique
used. The differences in morphological, structural, chemical and optoelectronic properties
reported in the literature are attributed to the vacant interstitial sites occupied by atomsother than Cu and the chemical interactions between them [
20
]. That is why the
knowledge of its local atomic structure is of great importance; it permits the determinationofthemostsuitablematerialthatfitsthedevicerequirements. Inaddition, bymodifyingthedeposition conditions, both the lattice constant and the band gap energies can also be easily
tuned, from 0.3815 to 0.3885 nm [
], and in a surprisingly wide range of values from 0.23
to 1.90 eV [
], respectively. Many reports have shown that the abovementioned materialhas been fabricated by different approaches, such as pulsed laser deposition (RPLD) [
],
5
 
 Materials
 2023
,
 16
, 1508 3 of 18
molecular beam epitaxy (MBE) [
26
], atomic layer deposition (ALD) [
], ion-assisteddeposition [
26
], direct current (DC) triode sputtering [
28
] and magnetron sputtering [
].Among them, one of the most commonly used approaches is reactive radio-frequency(RF) magnetron sputtering due to the ease with which this technique modifies the film
properties by changing the deposition parameters. It is also preferred because of its great
capability of scalability to large areas [
4
]; it is a reliable, robust and cost-effective technique
that can be easily incorporated into an industrial production chain. In this sense, Cu
3
Nfilms with different structural, optoelectronic properties have been already reported bymodifying parameters such as the nitrogen fraction in the gas mixture, the power value,
the type of substrate, the substrate-to-target distance and the substrate temperature, among
others [23,2931].
Finally, the thermal decomposition of Cu
3
N is a parameter that should also be con-sidered because it occurs at a temperature that is not relatively high, from 100 to 470
 
C.
Okrasa et al. [
] reported that, at a high temperature of 200
 
C, structural changes related
to the release of pure Cu could be observed.
This work presents the fabrication of a Cu
3
N binary compound by reactive RF mag-
netron sputtering at RT in a pure N
2
 environment by modifying the RF power and the total
working pressure. In our previous works, we observed that the use of an argon (Ar)-freeenvironment during the sputtering process led to films with better structural quality, the(100) plane being the preferential orientation, and smoother surfaces when Ar was usedduring the sputtering process [
]. Considering that the design of desired functional ma-terial properties by controlling deposition parameters is a key technological topic, in this
work, we establish the crystal nature, morphology, electrical and optical properties of the
deposited Cu
3
N films as functions of the preparation conditions. Our main goal is to study
the impact of both the RF power and the gas pressure on the thin film properties, and tothen adjust them according to the device requirements. Thanks to the possibility of suchtuning, an ad hoc material could be fabricated with suitable properties to be used as a
potential substitute absorber of silicon in PV devices. Finally, the sputtering conditions that
lead to a best thin film performance are discussed.
2. Materials and Methods
The deposition of the Cu
3
N films was carried out using a commercial MVSystem LLC(Golden, CO, USA) mono-chamber sputtering system, where the gun was radio-frequency
(RF) operated and vertically adjustable. The substrates used were Corning glass 1737F(Corning Inc., New York, NY, USA) and <100> polished n-type floating zone crystallinesilicon (c-Si) wafers. The 3-inch diameter and 6-mm-thick Cu target was from LeskerCompany (St. Leonards-on-Sea, East Sussex, UK) and had a purity of 99.99%. Prior toloading the substrates into the sputtering chamber, native silicon dioxide was removedfrom the surface silicon wafer with a 1% HF solution in deionized water and ethanol for5 min. On the other hand, the glass was ultrasonically cleaned for 10 min with ethanoland deionized water, and it was finally immersed in isopropyl alcohol. Afterwards, all
substratesweredriedbyblowingnitrogenoverthem. Thesputteringchamberwaspumpedto a base pressure of 10
5
Pa, whereas the sputtering process was carried out in an Ar-freeenvironment with the total working pressure set to 3.5 and 5 Pa, respectively, adjusted with
a “butterfly” valve. The process gas was nitrogen (N
2
) (99.999%), with a flow rate of 20
sccm controlled by a mass flow controller (MFC) (MKS Instruments, Andover, MA, USA).The distance between the target and the substrate was set to 10 cm. The RF power varied
from 25 to 200 W, and the deposition time was modified from 420 to 1800 s to offset the
effect of the power. In all of the experiments, the substrate was not intentionally heated.
The thickness, the structure, the morphology and the chemical composition of the
Cu
3
N films were studied by profilometry, X-ray diffraction (XRD), atomic force microscopy
(AFM), Energy-Dispersive X-ray Spectroscopy (EDS), Raman and Fourier transform in-frared spectroscopy (FT-IR) in transmission mode, respectively. The thickness was mea-sured using a Veeco Dektak 8 Stylus profiler. The crystal structure, the grain size and the
5
 
 Materials
 2023
,
 16
, 1508 4 of 18
lattice parameters were determined from the XRD diffraction pattern, measured with a
Panalytical
 powder diffractometer, model X’Pert MPD/MRD, using the Cu-k
α 
 radiation(
λ
 = 0.15406 nm). The scanned 2
θ
 range was 10–60
at a step size of 0.1
. The topogra-phy of the Cu
3
N was evaluated with a multimode nanoscope AFM model III A (SPM,
Veeco-Digital instrument) in tapping mode, using silicon nitride AFM tips (OTR8, Veeco).
The roughness of the samples was quantified by root mean square (RMS); meanwhile, the
grain size was also determined, in both 1
 ×
 1
 µ 
m
2
2D AFM images using a commercial
software (Gwyddion software 2.61, http://gwyddion.net/) (accessed on 31 January 2023).
A semi-quantitative elemental chemical analysis was carried out using an Oxford detec-tor energy dispersive X-ray spectroscopy (EDS) system equipped with a high-resolutionfield emission SEM (JEOL JSM 6335F). Determinations of the molecular structure wereperformed with a dispersive spectrometer Horiba Jobin-Yvon LabRam HR 800 coupled
to an optical microscope Olympus BXFM, using a solid-state laser as an excitation source
emitting at 532 nm. Furthermore, the laser power at the sample was around 5 mW andthe used microscope objectives were 100
×
 and 50
×
. A Perkin Elmer Spectrum 100 FT-IRwas also used to complement the information obtained from the Raman measurements.
The spectra were measured in the transmittance mode in the wave number, which ranged
 between 400–4000 cm
1
. Electrical properties, such as sheet resistance and resistivity, wereobtained with a commercial 4-point probe measurement system (Signatone, EEUU). It was
estimated that the errors in the measured parameters were around 2%. Finally, the optical
transmittance spectra were measured at normal incidence using a UV/VIS/NIR
 Perkin
Elmer
 Lambda 1050 spectrophotometer. From these spectra, the optical band gap energies
E
g
 from the indirect and direct transitions were calculated to determine the most favourable
sputtering conditions, which obtained a suitable solar absorber. The glass substrate wasused to characterize the films by profilometry, XRD, AFM and optical properties. On the
other hand, the silicon substrate was used to carry out the Raman and FTIR measurements.
3. Results and Discussion
All of the Cu
3
N films presented in this work were physically stable with good adher-
ence to the substrates after exposure to ambient air. No cracking or peel-off was observed
after the deposition. Table 1 describes the sputtering deposition conditions used and the
measured film thickness of the samples.
Table 1.
 Sputtering deposition conditions used to fabricate the Cu
3
N films at RT in N
2
 pure atmo-
sphere at different RF powers and gas pressures. Film thickness is also included.
Total Gas Pressure: 3.5 Pa
RF Power (W) 25 50 75 100 150 200Film thickness (nm) 45 85 95 100 175 215Deposition time (s) 1800 1800 900 900 600 420
Total Gas Pressure: 5.0 Pa
RF Power (W) 25 50 75 100 150 200Film thickness (nm) 40 85 90 100 155 210Deposition time (s) 1800 1800 900 900 600 420
Figure 2 shows the deposition rate as a function of the RF power values used. This
parameterwascalculatedfromtheCu
3
Nfilmthickness,dataobtainedwiththeprofilometer
and the deposition time (see data in Table 1). In spite of trying to adjust the deposition
time to obtain similar thicknesses, film thicknesses ranged from 40 to 215 nm. As expected,
the deposition rate increased with the RF power. This tendency can be attributed to the
enhancement of the Cu atoms flow obtained as RF power rises at a constant gas pressure.
It should be pointed out that a linear trend with two different slopes was observed as afunction of the RF power value. This indicates that two different growth regimes weretaking place, which had an effect on the film properties. On the other hand, there was
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5

Reward Your Curiosity

Everything you want to read.
Anytime. Anywhere. Any device.
No Commitment. Cancel anytime.
576648e32a3d8b82ca71961b7a986505