Access to this full-text is provided by AIP Publishing.
Content available from APL Materials
This content is subject to copyright. Terms and conditions apply.
APL MATERIALS 4, 046106 (2016)
Transparent ferrimagnetic semiconducting CuCr2O4thin
films by atomic layer deposition
T. S. Tripathi,1C. S. Yadav,2and M. Karppinen1,a
1Department of Chemistry, Aalto University, P.O. Box 16100, FI-00076 Aalto, Finland
2School of Basic Sciences, Indian Institute of Technology Mandi, Mandi, H.P. 175001, India
(Received 27 February 2016; accepted 4 April 2016; published online 14 April 2016)
We report the magnetic and optical properties of CuCr2O4thin films fabricated by
atomic layer deposition (ALD) from Cu(thd)2, Cr(acac)3, and ozone; we deposit
200 nm thick films and anneal them at 700 ◦C in oxygen atmosphere to crystal-
lize the spinel phase. A ferrimagnetic transition at 140 K and a direct bandgap
of 1.36 eV are determined for the films from magnetic and UV-vis spectrophoto-
metric measurements. Electrical transport measurements confirm the p-type semi-
conducting behavior of the films. As the ALD technique allows the deposition
of conformal pin-hole-free coatings on complex 3D surfaces, our CuCr2O4films
are interesting material candidates for various frontier applications. C2016 Au-
thor(s). All article content, except where otherwise noted, is licensed under a Creative
Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
[http://dx.doi.org/10.1063/1.4946884]
Atomic layer deposition (ALD) is a state-of-the-art thin-film deposition technique particu-
larly advantageous to applications where precisely thickness-controlled and pin-hole-free nanoscale
coatings are required on high-aspect-ratio substrates. The unique atomic layer-by-layer growth
mechanism via self-saturative surface reactions makes it attractive for the fabrication of modern
semiconducting devices with complex 3D architectures.1–4Here, we use the technique to deposit
high-quality transparent semiconducting p-type CuCr2O4thin films. Spinel-structured chromium
oxides ACr2O4(A=Cu, Mn, Fe, Co, and Ni) exhibit a wide range of electronic, magnetic, and
optical properties through the variation of the A-site cation constituent.
At high temperatures, the ACr2O4compounds possess the normal cubic spinel structure with
space group Fd3m. Within this structure, cations on the A and B sites form a bipartite lattice
system. The A2+cations sit on the tetrahedral sites in a diamond sublattice formed by the oxygen
atoms, while the octahedrally coordinated B-site Cr3+cations form a pyrochlore-type sublattice
with edge-sharing oxygen octahedra.5The nature and the type of intra- and/or inter-site interactions
between the A- and B-site cations are then the main source of various exotic ground states in this
material family.6,7In particular, the strong preference of Cr3+cations towards octahedral coordina-
tion minimizes the cation site disorder in ACr2O4spinels. They are thus ferrimagnetic in nature and
have been identified as, e.g., novel spin filter materials for magnetic junctions.8,9Most importantly,
it has been observed that the isostructural interface between half-metallic Fe3O4electrodes and the
spinel barrier layer gives rise to a significant increase in junction magnetoresistance that surpasses
the values previously reported for Fe3O4-based junctions. Very recently, an optical resistive switch-
ing behavior was reported for the A =Cu member of the ACr2O4family in Ag/CuCr2O4/FTO
devices.10 Besides, the same CuCr2O4phase with the unique tetragonally distorted normal spinel
structure with c/a<111 has also been highlighted as a prominent catalyst in, e.g., various oxidation,
hydrogenation, and alkylation reactions of large-scale industrial relevance.12–14 A more recent appli-
cation area is the solid oxide fuel cell where spinel oxides including CuCr2O4are being explored
amaarit.karppinen@aalto.fi
2166-532X/2016/4(4)/046106/7 4, 046106-1 ©Author(s) 2016.
046106-2 Tripathi, Yadav, and Karppinen APL Mater. 4, 046106 (2016)
as interconnect materials that need to show both high corrosion resistance and high electrical
conductivity.15 Moreover, CuCr2O4and related materials have found usable as burn rate modifiers
in solid propellant processing for space launch vehicles.16–18 The various application possibilities
and corresponding fabrication challenges were recently summarized in the review article of Prasad
and Singh,19 underlining the importance of further endeavors to explore CuCr2O4for its improved
functionalities.
Normally, the spinel oxides are insulating in nature due to their prevailing ionic bonding.
However, there are few unambiguous reports regarding the electrical conductivity of CuCr2O4.14,20
In some reports, it has been found to be an insulator while in others a p-type narrow-bandgap semi-
conductor.21 Magnetically, it is ferrimagnetic and crystal structure-wise a tetragonally distorted (a
=b=6.03 Å, c =7.78 Å, c/a=1.29) normal spinel at room temperature.22,23 The tetragonal
distortion is a consequence of the Jahn–Teller effect of Cu2+cations that removes the ground-state
degeneracy resulting in the flattening of the CuO4tetrahedra to the lower-symmetry tetragonal
phase. However, upon heating, it transforms to the cubic spinel phase near 600 ◦C, where the
compression of the CuO4tetrahedra is removed by orbital melting (transition from an orbital-
ordered to orbital-disordered state).24
There are very few thin-film studies for CuCr2O4. An enhanced magnetization in excess of
200% of the accepted bulk value was reported due to substrate strain for pulsed laser deposited
(PLD) epitaxial CuCr2O4thin films on single-crystal (110) MgAl2O4substrates.25 The spin-coating
method was employed for depositing CuCr2O4films for resistive switching10 and the phase was
also found as a co-product in some rf-sputtered Cu–Cr–O films.26–28 There is one report where the
metal organic chemical vapor deposition (MOCVD) technique was used, with copper and chro-
mium acetyl acetonate as precursors.29 To the best of our knowledge, there are no ALD processes
reported for this compound. Motivated by its attractive basic physical properties relevant to frontier
applications where the ALD technology could be expected to be highly beneficial, we decided
to develop an ALD process for CuCr2O4and characterize the resultant thin films for their basic
magnetic, optical, and electrical transport properties. As a starting point for the work, we utilize the
detailed ALD growth parameters we recently optimized for the deposition of delafossite CuCrO2
thin films.30
For the deposition of CuCr2O4thin films, copper 2,2,6,6-tetramethyl-3,5-heptanedionate
(Cu(thd)2) and chromium acetyl acetonate (Cr(acac)3) were used as metal precursors and ozone as
the oxygen source; the depositions were carried out in a commercial hot-wall flow-type F-120 ALD
reactor (ASM Microchemistry Ltd., Finland). The reactor was operated under a nitrogen (99.9995%)
pressure of 2–3 mbar produced with a NITROX UHPN 3000 nitrogen generator. Nitrogen gas was
used both as a carrier and purging gas. For the depositions, Cu(thd)2was prepared in-house from
copper acetate (Fluka; 98%) and 2,2,6,6-tetramethyl hepatane-3,5-dione (Fluka; >98%), whereas for
Cr(acac)3(97.5%) commercial powder (STREM Chemicals) was used. Ozone was produced from
oxygen (99.999%; Fischer model 502 laboratory ozone generator) and pulsed into the reactor through
a needle valve and a solenoid valve from the main ozone flow line.
The sublimation temperatures of the metal precursors, Cu(thd)2and Cr(acac)3, were 120 and
130 ◦C, respectively;30 the precursors were sublimed from open glass boats held inside the reactor.
The pulse times were fixed at 2 s for all the three precursors (two metal precursors and ozone),
followed by a 3 s long N2purge based on our work on delafossite CuCrO2films,30 and all
the depositions were carried out at 250 ◦C. To adjust the film composition to the desired copper
and chromium molar ratio of 1:2, the following super-cycle sequence was optimized: (Cu(thd)2
+O3)+4×(Cr(acac)3+O3). The correct elemental composition of the films was confirmed us-
ing wavelength-dispersive X-ray fluorescence spectroscopy (WD-XRF; PANanalytical AxiosmAX
microanalysis system equipped with SST-mAX X-ray tube that virtually eliminates instrument
drift). The aforementioned super-cycle was then repeated 300 times to yield CuCr2O4films with
the desired thickness of ca. 200 nm, i.e., with the growth rate of 0.67 nm per super-cycle. It is
commonly observed for ALD-grown ternary oxides31–34 that the as-deposited films are amorphous,
and a post-deposition annealing is required for crystallization. The present films were annealed
at 700 ◦C in a rapid thermal annealing (RTA) furnace (PEO 601; ATV Technologie GmbH) in an
oxygen gas flow.
046106-3 Tripathi, Yadav, and Karppinen APL Mater. 4, 046106 (2016)
FIG. 1. GIXRD patterns and AFM images for as-deposited and annealed CuCr2O4films. The red lines show the diffraction
peaks matched with JCPDS reference data for CuCr2O4.
The films were deposited on borosilicate glass as the substrate material to allow the electrical
transport and optical measurements. The UV-vis spectrophotometric measurements were carried
out in the wavelength range of 190–1100 nm (Hitachi-U 2000 spectrophotometer). Grazing inci-
dence X-ray diffraction measurement (GIXRD; PANalytical model X’pert Pro diffractometer, Cu
Kαradiation) was performed to identify the crystal structure of the post-deposition annealed films.
The same diffractometer was used for the thickness and density determination of the as-deposited
thin films from X-ray reflectivity (XRR) patterns. The thickness of the films after annealing was
confirmed from the cantilever tip jump using an atomic force microscope (AFM; TopoMetrix Ex-
plorer).35 It should be noted that within the estimated error limits (<5%), the thickness of the
films remained the same. For the physical property measurements, we deposited ca. 200 nm thick
films. The structural parameters such as grain size (D) and dislocation density (δ) were calculated
from the GIXRD patterns. The grain size of the thin films was calculated using Debye Scher-
rer’s formula, D =0.9λ/βcos θ, where D is the grain size, λis the X-ray wavelength used, β
046106-4 Tripathi, Yadav, and Karppinen APL Mater. 4, 046106 (2016)
is the full width at half-maximum (FWHM) intensity in radians, and θis Bragg’s angle. Addi-
tionally, the dislocation density (δ) was evaluated by the formula,36 δ=1/D2, where the larger
D and smaller δvalues indicate better crystallization of the films. The surface topography and
root-mean-square (RMS) roughness measurements were performed using the same atomic force
microscope. Magnetic measurements were performed with DynaCool physical property measure-
ment system (PPMS) from Quantum design equipped with vibration sample magnetometer (VSM).
The specimen dimension was 10 ×5 mm2.
In Figure 1, we show the GIXRD patterns for both as-deposited and O2-annealed films. The
as-deposited film is amorphous but, as expected, crystallizes upon the O2-annealing at 700 ◦C. The
annealing temperature 700 ◦C was chosen based on literature data revealing that the pure CuCr2O4
phase is formed above 450◦C, the degree of crystallinity enhancing with increasing temperature.37
All the peaks in the GIXRD pattern can be readily indexed according to the spinel CuCr2O4struc-
ture (JCPDS 05-0657). It is to be mentioned here that we have tried annealing the films in Ar
atmosphere also but the resultant films had mixed phases of CuCr2O4and CuCrO2. The grain size
estimated using the FWHM of the (211) peak at 2θ≈35.16 is ca. 30 nm. The dislocation density
estimated for our CuCr2O4films is 0.001 nm−2.
We also took AFM images to demonstrate the changes in surface topography and RMS rough-
ness upon the post-deposition annealing of our CuCr2O4thin films, see Figure 1. Congruent to the
GIXRD data, the RMS roughness value of 1.40 nm for the as-deposited film was found to increase
to 5.36 nm for the annealed crystalline film; this trend is common for many ALD-grown amorphous
films, which are extremely smooth after deposition and then upon the post-deposition heat treatment
crystallize and become rougher.
To confirm the ferrimagnetic phase transition reported in literature for CuCr2O4, we measured
the temperature dependence of magnetization at a magnetic field of 20 kOe for our crystalline
CuCr2O4thin film, see Figure 2. The data indicate that the Curie temperature (TC) is approximately
140 K, which is in close agreement with the TCvalues reported for epitaxial film25 and bulk powder
sample23 of CuCr2O4. In the inset of Figure 2, we plot the magnetic hysteresis data measured at
10 K (after subtracting the diamagnetic contribution from the substrate). An average saturation
magnetic moment of 0.28 µB/f.u.was estimated from the data using the calculated density 3.9
g/cc from the XRR fittings. The calculated density is, however, lower than the value of 5.4 g/cc
reported in literature for bulk samples.38 The estimated saturation magnetization of 0.28 µB/f.u is
less than the values of 0.5039 and 0.725 µB/f.u40 reported in literature for single-crystal samples.
FIG. 2. Temperature dependence of magnetization measured for an annealed CuCr2O4film in an applied magnetic field of
20 kOe; the inset shows the hysteresis loop at 10 K.
046106-5 Tripathi, Yadav, and Karppinen APL Mater. 4, 046106 (2016)
FIG. 3. UV-vis spectra measured as transmittance and reflectance for an annealed CuCr2O4film and for the borosilicate
substrate for reference; the lower panel demonstrates the determination of the direct energy bandgap. The inset shows the
absorption coefficient as a function of wavelength.
The discrepancy in saturation moments may be related to complex non-collinear23,39,40 nature of
the magnetic structure with increased tetragonality due to an increase in the angle between the
Cr3+–Cr3+spins. The increased angle results in decreased collinear projection along the Cu2+spins,
yielding the decreased net moment.25
In Figure 3, we present UV-vis spectra measured for the crystalline CuCr2O4film. High
transmittance (>60%) is observed in the wavelength range 800–1000 nm. As for the bandgap of
CuCr2O4, there are very few reports in literature.21 The bandgap and the absorption coefficient
(α) at each wavelength have been calculated as reported by us in Ref. 30. The bandgap energy is
then obtained as shown in the lower panel of Figure 3. The estimated bandgap 1.36 eV is in good
agreement with values recently reported by Bajaj et al.21 for BiVO4/CuCr2O4composites.
Finally in Figure 4, we show the Seebeck coefficient (S) and electrical resistivity (ρ) data for
CuCr2O4. The p-type conductivity and the semiconducting-type behavior with temperature are seen
from the positive Seebeck values and the negative dρ/dT slope, respectively. The inset shows the
well-known exponential law between resistivity and temperature
ρ=ρ0exp(∆E/kBT),(1)
046106-6 Tripathi, Yadav, and Karppinen APL Mater. 4, 046106 (2016)
FIG. 4. Electrical transport properties measured as Seebeck coefficient and electrical resistivity as a function of temperature.
The inset shows the log ρvs 1/T plot for the calculation of electrical bandgap or activation energy.
where ρis resistivity, ρ0is constant, ∆E is bandgap or activation energy, kBis Boltzmann constant,
and T is absolute temperature. The relation is obeyed strictly in the measured temperature range,
which confirms the intrinsic semiconducting nature of the film. The value of ∆E=0.17 eV is in
good agreement with values reported for similar oxide spinels in literature.41
In the present work, we introduced an ALD process to fabricate CuCr2O4thin films based
on Cu(thd)2, Cr(acac)3, and ozone precursors. The as-deposited films were extremely smooth but
amorphous. When annealed at 700 ◦C in O2then pure crystalline CuCr2O4films were obtained with
high transmittance values greater than 60% in the visible range with a direct bandgap of 1.36 eV.
Magnetic measurements confirmed the ferrimagnetic phase transition with TCclose to 140 K. The
estimated average saturation moment of 0.36 µB/f.u.is in good agreement with the value reported
in literature for single crystal samples. Electrical transport measurements confirmed the p-type
semiconducting behavior of the films.
The present work has received funding from the European Research Council under the Euro-
pean Union’s Seventh Framework Programme (No. FP/2007-2013)/ERC Advanced Grant Agree-
ment (No. 339478) and also from the Aalto Energy Efficiency Research Programme.
1Atomic Layer Deposition for Semiconductors, edited by C. Seong Hwang (Springer Science & Business Media, 2013).
2S. M. George, Chem. Rev. 110, 111–131 (2010).
3R. W. Johnson, A. Hultqvist, and S. F. Bent, Mater. Today 17, 236 (2014).
4F. Gao, S. Arpiainen, and R. L. Puurunen, J. Vac. Sci. Technol., A 33, 010601 (2015).
5V. Kocsis, S. Bordács, D. Varjas, K. Penc, A. Abouelsayed, C. A. Kuntscher, K. Ohgushi, Y. Tokura, and I. Kézsmárki, Phys.
Rev. B 87, 064416 (2013).
6Handbook of Magnetic Materials, edited by E. P. Wohlfarth (North-Holland Publishing Company, 1982), Vol. 3.
7M. C. Kemei, S. L. Moffitt, L. E. Darago, R. Seshadri, M. R. Suchomel, D. P. Shoemaker, K. Page, and J. Siewenie, Phys.
Rev. B 89, 174410 (2014).
8G. Hu and Y. Suzuki, Phys. Rev. Lett. 89, 276601 (2002).
9M. G. Chapline and S. X. Wang, Phys. Rev. B 74, 014418 (2006).
10 B. Sun, J. Wu, X. Jia, F. Lou, and P. Chen, J. Sol-Gel Sci. Technol. 75, 664–669 (2015).
046106-7 Tripathi, Yadav, and Karppinen APL Mater. 4, 046106 (2016)
11 P. S. Sathiskumar, C. R. Thomas, and G. Madras, Ind. Eng. Chem. Res. 51, 10108–10116 (2012).
12 H. Adkins and R. Connor, J. Am. Chem. Soc. 53, 1091–1095 (1931).
13 R. Connor, K. Folkers, and H. Adkins, J. Am. Chem. Soc. 54, 1138–1145 (1932).
14 S. Roy and J. Ghose, Mater. Res. Bull. 34, 1179–1186 (1999).
15 A. Petric and H. Ling, J. Am. Ceram. Soc. 90, 1515 (2007).
16 A. M. Kawamoto, L. C. Pardini, and C. Rezende, Aerosp. Sci. Technol. 8, 591–598 (2004).
17 P. W. M. Jacobs and H. M. Whitehead, Chem. Rev. 69, 551–590 (1969).
18 R. W. Armstrong, B. Baschung, and D. W. Booth, Nano Lett. 3, 253–255 (2003).
19 R. Prasad and P. Singh, Bull. Chem. React. Eng. Catal. 6(2), 63 (2011).
20 K. S. De, J. Ghose, and K. S. R. C. Murthy, J. Solid State Chem. 43, 261 (1982).
21 R. Bajaj, M. Sharma, and D. Bahadur, Dalton Trans. 42, 6736–6744 (2013).
22 N. Padmanaban, B. N. Avasthi, and J. Ghose, J. Solid State Chem. 81, 250 (1989).
23 E. Prince, Acta Crystallogr. 10, 554 (1957).
24 B. J. Kennedy and Q. Zhou, J. Solid State Chem. 181, 2227 (2008).
25 J. M. Iwata, R. V. Chopdekar, F. J. Wong, B. B. Nelson-Cheeseman, E.Arenholz, and Y. Suzuki, J. Appl. Phys. 105, 07A905
(2009).
26 R. S. Yu and C. M. Wu, Appl. Surf. Sci. 282, 92–97 (2013).
27 H. Sun, M. A. P. Yazdi, P. Briois, J. F. Pierson, F. Sanchette, and A. Billard, Vacuum 114, 101–107 (2015).
28 A. Barnabé, Y. Thimont, M. Lalanne, L. Presmanes, and P. Tailhades, J. Mater. Chem. C 3, 6012–6024 (2015).
29 Y. Chang, C. Lin, and B. Lee, in ECS Proceedings of the Electrochemical Society 203rd Meeting - CVD XVI and
EUROCVD 14, edited by M. Allendorf, F. Maury, and F. Teyssandier (Electrochemical Society, Pennington, NJ, 2003),
pp. 1492–1499.
30 T. S. Tripathi, J.-P. Niemelä, and M. Karppinen, J. Mater. Chem. C 3, 8364–8371 (2015).
31 E. Ahvenniemi, M. Matvejeff, and M. Karppinen, Dalton Trans. 44, 8001 (2015).
32 K. Uusi-Esko, E.-L. Rautama, M. Laitinen, T. Sajavaara, and M. Karppinen, Chem. Mater. 22, 6297–6300 (2010).
33 K. Uusi-Esko and M. Karppinen, Chem. Mater. 23, 1835–1840 (2011).
34 M. Lie, O. Nilsen, H. Fjellvåg, and A. Kjekshus, Dalton Trans. 3, 481 (2009).
35 L. K. Tan, B. Liu, J. H. Teng, S. Guo, H. Y. Lowd, and K. P. Loh, Nanoscale 6, 10584 (2014).
36 W. D. Callister, Materials Science and Engineering—An Introduction (John Wiley and Sons, New York, 1997).
37 Z. Ma, Z. Xiao, J. A. van Bokhoven, and C. Liang, J. Mater. Chem. 20, 755–760 (2010).
38 Handbook of Inorganic Compounds, 2nd ed., edited by D. L. Perry (CRC Press, 2011).
39 K. Ohgushi, Y. Okimoto, T. Ogasawara, S. Miyasaka, and Y. Tokura, J. Phys. Soc. Jpn. 77, 034713 (2008).
40 M. R. Suchomel, D. P. Shoemaker, and L. Ribaud, Phys. Rev. B 86, 054406 (2012).
41 S. T. Kshirsagar and C. D. Sabane, Jpn. J. Appl. Phys., Part 1 10, 794 (1971).
Available via license: CC BY 4.0
Content may be subject to copyright.
