materials

Article
Luminescence Properties and Mechanisms of CuI
Thin Films Fabricated by Vapor Iodization of
Copper Films
Guochen Lin, Fengzhou Zhao, Yuan Zhao, Dengying Zhang, Lixin Yang, Xiaoe Xue,
Xiaohui Wang, Chong Qu, Qingshan Li and Lichun Zhang *
School of Physics and Optoelectronic Engineering, Ludong University, Yantai 264025, China;
guochllin@163.com (G.L.); fzzhao@ldu.edu.cn (F.Z.); zhaoyuan062312514@163.com (Y.Z.);
dengying_zhang@163.com (D.Z.); lixin_yang0131@163.com (L.Y.); xxe9898@163.com (X.X.);
isabellaxiaohui@163.com (X.W.); quchongbjtu@126.com (C.Q.); qsli@ldu.edu.cn (Q.L.)
* Correspondence: phyzlc@163.com; Tel.: +86-535-667-3397

Academic Editor: Giorgio Biasiol

Received: 15 September 2016; Accepted: 22 November 2016; Published: 7 December 2016

Abstract: Copper iodide (CuI) thin films were grown on Si(100) substrates using a copper film
iodination reaction method. It was found that γ-CuI films have a uniform and dense microstructure
with (111)-orientation. Transmission spectra indicated that CuI thin films have an average
transmittance of about 60% in the visible range and the optical band gap is 3.01 eV. By checking
the effect of the thickness of the Cu films and annealing condition on the photoluminescence (PL)
character of CuI films, the luminescence mechanisms of CuI have been comprehensively analyzed,
and the origin of different PL emissions are proposed with Cu vacancy and iodine vacancy as
defect levels.

Keywords: CuI thin films; photoluminescence; vapor iodization; iodine vacancy

1. Introduction
Copper iodide (CuI) is a binary metal halide semiconductor with a cubic zinc-blende structure.
Due to the wide direct bandgap of 3.1 eV and a large exciton binding energy (62 meV), CuI has
been considered as a promising candidate material for optoelectronic applications, especially for
short-wavelength optoelectronic detectors and light emitting devices [1]. Furthermore, CuI has
a number of merits, such as p-type conductivity arising from the presence of Cu vacancies, high
hole mobility owing to its light hole effective mass, and high transmittance in the visible spectrum.
By virtue of these advantageous properties, CuI has been successfully applied in various fields such
as transparent conductive films [2], solid-state dye solar cells [3,4], and light-emitting devices [5] for
several decades.
Recently, there is a renewed interest in CuI thin films due to their attractive prospects for
heterojunction optoelectronic device applications [6–8], which has led to a large number of studies
of CuI thin films, including growth methods, quality of epitaxial thin films, optical and electrical
properties, and device applications [9–15]. However, in spite of these previous efforts and results,
some of the basic properties, such as the defect behavior and the process of carrier recombination in
CuI films, still remain unclear. Moreover, the luminescence origin and mechanisms of CuI still needs
further investigation.

Materials 2016, 9, 990; doi:10.3390/ma9120990 www.mdpi.com/journal/materials

Materials 2016, 9, 990 2 of 8

Different synthesis techniques can be used to prepare CuI thin films [2,11–16], such as pulse
laser deposition, magnetron sputtering, iodination reaction, vacuum evaporation, and hydrothermal
evaporation. Among them, the iodination reaction technique using a copper thin film has attracted
enormous attention because of the potential for large area uniform films, low growth temperature,
and inexpensive cost [6]. In this paper, we report the fabrication of CuI thin films on Si(100)
substrates under different conditions by the copper film iodination reaction technique. Comprehensive
photoluminescence (PL) spectral studies, including the effect of the thickness of Cu films and annealing
environment, were implemented regarding these different emissions. The defect behavior and
luminescence mechanism of CuI thin films are also discussed.

2. Experiments
The growth of CuI thin films was carried out in a two-step process. Firstly, the copper thin
films were deposited on Si(100) substrates by electron beam evaporation using high purity Cu pellets
(99.999%, Kurt J. Lesker Company, Clairton, PA, USA) as the evaporation source. Before depositing the
Cu film, the Si(100) wafer was initially cleaned with acetone and ethanol to remove organic grease and
then etched with a dilute aqueous HF solution (5.0 mol·L−1 ) for 10 min to remove the native oxide.
The growth chamber was evacuated using a turbo molecular pump at 3 × 10−4 Pa, and the growth rate
of the copper thin films was kept at ~2 Å/s at room temperature. Secondly, the as-fabricated Cu films
were transferred to a Petri dish and fixed by a Teflon holder. Iodine particles (99.999%, Alfa Aesar,
Ward Hill, MA, USA) were put into the Petri dish and then heated to 150 ◦ C for 1 h by a hotplate.
The thickness of the Cu thin films was determined by a Dektak XT profilometer (Bruker, Karlsruhe,
Germany). A NanoNova 450 scanning electronic microscope (SEM, FEI, Hillsboro, OR, USA) was
used to examine the morphology of the samples. The X-ray diffraction (XRD) measurements were
performed on Rigaku D/MAX2500V diffractometer (Rigaku, Osaka, Japan) with Cu Kα radiation
(λ = 0.15405 nm). Photoluminescence (PL) spectra were excited by a 325 nm He-Cd laser (Kimmon
Electric, Ltd., Saitama, Japan) at room temperature. A monochromator (SR-500i-A, Andor Technology,
Belfast, UK) was employed for collecting the PL emission spectra. All the results were measured at
room temperature.

3. Results and Discussion

Figure 1a shows the surface SEM images of the Cu thin films grown on Si(100) substrates.
The thickness of the Cu thin films was determined by a Bruker Dektak XT profilometer to be about
80 nm. It can be observed that the Cu films grown by electron beam evaporation possess the crystalline
size of nanometer order. The surface SEM images in Figure 1b–d illustrate CuI thin films grown under
different thickness of Cu films (80 nm, 200 nm, 500 nm, respectively). The corresponding cross-sectional
SEM images are shown in the insets of Figure 1b–d, respectively. Compared with the surface SEM
images of the Cu films (Figure 1a) and CuI films (Figure 1b), evident change are easily observed. All
of the CuI films exhibit a very rough surface morphology, consisting of many small crystallites with
facets in different growth orientations [6]. As shown in the cross-sectional SEM images, microcolumnar
structures are vertically arranged on the Si(100) substrate. In comparison with the Cu thin film, the
thickness of all of the CuI films increased significantly (nearly five times), resulting from the theoretical
volume ratio of CuI and Cu (4.76) [6]. A typical atomic force microscopy (AFM) scan is depicted in
Figures S1 and S2, revealing a very rough surface morphology for the CuI thin films grown using
200 nm thickness of the Cu films (see Supplementary Materials).
Materials 2016, 9, 990 3 of 8
Materials 2016, 9, 990 3 of 8

Figure 1. (a) Surface SEM images of the Cu films. Surface and cross-sectional (inset) view SEM
Figure 1. (a) Surface SEM images of the Cu films. Surface and cross-sectional (inset) view SEM images
images of the CuI thin films grown with different thickness of Cu films (b) 80 nm; (c) 200 nm; (d) 500
of the CuI thin films grown with different thickness of Cu films (b) 80 nm; (c) 200 nm; (d) 500 nm.
nm.

As
As the
theCuI
CuIgrown
grownunder under 200200nmnmCu Cu films had had
films a uniform surface
a uniform texture,
surface it was it
texture, usedwasfor the growth
used for the
of
growth of CuI for XRD measurements. Figure 2a shows XRD patterns of Cu and CuI thin films,the
CuI for XRD measurements. Figure 2a shows XRD patterns of Cu and CuI thin films, in which in
thickness
which theof the Cu films
thickness of thewas Cu 200
filmsnm.
was The 200XRD
nm.result
The XRD of the as-deposited
result Cu film grown
of the as-deposited Cu filmon Si(100)
grown
substrate indicates that there were that
two diffraction peaks ◦ and 50.44◦ , corresponding to Cu(111)
on Si(100) substrate indicates there were twoat 43.24
diffraction peaks at 43.24° and 50.44°,
and Cu(200), respectively.
corresponding to Cu(111)The andXRD pattern
Cu(200), of the as-grown
respectively. The CuI
XRDfilms
patternexhibits peaks
of the corresponding
as-grown CuI films to
the (111), (220), (311), and (222) planes of CuI in zinc-blende structure (JCPDS
exhibits peaks corresponding to the (111), (220), (311), and (222) planes of CuI in zinc-blende card No. 060246) [2,10,17].
Figure
structure 2b (JCPDS
shows XRD card patterns
NO. 060246) of CuI thin films
[2,10,17]. with2b
Figure diferent
shows annealing
XRD patterns treatments.
of CuI thinThefilms
CuI films
with
were grown under the same condition (the thickness of the Cu films was
diferent annealing treatments. The CuI films were grown under the same condition (the thickness of 200 nm) and the annealing
treatment
the Cu films waswasperformed
200 nm) in aand
quartz
the tube under treatment
annealing different conditions
was performed(air, vacuum, and iodine
in a quartz tubevapors)
under
at 300 ◦ C for 2 h. For CuI thin films annealed in air, vacuum, and iodine vapors [10], the intensity and
different conditions (air, vacuum, and iodine vapors) at 300 °C for 2 h. For CuI thin films annealed
full width
in air, at half
vacuum, maximum
and iodine vapors(FWHM) [10],ofthethe CuI(111)
intensity andwere
fullchanged.
width at Under different(FWHM)
half maximum environments, of the
the FWHM of the CuI(111) decreases from 0.213 ◦ (as grown) to 0.207 ◦ (air), 0.163 ◦ (vacuum), and
CuI(111) were changed. Under different environments, the FWHM of the CuI(111) decreases from
0.157 ◦ (iodine vapors), respectively. It is indicated that the crystallization quality of CuI thin films
0.213° (as grown) to 0.207° (air), 0.163° (vacuum), and 0.157° (iodine vapors), respectively. It is
has been improved
indicated through the annealing
that the crystallization quality oftreatment
CuI thin [18].
filmsHowever, in additionthrough
has been improved to the CuI(111), three
the annealing
additional peaks
treatment [18]. are also found
However, in the to
in addition XRD thespectrum
CuI(111),ofthree
CuI films annealed
additional peaksin air,
are which correspond
also found in the
to CuO(110), (002), and (111) (JCPDS card No 45-0937) [19]. This can
XRD spectrum of CuI films annealed in air, which correspond to CuO(110), (002), and (111) be attributed to the fact (JCPDS
that an
oxidation reaction[19].
card No 45-0937) occurred
This can between the CuI films
be attributed to theand
factoxygen.
that an oxidation reaction occurred between
Quantitative analysis
the CuI films and oxygen. of the chemical compositions of the CuI thin films was further performed by
surface scanning EDXA (Energy Dispersive X-ray
Quantitative analysis of the chemical compositions of the Analysis). As shown
CuI thin in films
Figurewas 2b, further
the elemental
performedratio
(at
by %) of the scanning
surface Cu/I is 1.06:1,
EDXA which is nearDispersive
(Energy the standard stoichiometric
X-ray Analysis). composition
As shown in [17]. Furthermore,
Figure 2b, the
other elements have also been found in the CuI thin films, including Al,
elemental ratio (at %) of the Cu/I is 1.06:1, which is near the standard stoichiometric compositionSi, C, and O. It was deduced
that
[17]. Si comes fromother
Furthermore, the Si(100)
elements substrate
have also andbeen
the found
Al signal observed
in the CuI thin infilms,
the spectrum
includingarisesAl, Si,from
C, andAl
contamination
O. It was deduced in thethat
vacuum chamber.
Si comes fromThe theappearance of O andand
Si(100) substrate C might
the Albesignal
attributed to H2 O,
observed in O 2,
the
spectrum arises from Al contamination in the vacuum chamber. The appearance of O and C might
be attributed to H2O, O2, and CO2 adsorbed on the surface of the CuI crystals in air [17]. Though the
Materials 2016, 9, 990 4 of 8

Materials 2016, 9, 990 4 of 8

Materials 2016, 9, 990 4 of 8
and CO2 adsorbed on the surface of the CuI crystals in air [17]. Though the EDXA for this kind of
EDXA for this kind of quantitative analysis is shown, X-ray photoemission spectroscopy (XPS) will
quantitative analysis is shown, X-rayanalysis
photoemission
is shown,spectroscopy (XPS) will be further (XPS)
explored
will in
be EDXA
furtherfor this
exploredkind
inof quantitative
subsequent works. X-ray photoemission spectroscopy
subsequent
be furtherworks.
explored in subsequent works.

(a)
(a) (b)
(b)

(c)
(c)
Figure 2. (a) XRD patterns of Cu films and CuI films; (b) XRD patterns of CuI films with and without
Figure
Figure 2. (a) XRD
2. (a) XRD patterns
patterns of
of Cu
Cu films
films and
and CuI
CuI films;
films; (b)
(b) XRD
XRD patterns
patterns of
of CuI
CuI films
films with
with and
and without
without
annealing treatment; (c) EDXA result of the CuI films.
annealing treatment; (c) EDXA result of the CuI films.
annealing treatment; (c) EDXA result of the CuI films.
Figure 3 shows the transmission spectra of the CuI thin films grown under 80 nm thickness Cu
Figure
films 33 shows
(glass shows the
thetransmission
substrate). The optical spectra
transmission spectraofof
transmittancethetheCuI thin
ofCuI
CuI films
thin
thin filmsgrown
films grown under
under
is about 80 nm
60% thickness
80innm visibleCu
the thickness
films
Cu (glass
films
wavelength substrate).
(glassrangesubstrate).The nm).
(400–800 optical
TheThe transmittance
optical
opticaltransmittance
band gap of (E
CuI
g) of
thin
was CuI films
thin isfilms
calculated about
by is 60%
using the in
about the visible
60% in the
relationship
wavelength
α
visible A range, (400–800
wavelength where nm).hνThe
rangeα(400–800
and are optical
nm).
the The band gap
optical
absorption (E g) was
band
coefficient gap calculated
and (Ethe
g ) was by using
calculated
photon energy, therespectively
byrelationship
using the
α [10],A and αA2 is


relationship ,=where
A − and
hv α
a constant. Eg ,hν arevalues
where
The the absorption
α and ofhνthe coefficient
aretheoretical
the absorption and
band the photon
coefficient
gap wereand energy,
the photon
determined respectively
byenergy,
the
[10], and A[10],
respectively
extrapolation is ofaandconstant.
linearA isregionsTheof values
a constant.
theThe
plots ofto the
valueszerooftheoretical
the theoretical
absorption bandband
(αhv gap
= 0). were
gap
The weredetermined
optical determined
band gap of bythethe
by
extrapolation
as-grown CuI of thin
linear regions
films, of thefrom
obtained plots zeroof
toinset
the zero absorption
absorption
Figure 3, was (αhv3.01
(αhv = 0).eV,
The optical
which band gapwith
is consistent of the
the results
as-grown CuIinthin
CuI previous
thin films,works
films, obtained[2,10].
obtained fromthe
from theinset
insetofofFigure
Figure 3, 3,
waswas 3.013.01
eV,eV, which
which is consistent
is consistent withwith
the
the results
results in previous
in previous works works [2,10].
[2,10].

Figure 3. Transmission spectra of the CuI thin films, the inset shows the optical band gap of CuI thin
films grown on glass substrate.
Figure 3. Transmission
Figure 3. spectra of
Transmission spectra of the
the CuI
CuI thin
thin films,
films, the
the inset
inset shows
shows the
the optical
optical band
band gap
gap of
of CuI
CuI thin
thin
films
films grown
Figure
grown on
4aon glass
shows substrate.
the PL spectra of the as-grown CuI samples, which have been iodinated by
glass substrate.
different thicknesses of Cu films under the same conditions. All of them exhibit a strong blue
Figure 4a shows the PL spectra of the as-grown CuI samples, which have been iodinated by
different thicknesses of Cu films under the same conditions. All of them exhibit a strong blue
Materials 2016, 9, 990 5 of 8

Figure 4a shows the PL spectra of the as-grown CuI samples, which have been iodinated
Materials 2016, 9, 990
by
5 of 8
different thicknesses of Cu films under the same conditions. All of them exhibit a strong blue emission
band andband
emission a weak and broad
a weakred broad
emissionredband.
emission In order
band. to In
better
order understand the origin ofthe
to better understand theorigin
PL emission,
of the
PL emission, the Gaussian fits of the PL spectra are shown in Figure 4a. It can be seen that the band
the Gaussian fits of the PL spectra are shown in Figure 4a. It can be seen that the blue emission blue
can be divided
emission band can intobetwo luminescent
divided into twopeaks, which are
luminescent centered
peaks, whichatare ~410 nm (Peak
centered 1) and
at ~410 nm ~419
(Peaknm 1)
(Peak 2), respectively. This phenomenon can be observed directly
and ~419 nm (Peak 2), respectively. This phenomenon can be observed directly in the PL spectra of in the PL spectra of CuI grown using
500 nm
CuI grownCu films
usingand 500other
nm previous
Cu films reports
and other [9,10,18].
previousGenerally
reports speaking,
[9,10,18]. theGenerally
emission speaking,
of Peak 1 with the
3.02 eV isofreasonably
emission Peak 1 with ascribed
3.02 eV to is
radiative
reasonably recombination
ascribed to of free excitons
radiative [10,13]. However,
recombination there is
of free excitons
still much
[10,13]. debate on
However, the is
there origin of the debate
still much emission onofthePeak 2 with
origin 2.96emission
of the eV and the broad2red
of Peak with emission
2.96 eVband and
the broad red emission band around 715 nm (Peak 3). One theory claimed that the emissionattributed
around 715 nm (Peak 3). One theory claimed that the emission of Peak 2 with 2.96 eV is of Peak 2
to the2.96
with excessive iodine accumulated
eV is attributed on the CuI
to the excessive iodinesurface [20,21], while
accumulated on the others
CuI have postulated
surface that it
[20,21], while
originated from the bulk iodide vacancies [9]. Meanwhile, the intrinsic
others have postulated that it originated from the bulk iodide vacancies [9]. Meanwhile, the intrinsic Cu vacancies have also been
Cu vacancies have also been speculated as the potential sources for the near-band emission (Peakthe
speculated as the potential sources for the near-band emission (Peak 2) [10,22]. For the origin of 2)
red emission (Peak 3), some have argued that it is induced by the
[10,22]. For the origin of the red emission (Peak 3), some have argued that it is induced by the oxygen oxygen vacancies [15], while the
majority still
vacancies [15],consider
while the it to be related
majority stillto the iodine
consider it tovacancy
be related [10,23].
to the iodine vacancy [10,23].
In order
In order to to investigate
investigate the the origin
origin of of the
the Peak
Peak 22 and and Peak
Peak 33 emissions
emissions in in CuI
CuI films,
films, Figure
Figure 4b 4b
shows the intensity ratio of Peak 2 and Peak 3 relative to Peak 1.
shows the intensity ratio of Peak 2 and Peak 3 relative to Peak 1. The intensity ratio of Peak 2 to Peak The intensity ratio of Peak 2 to
Peak 1 decreased substantially with the increasing thickness of Cu
1 decreased substantially with the increasing thickness of Cu films, whereas the intensity ratio of films, whereas the intensity ratio
of Peak
Peak 3 to3 to
PeakPeak 1 increased
1 increased slightly.
slightly. It Itisisimportant
importanttotonote notethatthatthetheiodination
iodinationreaction
reactionoccurred
occurred in in
iodine-enriched environments, and the Cu vacancies appeared
iodine-enriched environments, and the Cu vacancies appeared frequently [10], acting as an acceptorfrequently [10], acting as an acceptor
level in
level in CuI.
CuI. AsAs aa result,
result, the
the electron
electron transfer
transfer from from the the conduction
conduction band band minimum
minimum (CBM) (CBM) to to the
the CuCu
vacancy acceptor
vacancy acceptor level
level and
and thethe emission
emission band band centered
centered at at ~419
~419 nm nm (Peak
(Peak 2) 2) was
was observed.
observed. As As can
can bebe
seen from Figure 4b, since the concentration of Cu vacancy decreased
seen from Figure 4b, since the concentration of Cu vacancy decreased with the increasing thickness with the increasing thickness
of Cu
of Cu films,
films, the
the intensity
intensity of of Peak
Peak 22 reduced.
reduced. However, However, the the change
change of of the
the intensity
intensity ratio
ratio ofof Peak
Peak 33 to to
Peak 1 is too slight to confirm whether the origin of the red emission
Peak 1 is too slight to confirm whether the origin of the red emission is related to the iodine vacancy is related to the iodine vacancy or
thethe
or oxygen
oxygen vacancy.
vacancy.

(a) (b)
Figure
Figure 4.
4. (a)
(a) Photoluminescence (PL) spectra
Photoluminescence (PL) spectra of
ofthe
theas-grown
as-grownCuI
CuIsamples
samplesunder
underdifferent
different thickness
thickness of
of Cu films; (b) The intensity ratios of the different PL peaks.
Cu films; (b) The intensity ratios of the different PL peaks.

To get a comprehensive understanding of the origin of the red emission, annealing treatment
To get a comprehensive understanding of the origin of the red emission, annealing treatment for
for CuI films grown under the same conditions (the thickness of Cu films was 200 nm) was
CuI films grown under the same conditions (the thickness of Cu films was 200 nm) was performed
performed under air, vacuum, and iodine vapor conditions, respectively. Figure 5 displays the PL
under air, vacuum, and iodine vapor conditions, respectively. Figure 5 displays the PL spectra of
spectra of the CuI thin films annealed in different environments, while the peak-deconvolution
the CuI thin films annealed in different environments, while the peak-deconvolution with Gaussian
with Gaussian function was applied to each spectrum. In comparison, the PL parameters of the CuI
function was applied to each spectrum. In comparison, the PL parameters of the CuI thin films with
thin films with and without annealing treatment, such as the FWHM of Peak 1 and the intensity
and without annealing treatment, such as the FWHM of Peak 1 and the intensity ratio of different PL
ratio of different PL peaks, are shown in Table 1.
peaks, are shown in Table 1.
Materials 2016, 9, 990 6 of 8
Materials 2016, 9, 990 6 of 8

Figure
Figure 5.
5. PL
PL spectra
spectra of
of CuI
CuI thin
thin films
films annealed
annealed in
in different
different environments.
environments.

Table
Table 1.
1. PL
PL parameters
parameters of
of CuI
CuI thin
thin films
films with
with and
and without
without annealing
annealing treatment.
treatment.

Sample (Annealing Treatment) FWHM of Peak 1 Ipeak2/Ipeak1 Ipeak3/Ipeak1

Sample (Annealing Treatment) FWHM of Peak 1 Ipeak2 /Ipeak1 Ipeak3 /Ipeak1
as-grown 5.586 1.262 0.106
as-grown
Air 5.586 6.361 1.262
0.219 0.106
0.859
Air 6.361 0.219 0.859
vacuum
vacuum 5.366 5.366 0.438
0.438 0.598
0.598
iodine
iodine vaporvapor 5.243 5.243 3.582
3.582 0.008
0.008

As can be
As can beseen
seenin in Figure
Figure 5 and
5 and TableTable
1, the1,FWHM
the FWHM
of Peakof1 decreased
Peak 1 decreased after annealing
after annealing treatment
treatment
in vacuuminand vacuumiodineand iodine
vapor, vapor,
which which indicated
indicated that the annealing
that the annealing environmentenvironment
of CuI thinof CuI thin
films in
films in vacuum and iodine vapor can effectively reduce film defects [10,18,24].
vacuum and iodine vapor can effectively reduce film defects [10,18,24]. However, for the CuI thin films However, for the
CuI thin films
annealed in air,annealed
the FWHM in air, the FWHM
of Peak 1 increasedof Peak 1 increased
in comparison to in
thecomparison
as-grown CuI to the
thinas-grown CuI
films without
thin films without annealing treatment, which is probably due to the oxidation
annealing treatment, which is probably due to the oxidation reaction occurring in the annealing reaction occurring in
the annealing treatment
treatment under air. under air.
According
According to to the
the changes
changesofofthe theintensity
intensityratio
ratio
ofof Peak
Peak 2 to
2 to Peak
Peak 1, the
1, the value
value decreased
decreased when when
the
the CuI thin films were annealed in air and vacuum, but was dramatically
CuI thin films were annealed in air and vacuum, but was dramatically enhanced by iodine annealing. enhanced by iodine
annealing. As the mobility of iodine ions in the crystal lattice of CuI becomes more active at high
As the mobility of iodine ions in the crystal lattice of CuI becomes more active at high annealing
annealing temperatures, iodine can escape from the material, leading to a reduction of Cu vacancy
temperatures, iodine can escape from the material, leading to a reduction of Cu vacancy [10]. As a
[10]. As a result, the intensity of Peak 2 related to the Cu vacancy acceptor level decreased when
result, the intensity of Peak 2 related to the Cu vacancy acceptor level decreased when annealing in air
annealing in air and vacuum. Conversely, iodine annealing can increase the amount of Cu
and vacuum. Conversely, iodine annealing can increase the amount of Cu vacancies, and the intensity
vacancies, and the intensity of the Peak 2 improved.
of the Peak 2 improved.
Finally, the origin of the red emission can be deduced by examining the changes of Ipeak3/Ipeak1.
Finally, the origin of the red emission can be deduced by examining the changes of Ipeak3 /Ipeak1 .
Compared with the as-grown CuI thin films, the intensity of Peak 3 obviously enhanced when
Compared with the as-grown CuI thin films, the intensity of Peak 3 obviously enhanced when annealed
annealed in air, which implies that the red emission should not be related to the oxygen vacancy.
in air, which implies that the red emission should not be related to the oxygen vacancy. However,
However, since iodine can escape from CuI, and Cu should be oxidized when annealed in air, it can
since iodine can escape from CuI, and Cu should be oxidized when annealed in air, it can be inferred
be inferred that the red emission could be induced by the iodine vacancy. Furthermore, the value of
that the red emission could be induced by the iodine vacancy. Furthermore, the value of Ipeak3 /Ipeak1
Ipeak3/Ipeak1 also improved by vacuum annealing without oxygen participation, which is further proof
also improved by vacuum annealing without oxygen participation, which is further proof of the
of the relationship between the iodine vacancy and the red emission [10,23]. Additionally, the red
relationship between the iodine vacancy and the red emission [10,23]. Additionally, the red emission
emission almost disappeared after iodine annealing, indicating that the lack of iodine in the
almost disappeared after iodine annealing, indicating that the lack of iodine in the as-grown CuI thin
as-grown CuI thin films can be supplemented through annealing the films in iodine vapors [10]. In
films can be supplemented through annealing the films in iodine vapors [10]. In conclusion, it can be
conclusion, it can be confirmed that the red emission in CuI films can be attributed to the iodine
confirmed that the red emission in CuI films can be attributed to the iodine vacancy.
vacancy.
According to the abovementioned results about the PL properties of CuI thin films [25], we
According to the abovementioned results about the PL properties of CuI thin films [25], we
propose their origin as shown in Figure 6. The emission of Peak 1 near the band edge with 3.02 eV
propose their origin as shown in Figure 6. The emission of Peak 1 near the band edge with 3.02 eV is
is reasonably ascribed to free exciton recombination [10,13]. The emission of Peak 2 with 2.96 eV is
reasonably ascribed to free exciton recombination [10,13]. The emission of Peak 2 with 2.96 eV is
ascribed to the electron transition from the CBM to Cu vacancy acceptor level, and the intensity of
ascribed to the electron transition from the CBM to Cu vacancy acceptor level, and the intensity of
Peak 2 can be improved by iodine annealing [10]. According to the experimental results in this paper
Peak 2 can be improved by iodine annealing [10]. According to the experimental results in this paper
and other previous reports [10,23], the red emission band can be attributed to the transition from
Materials 2016, 9, 990 7 of 8

Materials 2016, 9, 990 7 of 8

and other previous reports [10,23], the red emission band can be attributed to the transition from
iodine
iodine vacancy
vacancy deep
deepdonor
donorlevel
leveltotothe
thevalue
valueband maximum
band maximum (VBM). These
(VBM). defects
These andand
defects PL bands can
PL bands
be
canformed during
be formed the air
during theor
airvacuum
or vacuum annealing process.
annealing process.

Figure 6. Proposed mechanisms of PL in CuI thin films.

Figure 6. Proposed mechanisms of PL in CuI thin films.

4. Conclusions
4. Conclusions
In summary, CuI thin films were deposited onto Si(100) substrates via iodination reaction with
In summary, CuI thin films were deposited onto Si(100) substrates via iodination reaction with
Cu films. The morphology, structure, composition, and luminescence characteristics of the CuI thin
Cu films. The morphology, structure, composition, and luminescence characteristics of the CuI thin
films have been investigated by SEM, XRD, EDXA, and PL. It is shown that CuI thin films have a
films have been investigated by SEM, XRD, EDXA, and PL. It is shown that CuI thin films have a
(111) preferred orientation, and a 60% average transmittance in the visible range can be obtained. A
(111) preferred orientation, and a 60% average transmittance in the visible range can be obtained.
comprehensive interpretation for the origin of the different PL emission bands has been obtained by
A comprehensive interpretation for the origin of the different PL emission bands has been obtained by
analyzing the effect of thickness of the Cu films and the annealing environment. It is believed that
analyzing the effect of thickness of the Cu films and the annealing environment. It is believed that the
the 419 nm emission is ascribed to the electron transition from the CBM to the Cu vacancy acceptor
419 nm emission is ascribed to the electron transition from the CBM to the Cu vacancy acceptor level,
level, whereas the red emission band is related to iodine vacancies.
whereas the red emission band is related to iodine vacancies.
Supplementary Materials:
Supplementary Materials:The
Thefollowing areare
following available online
available at www.mdpi.com/1996-1944/9/12/990/s1.
online Figure S1:
at www.mdpi.com/1996-1944/9/12/990/s1.
AFM image of the CuI thin film, Figure S2: Roughness analysis for the AFM image of the CuI thin film.
Figure S1: AFM image of the CuI thin film, Figure S2: Roughness analysis for the AFM image of the CuI thin film.
Acknowledgments: This Thiswork
work is supported
is supported by theby the financial
financial support
support of of the
the National National
Natural Natural
Science Science
Foundation of
China (No. 11504155),
Foundation the Natural
of China (No. Science
11504155), theFoundation of Shandong
Natural Science Province
Foundation (No. ZR2015PA011,
of Shandong No. ZR2014AM027),
Province (No. ZR2015PA011,
the
No.Shandong Provincethe
ZR2014AM027), Higher Educational
Shandong Science
Province and Technology
Higher EducationalProgram
Science (No.
and J14LJ04), and the
Technology Innovation
Program (No.
Project of Ludong University (No. LY2014006).
J14LJ04), and the Innovation Project of Ludong University (No. LY2014006).
Author Contributions: Guochen Lin and Lixin Yang conceived and designed the experiments; Xiaoe Xue and
Author Contributions:
Xiaohui Wang performed Guochen Lin and Lixin
the experiments; Yang Zhang
Dengying conceived
and and
Chongdesigned the experiments;
Qu analyzed Xiaoe XueZhao,
the data; Fengzhou and
Xiaohui
Yuan Wang
Zhao, andperformed
Qingshan the experiments;
Li contributed Dengying Zhang and Chong
reagents/materials/analysis Qu analyzed
tools; the data;
Lichun Zhang wroteFengzhou Zhao,
the paper.
Yuan Zhao, and Qingshan Li contributed reagents/materials/analysis
Conflicts of Interest: The authors declare no conflict of interest. tools; Lichun Zhang wrote the paper.

Conflicts of Interest: The authors declare no conflict of interest.

References
References
1. Grundmann, M.; Schein, F.L.; Lorenz, M.; Böntgen, T.; Lenzner, J.; Wenckstern, H.V. Cuprous iodide: A p-type
1. transparent semiconductor,
Grundmann, M.; Schein, F.L.; history,
Lorenz, and
M.;novel applications.
Böntgen, Phys.
T.; Lenzner, Status Solidi H.V.
J.; Wenckstern, A 2013, 210, 1671–1703.
Cuprous iodide: A
[CrossRef]
p-type transparent semiconductor, history, and novel applications. Phys. Status Solidi A 2013, 210, 1671–
2. Zhu, B.L.; Zhao, X.Z. Transparent conductive CuI thin films prepared by pulsed laser deposition. Phys. Status
1703.
2. Solidi A 2011,
Zhu, B.L.; 208,X.Z.
Zhao, 91–96. [CrossRef]conductive CuI thin films prepared by pulsed laser deposition. Phys.
Transparent
3. Sankapal,
Status B.R.;
Solidi Ennaoui,
A 2011, A.; Guminskaya, T.; Dittrich, T.; Bohne, W.; Rohrich, J.; Strub, E.; Lux-Steiner, M.C.
208, 91–96.
3. Characterization
Sankapal, of p-CuI
B.R.; Ennaoui, A.;prepared
Guminskaya,by the SILAR technique
T.; Dittrich, T.; Bohne, on
W.;Cu-tape/n-CuInS forLux-Steiner,
Rohrich, J.; Strub,2 E.; solar cells.
Thin Solid Films 2005, 480, 142–146. [CrossRef]
M.C. Characterization of p-CuI prepared by the SILAR technique on Cu-tape/n-CuInS2 for solar cells. Thin
4. Cheng,
Solid C.H.;
Films Wang,
2005, 480,J.;142–146.
Du, G.T.; Shi, S.H.; Du, Z.J.; Fan, Z.Q.; Bian, J.M.; Wang, M.S. Organic solar cells with
4. remarkable
Cheng, C.H.;enhanced
Wang, J.;efficiency
Du, G.T.;by using
Shi, a CuI
S.H.; Du, buffer to control
Z.J.; Fan, the molecular
Z.Q.; Bian, orientation
J.M.; Wang, and modify
M.S. Organic the
solar cells
Appl. Phys.enhanced
anode.remarkable
with Lett. 2010,efficiency
97, 083305.by[CrossRef]
using a CuI buffer to control the molecular orientation and
modify the anode. Appl. Phys. Lett. 2010, 97, 083305.
5. Lee, J.H.; Leem, D.S.; Kim, H.J.; Kim, J.J. Effectiveness of p-dopants in an organic hole transporting
material. Appl. Phys. Lett. 2009, 94, 123306.
Materials 2016, 9, 990 8 of 8

5. Lee, J.H.; Leem, D.S.; Kim, H.J.; Kim, J.J. Effectiveness of p-dopants in an organic hole transporting material.
Appl. Phys. Lett. 2009, 94, 123306. [CrossRef]
6. Schein, F.L.; Wenckstern, H.V.; Grundmann, M. Transparent p-CuI/n-ZnO heterojunction diodes.
Appl. Phys. Lett. 2013, 102, 092109. [CrossRef]
7. Ding, K.; Hu, Q.C.; Chen, D.G.; Zheng, Q.H.; Xue, X.G.; Huang, F. Fabrication and Energy Band Alignment
of n-ZnO/p-CuI Heterojunction. IEEE Electr. Device Lett. 2012, 33, 1750–1752. [CrossRef]
8. Yang, C.; Kneiß, M.; Schein, F.L.; Lorenz, M.; Grundmann, M. Room-temperature domain epitaxy of Copper
Iodide thin films for transparent CuI/ZnO heterojunctions with high rectification ratios larger than 109 .
Sci. Rep. 2016, 6, 21937. [CrossRef] [PubMed]
9. Zheng, Z.; Liu, A.; Wang, S.; Huang, B.; Wong, K.W.; Zhang, X.; Hark, S.K.; Lau, W.M. Growth of highly
oriented (110) γ-CuI film with sharp exciton band. J. Mater. Chem. 2008, 18, 852–854. [CrossRef]
10. Xia, M.; Gu, M.; Liu, X.; Liu, B.; Huang, S.; Ni, C. Luminescence characteristics of CuI film by iodine
annealing. J. Mater. Sci. Mater. Electron. 2015, 26, 5092–5096. [CrossRef]
11. Kosta, I.; Azaceta, E.; Yate, L.; Cabañero, G.; Grande, H.; Tena-Zaera, R. Cathodic electrochemical deposition
of CuI from room temperature ionic liquid-based electrolytes. Electrochem. Commun. 2015, 59, 20–23.
[CrossRef]
12. Nejand, B.A.; Ahmadi, V.; Shahverdi, H.R. Growth of plate like γ-CuI nanostructure on copper substrate by
hydrothermal evaporation of solution. Mater. Lett. 2014, 132, 138–140. [CrossRef]
13. Ma, Y.; Gu, M.; Huang, S.; Liu, X.; Liu, B.; Ni, C. Colloidal synthesis of uniform CuI nanoparticles and their
size dependent optical properties. Mater. Lett. 2013, 100, 166–169. [CrossRef]
14. Johan, M.R.; Wen, K.S.; Hawari, N.; Aznan, N.A.K. Synthesis and Characterization of Copper (I) Iodide
Nanoparticles via Chemical Route. Int. J. Electrochem. Sci. 2012, 7, 4942–4950.
15. Sirimanne, P.M.; Rusop, M.; Shirata, T.; Soga, T.; Jimbo, T. Characterization of CuI thin films prepared by
different techniques. Mater. Chem. Phys. 2003, 80, 461–465. [CrossRef]
16. Sirimanne, P.M.; Rusop, M.; Shirata, T.; Soga, T.; Jimbo, T. Characterization of transparent conducting CuI
thin films prepared by pulse laser deposition technique. Chem. Phys. Lett. 2002, 366, 485–489. [CrossRef]
17. Gu, M.; Gao, P.; Liu, X.L.; Huang, S.M.; Liu, B.; Ni, C.; Xu, R.K.; Ning, J. Crystal growth and characterization
of CuI single crystals by solvent evaporation technique. Mater. Res. Bull. 2010, 45, 636–639. [CrossRef]
18. Gruzintsev, A.N.; Zagorodnev, W.N. Effect of Annealing on the Luminescence of p-CuI Crystals.
Semiconductors 2012, 46, 149–154. [CrossRef]
19. Volanti, D.P.; Keyson, D.; Cavalcante, L.S.; Simoes, A.Z.; Joya, M.R.; Longo, E.; Varela, J.A.; Pizani, P.S.;
Souza, A.G. Synthesis and characterization of CuO flower-nanostructure processing by a domestic
hydrothermal microwave. J. Alloys Compd. 2008, 459, 537–542. [CrossRef]
20. Perera, V.P.S.; Tennakone, K. Recombination processes in dye-sensitized solid-state solar cells with CuI as
the hole collector. Sol. Energy Mater. Sol. Cells 2003, 79, 249–255. [CrossRef]
21. Yang, Y.; Ga, Q. Growth and photoluminescence characterization of highly oriented CuI/β-Cyclodextrin
hybrid composite film. Langmuir 2005, 21, 6866–6871. [CrossRef] [PubMed]
22. Chen, D.; Wang, Y.; Zhang, L.; Huang, J.; Chen, X.; Pan, D.; Huang, F. Growth strategy and physical properties
of the high mobility p-type CuI crystal. Cryst. Growth Des. 2010, 10, 2057–2060. [CrossRef]
23. Gao, P.; Gu, M.; Liu, X.L.; Liu, B.; Huang, S.M. X-ray excited luminescence of cuprous iodide single crystals:
On the nature of red luminescence. Appl. Phys. Lett. 2009, 95, 221904. [CrossRef]
24. Gao, P.; Gu, M.; Liu, X.; Zheng, Y.Q.; Shi, E.W. Photoluminescence study of annealing effects on CuI crystals
grown by evaporation method. Cryst. Res. Technol. 2012, 47, 707–712. [CrossRef]
25. Meng, L.; Mo, R.; Zhou, H.; Wang, G.; Chen, W.; Wang, D.; Peng, Q. Synthesis of luminescent cubic phase
one-dimensional CuI nanostructures in solution. Cryst. Growth Des. 2010, 10, 3387–3390. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).