We report the development of filterless deep ultraviolet photoconductive detectors using cerium fluoride (CeF3) thin films fabricated by pulsed laser deposition (PLD). By varying the PLD laser power during thin film growth, we observed that CeF3 breaks down to CeF2 at PLD laser powers greater than 100 mW. This consequently leads to the formation of fluorine defects that effectively narrowed the optical bandgap of the thin films, resulting in the decreased resistivity of the photoconductive detector. Under ultraviolet irradiation, the detector using a thin film grown at 5 mW PLD laser power exhibited close to four orders of magnitude increase in photocurrent compared to the dark current. The spectral response of the photoconductive detectors can be tuned from 300 nm to 400 nm when PLD laser powers ranging from 5 mW to 400 mW are used to fabricate the thin films. The filterless nature of the detectors simplifies their production, and their tunability can extend their use to a wider range of applications.

Ultraviolet (UV) light is being used in many applications over a wide range of fields, including optical surface cleaning,1,2 forensic investigations,3 fuel identification,4 and atmospheric observation.5 These applications drive the development of UV light sources and its associated detector technologies. The development of a vacuum UV field emission lamp utilizing KMgF3 thin film phosphor emitting at a wavelength as short as 155 nm6 is an example of a breakthrough in the development of short wavelength light sources. Development of appropriate detectors should follow suit. Existing detectors and recent research focused on using wide bandgap semiconductors that are based on oxides, nitrides, and diamond to develop photoconductive detectors in the deep ultraviolet (DUV) region.7–11 However, when used to monitor DUV radiation, these detectors require filters to block out unwanted UV radiation that will otherwise lead to poor DUV/UV contrast. The use of filters also results in a significant loss in the effective sensing area, aside from adding to the complexity of the detector. In order to eliminate these filters, the photoconductive detectors must use materials with extremely wide bandgaps since the response wavelength of photoconductive detectors is mainly influenced by the bandgap of the material used. Fluoride compounds have extremely wide bandgaps that make the detector intrinsically transparent to background UV radiation, without having to use filters. Previously, we have demonstrated a filterless deep UV (DUV) detector based on CeF3 thin films grown by pulsed laser deposition (PLD).14 The bandgap of CeF3 is 4.9 eV,15 making the detector sensitive only to wavelengths below 253 nm even without using filters. If we are able to tune the wavelength response of the CeF3 thin film detector, flexibility in the design and optimization of the photoconductive detector can be achieved and this will open up new applications for this detector. Several groups have demonstrated the ability to engineer the bandgap of semiconductor thin films by controlling the chemical compositions of the thin films during fabrication.16–19 Moreover, the density of defects in the semiconductor thin films has a strong influence on the films’ absorption coefficient and photoconductivity.20 In this work, we investigate the influence of experimental conditions, specifically the laser power, on the fluorine defects of CeF3 thin films fabricated by pulsed laser deposition (PLD). The small difference in chemical composition between the source target and the deposited film and the easy evaporation of materials with a high melting point are features that make PLD suitable for fabricating fluoride thin films.12,13 However, much is not known about the parameters that affect the properties of fluoride thin films fabricated through PLD. By carrying out such investigations, we have successfully fabricated a filterless DUV photoconductive detector using CeF3 thin films. We further demonstrate that the response wavelength of the detector depends on the PLD laser power used during deposition.

The thin films were fabricated by irradiating a sintered CeF3 ceramic target with the focused laser pulses of a Nd:YAG laser operating at 532 nm wavelength (second harmonics), 20 ns pulse width, and 10 Hz repetition rate. Five thin films were deposited onto a fused SiO2 substrate at 670 K under a high vacuum condition of ∼6 × 10−6 Pa. The laser powers used for each of the five thin films were 5 mW, 10 mW, 100 mW, 200 mW, and 400 mW. The deposition times were 6 h, 4 h, 2 h, 1.5 h, and 0.75 h, respectively. UV photoconductive detectors were then fabricated by depositing a pair of interdigitated aluminum electrodes on the surface of the CeF3 thin films using vacuum vapor deposition. The gap of the electrodes was 4.8 mm, while the distance between the electrodes was 0.4 mm. The length of the electrodes was 10 mm. The patterned area of the interdigitated electrodes was 10 × 5 mm2. The electrical resistance was measured by applying voltage between the electrodes. The resistivity was then calculated on the assumption that current flows only between the electrodes. The schematic diagram of the photoconductive detector is shown in Fig. 1. The current–voltage (I–V) characteristic curves of the detectors were evaluated by using a deuterium lamp (L10366 series, Hamamatsu Photonics K. K.). The current under irradiation of UV light (photocurrent) and under unirradiation (dark current) was measured for various applied voltages up to 100 V.

FIG. 1.

Schematic diagram of the photoconductive detector fabricated from the CeF3 thin film. The detector’s resistivity was determined by assuming that current flows only between the aluminum (Al) electrodes.

FIG. 1.

Schematic diagram of the photoconductive detector fabricated from the CeF3 thin film. The detector’s resistivity was determined by assuming that current flows only between the aluminum (Al) electrodes.

Close modal

Figure 2(a) shows the x-ray diffraction (XRD) patterns of the thin films fabricated with different PLD laser powers, and Fig. 2(b) shows the enlarged view of results from 25.5° to 28.5°. Diffraction peaks appearing at 2θ = 24.40, 25.10, and 27.90 correspond to the (0 0 2), (1 1 0), and (1 1 1) planes of CeF3, respectively. Another peak appears at 2θ = 26.50, and this peak becomes more prominent as the PLD laser power is increased, which is hardly visible for 5 mW and 10 mW irradiated samples [Fig. 2(b)]. The peak at 26.50 is characteristically from the (1 1 1) plane of CeF2,21 which indicates that CeF3 breaks down into CeF2 as the power of the PLD irradiating laser increases. The presence of CeF2 could introduce fluorine defects that would alter the spectral response of the photoconductive detectors. In order to confirm this, we obtained the absorption spectra of the thin films. The absorption spectra in Fig. 3 show that all thin films exhibited a dominant absorption peak at 251 nm (4.94 eV). This corresponds to the band gap energy of CeF3.15 An additional absorption flat-roof peak from 288 nm to 306 nm maybe attributed to the energy transfer from the regular Ce3+ to the perturbed Ce3+, which is associated with a vacancy defect due to the absence of fluorine.22,23 Another much broader peak from 450 nm to 650 nm was observed from thin films fabricated with PLD laser powers of at least 100 mW. This broader band peak may be attributed to the formation of deep traps which can be caused by ionization radiation excitation.23 The appearance of these additional peaks confirms that defect levels were formed within the bandgap of CeF3. The defect levels consist of a narrow band at 4.08 eV (304 nm) and a broad band between 1.91 eV (650 nm) and 2.76 eV (450 nm), as shown in Fig. 4. SEM images of the thin films were obtained to analyze the average droplet size for different PLD laser powers. A sample SEM image for the thin film grown with 200 mW PLD laser power is shown in Fig. 5. As Table I shows, the average droplet size does not significantly change as a function of PLD laser power.

FIG. 2.

(a) X-ray diffraction patterns of CeF3 thin films fabricated with different PLD laser powers. For low power (5 mW and 10 mW) irradiated samples, diffraction peaks observed at 21° attribute to the SiO2 substrate due to the longer permeation depth of the x-ray than the thickness of thin films. (b) The enlarged view from 25.5° to 28.5°.

FIG. 2.

(a) X-ray diffraction patterns of CeF3 thin films fabricated with different PLD laser powers. For low power (5 mW and 10 mW) irradiated samples, diffraction peaks observed at 21° attribute to the SiO2 substrate due to the longer permeation depth of the x-ray than the thickness of thin films. (b) The enlarged view from 25.5° to 28.5°.

Close modal
FIG. 3.

Absorption spectra of CeF3 thin films fabricated with different PLD laser powers. The dominant absorption peak at 251 nm (4.94 eV) corresponds to the band gap energy of CeF3.

FIG. 3.

Absorption spectra of CeF3 thin films fabricated with different PLD laser powers. The dominant absorption peak at 251 nm (4.94 eV) corresponds to the band gap energy of CeF3.

Close modal
FIG. 4.

Energy level diagram showing the fluoride defect states that are formed at high PLD laser powers.

FIG. 4.

Energy level diagram showing the fluoride defect states that are formed at high PLD laser powers.

Close modal
FIG. 5.

SEM images of the thin film grown with 200 mW PLD laser power.

FIG. 5.

SEM images of the thin film grown with 200 mW PLD laser power.

Close modal
TABLE I.

Average droplet size for different PLD laser powers.

Laser power (mW) 10 100 200 400 
Deposition rate (nm/h) 4.43 6.83 64.4 49.1 59.6 
Average droplet size (μm) 0.22 0.22 0.33 0.36 0.34 
Laser power (mW) 10 100 200 400 
Deposition rate (nm/h) 4.43 6.83 64.4 49.1 59.6 
Average droplet size (μm) 0.22 0.22 0.33 0.36 0.34 

Figure 6 shows the resistivity of each fabricated thin film. The thin film fabricated with the lowest PLD laser power (5 mW) exhibited the highest resistivity. Interestingly, as the irradiating laser power increases, the resistivity decreases. This means that the resistivity of the photoconductive detector fabricated using fluoride thin films can be manipulated by controlling the laser power used during PLD. The decrease in resistivity could be due to changes in the composition of the thin films as CeF3 breaks down into CeF2. We estimated that CeF2 are formatted due to the absence of fluorine (vacancy defect), which is confirmed by the XRD results. Consequently, the regular Ce3+ will change to a perturbed state, which is in a higher energy state away from the conduction band, i.e., the bandgap of the thin films was decreased by the presence of CeF2 leading to the observed decrease in resistivity for thin films grown with high PLD laser powers.

FIG. 6.

Resistivity of CeF3 thin films fabricated with different PLD laser powers. Resistivity decreases as the PLD laser power increases.

FIG. 6.

Resistivity of CeF3 thin films fabricated with different PLD laser powers. Resistivity decreases as the PLD laser power increases.

Close modal

The I–V characteristic curves of the detectors are shown in Figs. 7(a) and 7(b). The red dots indicate the photocurrent value (current under UV irradiation) corresponding to photo excited carriers, while the black dots indicate the dark current value (current without UV irradiation). All thin films exhibited an increase in current after irradiation. At a bias voltage of 100 V, the dark current values of the detector fabricated using 5 mW and 400 mW PLD laser powers are 0.2 pA and 9.6 pA, respectively, while the photocurrent values are 0.41 nA and 0.35 nA, respectively. Figure 7(c) shows the relationship between PLD laser power and sensitivity at a voltage of 100 V, which is defined as the ratio of photocurrent to dark current (photocurrent/dark current). The thin film fabricated with the lowest PLD laser power clearly exhibited the highest signal-to-noise ratio. As discussed above, defect levels found in the thin films were grown at laser powers of 100 mW or higher (Figs. 2 and 3). Figure 7(c) confirms that fewer fluorine defects result in higher sensitivity. Therefore, the sensitivity of the thin film photoconductive detector can be optimized by lowering the laser power during the PLD process although sufficient contrast between the photo and dark currents (about four orders of magnitude) is still possible even at higher PLD laser powers [Fig. 7(b)]. These results indicated that when there are more vacancy defects, the excited free photo-carriers will be banded easier and there will be weaker photocurrent. Consequently, the detector sensitivity will be lower. We evaluated the spectral sensitivity of the detectors under UV irradiation from a xenon lamp (SPEX Fluorolog 3-21, HORIBA, Ltd.), and the results are shown in Fig. 8. The spectral response of the detector shifts to longer wavelengths as the PLD laser power increases. The spectral response of the detector can, therefore, be tuned between 300 nm and 400 nm by controlling the PLD laser power between 5 mW and 400 mW, respectively. The transmission edge also follows the same trend. Specifically, the transmission edge of the thin film grown at 5 mW PLD laser power is 260 nm, while the transmission edge of the thin film grown at 400 m W is 315 nm. The observed red shift in the spectral response and transmission edge is consistent with the formation of CeF2, whereby the additional fluorine defect levels decrease the effective optical bandgap of the thin films as electrons can transition to the higher-level defect states. With filterless DUV photoconductive detectors, our results suggest that a low PLD laser power is beneficial for shifting the spectral response away from unwanted UV background radiation.

FIG. 7.

Current–voltage characteristic curves under UV irradiation (red dots) and without irradiation (black dots) of the detectors fabricated from thin films deposited at (a) 5 mW and (b) 400 mW laser powers. (c) Relationship between PLD laser power and sensitivity (photocurrent/dark current).

FIG. 7.

Current–voltage characteristic curves under UV irradiation (red dots) and without irradiation (black dots) of the detectors fabricated from thin films deposited at (a) 5 mW and (b) 400 mW laser powers. (c) Relationship between PLD laser power and sensitivity (photocurrent/dark current).

Close modal
FIG. 8.

Spectral responses of the detectors under UV irradiation. The detectors were fabricated from CeF3 thin films deposited using PLD laser powers of 5 mW, 10 mW, 100 mW, 200 mW, and 400 mW. The spectral response and transmittance of the detector shift to longer wavelengths as the PLD laser power increases.

FIG. 8.

Spectral responses of the detectors under UV irradiation. The detectors were fabricated from CeF3 thin films deposited using PLD laser powers of 5 mW, 10 mW, 100 mW, 200 mW, and 400 mW. The spectral response and transmittance of the detector shift to longer wavelengths as the PLD laser power increases.

Close modal

In conclusion, CeF3 thin films were fabricated using pulsed laser deposition (PLD). The PLD laser power was varied from 5 mW to 400 mW. CeF3 was observed to break down to CeF2, leading to the formation of fluorine defects when the PLD laser power was increased. These defects manifested as extra peaks at 304 nm and 450–650 nm in the absorption spectra. Consequently, the resistivity of the thin films was observed to be lower at higher PLD laser powers. The I–V characteristic curves under UV irradiation reveal that the thin film grown at 5 mW PLD laser power has a higher sensitivity characterized by a high photocurrent-to-dark current ratio. Irradiating the thin films with UV radiation, without using any filters, revealed that the spectral response and transmission edge of the detector shifts to longer wavelengths as the PLD laser power increases. The response wavelength can be tuned from 300 nm to 400 nm depending on the PLD laser power used during the fabrication process. The filterless nature of the photoconductive thin film detectors and its tunability make the detectors simpler and more flexible to design.

This work was supported by a Grant-in-Aid for Scientific Research (C) (Grant No. 16K04961) from the Japan Society for the Promotion of Sciences (JSAP) and Strategic Foundational Technology Improvement Support Operation, Japan. M.C.-R. gratefully acknowledges the Massey University Research Fund (No. MURF 1000021467) and the Strategic Research Excellence Fund (SREF) for financial support (Grant No. 1000022242).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
D. A.
Bolon
and
C. O.
Kunz
,
Polym. Eng. Sci.
12
,
109
(
1972
).
2.
J. R.
Vig
,
J. Vac. Sci. Technol., A
3
,
1027
(
1985
).
3.
T.
Krauss
and
S.
Warlen
,
J. Forensic Sci.
30
,
10991J
(
1985
).
4.
A.
D’Alessio
,
A.
D’Anna
,
G.
Gambi
, and
P.
Minutolo
,
J. Aerosol Sci.
29
,
397
(
1998
).
5.
U.
Schühle
and
J.-F.
Hochedez
,
Observing Photons in Space
, ISSI Scientific Report Series Vol. 9 (
Springer
,
New York, NY
,
2010
), pp.
467
477
.
6.
M.
Yanagihara
,
M. Z.
Yusop
,
M.
Tanemura
,
S.
Ono
,
T.
Nagami
,
K.
Fukuda
,
T.
Suyama
,
Y.
Yokota
,
T.
Yanagida
, and
A.
Yoshikawa
,
APL Mater.
2
,
046110
(
2014
).
7.
L. S.
Pan
,
D. R.
Kania
,
P.
Pianetta
, and
O. L.
Landen
,
Appl. Phys. Lett.
57
,
623
(
1990
).
8.
M. A.
Plano
,
S.
Zhao
,
C. F.
Gardinier
,
M. I.
Landstrass
,
D. R.
Kania
,
H.
Kagan
,
K. K.
Gan
,
R.
Kass
,
L. S.
Pan
,
S.
Han
,
S.
Schnetzer
, and
R.
Stone
,
Appl. Phys. Lett.
64
,
193
(
1994
).
9.
F.
Foulon
,
P.
Bergonzo
,
C.
Borel
,
R. D.
Marshall
,
C.
Jany
,
L.
Besombes
,
A.
Brambilla
,
D.
Riedel
,
L.
Museur
,
M. C.
Castex
, and
A.
Gicquel
,
J. Appl. Phys.
84
,
5331
(
1998
).
10.
M.
Misra
,
D.
Korakakis
,
H. M.
Ng
, and
T. D.
Moustakas
,
Appl. Phys. Lett.
74
,
2203
(
1999
).
11.
W.
Yang
,
R. D.
Vispute
,
S.
Choopun
,
R. P.
Sharma
,
T.
Venkatesan
, and
H.
Shen
,
Appl. Phys. Lett.
78
,
2787
(
2001
).
12.
X. L.
Tong
,
Q. G.
Zheng
,
S. L.
Hu
,
Y. X.
Qin
, and
Z. H.
Ding
,
Appl. Phys. A
79
,
1959
(
2004
).
13.
T.
Ishimaru
,
M.
Ieda
,
S.
Ono
,
Y.
Yokota
,
T.
Yanagida
, and
A.
Yoshikawa
,
Thin Solid Films
534
,
12
(
2013
).
14.
M.
Ieda
,
T.
Ishimaru
,
S.
Ono
,
N.
Kawaguchi
,
K.
Fukuda
,
T.
Suyama
,
Y.
Yokota
,
T.
Yanagida
, and
A.
Yoshikawa
,
Jpn. J. Appl. Phys., Part 1
51
,
062202
(
2012
).
15.
C. G.
Olson
,
M.
Piacentini
, and
D. W.
Lynch
,
Phys. Rev. B
18
(
10
),
5740
(
1978
).
16.
Y.-H. A.
Wang
,
X.
Zhang
,
N.
Bao
,
B.
Lin
, and
A.
Gupta
,
J. Am. Chem. Soc.
133
,
11072
(
2011
).
17.
W. S.
Choi
,
M. F.
Chisholm
,
D. J.
Singh
,
T.
Choi
,
G. E.
Jellison
, Jr.
, and
H. N.
Lee
,
Nat. Commun.
3
,
689
(
2012
).
18.
T.
Xie
 et al,
ACS Appl. Mater. Interfaces
7
,
9660
(
2015
).
19.
C.
Ning
,
L.
Dou
, and
P.
Yang
,
Nat. Rev. Mater.
2
,
17070
(
2017
).
20.
C. R.
Wronski
,
Z. E.
Smith
,
S.
Aljishi
,
V.
Chu
,
K.
Shepard
,
D.-S.
Shen
,
R.
Schwarz
,
D.
Slobodin
, and
S.
Wagner
,
AIP Conf. Proc.
157
,
70
(
1987
).
21.
S. S.
Batsanov
,
V. A.
Egorov
, and
Yu. B.
Khvostov
,
Dokl. Proc. Acad. Sci. USSR, Dokl. Chem.
227
,
251
(
1976
).
22.
M.
Nikl
,
J. A.
Mares
,
E.
Mihokova
,
A.
Beitlerova
,
K.
Blazek
, and
J.
Jindra
,
Solid State Commun.
87
,
185
(
1993
).
23.
C.
Shi
,
J.
Deng
,
T.
Wei
, and
G.
Zhang
,
Chin. Phys. Lett.
17
,
532
(
2000
).