Epitaxial Fe(Se,Te) thin films were prepared by pulsed laser deposition on (La0.18Sr0.82)(Al0.59Ta0.41)O3 (LSAT), CaF2-buffered LSAT and bare CaF2 substrates, which exhibit an almost identical in-plane lattice parameter. The composition of all Fe(Se,Te) films were determined to be FeSe0.7Te0.3 by energy dispersive X-ray spectroscopy, irrespective of the substrate. Albeit the lattice parameters of all templates have comparable values, the in-plane lattice parameter of the FeSe0.7Te0.3 films varies significantly. We found that the superconducting transition temperature (Tc) of FeSe0.7Te0.3 thin films is strongly correlated with their a-axis lattice parameter. The highest Tc of over 19 K was observed for the film on bare CaF2 substrate, which is related to unexpectedly large in-plane compressive strain originating mostly from the thermal expansion mismatch between the FeSe0.7Te0.3 film and the substrate.

Iron chalcogenides are attractive materials due to their very high superconducting transition temperature (Tc) above 100 K in the form of monolayer.1,2 Additionally, iron chalcogenides have the simplest crystal structure among the Fe-based superconductors (FBS) and the less toxic nature, which is favorable for fundamental studies as well as applications.3–6 In the case of FeSe1-xTex thin films, Tc varies considerably depending on growth condition and substrate material.7–15 However, the reason for such a large Tc variation in thin films is still under debate. Bellingeri et al. suggested that the lattice parameters and the superconducting properties of Fe(Se,Te) thin films show a non-trivial dependence on the in-plane lattice constant of the substrates on which they are deposited.7 Fe(Se,Te) thin films with high crystalline quality as well as excellent superconducting properties have been prepared on CaF2(001),8,9 CeO2-buffered single crystals (Y-stabilized ZrO2 and SrTiO3),10,11 and technical substrates due to the good matching of the a-axis length between the various templates and Fe(Se,Te) film (a = 3.801 Å),16 where the respective lattice parameters of CaF2 and CeO2 are a/2 = 3.862Å and a/2 = 3.82 Å.8,10 However, it was shown that the lattice parameter a of Fe(Se,Te) thin films on CaF2 is shorter than that of bulk materials,7–9,12–15 whereas a larger value would be expected for a coherent growth on this substrate. On the other hand, (La0.18Sr0.82)(Al0.59Ta0.41)O3 (LSAT) seems to be a good template, as the lattice parameter a of LSAT is 3.868 Å, which is almost the same value as that of the CaF2 substrate, resulting in a lattice mismatch of -1.76% for FeSe0.5Te0.5 films.16 However, Imai et al. did not observe epitaxial growth of Fe(Se,Te) on LSAT(001) and additionally no superconductivity.17–19 Hence, the question arises, why Fe(Se,Te) films on CaF2 and LSAT substrates show contrasting crystalline and superconducting properties, even though the in-plane lattice parameter of CaF2 is almost identical to that of LSAT. In this letter, FeSe0.7Te0.3 thin films were deposited on three different templates with similar in-plane lattice constant, i.e. bare LSAT, CaF2-buffered LSAT and bare CaF2 single crystal substrates in order to investigate the key factor for achieving high Tc. We show that the in-plane lattice parameter for FeSe0.7Te0.3 strongly depends on substrate and correlates with the superconducting transition temperature. The origin of this behavior is discussed in terms of interface properties and thermal expansion mismatch.

CaF2 buffer layers with different thickness between 20 nm and 80 nm were deposited on LSAT(001) substrates in a customer-designed molecular beam epitaxy (MBE) chamber at 400°C. Afterwards, Fe(Se,Te) films were deposited on the aforementioned templates by pulsed laser deposition (PLD) with a KrF excimer laser (wavelength: 248 nm, repetition rate: 7 Hz) under UHV conditions with a background pressure of 10-9 mbar as described in our previous report.14 The nominal FeSe0.5Te0.5 target prepared by a melting process was used for the deposition process. The detailed target preparation was found in Ref. 20. The substrate temperature was fixed at 360°C during deposition.

The structural properties of the films were investigated by X-ray diffraction (XRD) in θ-2θ geometry at a Bruker D8 Advance with Co-Kα radiation (wavelength: 0.178896 nm) in order to reduce the fluorescence and at a texture goniometer Philips X’Pert with Cu-Kα radiation (wavelength: 0.15418 nm). The c-axis lattice parameters of Fe(Se,Te) were determined by plotting the Nelson Riley function versus the lattice parameter calculated for each peak of the θ-2θ scans.21 The lattice parameter a was derived from reciprocal space maps (RSM) measured with a Panalytical X’pert Pro system with Cu-Kα radiation. Transmission electron microscopy (TEM) investigations of the films were performed in a FEI Tecnai-T20 TEM operated at 200 kV acceleration voltage. TEM lamellae were prepared by focused ion beam technique (FIB) in a FEI Helios 600i using an acceleration voltage of 3 kV in the last FIB step. The film thickness was confirmed to be 90 ± 5 nm. The composition of the samples was determined by energy-dispersive X-ray spectroscopy (EDX) with an Edax EDAMIII spectrometer. EDX linescans across the cross-section of the films confirm the stoichiometry to be homogeneous over the film thickness. It was found that the composition of the films is FeSe0.7Te0.3 for all studied substrates, irrespectively, if a sintered or single crystalline FeSe0.5Te0.5 target was used. The stoichiometry of the films might originate from a loss of Te during the PLD process due to its high vapor pressure and simultaneous Fe enrichment of the target surface due to the used low energy density of the laser beam. Electrical transport properties were measured in a Physical Property Measurement System [(PPMS) Quantum Design] by a standard four-probe method.

Figure 1 summarizes the structural characterization of the FeSe0.7Te0.3 films deposited on the different templates by XRD. In Fig. 1(a), only sharp 00l peaks of the FeSe0.7Te0.3 films, the substrates and the CaF2 buffer layers are present, indicating high phase purity with c-axis alignment of the films. The φ-scans of the 101 reflection for all FeSe0.7Te0.3 films, Fig. 1(b), show a fourfold symmetry, indicating that the films are epitaxially grown. Here, the respective epitaxial relation for the films on CaF2, CaF2-buffered LSAT and LSAT are identified as (001)[100]FeSe0.7Te0.3||(001)[110]CaF2, (001)[100]FeSe0.7Te0.3||(001)[110]CaF2||(001)[100]LSAT and (001)[100]FeSe0.7Te0.3||(001)[100]LSAT. However, the FeSe0.7Te0.3 on bare LSAT shows a significant larger full width at half maximum FWHM value, Δφ, of 8.12°, indicating that the crystalline quality of the film deposited on bare LSAT is inferior to the other films. On the other hand, the FeSe0.7Te0.3 film deposited on CaF2-buffered LSAT has a Δφ of around 0.7°, similar to the one on CaF2 single crystals, regardless of CaF2-buffer layer thickness.

FIG. 1.

XRD patterns for FeSe0.7Te0.3 films deposited on bare LSAT, CaF2-buffered LSAT (80 nm of CaF2 buffer) and bare CaF2: (a) θ/2θ scan and (b) φ-scans using the 101 reflection of FeSe0.7Te0.3. The asterisks mark reflections of the main substrate peak arising from secondary radiation of the x-ray tube.

FIG. 1.

XRD patterns for FeSe0.7Te0.3 films deposited on bare LSAT, CaF2-buffered LSAT (80 nm of CaF2 buffer) and bare CaF2: (a) θ/2θ scan and (b) φ-scans using the 101 reflection of FeSe0.7Te0.3. The asterisks mark reflections of the main substrate peak arising from secondary radiation of the x-ray tube.

Close modal

The cross-sectional scanning TEM images of the FeSe0.7Te0.3 thin films on different substrates are presented in Fig. 2. The large area view of the film on bare LSAT is shown in Fig. 2(a). The areas with different contrast indicate different crystal orientations in the ab-plane, which is consistent with the XRD φ-scans. As shown in Fig. 2(b), the interface between LSAT and the FeSe0.7Te0.3 layer is clean. The FeSe0.7Te0.3 layers seem to contain no correlated defects except the in-plane grain boundaries mentioned above. A cross-sectional TEM image of the film on LSAT with CaF2 buffer layer (80 nm) is shown in Fig. 2(c). A unique structure is observed at the interface between FeSe0.7Te0.3 and the CaF2 buffer layer. Many triangular shaped features with several tens of nanometers in length are observed at the interface, which reflects the preferred habit plane for CaF2 crystal (i.e., [111]). A similar observation is reported in Ref. 9. In spite of the rough interface between the CaF2 buffer layer and FeSe0.7Te0.3, the thin films exhibit a flat surface. An area near the interface between the FeSe0.7Te0.3 film and the CaF2 buffer layer is shown in Fig. 2(d). Surprisingly, the ordered atomic FeSe0.7Te0.3 planes immediately appear from the bottom of the valley of the CaF2 buffer layer. Additionally, the FeSe0.7Te0.3 layer neither contains extended defects nor high angle grain boundaries. In contrast, no peculiar triangular features are found at the interface between FeSe0.7Te0.3 and CaF2 substrate, as shown in Fig. 2(e). Nevertheless, a bright area of about 5 nm thick is observed at the interface between the CaF2 substrate and the FeSe0.7Te0.3 film, similar to the observations in previous reports.22,23 Presumably this is a reaction layer between film and substrate. The sharp boundary between FeSe0.7Te0.3 and reaction layer indicates that the reaction layer is likely formed on the CaF2 substrate side of the interface. The possible origin will be discussed below.

FIG. 2.

Cross-sectional TEM images of the FeSe0.7Te0.3 thin film on: (a) bare LSAT and (b) interface between LSAT substrate and FeSe0.7Te0.3 film, (c) 80 nm thick CaF2 buffered LSAT, (d) interface between CaF2 buffer and FeSe0.7Te0.3 superconducting film, (e) interface between CaF2 single crystal substrate and FeSe0.7Te0.3 superconducting film.

FIG. 2.

Cross-sectional TEM images of the FeSe0.7Te0.3 thin film on: (a) bare LSAT and (b) interface between LSAT substrate and FeSe0.7Te0.3 film, (c) 80 nm thick CaF2 buffered LSAT, (d) interface between CaF2 buffer and FeSe0.7Te0.3 superconducting film, (e) interface between CaF2 single crystal substrate and FeSe0.7Te0.3 superconducting film.

Close modal

The analysis of the RSM measurements shows a clear dependence of the in-plane lattice parameter on the substrate used. The films on bare LSAT substrates have the largest a-axis lattice parameter, with a value of around 3.79 Å. This agrees well with the bulk lattice parameter of this material indicating a relaxed growth due to the high number of grain boundaries in this sample observed by the TEM.24 By utilizing a CaF2 buffer layer, the a-axis value decreases to 3.775 Å, irrespective of the fact that the in-plane lattice parameter of the CaF2 buffer is a/2 = 3.87 Å almost identical to that of the LSAT substrate. Additionally, no influence of the CaF2 buffer thickness on the a-axis length of the FeSe0.7Te0.3 films is observed. Finally, the FeSe0.7Te0.3 films on bare CaF2 have the smallest a-axis length, ∼3.75 Å. In general, such small a-axis lengths of the FeSe0.7Te0.3 films on bare CaF2 substrates cannot be explained by the coherent epitaxial growth between the substrate and the film as the lattice parameter of CaF2 is larger than the a-axis of FeSe0.7Te0.3. However, the coefficients of linear thermal expansion of the substrates are 18.9 × 10−6 K−1 for CaF2,25 and 8.2 × 10−6 K−1 for LSAT,26 which are both higher than that of Fe(Se,Te),27 having the value of 4.2 × 10−6 K−1. Consequently, the larger shrinkage of the substrate in comparison to the film leads to a compressive strain,12,28,29 during the cooling process after deposition resulting in a shorter a-axis length of the film. The expected change is larger for the film on CaF2 single crystal in comparison to the LSAT-based films. We assume that this is the major contribution. Additionally, Ichinose et al. attributed the shrinkage of the a-axis lattice constant of Fe(Se,Te) film to an inter-diffusion of Se2- and F- at the interface.9 This might give an additional contribution to the lattice parameter change and would explain the difference in the a-axis for the FeSe0.7Te0.3 films on LSAT and CaF2-buffered LSAT, respectively. However, no clear evidence for Se diffusion was found in EDX experiments done in our TEM study of the interface between the FeSe0.7Te0.3 films and the CaF2.

The temperature dependencies of the normalized resistance for the FeSe0.7Te0.3 films grown on the different substrates are shown in Fig. 3. All films show superconductivity, however, the normal state behavior is different. The film on bare LSAT exhibits a crossover from metallic to semiconducting-like behavior with decreasing temperature prior to the superconducting transition. The resistive upturn just before the transition is suppressed for the films on CaF2 buffer layer (80 nm). The inset of Fig. 3 displays an enlarged view in the vicinity of the superconducting transition. Tc is defined as 90% of the resistance in the normal state, which is just before the superconducting transition. The lowest Tc of 4.9 K is measured for the film on bare LSAT with the poorest degree of texture. One of the most interesting features in Fig. 3 is the significant enhancement of Tc by employing a CaF2 buffer layer. Tc is increased to 12.4 K for the film with CaF2 buffer on LSAT. However, it is lower in comparison to the film on bare CaF2. In this case, FeSe0.7Te0.3 has the highest Tc.

FIG. 3.

Normalized resistive traces for FeSe0.7Te0.3 films on a variety of templates in zero magnetic field. The inset shows a close view on the superconducting transition.

FIG. 3.

Normalized resistive traces for FeSe0.7Te0.3 films on a variety of templates in zero magnetic field. The inset shows a close view on the superconducting transition.

Close modal

In order to emphasize the correlation between the structural parameters and transport properties, we tentatively plotted Tc as a function of lattice constants for a series of the FeSe0.7Te0.3 films on different substrates (Fig. 4). As shown in Fig. 4(a), the film on bare LSAT has the lowest Tc with the largest a value. Moreover, the film on bare CaF2 has the highest Tc with the shortest a-axis value. Despite the variation of the substrates, it is clear that the superconducting transition temperature Tc is observed to decrease linearly with increasing a-axis lattice parameter for the FeSe0.7Te0.3 films, which is consistent with the previous report.30 Therefore, the shrinkage of the a-axis lattice parameter, related to the compressive strain induced by the CaF2 substrate due to the large thermal misfit and an additional inter-diffusion layer, might be crucial for such a high Tc value. In contrast, the behavior of Tc is widely scattered against the c-axis length, as shown in Fig. 4(b). Therefore, these results strongly suggest that the a-axis length is the dominant factor, which affects the superconducting properties of FeSe0.7Te0.3 films.

FIG. 4.

The superconducting transition temperature, Tc as a function of (a) a-axis and (b) c-axis lattice parameters. The line is a guide for the eye.

FIG. 4.

The superconducting transition temperature, Tc as a function of (a) a-axis and (b) c-axis lattice parameters. The line is a guide for the eye.

Close modal

In summary, FeSe0.7Te0.3 thin films were deposited on a variety of templates by PLD. The film on bare LSAT has the lowest Tc with the largest a value. However, a CaF2 buffer layer significantly improves the crystalline quality and superconducting properties of the FeSe0.7Te0.3 films. Furthermore, for the film on bare CaF2, the a-axis lattice parameter shrinks due to the epitaxial compressive strain from CaF2 originating from the thermal expansion mismatch. Thus, the in-plane lattice mismatch between Fe(Se,Te) and the substrates (CaF2 and LSAT) is not the key factor for the a-axis lattice parameter of the film. Additionally, Tc of FeSe0.7Te0.3 films is dominantly affected by the a-axis length.

The authors thank S. Richter (IFW Dresden) for fruitful discussions, J. Scheiter for help with TEM lamellae preparation, and M. Kühnel and U. Besold for technical support. The work was partly supported by the National Science Foundation of China (Grant No. NSFC-U1432135,11674054 and 11611140101) and Open Partnership Joint Projects of JSPS Bilateral Joint Research Projects (Grant No. 2716G8251b), the JSPS Grant-in-Aid for Scientific Research (B) Grant Number 16H04646 and the DFG funded GRK1621. V. G. is grateful to the DFG (GR 4667/1-1) for financial support. The publication of this article was fund by the Open Access Fund of the Leibniz Association.

1.
J.-F.
Ge
,
Z.-L.
Liu
,
C.
Liu
,
C.-L.
Gao
,
D.
Qian
,
Q.-K.
Xue
,
Y.
Liu
, and
J.-F.
Jia
,
Nature Mater.
14
,
285
(
2015
).
2.
I.
Bozovic
and
C.
Ahn
,
Nature Phys.
10
,
892
(
2014
).
3.
F.-C.
Hsu
,
J.-Y.
Luo
,
K.-W.
Yeh
,
T.-K.
Chen
,
T.-W.
Huang
,
P. M.
Wu
,
Y.-C.
Lee
,
Y.-L.
Huang
,
Y.-Y.
Chu
,
D.-C.
Yan
, and
M.-K.
Wu
,
Proc. Natl. Acad. Sci. USA
105
,
14262
(
2008
).
4.
Y.
Kamihara
,
T.
Watanabe
,
M.
Hirano
, and
H.
Hosono
,
J. Am. Chem. Soc.
130
,
3296
(
2008
).
5.
M.
Rotter
,
M.
Tegel
, and
D.
Johrendt
,
Phys. Rev. Lett.
101
,
107006
(
2008
).
6.
X.
Wang
,
Q.
Liu
,
Y.
Lv
,
W.
Gao
,
L.
Yang
,
R.
Yu
,
F.
Li
, and
C.
Jin
,
Solid State Commun.
148
,
538
(
2008
).
7.
E.
Bellingeri
,
S.
Kawale
,
V.
Braccini
,
R.
Buzio
,
A.
Gerbi
,
A.
Martinelli
,
M.
Putti
,
I.
Pallecchi
,
G.
Balestrino
,
A.
Tebano
, and
C.
Ferdeghini
,
Supercond. Sci. Technol.
25
,
084022
(
2012
).
8.
I.
Tsukada
,
M.
Hanawa
,
T.
Akiike
,
F.
Nabeshima
,
Y.
Imai
,
A.
Ichinose
,
S.
Komiya
,
T.
Hikage
,
T.
Kawaguchi
,
H.
Ikuta
, and
A.
Maeda
,
Appl. Phys. Express
4
,
053101
(
2011
).
9.
A.
Ichinose
,
F.
Nabeshima
,
I.
Tsukada
,
M.
Hanawa
,
S.
Komiya
,
T.
Akiike
,
Y.
Imai
, and
A.
Maeda
,
Supercond. Sci. Technol.
26
,
075002
(
2013
).
10.
W.
Si
,
S. J.
Han
,
X.
Shi
,
S. N.
Ehrlich
,
J.
Jaroszynski
,
A.
Goyal
, and
Q.
Li
,
Nature Commun.
4
,
1347
(
2013
).
11.
T.
Ozaki
,
L.
Wu
,
C.
Zhang
,
J.
Jaroszynski
,
W.
Si
,
J.
Zhou
,
Y.
Zhu
, and
Q.
Li
,
Nature Commun.
7
,
13036
(
2016
).
12.
A.
Ichinose
,
I.
Tsukada
,
F.
Nabeshima
,
Y.
Imai
,
A.
Maeda
,
F.
Kurth
,
B.
Holzapfel
,
K.
Iida
,
S.
Ueda
, and
M.
Naito
,
Appl. Phys. Lett.
104
,
122603
(
2014
).
13.
Y.
Imai
,
Y.
Sawada
,
F.
Nabeshima
, and
A.
Maeda
,
Proc. Natl. Acad. Sci. USA
112
,
1937
(
2015
).
14.
F.
Yuan
,
K.
Iida
,
M.
Langer
,
J.
Hänisch
,
A.
Ichinose
,
I.
Tsukada
,
A.
Sala
,
M.
Putti
,
R.
Hühne
,
L.
Schultz
, and
Z.
Shi
,
Supercond. Sci. Technol.
28
,
065005
(
2015
).
15.
I.
Tsukada
,
A.
Ichinose
,
F.
Nabeshima
,
Y.
Imai
, and
A.
Maeda
,
AIP Adv.
6
,
095314
(
2016
).
16.
B. C.
Sales
,
A.
Sefat
,
M. A.
Mcguire
,
R.
Jin
,
D.
Mandrus
, and
Y.
Mozharivskyj
,
Phys. Rev. B
79
,
094521
(
2009
).
17.
Y.
Imai
,
T.
Akiike
,
M.
Hanawa
,
I.
Tsukada
,
A.
Ichinose
,
A.
Maeda
,
T.
Hikage
,
T.
Kawaguchi
, and
H.
Ikuta
,
Appl. Phys. Express
3
,
043102
(
2010
).
18.
M.
Hanawa
,
A.
Ichinose
,
S.
Komiya
,
I.
Tsukada
,
T.
Akiike
,
Y.
Imai
,
T.
Hikage
,
T.
Kawaguchi
,
H.
Ikuta
, and
A.
Maeda
,
Jpn. J. Appl. Phys.
50
,
053101
(
2011
).
19.
M.
Hanawa
,
A.
Ichinose
,
S.
Komiya
,
I.
Tsukada
,
Y.
Imai
, and
A.
Maeda
,
Jpn. J. Appl. Phys.
51
,
010104
(
2011
).
20.
A.
Palenzona
,
A.
Sala
,
C.
Bernini
,
V.
Braccini
,
M.
Cimberle
,
C.
Ferdeghini
,
G.
Lamura
,
A.
Martinelli
,
I.
Pallecchi
, and
G.
Romano
,
Supercond. Sci. Technol.
25
,
115018
(
2012
).
21.
R.
Deokate
,
S.
Pawar
,
A.
Moholkar
,
V.
Sawant
,
C.
Pawar
,
C.
Bhosale
, and
K.
Rajpure
,
Appl. Surf. Sci.
254
,
2187
(
2008
).
22.
P.
Mele
,
K.
Matsumoto
,
K.
Fujita
,
Y.
Yoshida
,
T.
Kiss
,
A.
Ichinose
, and
M.
Mukaida
,
Supercond. Sci. Technol.
25
,
084021
(
2012
).
23.
V.
Braccini
,
S.
Kawale
,
E.
Reich
,
E.
Bellingeri
,
L.
Pellegrino
,
A.
Sala
,
M.
Putti
,
K.
Higashikawa
,
T.
Kiss
, and
B.
Holzapfel
,
Appl. Phys. Lett.
103
,
172601
(
2013
).
24.
M.
Fang
,
H.
Pham
,
B.
Qian
,
T.
Liu
,
E.
Vehstedt
,
Y.
Liu
,
L.
Spinu
, and
Z.
Mao
,
Phys. Rev. B
78
,
224503
(
2008
).
25.
R.
Roberts
and
G.
White
,
J. Phys. C
19
,
7167
(
1986
).
26.
W.
Wang
,
H.
Yang
, and
G.
Li
,
CrystEngComm
15
,
2669
(
2013
).
27.
Y.
Xiao
,
Y.
Su
,
C.
Kumar
,
C.
Ritter
,
R.
Mittal
,
S.
Price
,
J.
Perßon
, and
T.
Brückel
,
Eur. Phys. J. B
82
,
113
(
2011
).
28.
Q.
Lei
,
M.
Golalikhani
,
D.
Yang
,
W.
Withanage
,
A.
Rafti
,
J.
Qiu
,
M.
Hambe
,
E.
Bauer
,
F.
Ronning
,
Q.
Jia
,
J.
Weiss
,
E.
Hellstrom
,
X.
Wang
,
X.
Chen
,
F.
Williams
,
Q.
Yang
,
D.
Temple
, and
X.
Xi
,
Supercond. Sci. Technol.
27
,
115010
(
2014
).
29.
K.
Iida
,
V.
Grinenko
,
F.
Kurth
,
A.
Ichinose
,
I.
Tsukada
,
E.
Ahrens
,
A.
Pukenas
,
P.
Chekhonin
,
W.
Skrotzki
,
A.
Teresiak
,
R.
Hühne
,
S.
Aswartham
,
S.
Wurmehl
,
I.
Mönch
,
M.
Erbe
,
J.
Hänisch
,
B.
Holzapfel
,
S.-L.
Drechsler
, and
D. V.
Efremov
,
Sci. Rep.
6
,
28390
(
2016
).
30.
E.
Bellingeri
,
I.
Pallecchi
,
R.
Buzio
,
A.
Gerbi
,
D.
Marrè
,
M. R.
Cimberle
,
M.
Tropeano
,
M.
Putti
,
A.
Palenzona
, and
C.
Ferdeghini
,
Appl. Phys. Lett.
96
,
102512
(
2009
).