Effects of magnetic impurities on upper critical fields in

the high-Tc superconductor La-doped CaFe2As2

Soon-Gil Jung1, Soohyeon Shin1, Harim Jang1, Pavlo Mikheenko2, Tom H Johansen2, and

Tuson Park1

1Center for Quantum Materials and Superconductivity (CQMS) and Department of Physics,

Sungkyunkwan University, Suwon 16419, Republic of Korea

2Department of Physics, University of Oslo, P.O. Box 1048 Blindern, 0316 Oslo, Norway

E-mal: tp8701@skku.edu

Fax: +82 31 290 7056

Abstract

We investigate the effects of magnetic impurities on the upper critical field (μ0Hc2) in La-doped

CaFe2As2 (LaCa122) single crystals. The magnetic field dependency of the superconducting transition

temperature (Tc) for LaCa122 is rapidly suppressed at low fields up to ~1 kOe despite its large μ0Hc2(0)

value on the order of tens of Tesla, resulting in a large positive curvature of μ0Hc2(T) near Tc. The

magnetization hysteresis (M − H) loop at temperatures above Tc shows a ferromagnetic signal and the

M(H) value rapidly increases with increasing magnetic field up to ~1 kOe. Taken together with the linear

suppression of Tc with the magnetization in the normal state, these results suggest that the large upward

curvature of μ0Hc2(T) near Tc in La-doped CaFe2As2 mainly originates from the suppression of

superconductivity due to the presence of magnetic impurities.

PACS numbers: 74.25.Op, 74.62.Dh, 74.70.Xa, 74.81.-g

keywords: La-doped CaFe2As2, upper critical field, magnetic impurity

1. Introduction

Since the discovery of Fe-based superconductors with a superconducting (SC) transition

temperature (Tc) of 26 K [1], interesting results have continued to be reported, such as high SC

transition temperature in an electron-doped FeSe monolayer (Tc > 100 K) [2] and in rare-earth

(RE) doped CaFe2As2 (Tc ~ 49 K) [3, 4]. Electron-doped CaFe2As2 (Ca122) compounds with RE

elements (RE: La, Ce, Sm, Pr, Nd) as dopants, show the highest Tc values among AFe2As2 (A:

Ca, Sr, Ba, Eu) compounds [5, 6]. The compounds have different Tc - dopant concentration

phase diagrams distinguishing between hole and electron doping, compared to high-Tc cuprates,

where hole doping usually induces a higher Tc than electron doping [7, 8]. The origin of the

higher Tc in electron-doped Ca122 is still under debate because of their large inhomogeneity and

non-bulk superconductivity [9-12].

Understanding the mechanism of the high Tc in Fe-based superconductors is critical to

achieving room temperature superconductivity. Together with the critical current and SC

transition temperature, the upper critical field (μ0Hc2) is one of the most important

superconducting parameters that can provide information on the pairing mechanism [13-14].

The μ0Hc2(T) of RE-doped Ca122 shows a remarkable upward curvature near Tc, which could be

explained by the two-band theory, Ginzburg-Landau phenomenology, and the Werthamer-

Helfand-Hohenberg (WHH) model [14-19]. However, its origin is still contradictory and needs

to be clarified. Despite the large μ0Hc2(T) on the order of tens of Tesla, a fast decrease of Tc at

low fields is difficult to explain [15, 16, 20-23]. It has been reported that there exists a large

inhomogeneity of superconducting phases and filamentary superconductivity in these materials

[9, 10, 12], which in turn could strongly affect the Tc.

Here, we report on the upper critical fields of the La-doped Ca122 single crystals and

discuss a plausible origin of the large suppression of Tc at low fields. Magnetotransport

measurements show a rapid decrease of Tc with magnetic fields up to ~1 kOe, which is

consistent with the ferromagnetic-like magnetization hysteresis (M − H) loops measured at

temperatures above Tc. M(H) shows a strong field dependence up to ~1 kOe and is then

saturated at higher magnetic fields, indicating that the rapid decrease of Tc at low fields is

closely associated with the destruction of superconductivity by ferromagnetic impurities in the

La-doped Ca122 system.

2. Experimental

La-doped Ca122 single crystals were synthesized by using the FeAs self-flux method [20, 22].

First, the binary FeAs precursor was prepared by reacting Fe powder (99.998%, Alfa Aesar)

with As lumps (99.999%, Alfa Aesar). Then, Ca (granules 99.5%, Alfa Aesar), La (rods 99.9%,

Alfa Aesar), and the FeAs precursor were put into an alumina crucible at a ratio of (Ca,

La):FeAs = (0.8, 0.2):4 and sealed in an evacuated quartz tube. Next, the ampoule was heat-

treated at 1,180 °C for 24 h and then slowly cooled to 960 °C at a rate of 4 °C/h. Finally, the

furnace was turned off and cooled down to room temperature naturally. Through this process,

we obtained ~18% La-doped Ca122 (LaCa122) single crystals with plate-like shapes. The

composition ratio of the synthesized LaCa122 crystals was determined by employing energy-

dispersive spectroscopy (EDS).

Two small pieces of LaCa122 single crystals, LaCa122-A and LaCa122-B, were used in this

study with sizes of 1.50 0.45 0.045 mm3 and 1.17 0.46 0.074 mm3, respectively. The

superconducting phase transition of LaCa122-A is quite sharp compared to that of LaCa122-B,

and LaCa122-A shows a clear two-step superconducting transition. The temperature

dependencies of the in-plane electrical resistivity in magnetic fields were measured by using a

standard four-probe technique and the measurements were performed using a physical property

measurement system (PPMS 9T, Quantum Design). Magnetization hysteresis (M − H) loops

were measured by using a magnetic property measurement system (MPMS 5T, Quantum Design).

In order to probe the homogeneity of the crystals, magneto-optical imaging was performed at 4 K

after field cooling for the relatively large-sized La-Ca122 crystals.

3. Results and discussion

Figures 1(a) and 1(b) show the temperature dependencies of the in-plane electrical resistivity

(ρab) and magnetization (M) for the LaCa122-A and LaCa122-B crystals, respectively.

LaCa122-A sample reveals a shaper SC transition than LaCa122-B in the ρab(T), as presented in

figure 1(a), where ρab(T) is normalized by its value at 46 K for comparison for each sample. The

SC transition widths (ΔTc) for LaCa122-A and -B are 5.0 and 9.3 K, respectively, where the

width was determined from a resistivity decrease from 90 to 10% of the normal state value at Tc.

When plotted on a semi-logarithmic scale, as shown in the inset in figure 1(a), the two-step SC

transition is clearly observed in LaCa122-A, which is commonly seen in RE-doped Ca122

superconductors [9-12, 20, 24].

Difference is also observed in the Meissner effect for LaCa122-A and -B samples as shown

by the temperature dependencies of magnetization (M) in figure 1(b). The stronger temperature

dependency of the zero-field cooled (ZFC) M values for LaCa122-B indicates more rapid

suppression of the SC volume with increasing temperature compared to LaCa122-A. An

increase of the field cooled (FC) M with decreasing temperature and the positive magnetization

values at low temperatures imply that LaCa122-B contains ferromagnetic-like impurities [25,

26].

The temperature dependencies of the in-plane resistivity (ρab) of LaCa122-A and -B in

different magnetic fields are shown in figures 2(a) and (c), respectively, where the magnetic

fields were field was applied perpendicular to the ab plane. The SC transition for both samples

becomes much broader even in a small magnetic field of 100 Oe. This rapid suppression of

superconductivity by small fields is observed up to 0.1 T and then, the SC suppression rate is

reduced as the magnetic field increases towards 9 T. Figures 2(b) and (d) present the field

dependencies of ρab(T) on a semi-logarithmic scale, revealing a long tail with another SC

transition for both LaCa122-A and -B. The two-step SC phase transitions can be attributed to the

inhomogeneously distributed high-Tc phases that are surrounded by the low-Tc or non-SC

phases [9-12].

Figures 3(a) and 3(b) show the upper critical fields (μ0Hc2) as a function of temperature for

LaCa122-A and B, respectively. Because both samples have a broad SC transition, μ0Hc2(T)

was obtained by using three different criteria of ρ90%, ρ50%, and ρ10% of the normal-state

resistivity value near the Tc for each magnetic field. Significant positive curvature near Tc is

observed for all cases and is linked to the rapid decrease of Tc in low magnetic fields. The two-

band or multiband theory has been broadly used to describe the positive curvature in μ0Hc2(T)

[13, 14, 19, 27]. The Werthamer-Helfand-Hohenberg (WHH) model has been previously

applied to estimate μ0Hc2(0), for which the upward curvature was not taken into account [16, 21-

23]. In this study, the μ0Hc2(T) values for both samples were fitted by using the empirical

formula μ0Hc2(T) = μ0Hc2(0)[1 - (T/Tc)2]n, where the best fits (solid lines) were obtained with

similar n values for both compounds: n(ρ90%) = 2.48 and 2.45, n(ρ50%) = 2.84 and 2.90, and

n(ρ10%) = 3.28 and 3.21 for LaCa122-A and B, respectively. The large n values for the ρ10%

criterion are related to the fast decrease of Tc as the magnetic field increases. High-field

experiments are required for more precise investigation of μ0Hc2(T) for RE-doped Ca122

compounds.

The upper critical field of LaCa122 is plotted against 1−(T/Tc)2 on a logarithmic scale in

figure 4(a), where μ0Hc2 shows a slight deviation from the linear dependence below 1 kOe. As

shown in figure 4(b), the SC transition width defined as the difference between 90% and 10% of

the normal-state resistivity value at Tc, ΔTc = Tc(ρ90%) –Tc( ρ10%), rapidly increases up to ~1 kOe.

Both crystals show similar field dependencies of ΔTc(H) although their ΔTc values at zero field

are very different: 5.0 K and 9.3 K for LaCa122-A and -B, respectively. Figures 4(c) and 4(d)

representatively show the M − H loops at 2 and 50 K for LaCa122-A and -B, respectively. At

50 K, which is higher than Tc, both compounds show ferromagnetic (FM) behavior. The FM

signal for LaCa122-A is relatively small compared to the large SC signal at 2 K. In contrast, the

FM signal for LaCa122-B is comparable to the SC signal at 2 K and strongly distorts measured

magnetization loop, which is a superposition of the FM and SC signals [28-30]. The field-

dependent M value at 50 K for LaCa122-B rapidly increases with magnetic field up to ~1 kOe

and then is almost saturated. This FM-like behavior is probably responsible for the small hump

developed at 1 kOe in the M − H curve at 2 K, as indicated by the arrows [31]. These results

suggest that the significant upward curvature in μ0Hc2(T) near Tc is mainly caused by the FM ic

impurities related to Fe or the undoped Ca122 phase, because ferromagnetic ordering can act as

a strong pair-breaking source in superconductors [10, 32-34].

Support of the importance of magnetic impurities for understanding the unusual behaviour

of upper critical field with temperature is given by figure 5(a). It shows that Tc’s for both

crystals are almost linearly decreases as function of the magnetization value at 50 K. Here we

used M at 50 K because it represents magnetic impurities only. Even though the relative

impurities contribution to the M − H loops for LaCa122-A is much smaller than for LaCa122-B

(see figure 4), the widening of superconducting transition with field is similar for both crystals.

Figure 5(b) shows the SC transition width ΔTc as a function of the normalized magnetization on

a logarithmic scale. The field-induced increase in the transition width ΔW follows a power-law

with the same exponent α = 0.520.05 for both crystals, ΔW = ΔTc(H) − ΔTc(0) ∝ Mα. Further

experimental and theoretical study is necessary to properly understand if the power-law

behavior is accidental or reflects the nature of the competition between superconductivity and

magnetism of impurities.

Magneto-optical imaging (MOI) was performed to visualize magnetism of the LaCa122

crystals. Figures 6(a) and (c) show optical images for both faces of the LaCa122 crystal and (b)

and (d) show MOI images of the crystal in the superconducting state, respectively. The latter are

remnant-state MOI images at 4 K after a field cooling procedure, and the brightness of in the

image represents the local density of trapped by superconductor flux. The images demonstrate

that the crystals are not uniform. The distinct nearly rectangular dark area in the right part of the

MO images shows a region, in which flux trapping is very weak. Note that the MO image

presented in figure 6(d), which is the image taken from the other side of the crystal shown in

figure 6(b), has the same dark region resulting likely from magnetic impurities. Apparently, this

particular region is weakly or non-superconducting throughout the entire sample thickness.

These MOI results are consistent with earlier reports in which RE-doped Ca122

superconductors were inhomogeneous and even that the high Tc could not be of bulk origin [9-

11, 35]. Our MOI study confirms bulk origin of superconductivity. Moreover, areas of

different contrast could be identified as having magnetic impurities. The similar weak contrast

is present at elevated temperatures, where superconductivity is suppressed (not shown). This is

in agreement with the role of magnetic impurities revealed by the M − H results.

4. Conclusions

In conclusion, we studied the upper critical field of La-doped Ca122 single crystals, in

which a significant positive curvature in μ0Hc2(T) is observed near Tc. The regime (μ0H 1 kOe),

where superconducting transition is strongly influenced by magnetic field, corresponds to that in

the magnetization in the normal state, where the M − H loops show ferromagnetic-like behavior

and M is almost saturated for μ0H ≥ 1 kOe. The observation of the linear relationship between

characteristic temperatures of superconducting transition and the normal-state magnetization

emphasizes that the fast increase of the width of superconducting transition at small magnetic

fields, which is typically ascribed to multiband effects in other Fe-based superconductors, may

originate from the magnetic impurities that suppress the Cooper electron pairing in the high-Tc

phase of the La-doped Ca122 superconductors. The results of magneto-optical study are in

agreement with these conclusions.

Acknowledgements

This work was supported by the National Research Foundation (NRF) of Korea by a grant

funded by the Korean Ministry of Science, ICT and Planning (No. 2012R1A3A2048816). S.-G.

Jung was also supported by the Basic Science Research Program through the National Research

Foundation of Korea (NRF) funded by the Ministry of Education (NRF-

2015R1D1A1A01060382).

Figure Captions

Figure 1. (a) Temperature dependence of the in-plane resistivity (ρab) of LaCa122-A (black) and

B (red), where ρab(T) is normalized by the resistivity value at 46 K for comparison. The semi-

logarithmic plot of (a) presented in the inset shows broad and clearly two-step SC transitions in

LaCa122-B. (b) Zero-field cooled (ZFC) and field cooled (FC) dc magnetization (M) at 2 Oe for

LaCa122-A and -B, where M(T) curves are normalized by the ZFC magnetization value at 2 K.

0 10 20 30 40 50

-0.9

-0.6

-0.3

0.0

0 10 20 30 40 500.0

0.2

0.4

0.6

0.8

1.0

FC

ZFC

M(T

)/M

(2 K

)

T (K)

LaCa122-A

LaCa122-B

2 Oe

LaCa122-A

LaCa122-B

(b)

ab(T

)/

ab(4

6 K

)

T (K)

(a)

0 10 20 30 40 50

1E-3

0.01

0.1

1

ab(T

)/

ab(4

6 K

)

T (K)

Figure 2. Temperature dependence of the in-plane resistivity (ρab) of LaCa122-A ((a) and (b))

and LaCa122-B ((c) and (d)) at different magnetic fields. The resistivitiy is plotted in (a) and (c)

on a linear scale and in (b) and (d) on a semi-logarithmic scale. Even though the upper critical

fields are large in both crystals, the SC transition is strongly suppressed by relatively low

magnetic field on the order of 0.1 T.

0.1

1

10

100

0 10 20 30 40 50

0.1

1

10

100

ab (c

m)

9 T

0 T

9 T

0 T

0 T

0.01 T

0.05 T

0.1 T

0.2 T

0.5 T

1 T

3 T

5 T

7 T

9 T

0 T

0.01 T

0.05 T

0.1 T

0.2 T

0.5 T

1 T

3 T

5 T

7 T

9 T

(c) LaCa122-B

(a) LaCa122-A

0 T

T (K)

9 T

9 T

(d) LaCa122-B

(b) LaCa122-A

ab (c

m)

0 T

0

20

40

60

80

ab (c

m)

0 10 20 30 40 500

60

120

180

ab (c

m)

T (K)

Figure 3. Temperature dependencies of the upper critical field (μ0Hc2) for (a) LaCa122-A and (b)

LaCa122-B. The upper critical field μ0Hc2(T) values were obtained using three different criteria

of resistivity: ρ90%, ρ50%, and ρ10%, which correspond to 90, 50, and 10% of the normal-state

resistivity value near Tc, respectively. Irrespective of the criteria, both crystals show a large

upward curvature in Hc2(T) near Tc. The solid lines are plotted based on the phenomenological

relationship of μ0Hc2(T) [1 − (T/Tc)2]n.

0 10 20 30 400

5

10

15

20

0

5

10

15

20

90%

50%

10%

0H

c2 (

T)

T (K)

(b)

LaCa122-B

LaCa122-A

90%

50%

10%

0H

c2 (

T)

(a)

Figure 4. (a) The upper critical field, μ0Hc2(T), of LaCa122-A (open circles) and -B (solid circles)

plotted as a function of 1−(T/Tc)2 on a logarithmic scale. (b) The SC transition width (ΔTc(H) −

ΔTc(0)) is shown as a function of the magnetic field, where the width rapidly increases at low

magnetic fields up to ~1 kOe. Here, ΔTc(H) = ρ90%(H) − ρ10%(H). The M − H loops for

LaCa122-A and -B are shown in (c) and (d), respectively. Ferromagnetic-like behavior for both

samples is observed in the M − H curves at 50 K, which is more clear for LaCa122-B.

0.01 0.1 1

0.01

0.1

1

10

0 2 4 60

2

4

6

8

-15 -10 -5 0 5 10 15

-2

-1

0

1

2

-15 -10 -5 0 5 10 15

-1.0

-0.5

0.0

0.5

1.0

A B

90%

50%

10%

0H

c2 (

T)

1-(T/Tc)

2

LaCa122-A

LaCa122-B

T

c(H

) -

Tc(0

) (K

)

0H (T)

2 K

50 K

M (10

-4 e

.m.u

.)

0H (kOe)

1 kOe

2 K

50 K

(d) LaCa122-B(c) LaCa122-A

(b) (a) LaCa122

M (10

-4 e

.m.u

.)

0H (kOe)

1 kOe

Figure 5. Relationship between width of superconducting transition and magnetization (M) of

the sample in normal state. (a) The decrease of characteristic temperature at which resistance is

50% of its value at Tc with corresponding magnetization value at 50 K. The open and solid

circles are for LaCa122-A and -B, respectively. The magnetization value is normalized by its

value at 5 kOe, M(H)/M(5kOe). Dashed lines show linear fits to the points. (b) The SC

transition width ΔW as a function of normalized magnetization at 50 K plotted in double

logarithmic scale. Here ΔW = ΔTc(H) − ΔTc(0). ΔW(M) follows a power-law with the same

exponent of 0.52 for both compounds, i.e., ΔW∝M0.520.05. Solid lines are results from the

power-law fits.

Figure 6. Optical ((a) and (c)) and magneto-optical ((b) and (d)) images of LaCa122 crystals.

The magneto-optical (MO) image in (b) corresponds to the optical image in (a), while the MO

image in (d) corresponds to the optical image in (c). The MO images were obtained at 4 K after

field cooling and the degree of brightness represents the local density of flux trapped in

superconductor, indicating that the crystal is inhomogeneous to some extent.

References

[1] Kamihara Y, Watanabe T, Hirano M and Hosono H 2008 Iron-based layered superconductor

La[O1-xFx]FeAs (x = 0.05 - 0.12) with Tc = 26 K J. Am. Chem. Soc. 130 3296

[2] Ge J- F et al 2015 Superconductivity above 100 K in single-layer FeSe films on doped

SrTiO3 Nat. Mater 14 285

[3] Lv B et al 2011 Unusual superconducting state at 49 K in electron-doped CaFe2As2 at

ambient pressure. Proc. Nat. Acad, Sci. 108 15705

[4] Saha S R et al 2012 Structural collapse and superconductivity in rare-earth-doped CaFe2As2.

Phys. Rev. B 85 024525

[5] Kuroki K and Hosono H 2015 Iron-based superconductors: Current status of materials and

pairing mechanism Physica C 514 399

[6] Lin C T and Chen D P 2014 The growth of 122 and 11 iron-based superconductor single

crystals and the influence of doping Supercond. Sci. Technol. 27 103002

[7] Rybicki D, Jurkutat M, Reichardt S, Kapusta C and Haase J 2016 Perspective on the phase

diagram of cuprate high-temperature superconductors Nat. Commun. 7 11413

[8] Basov D N and Chubukov A V 2011 Manifesto for a higher Tc Nat. Phys. 7 272

[9] Gofryk K et al 2014 Local inhomogeneity and filamentary superconductivity in Pr-Doped

CaFe2As2 Phys. Rev. Lett. 112 047005

[10] Deng L Z et al 2016 Evidence for defect-induced superconductivity up to 49 K in (Ca1-

xRx)Fe2As2 Phys. Rev. B 93 054513

[11] Chu C W et al 2013 The rise of Tc: A promising paradigm via interfacial mechanism.

Journal of Physics: Conference Series 449 012014

[12] Gao B et al 2014 Phase diagram and weak-link behavior in Nd-doped CaFe2As2 New J.

Phys. 16 113024

[13] Kogan V G and Prozorov R 2012 Orbital upper critical field and its anisotropy of clean

one- and two-band superconductors Rep. Prog. Phys. 75 114502

[14] Lee H- S et al 2009 Effects of two gaps and paramagnetic pair breaking on the upper

critical field of SmFeAsO0.85 and SmFeAsO0.8F0.2 single crystals Phys. Rev. B 80 144512

[15] Chen D- Y et al 2016 Superconductivity in Sm-doped CaFe2As2 single crystals Chin. Phys.

B 25 067403

[16] Zhou W et al 2013 Upper critical fields and anisotropy of Ca1-xLaxFe2As2 single crystals

Supercond. Sci. Technol. 26 095003

[17] Xiao H et al 2012 Filamentary superconductivity across the phase diagram of

Ba(Fe,Co)2As2 Phys. Rev. B 86 064521

[18] Zhang J- L, Jiao L and Yuan H- Q 2011 Universal behavior of the upper critical field in

iron-based superconductors Front. Phys. 6 463

[19] Gurevich A 2003 Enhancement of the upper critical field by nonmagnetic impurities in

dirty two-gap superconductors Phys. Rev. B 67 184515

[20] Sun Y et al 2013 Evidence of two superconducting phases in Ca1-xLaxFe2As2 AIP Adv. 3

102120

[21] Tamegai T, Ding Q P, Ishibashi T and Nakajima Y 2013 Superconducting properties of

Ca1-xRExFe2As2 (RE: Rare Earths) Physica C 484 31

[22] Gao Z et al 2011 Synthesis and properties of La-doped CaFe2As2 single crystals with

Tc=42.7 K EPL 95 67002

[23] Qi Y et al 2012 Transport properties and anisotropy in rare-earth doped CaFe2As2 single

crystals with Tc above 40 K Supercond. Sci. Technol. 25 045007

[24] Saha S R et al 2014 Segregation of antiferromagnetism and high-temperature

superconductivity in Ca1−xLaxFe2As2 Phys. Rev. B 89 134516

[25] Deng Q et al 2014 The effect of impurity and the suppression of superconductivity in

Na(Fe0.97−xCo0.03Tx )As (T = Cu, Mn) New J. Phys. 16 063020

[26] Bawa A, Gupta A, Singh S, Awana V P S and Sahoo S 2016 Ultrasensitive interplay

between ferromagnetism and superconductivity in NbGd composite thin films Sci. Rep. 6 18689

[27] Askerzade İ N, Gencer A and Güҫlü N 2002 On the Ginzburg-Landau analysis of the upper

critical field Hc2 in MgB2 Supercond. Sci. Technol. 15 L13

[28] Lee S F, Liou Y, Yao Y D, Shih W T and Yu C 2000 Magnetic flux penetration depth study

in Nb/Co system J. Appl. Phys. 87 5564

[29] Kobayashi S, Oike H, Takeda M and Itoh F 2002 Central peak position in magnetization

hysteresis loops of ferromagnet/superconductor/ferromagnet trilayered films Phys. Rev. B 66

214520

[30] Patel U et al 2009 Growth and superconductivity of FeSex crystals Appl. Phys. Lett. 94

082508

[31] Uchino T, Uenaka Y, Soma H, Sakurai T and Ohta H 2014 Magnetic hysteresis behavior

and magnetic pinning in a d0 ferromagnet/superconductor nanostructure J. Appl. Phys. 115

063910

[32] Li W et al 2012 Phase separation and magnetic order in K-doped iron selenide

superconductor Nat. Phys. 8 126

[33] Saparov B et al 2014 Complex structures of different CaFe2As2 samples Sci. Rep. 4 4120

[34] Cui J et al 2016 Coexistence of antiferromagnetic and ferromagnetic spin correlations in

Ca(Fe1−xCox)2As2 revealed by 75As nuclear magnetic resonance Phys. Rev. B 94 174512

[35] Xiao H et al 2012 Evidence for filamentary superconductivity nucleated at antiphase

domain walls in antiferromagnetic CaFe2As2 Phys. Rev. B 85 024530