Study of Ni-doping effect of specific heat and transport properties for LaFe1-yNiyAsO0.89F0.11

Article abstract of DOI:10.1143/JPSJ.80.084720

Specific heats and transport quantities of the LaFe _1-yNiyAsO_0.89F_0.11 system have been measured, and the results are discussed together with those reported previously by our group mainly for LaFe_1-yCoyAsO₀

Full text of DOI:10.1143/JPSJ.80.084720

Journal of the Physical Society of Japan 80 (2011) 084720
DOI: 10.1143/JPSJ.80.084720
FULL PAPERS
Study of Ni-Doping Effect of Specific Heat and Transport Properties
for LaFe1ÀyNiyAsO0:89F0:11
ꢀ
1;3
1;3
Takayuki KAWAMATA
, Erika SATOMI1;3, Yoshiaki KOBAYASHI
,
y
Masayuki ITOH1;3, and Masatoshi SATO
1;2;3
1Department of Physics, Division of Material Science, Nagoya University, Nagoya 464-8602, Japan
2Toyota Physical and Chemical Research Institute, Nagakute, Aichi 480-1192, Japan
3JST, TRIP, Nagoya University, Nagoya 464-8602, Japan
(Received May 5, 2011; accepted June 6, 2011; published online August 10, 2011)
Specific heats and transport quantities of the LaFe1ꢁyNiyAsO0:89F0:11 system have been measured, and the results
are discussed together with those reported previously by our group mainly for LaFe1ꢁyCoyAsO0:89F0:11 and
LaFeAsO0:89ꢁxF0:11þx systems. The y dependence of the electronic specific heat coefficient ꢀ can basically be
understood by using the rigid-band picture, where Ni ions provide two electrons to the host conduction bands and
behave as nonmagnetic impurities. The superconducting transition temperature Tc of LaFe1ꢁyNiyAsO0:89F0:11 becomes
zero as the carrier density p (¼ 2y þ 0:11) doped to LaFeAsO reaches its critical value, pc ꢂ 0:2. This pc value of ꢂ0:2
is commonly observed for LaFe1ꢁyCoyAsO0:89F0:11 and LaFeAsO0:89ꢁxF0:11þx systems, in which the relations p ¼
x þ 0:11 and p ¼ y þ 0:11 hold, respectively. As we pointed out previously, the critical value corresponds to the
disappearance of the hole-Fermi surface. These results indicate that the carrier number solely determines the Tc value.
We have not observed appreciable effects of pair breaking, which originates from the nonmagnetic impurity scattering
of conduction electrons and strongly suppresses Tc values of systems with sign reversal of the order parameter over the
Fermi surface(s). On the basis of the results, the so-called s symmetry of the order parameter with sign reversal is
ꢃ
excluded.
KEYWORDS: Fe pnictites, superconducting symmetry, rigid-band picture, impurity effect
to be common and to correspond to the electron number at
1. Introduction
which the doped electrons (almost) completely bury the hole
For newly found Fe pnictide superconductors with rather Fermi surfaces around the ꢁ point in the reciprocal space.5,6)
high transition temperature Tc,1) there has been considerable The result is important, because it implies that the carrier
interest in the symmetry of the superconducting order number solely determines the Tc values and that there is no
parameter ꢀ, because the symmetry gives us important appreciable effect of the pair breaking that originates, in
information on the pairing mechanism. If the spin fluctuation systems with sign-reversing ꢀ’s (around ꢁ and M points
is relevant to the pairing, we expect the so-called s
in the case of the s symmetry), from conduction electron
ꢃ
ꢃ
symmetry,2,3) which has a sign reversal between ꢀ’s on two scattering by nonmagnetic impurities and brings about the
kinds of Fermi surfaces around ꢁ and M points in the significant difference in the Tc-suppression rates or the pc
reciprocal space. However, as we have been pointing out values between the Co- and F-doped systems with and
from the early stage of the studies,4–8) the rates of the without the scattering centers in the conducting FeAs planes,
Tc suppression observed for Co- and Ru-doped Ln1111 respectively. Similar behaviors can also be observed in the
systems, LnFe1ꢁyMyAsO0:89F0:11 (Ln = La and Nd; M ¼ Co data reported for the BaFe2As2 system doped with various
and Ru), are too small to explain as the pair-breaking effect species of transition metal atoms to Fe sites.11,12)
for systems with the s symmetry originating from the
ꢃ
Here, we measured specific heats and transport quantities
conduction electron scattering by nonmagnetic impurities. of the Ni-doped La1111 system to obtain further information
Because this result is a strong indication that the systems do on the impurity effects. First, we examined the validity of
not have s symmetry, we may have to consider a new the rigid-band picture for Ni-doped La1111 to establish the
ꢃ
pairing mechanism involving the orbital degrees of freedom, following. (i) A doped Ni atom donates two electrons to the
for example.9,10)
host bands and does not carry localized magnetic moment.
In addition to the above finding, our studies4–6,8) have (Co and Ru atoms donate one and zero electron, respec-
also revealed that for LaFe1ꢁyCoyAsO0:89F0:11 and LaFeAs- tively.6)) (ii) The electronic band structure is not so severely
O
0:89ꢁxF0:11þx, the superconductivity disappears at a com- affected by the Ni impurities doped to the conducting planes.
mon critical values of x and y. Then, as long as the band (It is also true for the Co-doped system,6) while for the
shape is not severely changed by the doping, we can expect Ru-doped system, the band width becomes larger with
that the critical value (pc) of the electron numbers p increasing y in the relatively large y region.7,8)
)
If the above picture is valid for the present system, we can
[¼ ð y þ 0:11Þ and (x þ 0:11) for LaFe1ꢁyCoyAsO0:89F0:11
and LaFeAsO0:89ꢁxF0:11þx, respectively] is also considered separately argue, at least in the region of small y, effects of
the carrier-number change and conduction-electron scatter-
ing. Actually, the observed characteristics of the electronic
ꢀPresent address: Department of Applied Physics, Tohoku University,
Sendai 980-8578, Japan.
specific heat coefficient ꢀ of LaFe1ꢁyCoyAsO0:89F0:11 can
basically be understood in terms of the rigid band picture,13)
where the Ni atom donates two electrons to the La1111
yPresent address: Research Center for Neutron Science and Technology,
Comprehensive Research Organization for Science and Society, Tokai,
Ibaraki 319-1106, Japan.
084720-1
#2011 The Physical Society of Japan

J. Phys. Soc. Jpn. 80 (2011) 084720
FULL PAPERS
T. KAWAMATA et al.
system and behaves as a nonmagnetic impurity. The rate of
Tc suppression by Ni impurities can be understood by the
above idea that the electron number solely determines the Tc
values, where a pc value common to those of the Co- and
F-doped systems is found. It confirms that the effect of pair
8.8
8.7
8.6
8.5
8.4
8.3
8.2
4.14
4.12
4.10
4.08
4.06
4.04
4.02
(a)
LaFe1-yNiyAsO0.89F0.11
c-axis
breaking by nonmagnetic impurities expected for the s
symmetry of ꢀ does not exist. The result strongly supports
ꢃ
a-axis
that the system has the s symmetry without sign reversal
þþ
of the order parameters.
Experimental details are described in x2. Then, the
specific heat data and transport data are presented in x3.1
and x3.2, respectively, to show how the rigid band picture
describes them, where discussion is also given on whether
the pair breaking effect exists or not. In that section, we also
discuss briefly a randomness effect on the superconductivity
induced by the impurity doping, and in x4, the summary is
given.
8.1
1.0
0.0
0.2
0.4
0.6
0.8
y (Ni)
0.05
0
-5
0.045
0.04
0.02
0.03
(b)
0.00
0.035
0.005
2. Experiments
0.01
-10
-15
0.015
ZFC
Polycrystalline samples of LaFe1ꢁyNiyAsO0:89F0:11 were
prepared from initial mixtures of La, La2O3, LaF3, FeAs,
and NiAs at nominal molar ratios. The NiAs powder was
obtained by annealing mixtures of Ni- and As-powders in an
evacuated quartz tube at 850 ꢄC. Details of the preparations
are given in our previous paper.4) The X-ray powder patterns
were obtained with Cu Kꢁ radiation. We observed an
impurity phase, LaOF, in several samples, where typical
molar concentration was ꢂ0:03. We found that Ni doping
was basically successful, as can be judged from the y
dependence of the lattice parameters a and c shown in
Fig. 1(a). These results are consistent with those for the
LaFe1ꢁyNiyAsO system.14,15) The superconducting diamag-
χ
H = 10G
LaFe Ni AsO
F
1-y
20
y
0.89 0.11
0
10
30
40
T (K)
Fig. 1. (Color online) (a) Lattice parameters a and c of LaFe1ꢁyNiy-
AsO0:89F0:11 are plotted against y. (b) T dependence of the magnetic
susceptibility obtained under the condition of zero-field cooling (ZFC) for
LaFe1ꢁyNiyAsO0:89F0:11. The numbers attached to each curve indicate the
corresponding y value.
netic moments were measured using a Quantum Design magnitudes ꢀC at Tc and calculated the ꢀ values, simply
SQUID magnetometer with a magnetic field H of 10 G assuming the relation ꢀC=Tc ¼ 1:43ꢀ. [In Fig. 2(b), the
under both zero-field-cooling (ZFC) and field-cooling (FC) jump at Tc observed for the Ni-doped sample is shown, and
conditions. Only the ZFC data are shown in Fig. 1(b). similar data for the Co-doped samples are shown in Fig. 8(a)
Electrical resistivities were measured by a four-terminal of ref. 6.]
method with an ac-resistance bridge. Tc values were
Figure 3(b) shows the p dependence of ꢀ obtained
determined as described previously,4) using the data of the experimentally for LaFe1ꢁyNiyAsO0:89F0:11 together with
superconducting diamagnetic susceptibilities and electrical those for LaFe1ꢁyMyAsO0:89F0:11 (M ¼ Co, Mn) systems.6)
resistivities ꢂ. The two kinds of Tc values are in good In this figure, we used equations p ¼ 2y þ 0:11, p ¼
agreement. Hall coefficients RH of the polycrystalline y þ 0:11, and p ¼ ꢁy þ 0:11 for the Ni-, Co-, and Mn-
samples were measured with a stepwise increase in the doped systems, respectively. Actually, comparing the results
temperature T at a magnetic field H of 7 T, where the sample of the band calculation for LaFeAsO shown in Fig. 3(a),
plates were rotated around the axis perpendicular to the field, which shows the electronic density of states NðEÞ against
and the thermoelectric power S was measured by methods the electron energy E,13) we find that characteristics of the
described previously.16,17)
electronic p or E dependence of ꢀ or NðEÞ are very similar,
although we find two discrepancies between the observed
and calculated results. The first one is that there is no sharp
peak at p ꢂ 1:0 for the Ni-doped system. This is possibly
3. Results and Discussion
3.1 Rigid-band picture
In Fig. 2(a), data of the specific heat are shown in the due to the smearing of NðEÞ by the effect of severe
form of C=T–T2 for several samples of LaFe1ꢁyNiyAsO0:89
0:11. Several nonsuperconducting samples exhibit an upturn the other hand, because the peak position almost corresponds
-
randomness introduced by Ni impurities. In Co-doping, on
F
with decreasing T owing to the magnetically active nature to the pure Co system [see Fig. 3(a), where the chemical
of the systems at low temperatures and superconducting potentials for LaCoAsO and LaNiAsO are shown by the
samples exhibit a jump at Tc. For the nonsuperconducting arrows.13,18)], it is not smeared out, and a ferromagnetic
samples, the ꢀ values were estimated by fitting the equation phase transition originating from the large NðEÞ value takes
C=T ¼ ꢀ þ ꢃT2 þ ꢄT4 to the observed data. In this fitting, place at Tc ꢂ 60 K.19,20) The second discrepancy is that in
we neglected the upturn of the specific heat observed at low the y region of the Ni-doped system, [ð2y þ 0:11Þ > 1:3], the
temperature with decreasing T. For the superconducting observed ꢀ begins to decrease with increasing p or E,
samples, on the other hand, we estimated the jump whereas the calculated NðEÞ does not. The decrease may be
084720-2
#2011 The Physical Society of Japan

J. Phys. Soc. Jpn. 80 (2011) 084720
FULL PAPERS
T. KAWAMATA et al.
200
150
100
50
LaFe1-yNiyAsO0.89 0.11
F
0.95
0.80
0.60
0.50
LaFe0.99Ni0.01AsO0.89F0.11
240
200
Tc
0.45
0.41
0.35
0.22
0.05
160
120
80
ΔC/Tc
0.01
(a)
(b)
0
0
50 100 150 200 250 300 350 400
18
20
22
24
26
28
2
2
T (K)
T (K )
Fig. 2. (Color online) (a) Specific heat C divided by temperature T is plotted against T2 for several samples of LaFe1ꢁyNiyAsO0:89F0:11. The number
attached to each curve indicates the corresponding y value. (b) ꢀC=T is plotted against T for the sample of LaFe1ꢁyNiyAsO0:89F0:11 with y ¼ 0:01. From the
jump in ꢀC=T at Tc, the ꢀ value was estimated using the relation ꢀC=T ¼ 1:43ꢀ.
10
(a)
LaFe1-yNiyAsO0.89 0.11
F
pNi
3
2
1
0
0
0.03
0.02
5
pCo
1.0
E (eV)
0.04
0.01
0
0.0
0.5
1.5
2.0
0.20
40
0.30
0.05
0.10
Ni
Co
Mn
M=
LaFe1-yMyAsO0.89F0.11
0.40
(b)
30
20
10
0
ρ
0.50
0.60
0.80
1.00
γ
0
50
100 150 200 250 300
T (K)
0.0
0.5
1.0
1.5
2.0
2y+0.11 (Ni), y+0.11 (Co), -y+0.11 (Mn)
Fig. 4. (Color online) T dependence of the resistivity ꢂ of LaFe1ꢁy
NiyAsO0:89F0:11 The number attached to each curve indicates the
corresponding y value.
-
.
Fig. 3. (Color online) (a) Electronic density of states calculated for
LaFeAsO against the electron energy.13) (b) The electron-number (p)
dependence of the electronic specific heat coefficient ꢀ for LaFe1ꢁy
NiyAsO0:89F0:11
-
.
becomes larger. The anomalous behavior has also been
observed in the thermoelectric power S and Hall coefficient
RH, as shown in Figs. 5(a) and 5(c), respectively. For
due to the decrease of the electron effective mass, which has samples with y ꢅ 0:05 (p ꢅ 0:2), both S and RH exhibit
actually been observed in the behaviors of the resistivity and strong T dependence similar to those observed in high-Tc Cu
magnetic susceptibility, as the system approaches LaNiAsO. oxides.16,17) These anomalous behaviors have been observed
In the present studies, it has been shown how the rigid in various Fe pnictide systems.4–8) They indicate that the
band picture is applicable to the present system, where Ni system is magnetically active, similarly to the superconduct-
impurity can be considered to be nonmagnetic and to donate ing Cu oxides.16,17,21) The observed decrease in jSj and jRHj
two electrons to the host bands.
with increasing y can be understood by considering the
weakening of the magnetic fluctuation owing to the
reduction in the area of the hole Fermi surface. The
3.2 Transport properties
Figure 4 shows the temperature dependence of ꢂ for decreasing rates, djSj=dy and djRHj=dy, observed in the
several samples of LaFe1ꢁyNiyAsO0:89F0:11. The super- relatively low-T region (but above Tc) for the Ni-doped
conducting transition appears in the region of y . 0:04, system are roughly two times larger than those for the Co-
and samples with y & 0:05 exhibit nonsuperconducting doped system.6) This result also supports the idea that
metallic behavior. For y ¼ 0, ꢂ exhibits the S-shape T
Co- and Ni-atoms donate one and two electrons to the
dependence. This behavior seems to become weaker as y conduction band, respectively.
084720-3
#2011 The Physical Society of Japan

J. Phys. Soc. Jpn. 80 (2011) 084720
FULL PAPERS
T. KAWAMATA et al.
0
0
0.30
0.45
0.30
0.20
0.10
0.40
0.50
0.20
-50
-50
-100
-150
0.10
0.05
0.045
0.05
0.0125
0.03
0.02
-100
0.00
0.00
-150
0
LaFe1-yNiyAsO0.89 0.11
F
LaFe1-yCoyAsO0.89F0.11
(a)
(b)
0
50
100 150 200 250 300
50
100 150 200 250 300
T (K)
T (K)
1.00
0.10
0.50
0.50
0
0
0.20
0.40
0.10
0.00
0.05 0.04
0.01
0.0025
0.03
-10
-10
-20
0.00
0.01
0.005
0.00
0.00
0.02
-20
0
(c)
(d)
LaFe1-yNiyAsO0.89F0.11
LaFe1-yCoyAsO0.89F0.11
50
100 150 200 250 300
0
50
100 150 200 250 300
T (K)
T (K)
6)
Fig. 5. (Color online) T dependences of the thermoelectric power S of (a) LaFe1ꢁyNiyAsO0:89F0:11 and (b) LaFe1ꢁyCoyAsO0:89F0:11
.
T dependences of
Hall coefficient RH of (c) LaFe1ꢁyNiyAsO0:89F0:11 and (d) LaFe1ꢁyCoyAsO0:89F0:11
.
6) The number attached to each curve indicates the corresponding y value.
35
LaFe1-yMyAsO0.89F0.11
LaFe1-yMyAsO0.89-xF0.11+x
2
30
M = Ru
x = 0
25
M = Ni
1
20
ρ
F-doped
y = 0
15
M = Co
M = Co
0
x = 0
0.00
0.05
0.10
y (Ni, Co)
0.15
0.20
0.25
10
M = Ni
x = 0
5
Fig. 6. (Color online) Residual resistivities ꢂ0 of LaFe1ꢁyNiyAsO0:89F0:11
and LaFe1ꢁyCoyAsO0:89F0:11 are plotted against y. The broken line shows
the initial rate of ꢂ0 increase with y.
6)
0
0.1
0.2
0.3
x+2y+0.11 (Ni)
x+y+0.11 (Co, Ru, F)
Figure 6 shows the y dependence of ꢂ0 for LaFe1ꢁyNiyAs-
6)
O
0:89F0:11 together with that for LaFe1ꢁyCoyAsO0:89F0:11
.
Fig. 7. (Color online) Tc values determined from the resistivity meas-
urements are plotted for LaFe1ꢁyNiyAsO0:89F0:11 against x þ 2y þ 0:11 and
for LaFe1ꢁyMyAsO0:89ꢁxF0:11þx against x þ y þ 0:11 (x ¼ 0 for M ¼ Co6)
and Ru,8) and y ¼ 0 for the sample with nonzero x8)).
The values of ꢂ0 were estimated by extrapolating the ꢂ–T
curve to T ¼ 0 from the T region where the resistivity
upturn with decreasing T is not significant. The initial value
of dꢂ0=dy is nearly equal to that for the Co-doped system,6)
indicating that the scattering strength of Ni impurities is
similar to that of Co atoms.
systems, the horizontal axis indicates the electron numbers p
In Fig. 7, the Tc values of the Ni-doped system are shown doped into LaFeAsO. For the Ru-doped system, on the other
against x þ 2y þ 0:11. In the figure, those of Co-, Ru-, and hand, the horizontal axis merely indicates the Ru atom
F-doped systems6,8) are also plotted against x þ y þ 0:11. (In concentration (note that the electron number does not change
these figures, x ¼ 0 for Co-, Ni-, and Ru-doped systems, and with y). In the figure, the initial slope of the curve shown for
y ¼ 0 for the F-doped system). For Ni-, Co-, and F-doped the Ni-doped system is almost equal to those for the Co- and
084720-4
#2011 The Physical Society of Japan

J. Phys. Soc. Jpn. 80 (2011) 084720
FULL PAPERS
T. KAWAMATA et al.
F-doped systems, and Tc disappears at the critical value pnictide superconductors. We also studied the symmetry of
pc ꢂ 0:2, which corresponds to the disappearance of the hole the superconducting order parameter. We found that the
Fermi surface. For the Ru-doped system, the initial absolute characteristics of the electric specific coefficients ꢀ of the
slope is very small, which supports that Tc basically depends Ni- and Co-doped6) systems can be understood, (at least in
only on the carrier number.
If superconductors have the s symmetry, the initial rate calculation reported for the LaFeAsO system,13) showing the
of Tc suppression by nonmagnetic impurities can be validity of the rigid band picture. On the basis of this result,
the y region described by ½ð2y þ 0:11Þ < 1:3ꢇ), by the band
ꢃ
estimated by a simple pair breaking equation,
we can consider that a Ni dopant donates two electrons to
this system and also behaves as a nonmagnetic impurity. The
Tc suppression can be understood on the basis of the idea
that the carrier number solely determines the Tc values. No
ꢀ
ꢁ
ꢀ
ꢁ
ꢀ ꢁ
Tc0
Tc
1
2
ꢁ
1
ln
¼ ¼
þ
ꢁ ¼
;
2t
2
where ¼ðzÞ is the digamma function defined as ¼ðzÞ ꢆ effect of pair breaking due to conduction electron scattering
ln½dꢁðzÞ=dz=ꢁðzÞꢇ, Tc0 is the superconducting Tc of the by nonmagnetic impurities has been observed. The results
nondoped system, and t ¼ Tc=Tc0. The pair breaking strongly exclude the s symmetry of the order parameter.
ꢃ
parameter ꢁ ꢆ h=ð2ꢅk T ꢆÞ can be calculated from the
ꢂ
B
c0
Acknowledgment
initial rate of ꢂ0 and the carrier number estimated from the
value of RH by a method described in detail in a previous
The authors thank Professor H. Kontani for fruitful
paper.6) For the present system, Tc must disappear at a y discussion. The work is supported by Grants-in-Aid for
value smaller than 0.0035, which should be compared with Scientific Research from the Japan Society for the Promo-
the observed value of ꢂ0:04{0:05. This argument indicates tion of Science (JSPS) and JST, TRIP.
that not only for Co- and Ru-doped systems but also for the
Ni-doped system, the observed Tc-suppression rates cannot
be explained in terms of pair-breaking effect, that is, the
1) Y. Kamihara, T. Watanabe, M. Hirano, and H. Hosono: J. Am. Chem.
Soc. 130 (2008) 3296.
system does not have the s symmetry of the order
ꢃ
parameter.
Here, we present a brief comment on the relatively large
2) I. I. Mazin, D. J. Singh, M. D. Johannes, and M. H. Du: Phys. Rev.
Lett. 101 (2008) 057003.
decrease in Tc reported by Guo et al. for the Zn-doped
La1111 system.22) As we discussed in detail previously,6)
their results obtained for samples prepared under high
pressure can be naturally explained by the well-known effect
3) K. Kuroki, S. Onari, R. Arita, H. Usui, Y. Tanaka, H. Kontani, and H.
Aoki: Phys. Rev. Lett. 101 (2008) 087004.
4) A. Kawabata, S. C. Lee, T. Moyoshi, Y. Kobayashi, and M. Sato:
J. Phys. Soc. Jpn. 77 (2008) 103704.
5) S. C. Lee, A. Kawabata, T. Moyoshi, Y. Kobayashi, and M. Sato:
J. Phys. Soc. Jpn. 78 (2009) 043703.
of the electron localization: when the sheet resistance R
ꢀ
defined, by using the lattice parameter c, as R ꢆ ꢂ =c
ꢀ
6) M. Sato, Y. Kobayashi, S. C. Lee, H. Takahashi, E. Satomi, and Y.
Miura: J. Phys. Soc. Jpn. 79 (2010) 014710.
0
exceeds h=4e2 ꢂ 6:45 kꢃ, Tc disappears. It is also stressed
that for the present multiband system, the absolute slope
jdTc=dyj, due to the electron scattering by nonmagnetic
7) S. C. Lee, E. Satomi, Y. Kobayashi, and M. Sato: J. Phys. Soc. Jpn. 79
(2010) 023702.
8) E. Satomi, S. C. Lee, Y. Kobayashi, and M. Sato: J. Phys. Soc. Jpn. 79
(2010) 094702.
9) H. Kontani and S. Onari: Phys. Rev. Lett. 104 (2010) 157001.
impurities, is significantly larger, if it has the s symmetry,
ꢃ
than that expected for single band Cu oxide superconductors,
even though the latter systems also have the sign-reversing
ꢀ. It is because impurities such as Zn atoms doped into the
conducting planes of the single-band superconductors act as
the unitary scatterers, for which the Tc suppression effect is
greatly reduced.23) Therefore, for Fe pnictide systems, even
if the observed jdTc=dyj is as large as that observed for Zn-
doped Cu oxide superconductors, it does not indicate the
existence of the pair breaking effect.
¯
10) Y. Yanagi, Y. Yamakawa, and Y. Ono: Phys. Rev. B 81 (2010)
054518.
11) P. C. Canfield, S. L. Bud’ko, N. Ni, J. Q. Yan, and A. Kracher: Phys.
Rev. B 80 (2009) 060501(R).
12) N. Ni, A. Thaler, J. Q. Yan, A. Kracher, E. Colombier, S. L. Bud’ko,
P. C. Canfield, and S. T. Hannahs: Phys. Rev. B 82 (2010) 024519.
13) D. J. Singh and M.-H. Du: Phys. Rev. Lett. 100 (2008) 237003.
14) G. Cao, S. Jiang, X. Lin, C. Wang, Y. Li, Z. Ren, Q. Tao, C. Feng, J.
Dai, Z. Xu, and F.-C. Zhang: Phys. Rev. B 79 (2009) 174505.
15) S. Jiang, C. Wang, M. He, J. Yu, Q. Tao, J. Dai, Z. Xu, and G. Cao:
Physica C 470 (2010) S456.
As we reported previously, jdTc=dyj observed for
LaFe1ꢁyMnyAsO0:89F0:11 is also rather large.6) However,
for this system, the resistivity increases very rapidly with
increasing y, indicating that the electron localization takes
place at a very small value of y, and the system rapidly
exhibits the magnetic nature. We cannot distinguish which
of the magnetism and electron localization is of primary
importance for the Tc suppression.
16) T. Nishikawa, J. Takeda, and M. Sato: J. Phys. Soc. Jpn. 63 (1994)
1441.
17) J. Takeda, T. Nishikawa, and M. Sato: Physica C 231 (1994) 293.
18) G. Xu, W. Ming, Y. Yao, X. Dai, S.-C. Zhang, and Z. Fang: Europhys.
Lett. 82 (2008) 67002.
19) H. Yanagi, R. Kawamura, T. Kamiya, Y. Kamihara, M. Hirano, T.
Nakamura, H. Osawa, and H. Hosono: Phys. Rev. B 77 (2008) 224431.
20) H. Ohta and K. Yoshimura: Phys. Rev. B 79 (2009) 184407.
21) H. Kontani, K. Kanki, and K. Ueda: Phys. Rev. B 59 (1999) 14723.
22) Y. F. Guo, Y. G. Shi, S. Yu, A. A. Belik, Y. Matsushita, M. Tanaka, Y.
Katsuya, K. Kobayashi, I. Nowik, I. Felner, V. P. S. Awana, K.
Yamaura, and E. Tayama-Muromachi: Phys. Rev. B 82 (2010)
054506.
4. Summary
We carried out measurements of the specific heats and
transport properties of LaFe1ꢁyNiyAsO0:89F0:11 and exam-
ined whether the rigid-band picture can be applied to the Fe
23) S. Onari and H. Kontani: Phys. Rev. Lett. 103 (2009) 177001.
084720-5
#2011 The Physical Society of Japan