SingleSingle-Particle and SDW Charge -Particle and SDW Charge DynamicsDynamics In In

andand : A Comparative : A Comparative OverviewOverview(TMTSF)2AsF6

T.Vuletić1, D.Herman1, N.Biškup1, M.Pinterić1,2, A.Omerzu1, S.Tomić1, M.Nagasawa3

contact e-mail: tvuletic@ifs.hr, vuletic@lps.u-psud.fr; stomic@ifs.hr

1 Institute of Physics, Zagreb, Croatia

2 Faculty of Civil Engineering, University of Maribor, Maribor, Slovenia

3 Dept. Of Physics, Hokkaido University, Sapporo, Japan

A Comparative Overview of DC and AC (100 mHz – 1MHz) electrical

transport measurements, in low and high electric fields, which we performed

during the last four years, of the spin-density wave (SDW) state of the

Bechagaard salts (TMTSF)2PF6 and (TMTSF)2AsF6 [1].

We argue that a degree of complex structure of the SDW ground state

which is unfolded in a particular experiment depends strongly on the chosen

experimental probe and the crystal measured.

INDICATIONS of the complex structure of the SDW ground state

NMR measurements of the spin-lattice relaxation rate (1/T1) have shown

a changeover at 3.5 K from a temperature independent behaviour at higher

temperatures to an activated behaviour at lower temperatures, with an activation

energy lower than the free carrier resistivity one [2].

Another anomaly in 1/T1 was detected at about 2K [2].

Calorimetric transition at 3.5 K with large hysteretic phenomena in the

temperature range 2.5 K – 4 K caused by the sample history has been observed

[3]. Measurements of the dielectric response suggested the critical slowing

down behaviour indicating a glass transition around 2 K [3].

Further, it was found that anomaly in 1/T1 follows Arrhenius behaviour

with free carrier activation energy [4]. This result contradicts the critical

slowing down suggested on the basis of the dielectric relaxation. It was pointed

out that the phason fluctuations are responsible for the 1/T1 relaxation and that

the observed NMR behaviour might result from a gradual slowing down of the

phason fluctuations. Hence, no need for the existence of a glass transition is

invoked [4].

Even more publications have been dedicated to this subject and the

above differences have not been clarified yet, leaving open the intriguing

question of the complex nature of the SDW phase of (TMTSF)2PF6 (and of

(TMTSF)2AsF6).

(TMTSF)2PF6

Single - particle transport

RED ARROWS at about 2.5 K indicate change in the activation energy of the single-particle conduction channel. Same kind of plots obtained 9 years ago featured the same change, but the authors were unaware of it, at that time [5]. Recent magnetoresistance results for current direction along intermediate axis were first to reveal this change at about 4K [6].

Resistance is normalized on a minimum value. Some curves are shifted for clarity.Letters KB denote crystals prepared by K. Bechgaard and MN those prepared by M. Nagasawa.KB S and KB P denote PF6 crystals from the same

batches as those studied and reported in [3] and [5].For samples denoted PF6, KB and AsF6, MN1 & MN2

nonlinear transport and dielectric response data are also presented further below.Different resistivity ratios influenced the absolute values of the activation energy.

1/T (1/K)

0.0 0.2 0.4 0.6 0.8

T (K)

10 5 3 2.5 2 1.7 1.4

KB

MN1

MN2

TCO

Rn

106

1

108

104

102

(TMTSF)2AsF6

CONCLUSION We have shown that the features of both the collective conduction channel in the SDW ground state and the SDW low-frequency dielectric response are sample (batch) dependent. In the PF6 samples in which the SDW is pinned by impurities the electric threshold field which characterizes the sliding mechanism gradually weakens with temperature, concomitantly as the SDW dielectric relaxation gradually slows down. In contrast, in the PF6 and AsF6 samples in which the commensurability pinning prevails [5,7], ET displays the maximum below 3 K, and a critical slowing down of the SDW dielectric relaxation might be observed. It should be noted that the latter behaviour was only observed in the sample in which an extremely narrow ET peak was found. Keeping in mind that the SDW wave vector is found to be very close to commensurability [8], we propose that subtle variations of the disorder level in nominally pure samples might be responsible for the different behaviours observed.

SDW — nonlinear transport

COLOURED AREAS denote temperature range 3 K - 1.6 K where the low temperature(below 4 K) maximum of threshold field (ET) occurs.

The position and width of maximum are sample (batch)dependent. This maximum is clearly visible for all studied (TMTSF)2AsF6 samples and only for PF6 crystals with prevailing commensurability pinning (KB P). For PF6 crystals with dominant impurity pinning (KB S and KB) there is no firm evidence of maximum down to 1.5K [5,7].

orig. file sn1_4;;;1_8;;;2_2;;;3_2.ivXI. 1997 (TMTSF)2PF6

E (mV/cm)

0 10 20 30 40 50

0.0

0.1

0.2

0.3

0.4

0.5

1.4 K1.8 K2.2 K3.2 K

T (K)

0 1 2 3 4 5

ET (

T)/

ET(5

K)

0

1

2

3

4

KB SKB PKB

T (K)

0 1 2 3 4 50

2

4

6

8

10

12

14

16KBMN2MN1

orig.file tas1i012,021,023;T1as4_2.spw12. 3. 1998 (TMTSF)2AsF6

E (mV/cm)

0 10 20 30 40 50 60 70

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

1.5 K1.8 K2.3 K3.4 K

E (mV/cm)

0 2 4 6 8 10 12 14

0.00

0.02

0.04

0.06

0.08

0.10

1.6K2.2K

1.5K

(TMTSF)2AsF6, MN1

◊ KB ▲ MN2 ∆ MN1

(TMTSF)2AsF6 (TMTSF)2PF6

Non-ohmic conductivity (σ - σ0 )/ σ0 vs. electric field for some temperatures. Figures shown are concerning those samplesfor which low - frequency dielectric response data is presented further on.

1-

0.4

0.6

0.8

1.0

1.2

file

d:p

ygm

alio

n /t

om

isla

v/p

rije

fr/

EC

RY

S.jn

b

d:a

rju

na

/to

mis

lav/

EC

RY

S.jn

b

(T

MT

SF

) 2AsF

6

106

107

108

1/T (1/K)

0.2 0.3 0.4 0.5 0.6 0.7

0(s)

10-6

10-5

10-4

10-3

R(

)

102

103

104

105

MN2MN1

TCO

T (K)

4 3 2.5 2 1.7 1.4

(Hz)

101 102 103 104 105 106

"

106

107

108

1.4K1.6K

2.4K1.8K

KB

3.2K

1-

0.4

0.6

0.8

1.0

1.2

1/T (1/K)

0.2 0.3 0.4 0.5 0.6 0.7

0(s)

10-7

10-6

10-5

10-4

10-3

10-2

KBthree single crystalsfrom different batches ICSM'96

109

105

107

TCO

T (K)

4 3 2.5 2 1.7 1.4

Dielectric Response

1- determines width of relaxation mode Below 5 K, five PF6 samples with

dominant impurity pinning show relaxation mode broader than the Debye (1- = 0.8). (1-) vs. 1/T behaviour for these PF6

samples exerts TCO , (denoted with red

arrow), as found in R vs.1/T curves above. Only for the one AsF6 single crystal (denoted as MN1) the critical slowing down, accompanied by broadening of the response, was observed below 1.7 K. (as can be seen on related ε"(ω) plot below on the right).

Δε is the magnitude of relaxation mode (and corresponds to static dielectric susceptibility ε(0)) Main feature of Δε vs. 1/T behaviour is sample dependency.

τ0 is central relaxation time in Littlewood ’s model which features a distribution function for relaxation times. τ0 vs. 1/T behaviour shows thermal activation in a manner similar to the single-particle conductivity, but TCO, (denoted with red arrow), found in resistivity curves is clearly identified for only one AsF6 sample (MN2). Still, the crossover in τ0 to an almost saturated behaviour below 2 K is observed for the two PF6 samples.

(TMTSF)2AsF6 (TMTSF)2PF6

(Hz)

101 102 103 104 105 106

"

2x105

3x105

4x105

6x105

8x105

2x106

105

106 2.3K

1.7K

2.5K3.1K

2.0K

MN2

'(106)

0 4 8 12

''(1

05)

0

2

4

6

8

10

MN2

2.3K

1.7K

2.5K

3.1K

2.0K

Col 20 vs Col 21 2.0K2.3K2.5K3.1K

Plot 6

Plot 7

(Hz)

101 102 103 104 105 106

"

105

106

107 1.5K

1.7K1.8K

1.6K

MN1

▲ MN2 ∆ MN1

file pygmalion/d:/tomislav/prije fr/eps''.JNB arjuna/d:/tomislav/eps''.JNBFig. 8. PF6 neven doktorat, open hexagons

(Hz)

101 102 103 104 105 106

''

107

108

109

4.2K3.0K2.4K1.9K1.5K

ICSM '96

file pygmalion/d:/tomislav/prije fr/eps''.JNB arjuna/d:/tomislav/eps''.JNBFig. 16.(uz Fig. 8.) PF6 neven doktorat, open hexagons

COLE COLE

0.0 0.1 0.2 0.3 0.4

'' (

109 )

0.00

0.04

0.08

0.12

4.2K3.0K

1.9K1.5K

' (109)

0 1 2 3 4

'' (

109 )

0.0

0.4

0.8

1.2

ICSM '96

◊ KB ICSM ’96

ε"(ω), dielectric loss function - it is imaginary part of dielectric function ε(ω); Plots for four samples at certain temperatures are shown above.

We performed simultaneous fit of real and imaginary parts of ε(ω), which is necessary to obtain reliable values of τ0, Δε and 1-as they are fit parameters here.

Cole-Cole plots (on the left, for one sample of each material) relate real ε’(ω) and imaginary ε"(ω) part of dielectric function, thus improving the credibility of fit procedure results.

References[1] (a) N. Biškup, Ph. D. Thesis, University of Zagreb, unpublished (1996); (b) S. Tomić, N.Biškup and A. Omerzu, Synth. Metals 85, 1597 (1997) and (c) N. Biškup, T. Vuletić, D. Herman, S. Tomić, M. Nagasawa and K. Bechgaard, Proceedings of ICSM'98.[2] T. Takahashi, Y. Maniwa, H. Kawamura, K. Murata and G. Saito, Synth. Metals 19, 225 (1987).[3] J. C. Lasjaunias, K. Biljaković, F. Nad', P. Monceau and K. Bechgaard, Phys. Rev. Lett.72, 1283 (1994).[4] W. G. Clark, M. E. Hanson, W. H. Wong and B. Alavi, J. de Physique IV, C2, 3, 235 (1993). [5] S. Tomić, J. R. Cooper, W. Kang, D. Jerome and K. Maki, J. Phys. I France 1. 1603 (1991).[6] B. Korin-Hamzić, M. Basletić and K. Maki, Europhys. Lett. 43, 450 (1998) and B. Korin-Hamzić et al., These Proceedings.[7] M. Nagasawa, T. Sambongi, K. Nomura and H. Anzai, Solid State Commun. 93, 33 (1995).[8] T. Takahashi, Y. Maniwa, H. Kawamura and G. Saito, J. Phys. Soc. Jpn. 55, 1364 (1986).

1/T (1/K)

0.0 0.2 0.4 0.6 0.8

T (K)

Rn

10 5 3 2.5 2 1.7 1.4

KB P

KB S

KB

TCO

106

1

108

104

102

(TMTSF)2PF6