בירושלים העברית האוניברסיטהTHE HEBREW UNIVERSITY OF JERUSALEM The Racah Institute of Physics AC & DC PROPERTIES OF NIOBIUM IN A MAGNETIC FIELD by Gilad Masri A thesis submitted in partial fulfillment of the requirements for the degree of M.Sc. in Physics Advisors: Prof. Israel Felner , Dr. Menachem Tsindlecht The Hebrew University of Jerusalem January 2011
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
האוניברסיטה העברית בירושליםTHE HEBREW UNIVERSITY OF JERUSALEM
The Racah Institute of Physics
AC & DC PROPERTIES OF NIOBIUM IN A MAGNETIC FIELD
by
Gilad Masri
A thesis submitted in partial fulfillment of the requirements for the
degree of
M.Sc. in Physics
Advisors: Prof. Israel Felner , Dr. Menachem Tsindlecht
The Hebrew University of Jerusalem
January 2011
ABSTRACT
In Part I, the real and imaginary part of the magnetic susceptibility of a Niobium sample was
measured, in an AC magnetic field superimposed on a DC magnetic field, at various frequencies,
amplitudes and sweep rates. The results were compared to Fink’s model, which was found incorrect.
Key parameters of the susceptibility such as skin penetration depth and surface current were
calculated and their dependence on sweep rate was investigated.
In Part II, the magnetic field inside an electropolished and annealed Niobium tube was measured
when external DC and AC magnetic fields were applied. The dependence of the internal magnetic
field on temperature and sweep rate was investigated, as was the effect of the electropolishing and
annealing. The variation in time of internal magnetic field at constant external magnetic field was
measured and compared to previous experiments.
תקציר
בתדירויות , AC עם רכיב DCנמדד החלק הממשי והמדומה של הסוספטיביליות המגנטית של דגם מניוביום בשדה , בחלק הראשון
חושבו פרמטרים . ולא נמצאה התאמה, התוצאות הושוו למודל של פינק. באמפליטודות אקסיטציה שונות ובקצב סריקה שונה, שונות
. ונחקרה התלות שלהם בקצב הסריקה, סוספטיביליות המגנטית כגון עומק החדירה וזרם השפהחשובים של ה
צילינדר זה עבר . AC ו DCשלו רכיבי , נחקרה התלות של השדה המגנטי בתוך צילינדר חלול מניוביום בשדה החיצוני, בחלק השני
והשינוי בתלות זו , שדה המגנטי הפנימי בשדה המגנטי החיצוניונחקר האפקט של עיבוד זה על התלות של ה, חישולליטוש אלקטרוכימי ו
וזו הושוותה לניסויים , נחקרה התלות בזמן של השדה המגנטי הפנימי בשדה מגנטי חיצוני קבוע, כמו כן. עם השינוי בטמפרטורה
.קודמים
ii
TABLE OF CONTENTS
PART I.......................................................................................................................................................................................1 THEORETICAL BACKGROUND..........................................................................................................................................1
PART II ...................................................................................................................................................................................38 THEORETICAL BACKGROUND........................................................................................................................................38
SAMPLES AND EXPERIMENTAL METHODS ..................................................................................................................45 RESULTS ............................................................................................................................................................................48
RESULTS FOR UNPOLISHED SAMPLE ..........................................................................................................................48 RESULTS FOR POLISHED AND ANNEALED SAMPLE ...................................................................................................53
Referring to mh as the AC critical field amplitude, the induced magnetic field would be zero for 0 mh h< .
When 0 mh h> , a temporary induced magnetic field would emerge. They arrived at the theoretical
conclusion that the permeability should be related to the ratio 0/mh h through:
( ) ( ) ( )
( ) ( )
1
21
10
2 sin 2 sin 1 sin 11' ' n odd2 1 1
1 cos 1 1 cos 1sin 1'' '' 2sin1 1
n
mn
n nn n n
n n hn n n h
π θ θ θ θµ µ
π π
θ θθµ µ θπ π
−
− + + −= = − + −
− + − −= = − ≡ + −
In this model, the even harmonics are all zero, because of the cylindrical symmetry of the model. This
model is termed “Critical State Model”.
Earlier, in 1965 Maxwell & Robbins3, and in 1966 Schwartz & Maxwell4, found that the complex
susceptibility ' ''iχ χ χ= + is a function of the parameter
0hdH
dt
ωγ =
Here, the applied field has the following form:
( ) 00 0 cosdHH t H t h t
dtω= + +
Their results showed that γ is independent of the field ( )H t in the range 1 2C CH H H< < . Another
result was that 1
4χ
π= − , in this range of fields, but
only for 1γ < . For 1γ > , 'χ decreased while ''χ
increased (see picture on right). This continued until
3 E. Maxwell & W.P. Robbins, Phys. Lett. 19(8), 629-631 (1965)
4 B.B Schwartz & E. Maxwell, Phys. Lett. 22(1), 46-47 (1966)
14
a certain value of γ , at which both parts decreased. Theoretically, for any metal the susceptibility is
defined by:
( ) 4 1 4M H B H M Hχ π πχ= = + = +
For a superconductor 0B = , and therefore1χπ
= −4
. Comparing this to the experimental results,
Schwartz & Maxwell deduced that the imaginary part of the susceptibility ''χ is related to the energetic
losses due to normal currents in the superconductor which arise from the fast change in external
magnetic field. This may be seen as the metal’s inability to compensate for the external magnetic field.
Schwartz & Maxwell showed that the complex 1st harmonic susceptibility is given by:
( ) ( ) ( )
( ) ( )
1
3
2
1 320
1 41' '' 2
4 4
cos cos
H t i i
n n n n n
dMdH
i e d e d i
t
τ πτ τ
τ
πχ χ β τ τ β τ τ π
π γ π
τ ω β τ τ τ γ τ τ
+
+ = + + −
≡ ≡ − + −
∫ ∫
Here, the magnetic field cycle is divided as follows:
1. 10 t t≤ ≤ , where the magnetic field is increasing and the system follows the magnetization curve
( )M H .
2. 1 2t t t≤ ≤ , where the magnetic field is decreasing and the system follows a diamagnetic line with
slope 1
4dMdH π
= −
3. 2 3t t t≤ ≤ , where ( ) ( )1 3H t H t= and
0dMdH
> again.
15
4. 32t t πω
≤ ≤ , where the system continues along the magnetization curve ( )M H .
The diagram on the right, from the original article, shows this time division. It shows how the
magnetization ( )M t will behave (solid bold line) and how the induced field ( )B t will behave
according to this model.
AC Response of Type II superconductors: Fink’s Model
In 1967, H.J. Fink5 offered his own theoretical model, which applies for low frequencies. He found that
in addition to the parameter 1qγ
≡ , there is another parameter ( )0
0
Ch Hp
h≡ which defines the response
of the superconductor in a swept field. Here, Ch is a critical AC amplitude which depends on the swept
magnetic field 0H . He numerically calculated the first, second and third harmonics of the permeability as
function of p and q . Fink’s model assumes that relaxation effects can be ignored, meaning that 1ωτ ,
when τ is the relaxation time of the
superconductor.
In Fink’s model, there are 6 time-points, defined 1x
to 6x ( i ix tω= ). These points are shown on the
diagram on the right. In this model, B is the
5 H.J. Fink, Phys. Rev., 161(2), 417-422
16
average induced magnetic field, and Cb is the deviation from it during the cycle.
Fink was able to construct a function which describes ( )B t . It is similar to a sinus wave superimposed
on the increasing DC field. However, when the external field decreases, at point 2x , the surface sheath
will screen this change. It will continue to screen only if the deviation in the external field 0h is not too
large. If it is too large, and if 1q < , it may cause the induced field to decrease too. This is shown on the
upper-right side of the diagram, where the curve ( )B t changes from the points 1, 2, 3, 5A, 6 to 1, 2, 3,
4, 5, 6. The value of 0h which defines if the curve will change is given in the condition:
10 0 2 2 2 cos sin
2p p p x q x x qπ − = ≡ + − ≡
If 0p p> , than 0h is small enough and the screening will succeed (curve 1, 2, 5A, 6). On the contrary, if
0p p< the screening will not suffice (curve 1, 2, 3, 4, 5, 6). The point 2x is the point at which the total
external field is at its local maximum. The point 4x is the contrary one, where the external field is at its
local minimum. The point 3x is the point at which the screening currents in the sheath cease to screen
the external field. It is defined as:
( ) ( )3 4 2 CB x B x b− =
The function ( )B t was calculated using a numerical method and then decomposed to its Fourier
components. Assuming the following form for ( )B t :
( ) ( ) ( ) ( )0 0 0 01 1
cos sin ' cos '' sinn n n nn n
B t b a b n t a n t h c n t n t H tω ω µ ω µ ω µ∞ ∞
= =
= + + = + + =
∑ ∑
The results were plotted for the 1st, 2nd and 3rd harmonics (real and imaginary part), and are shown in the
original paper.
17
In this thesis, I will compare the results obtained for Niobium to the existing models.
SAMPLES AND EXPERIMENTAL METHODS
EXPERIMENTAL SETUP
The measurement system, as described in the diagram, is composed of a commercial MPMS SQUID
magnetometer, which has two superconducting coils capable of generating magnetic field. The DC coil
generates the DC field, and reaches a maximal field of 5 Tesla. There were two experimental setups
used:
1. In point-by-point setup, the magnetometer and the DC coil were controlled by the MPMS5
software.
2. In swept-field setup, the magnetic field was measured through a voltmeter connected in series
with a load resistor to the SQUID superconducting solenoid, and the DC coil was connected to a
power supply controlled by the Matlab R2008b software.
18
In DC measurements in point-by-point mode, the
sample was inserted to the sample chamber. There,
the pick-up coil inside the SQUID measured the
induced field inside the sample.
In AC measurements (point-by-point or swept field),
the samples were put in one of two pick-up coils
(made in the laboratory) that are connected in series
and have opposing screw-direction. These coils are
balanced so that the potential drop measured by
both of them, in the absence of a sample, is almost
zero. When a sample was inserted in one of these
coils and the driving AC coil was connected to a waveform generator (LFG), the magnetic flux variation
in this sample causes an induced voltage drop in the coils:
1 dc dt
ε Φ= −
This voltage drop was measured by three lock-in amplifiers (LCK), that were synchronized with the
waveform generator. One of them measured the first and third harmonics, and the other measured the
second harmonic. The waveform generator and the lock-in amplifiers, as well as the voltmeter
measuring the DC magnetic field and the power supply were controlled by the Matlab R2008b software.
The temperature in all measurements was 4.5ºK.
SAMPLE
The sample used in this part is a Niobium (Nb) single
crystal sample, with a measured Residual Resistivity Ratio
(RRR) of 200. This sample was mechanically polished, and
has the form of a parallelogram of sizes 1 mm by 2.4 mm
by 10 mm. The sample was put in a way that the AC and
10 mm
1 mm
2.4
H(t)
19
DC magnetic fields were parallel to its longest side, as shown in the picture.
MEASUREMENT TECHNIQUE
In order to get a broad picture of the behavior of the 1st, 2nd and 3rd harmonics (real and imaginary), the
waveform generator was set to give frequencies from 146.5Hz to 1465Hz, with excitation amplitudes
0.05Oe to 0.2Oe. The DC magnetic field was raised from slightly below zero to above HC3. The sweep-
rate of the DC field, dH/dt, was also varied from 0 Oe/sec (point-to-point measurements) to 30 Oe/sec.
RESULTS
The results will be divided to three parts:
1. The results for varying sweep rate at constant frequency and amplitude.
2. The results for varying frequency at constant sweep rate and amplitude.
3. The results for varying amplitude at constant sweep rate and frequency.
For each section, the two variables that remain constant were chosen at two different values. The results
shown here were normalized so that the real and imaginary parts of the susceptibility (first harmonic) at
zero DC field would be:
20
1 11' '' 0 at 0
4 DCHχ χπ
= − = =
Regarding the second and third harmonics, the absolute value of them will be plotted against the
magnetic DC field.
First, we show the magnetization graph for the sample:
0 1 2 3- 2 . 0
- 1 . 5
- 1 . 0
- 0 . 5
0 . 0
HC 1
Mag
netic
mom
ent (
emu)
M a g n e t i c f i e l d ( k O e )
Hc 2
M a g n e t ic m o m e n t V s . H D C , T = 4 . 5 K
Varying Sweep Rate
21
0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0
- 0 .0 8
- 0 .0 6
- 0 .0 4
- 0 .0 2
0 .0 0
0 .0 2
0 .0 4
χ 1' & χ
1''
H D C [O e ]
0 O e /s5 O e /s1 0 O e /s 2 0 O e /s 3 0 O e /s
χ1' & χ
1' ' V s . H D C fo r f = 1 4 6 .5 H z h A C = 0 .2 O e
H C 2H C 1
H C 3
0 1000 2000 3000 4000 5000 6000
0.000
0.002
0.004
0.006
0 Oe/sec 5 Oe/sec 10 Oe/sec 20 Oe/sec 30 Oe/sec
|χ2| Vs. HDC , hAC = 0.2 Oe, f = 146.5 Hz
|χ2 |
HDC
[Oe]0 1000 2000 3000 4000 5000 6000
0.0000
0.0005
0.0010
0.0015
|χ3| Vs. HDC , hac= 0.2 Oe, f = 146 Hz
0 Oe/sec 5 Oe/sec 10 Oe/sec 20 Oe/sec 30 Oe/sec
HDC [Oe]
|χ3|
22
0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0
-0 .0 8
-0 .0 6
-0 .0 4
-0 .0 2
0 .0 0
0 .0 2
0 .0 4
0 .0 6
H D C [O e ]
χ 1' & χ
1''
χ1' & χ
1' ' V s . H D C fo r f = 1 4 6 .5 H z h A C = 0 .0 2 O e
0 O e /s 5 O e /s 1 0 O e /s 2 0 O e /s 3 0 O e /s
0 1000 2000 3000 4000 5000 6000
0.000
0.001
0.002
0.003
|χ
2| Vs. H
DC , h
AC = 0.02 Oe, f = 146.5 Hz
0 Oe/s 5 Oe/ s 10 Oe/s 20 Oe/s 30 Oe/s
|χ2|
HDC [Oe]0 1000 2000 3000 4000 5000 6000
0.0000
0.0005
0.0010
0 Oe/s 5 Oe/s 10 Oe/s 20 Oe/s 30 Oe/s
|χ3| Vs. H
DC , f = 146.5 Hz h
AC = 0.02 Oe
|χ3|
HDC
[Oe]
23
0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0
-0 .0 8
-0 .0 7
-0 .0 6
-0 .0 5
-0 .0 4
-0 .0 3
-0 .0 2
-0 .0 1
0 .0 0
0 .0 1
0 .0 2
0 .0 3
0 .0 4
H D C [O e ]
χ 1' & χ
1''
0 O e /s5 O e /s1 0 O e /s 2 0 O e /s 30 O e /s
χ1' & χ
1' ' V s . H D C fo r f = 1 4 6 5 H z h A C = 0 .0 2 O e
0 1000 2000 3000 4000 5000 6000
0.0000
0.0002
0.0004
0.0006
0.0008
|χ2|
HDC [Oe]
0 Oe/s 5 Oe/s 10 Oe/s 20 Oe/s 30 Oe/s
|χ2| Vs. HDC , f=1465 Hz hAC=0.02 Oe
0 1000 2000 3000 4000 5000 6000
0.00000
0.00005
0.00010
0.00015
0.00020
0.00025
|χ3|
HDC [Oe]
0 Oe/s 5 Oe/s 10 Oe/s 20 Oe/s 30 Oe/s
|χ3| Vs. HDC , f=1465 Hz hAC=0.02 Oe
24
0 1000 2000 3000 4000 5000 6000
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0.00
0.01
0.02
0.03
0.04
H DC [O e]
χ 1' & χ
1''
χ1' & χ
1'' Vs. H DC for f = 1465 Hz hAC = 0.05 O e
0 Oe/s10 Oe/s 20 Oe/s30 Oe/s
0 1000 2000 3000 4000 5000 6000
0.0000
0.0005
0.0010
|χ2|
0 Oe/s 10 Oe/s 20 Oe/s 30 Oe/s
HDC [Oe]
|χ2| Vs. HDC , f = 1465 Hz hAC = 0.05 Oe
0 1000 2000 3000 4000 5000 6000
0.0000
0.0001
0.0002
0.0003
0.0004
HDC [Oe]
|χ3|
|χ3| Vs. HDC , f = 1465 Hz hAC = 0.05 Oe
0 Oe/s 10 Oe/s 20 Oe/s 30 Oe/s
25
Varying Frequency
0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0
-0 .0 8
-0 .0 6
-0 .0 4
-0 .0 2
0 .0 0
0 .0 2
0 .0 4
H D C [O e ]
χ 1' & χ
1''
χ1' & χ
1' ' V s . H D C , S R = 1 0 O e /s h A C = 0 .2 O e
1 4 6 .5 H z 2 9 3 H z 5 8 6 H z 8 7 9 H z 1 1 7 2 H z 1 4 6 5 H z
0 1000 2000 3000 4000 5000 6000
0
1
2
3
4
5
6
|χ2| Vs. H
DC , SR = 10 Oe/s h
AC=0.2 O e
146.5 Hz 293 Hz586 Hz 879 Hz1172 Hz 1465 Hz
|χ2| [
mkV
olt]
HDC [Oe]0 1000 2000 3000 4000 5000 6000
0
2
4
6
HDC [Oe]
|χ3| [
mkV
olt]
|χ3| Vs. HDC , SR = 10 Oe/s hAC=0.2 Oe
146.5 Hz 293 Hz586 Hz 879 Hz1172 Hz 1465 Hz
26
0 1000 2000 3000 4000 5000 6000
-0.08
-0.06
-0.04
-0.02
0.00
0.02
0.04
0.06
0.08
H DC [Oe]
χ 1' & χ
1''
χ1' & χ
1'' V s. H DC , SR = 30 O e/s hAC=0.1 O e
146.5 H z 293 H z 586 H z 879 H z 1172 H z 1465 H z
0 1000 2000 3000 4000 5000 6000
0
1
2
3
4
146.5 Hz 293 Hz 586 Hz 879 Hz 1172 Hz 1465 Hz
HDC [Oe]
|χ2| [
mkV
olt]
|χ2| Vs. HDC , SR = 30 Oe/s hAC=0.1 Oe
0 1000 2000 3000 4000 5000 6000
0
1
2
3
HDC [Oe]
|χ3| [
mkV
olt]
146.5 Hz 293 Hz 586 Hz 879 Hz 1172 Hz 1465 Hz
|χ3| Vs. H
DC , SR = 30 Oe/s h
AC=0.1 Oe
27
Varying Amplitude
0 1000 2000 3000 4000 5000 6000
-0.08
-0.06
-0.04
-0.02
0.00
0.02
0.04
χ 1' & χ
1''
χ1' & χ
1'' V s . H D C , S R = 10 O e/s f = 146 .5 H z
H DC [O e]
0 .05 O e 0 .1 O e 0 .15 O e0 .2 O e
0 1000 2000 3000 4000 5000 6000
0
1
2
3
4
5
H D C [O e]
|χ2| [
mkV
olt]
|χ2| Vs. H DC , SR = 10 O e/s f = 146.5 Hz
0.05 Oe0.1 O e 0 .15 Oe0.2 O e
0 1000 2000 3000 4000 5000 6000
0.0
0.5
1.0
1.5
2.0 0.05 Oe0.1 Oe 0.15 Oe0.2 Oe
HDC [Oe]
|χ3| [
mkV
olt]
|χ3| Vs. HDC , SR = 10 Oe/s f = 146.5 Hz
28
0 1000 2000 3000 4000 5000 6000
-0 .08
-0 .07
-0 .06
-0 .05
-0 .04
-0 .03
-0 .02
-0 .01
0 .00
0 .01
0 .02
0 .03
0 .04
0 .0 5 O e 0 .1 O e 0 .1 5 O e0 .2 O e
χ1' & χ 1'' V s . H D C , S R = 3 0 O e/s f = 8 79 H z
χ 1' & χ
1''
H D C [O e]
0 1000 2000 3000 4000 5000 6000
0
2
4
6
8
HDC [Oe]
|χ2| [
mkV
olt]
0.05 Oe0.1 Oe 0.15 Oe0.2 Oe
|χ2| Vs. H
DC , SR = 30 Oe/s f = 879 Hz
0 1000 2000 3000 4000 5000 6000
0
1
2
3
4
5
|χ3| [
mkV
olt]
H DC [Oe]
0.05 Oe0.1 Oe 0.15 Oe0.2 Oe
|χ3| Vs. HDC , SR = 30 Oe/s f = 879 Hz
29
DISCUSSION
Determination of the Critical Fields
From the DC magnetization curve and from the swept field measurements we can determine HC1 and HC2
:
1
2
1000 50 Oe2600 50 Oe
C
C
HH
= ±= ±
From the swept-field graphs we can determine HC3 :
3 4850 20 OeCH = ±
The swept-field graphs coincide with the magnetization graph, and show a rise at HC1 and a peak at HC2 .
From these critical field we can calculate the parameter κ , through the formula6 for the magnetization at
a field close to HC2:
( )2 0
24
1.16 2 1CH HMπ
κ−− =
−
From a fit to the magnetization curve close to HC2 , the slope was extracted and κ was calculated to be:
1.6 0.1κ = ±
This value is in accordance with the well-known value from literature. Another parameter that can be
calculated is the ratio 3 2/C CH H :
3
2
1.87 0.04C
C
HH
= ±
6 See Schmidt V.V., “The Physics of Superconductors: Introduction to Fundamentals and Applications” , Springer (1997), pp 110 - 113
30
This ratio is close to the theoretical value, which is 1.695.
The effect of swept DC field on measurements7
From the results for different sweep-rates, we can see that 1’’ exhibits a minimum at HC2, which means
that there is a change in the underlying mechanism that is responsible for dissipation in the vortex state
and in the surface superconducting state. This minimum is also apparent at 1’ and at the higher
harmonics. We notice that as the sweep rate rises, the signal from the first and second harmonics rises
also. However, the signal from the third harmonic decreases with increasing sweep rate.
To better see the effect that swept-field has on the susceptibility, the susceptibility at zero sweep rate
(point to point measurements) was subtracted from the susceptibility at other sweep-rates:
( ) ( )( ) ( )
1 1 1
1 1 1
' ' 0 ' 0
'' '' 0 '' 0
SR SR
SR SR
χ χ χχ χ χ
∆ = ≠ − =
∆ = ≠ − =
0 1 2 3 4 5
0 . 0 0
0 . 0 1
0 . 0 2
0 . 0 3
0 . 0 4
0 . 0 5
H D C [ k O e ]
∆ χ ' ; 5 O e / s ∆ χ ' ; 3 0 O e / s ∆ χ ' ' ; 5 O e / s ∆ χ ' ' ; 3 0 O e / s
H D C [ k O e ]
1 4 6 H z
∆ χ'
& ∆
χ''
0 1 2 3 4 5
0 . 0 0 0
0 . 0 0 5
0 . 0 1 0
8 7 9 H z
∆ χ1' & ∆ χ 1 ' ' V s . H D C , f = 1 4 6 . 5 H z & 8 7 9 H z , h A C = 0 . 1 O e
7 The results shown here and in the following pages are after the results and discussion in the article: Tsindlekht M.I., Genkin V.M, Leviev
G.I., Schlussel Y., Masri G., Tulin V.A., Berezin V.A., “Measurement of the AC conductivity of a Nb single crystal in a swept magnetic field”. (Submitted to Europhysics Letters)
31
These graphs show that as the frequency increases, the difference between point-to-point and swept-
field measurements decreases, both in 1’ as in 1’’. Near HC3 there is a region where the susceptibility at
swept-field is lower than in point-to-point measurements. This difference increases at increasing
frequency, and this is seen in the results. The range of magnetic field in which this happens is smaller
for higher amplitude. . We also note that there is a certain DC field above which this difference
decreases as the sweep rate increases. This phenomenon might be related to the sheath state
Comparison with Fink’s model
According to Fink’s model, the AC response of the superconducting sheath state in a swept field is a
function of only two parameters, p and q. To compare the results to this model, we compared data for
two different sets of frequencies, amplitudes and sweep-rates. We calculated the surface current, derived
in the following way.
Assuming that the thickness of the surface layer is small in comparison to the sample sizes, we can
separate the measured susceptibility to two parts: the susceptibility for 3DC cH H> is • , and the
susceptibility for 3DC cH H< . To find the surface current, we can write
( ) ( )0 00
41 4 1 4 14
SS
Jh h J
c hχ χππχ πχ
χ π
∞∞
∞
− + = + − ⇒ = +
For the first set, we took sweep-rate 5 Oe/s,
and frequency 146.5 Hz. For the second set,
we took sweep-rate 30 Oe/s, and frequency
879 Hz. The AC amplitude was 0.1 Oe for
bot sets.
The parameter p is equal in both sets since
the critical AC amplitude hC is independent
of the sweep rate and frequency and hAC is
the same. The parameter q is also the same,
since the product of the sweep rate and the
2.0 2.5 3.0 3.5 4.0 4.5 5.0-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
J' s/h
0
J'' s/h
0
HDC (kOe)
146.5 Hz; 5 Oe/s 879 Hz; 30 Oe/s
T = 4.5 Kh0 = 0.1 Oe
JS'/h
0 & J
S''/h
0 Vs. H
DC , T=4.5 K h
AC=0.1 Oe
32
frequency is the same. The result of this comparison is shown on the right. As can be seen, there is a
discrepancy between the two sets, and therefore Fink’s model is incorrect.
Calculation of the skin penetration depth
To find the skin penetration depth we must turn our attention to Maxwell’s equations (neglecting the
displacement current in Ampère’s law) combined with the definition of complex electric conductivity:
( )( )2
1 22
1 2
4 14
0
HH J E HH ic c tc t
J i E H
ππ σ σ
σ σ
∂∇× = ∇× = − ∂ ∇ = − +∂ ∂= + ∇ =
uurur uur ur ur ur uur
uur
ur ur ur uur
Assuming the magnetic field is in the z direction and oscillates in space and time, we arrive at the
following equation
( ) ( )2 2
1 2, , , Z
effi t Z ZZ Z eff
ih hH x y t h x y e h ix y c
ω π σ ωσ σ σ− 4∂ ∂= ⇒ + = ≡ +
∂ ∂
Here we have neglected the dependence of HZ on the z coordinate, since we assume that the sample
size in this direction is much larger than in the x and y directions. Hence, there is a small
demagnetization factor, which we will neglect. If we define the parameters and as:
2 1
2 2
c cλ δπσ ω πσ ω
≡ ≡
we arrive at the equation:
2 2
2 2 2 2
1 2 0Z ZZ
h h i hx y λ δ
∂ ∂ + − − = ∂ ∂
33
Taking the boundary conditions ( ) ( )0 0, ,Z x Z yh L y h h x L hα α± = ± = , when 2LX , 2LY are the sample’s x
and y lengths, we arrive at the following solution for the sample’s susceptibility:
( )
( )( )
22 2
1 2 21,3... 1 2 2
2 2 2
2 22 2 2
1,3...
tanh8 21 1 2
4tanh8
2
m ym
xm m y
m xm
m m x y
mk L k kZ LZ Zm k L ikq L mZ q k
m q L L
παπ
χπ λ δπ
π
=
=
≡ + =
+ − = ≡ − = ≡ +
∑
∑
The complex parameter takes into account the possible existence of the surface layer with
enhanced conductivity. This layer could provide the jump of an AC field at the sample surface. For
example, in the model with thin surface layer of thickness d and conductivity s effσ σ , one could
find ( )1/ 1 Uα = + , where
2 2' '' / 4 /s s sU U iU k d k k i cπ ωσ= + = = −
For the measurements where the sweep rate is zero (point-to-point), there is complete shielding for
0 2cH H< and we could conclude that the penetration depth is very small. In the swept field, we
observe an AC response, which differs from complete shielding, and it permits us to draw some
conclusion about the conductivity of the sample. We found that it is impossible to map adequately
the experimental values for in the swept field onto eff using the solution for :
( )1 2 1 / 4Z Zχ α α π= + −
In general, the two unknown complex quantities eff and could not be found using only one
complex quantity without any additional conditions. Hence, we assumed that the total surface
current has the smallest possible value because the current is localized in a thin surface layer. This
condition with other possible physical restrictions, such as 1 ' 0 '' 0U Uα ≤ ≥ ≤ permit us to find
the bulk eff and . From eff we calculated the skin penetration depth .
34
The results of these calculations allowed us to plot against HDC for different frequencies and
amplitudes:
1 . 0 1 . 5 2 . 0 2 . 5
1 0 - 2
1 0 - 1
M a g n e t i c f i e l d ( k O e )
δ eff (m
m)
5 O e / s e c 1 0 O e / s e c 2 0 O e / s e c 3 0 O e / s e c
h 0 = 0 . 0 2 O e
M a g n e t i c f i e l d ( k O e )
δ V s . H D C , f = 1 4 6 . 5 H z
1 . 0 1 . 5 2 . 0 2 . 5
1 0 - 3
1 0 - 2
1 0 - 1
h 0 = 0 . 2 O e
1 . 0 1 . 5 2 . 0 2 . 5
1 0 - 3
1 0 - 2
1 0 - 1
f = 1 4 6 5 H z
f = 1 4 6 H z δ eff (m
m)
M a g n e t i c f i e l d ( k O e )
δ V s . H D C , h A C = 0 . 2 O e
1 . 0 1 . 5 2 . 0 2 . 5
1 0 - 3
1 0 - 2
M a g n e t i c f i e l d ( k O e )
5 O e / s e c 1 0 O e / s e c 2 0 O e / s e c 3 0 O e / s e c
We can see from these graphs that in the plateau region between HC1 and HC2 the dissipative
component of the bulk conductivity 1 is considerably larger than 2 , i.e.δ λ . We can also see an
35
exponential rise in the range between 1000 Oe to 1500 Oe. This was confirmed by fitting an
exponential to this range, as shown below.
900 1000 1100 1200 1300
0.000
0.002
0.004
0.006
0.008
0.010
E qu atio n: y = A 1*exp(x/t1) + y0 C hi^2/D o F = 1.68 94E -8R ^2 = 0.99 767 y0 0 .000 27 ±0.0 000 6A 1 6 .138 6E -9 ±2.5 647 E-9t1 9 2.89 181 ±2.7 456 4
HDC
[O e ]
δ [m
m]
δ V s. H D C , f = 1465 H z h AC = 0 .2 O e S R = 30 O e/s
900 1000 1100 1200 1300
0 .00
0 .01
0 .02
0 .03
0 .04
0 .05
HDC
[Oe]
δ [m
m]
δ Vs. H DC , f = 1465 Hz hAC = 0.2 Oe SR = 10 Oe/s