Introductory Nanotechnology ~ Basic Condensed Matter Physics...
Transcript of Introductory Nanotechnology ~ Basic Condensed Matter Physics...
Department of Electronics
Introductory Nanotechnology ~ Basic Condensed Matter Physics ~
Atsufumi Hirohata
Quick Review over the Last Lecture
Quantum mechanics and classical dynamics
Quantum mechanics Classical dynamics
Schrödinger equation Equation of motion
: wave function A : amplitude
| |2 : probability A2 : energy
Quantum tunneling :
Optical absorption :
( transmittance ) + ( reflectance ) = 1
Abso
rption c
oef
fici
ent
Abso
rption c
oef
fici
ent
Valence band
Conduction band
Direct transition starts
Indrect transition starts
Direct transition starts
Conduction band
Valence band
Contents of Introductory Nanotechnology
First half of the course : Basic condensed matter physics
1. Why solids are solid ?
2. What is the most common atom on the earth ?
3. How does an electron travel in a material ?
4. How does lattices vibrate thermally ?
5. What is a semi-conductor ?
6. How does an electron tunnel through a barrier ?
Second half of the course : Introduction to nanotechnology (nano-fabrication / application)
7. Why does a magnet attract / retract ?
8. What happens at interfaces ?
Why Does a Magnet Attract / Retract ?
• Magnetic moment
• Magnetisation curve
• Origin of magnetism
• Curie temperature
• Types of magnets
• Magnetic anisotropy
• Magnetic domains
How Did We Find a Magnet ?
• 6th century BC, from Magnesia ( ) in Greece
Magnetite (Fe3O4)
• 3th century BC, from Handan area ( ) in China
found by shepherd Magnes ?
220 ~ 265 AD, first compass * http://www.wikipedia.org/
Magnet and Magnetism
Study on magnetism started by William Gilbert :
Magnete, Magneticisque Corporibus,
et de Magno Magnete Tellure (1600)
• The earth is a large magnet (compass).
• Fe looses magnetism by heating.
• N / S poles always appear in pair.
* http://www.wikipedia.org/
Development of Permanent Magnets
Permanent magnets are used in various applications :
* Corresponding pages on the web.
Only 4 elements are ferromagnetic at room temperature !
Which Elements are magnetic ?
In the periodic table,
Magnetic Moment
By dividing a magnet, N (+) and S (-) poles always appear :
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
• No magnetic monopole has been discovered !
• A pair of magnetic poles is the minimum unit : magnetic (dipole) moment.
m = m
m
Magnetisation :
• Vector sum of m per unit volume
Coulomb's Law
Force between two magnetic poles, m1 [Wb] and m
2, separated with r [m] is defined as
F =1
4 μ0
m1m2
r 2 [N]
Here, m2 receives magnetic force (= magnetic field) :
m1
m2 r
H
F = m2H H =1
4 μ0
m1
r 2 [N/Wb] = [A/m]
Magnetic flux density is proportional to the magnetic field :
Magnetic flux density at r from a magnetic pole m is B =m
4 r 2By comparing with H, B = μ0H
μ0 : magnetic permeability in a vacuum [H/m]
B
M
Under the presence of magnetisation, B = μ0 H +M( )
If the system is not in a vacuum, B = μH
By assuming ( : susceptibility), M = μ0( )Hμ = μ0 +
Magnetic Dipole Field and Magnetic Flux
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Magnetisation Curve
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
M
Initial permeability : μ i = (1 / μ0) (M / H)initial
Saturation magnetisation : MS
Residual magnetisation : Mr
Coercivity : Hc
Magnetic hysteresis
M
H
+ + +
- - - H
d
Demagnetising field : Hd = -NM
(N : demagnetising factor)
Magnetic Field Induced by an Electrical Current
Biot-Savart Law :
According to the right-handed screw rule,
dH = k1
r 2ids er( )
For an infinite straight wire, H =k
r 2i sin ds i
ds
dH
r (= rer)
a
By substituting = a / r and r = (a2 + s2) 1/2,
H = kia1
r 3ds = kia
1
a2 + s2( )3 2
ds dx
a2 + x 2( )3 2
=x
a2 a2 + x 2( )1 2
= kias
a2 a2 + s2( )1 2
=ki
a
s
a a2 + s2( )1 2
=2ki
a
By taking an integral along H,
Hdl =2ki
adl =
2ki
aad = 2ki d
0
2= 4 ki
Ampère’s law Hdl = i 4 k 1( ) H =i
2 a
Magnetic Field Induced by an Electrical Current (Cont'd)
For a circular current, using the right-handed screw rule,
H = dHx = dH cos
i ds
dH
r
a
By substituting the Biot-Savart law,
H =1
4
ids
r 2sin
2
cos =
i cos
4 r 2ds
=i cos
4 r 22 a =
ia cos
2r 2=ia2
2r 3
H
x
Magnetic Field Induced by a Magnetic Dipole
Potential at point P, which is separated r from the dipole :
=1
4 μ0
m
l1+
m
l2
=
m
4 μ0
l2 l1l1l2 r
Here,
l1 = r 2+
d
2
2
rd cos , l2 = r 2+
d
2
2
+ rd cos
l 1
-m
+m
l 2
P d
For r >> d, d 2 and higher terms can be neglected.
l1 r 1d
rcos r 1
d
2rcos
= r
d
2cos
l2 r +d
2cos
l2 l1 d cos
l1l2 r 2
Therefore, potential is calculated to be =
m
4 μ0
d cos
r 2=m r
4 μ0r3
H = =1
4 μ0
m
r 3r
=
1
4 μ0
m
r 33
r 4m r( )
rr
=
1
4 μ0r3m
3
r 2m r( )r
Magnetic field at P is
Magnetic Field Induced by a Magnetic Dipole (Cont'd)
In H, a component along m is
i ds
dH
r
a
H
x
H =1
4 μ0r3
m3
r 2mrr
=
m
2 μ0r3
Assuming m = μ0iA (A = a2),
H =μ0iA
2 μ0r3=ia2
2r 3
Same as H induced by a circular electrical current
Circular current i holds a magnetic moment of m = μ0iA.
Circular current i is equivalent to a magnetic moment.
Origin of Magnetism
Angular momentum L is defined with using momentum p :
L = r p
z component is calculated to be Lz = xpy ypx
0 p
r
L
In order to convert Lz into an operator, p
h
i q
Lz =h
ixy
yx
By changing into a polar coordinate system,
Lz =h
iSimilarly,
Lx =h
isin + cot cos
, Ly =
h
icos + cot sin
Therefore,
L2 = Lx2
+ Ly2
+ Lz2
= h2 1
sinsin
+
1
sin2
2
2
In quantum mechanics, observation of state = R is written as
L2 = h2 1
sinsin
+
ml2
sin2
R h
2K R( ) = l l +1( )h2
Origin of Magnetism (Cont'd)
Thus, the eigenvalue for L2 is
L2 = l l +1( )h2 L = l l +1( )h l = 1, 2, 3, K( )
azimuthal quantum number (defines the magnitude of L)
Lz = mlh ml = 0, ±1, ± 2, K( )
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Similarly, for Lz,
magnetic quantum number (defines the magnitude of Lz)
For a simple electron rotation,
L
Lz
Orientation of L : quantized
In addition, principal quantum number :
defines electron shells
n = 1 (K), 2 (L), 3 (M), ...
For l = 1, ml = 1, 0, -1
Orbital Moments
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Orbital motion of electron :
generates magnetic moment
m = μBL h
μB : Bohr magneton (1.165 10-29 Wb m)
Spin Moment and Magnetic Moment
Zeeman splitting :
For H atom, energy levels are split under H
dependent upon ml.
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Summation of angular momenta :
Russel-Saunders model J = L + S
Magnetic moment :
ml
2 1 0 -1 -2
1 0 -1
2
1
l
E = h
H = 0 H 0
Spin momentum :
L l ml = l, l 1, K, 0, K, l 2l +1( )
S s ms = s, s s =1
2
2
S = s s+1( )h =1
2
1
2+1
h
m = gμBJ h
g = 1 (J : orbital), 2 (J : spin)
z
S
msh =1
2h
M = Morb +Mspin = μB L+ 2S( ) h = gμBJ h
Exchange Energy and Magnetism
* K. Ota, Fundamental Magnetic Engineering I (Kyoritsu, Tokyo, 1973); ** http://www.wikipedia.org/ & http://www.bradley.edu/.
Exchange interaction between spins :
Eex = 2JexSiS jSi
Sj
Eex : minimum for parallel / antiparallel configurations
Jex : exchange integral
Atom separation [Å]
Exc
han
ge
inte
gra
l J e
x
antiferromagnetism
ferromagnetism
Dipole moment arrangement :
Paramagnetism
Antiferromagnetism
Ferromagnetism
Ferrimagnetism
Heisenberg
Ising
Paramagnetism
Applying a magnetic field H, potential energy of a magnetic moment with is
U = mH = mH cos m rotates to decrease U.
H
Assuming the numbers of moments with is n and energy increase with + d is + dU,
dn
n
1
TdU( ) lnn
U
T+ const. lnn =
U
kBT+ lnn0
n = n0 expU
kBT
Boltzmann distribution
Sum of the moments along z direction is between -J and +J
M = mzn = gμBMJ( )n0 expU
kBT
(MJ : z component of M)
Here, N = n = n0 expU
kBT
n0 = N exp
U
kBT
M =
N gμB MJ( ) expU
kBT
expU
kBT
= NgμB
b exp by( )
exp by( ) b MJ , y
gμBH
kBT
Paramagnetism (Cont'd)
Now, exp by( ) = e Jy + e J 1( )y+L+ eJy = e Jy + e Jyey +L+ eJy =
e Jy eJyey
1 ey
e y 2 exp by( ) =e y 2 e Jy eJyey( )e y 2 1 ey( )
=e Jye y 2 eJyey 2
e y 2 ey 2
Using sinha =1
2ea e a( ) eby =
eJ +1
2
y
eJ +1
2
y
ey 2 e y 2=
sinh J +1
2
y
sinhy
2Using
d
dyln eby( ) =
beby
eby
d
dyln eby( ) =
d
dyln
sinh J +1
2
y
sinhy
2
=d
dylnsinh J +
1
2
y lnsinh
y
2
=1
sinh J +1
2
y
cosh J +1
2
y J +
1
2
y
1
sinhy
2
coshy
2
1
2
= J +1
2
coth J +
1
2
y
1
2coth
y
2
=2J +1
2
coth
2J +1
2
a
1
2coth
a
2J a Jy( )
Paramagnetism (Cont'd)
Therefore,
M = NgμBJ2J +1
2J
coth
2J +1
2J
a
1
2Jcoth
a
2J
= NgμBJBJ a( ) a =
gμBJH
kBT
BJ (a) : Brillouin function
BJ a( ) = 11
Je a J
L 1 M = M 0 = NgμBJ
For a (H or T 0),
For J 0, M 0
For J (classical model),
2J +1
2J1
1
2Jcoth
a
2J=1
2Jcosh
a
2Jsinh
a
2J
1
2J1
a
2J
=1
a
B a( ) = coth a1
aL a( )
L (a) : Langevin function
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Ferromagnetism
Ferromagnetism
Weiss molecular field :
In paramagnetism theory, M = NgμBJBJ a( ), a =gμBJH
kBTH
m
Hm = wM (w : molecular field coefficient, M : magnetisation)
Substituting H with H + wM, and replacing a with x,
M = NgμBJBJ x( ), x =gμBJ
kBTH + wM( )
Spontaneous magnetisation at H = 0 is obtained as kBT = gμBJwM
Using M0 at T = 0,
M
M 0
= BJ x( )
M
M 0
=kBTx
Ng2μB2J 2w
For x << 1, BJ x( )J +1
3Jx
Assuming T = satisfies the above equations,
M
M 0
=J +1
3Jx =
kBNg2μB
2J 2wx
=Ng2μB
2J J +1( )w3kB
=Nm2
3kBw
* H. Ibach and H. Lüth, Solid-State Physics (Springer, Berlin, 2003).
(TC) : Curie temperature
Ferromagnetism Paramagnetism
Major Phases of Fe
Fe changes the crystalline structures with temperature / pressure :
-Fe (ferrite)
1184
-Fe (austenite)
1665
-Fe
-Fe
T [K]
p [hPa]
1808
Liquid-Fe
1043
( -Fe)
bcc
fcc
bcc
1
hcp
56 Fe :
Most stable atoms in the universe.
Phase change
Martensite Transformation :
’-Fe (martensite)
Ferromagnetism (Cont'd)
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997); ** http://www.wikipedia.org/.
For x << 1, BJ x( )J +1
3Jx
M = NgμBJBJ x( ) = NgμBJ J +1( )x
3
= Ng2μB2J J +1( )
1
3kBTH + wM( )
M = C1
TH + wM( ) C Ng2μB
2J J +1( ) 3kB = Nm2 3kB( )
M = CH
T CwTherefore, susceptibility is
=M
H=
C
T Cw=
C
T
(C : Curie constant)
Curie-Weiss law
Spin Density of States
* H. Ibach and H. Lüth, Solid-State Physics (Springer, Berlin, 2003).
Antiferromagnetism
By applying the Weiss field onto independent A and B sites (for x <<1),
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997); ** http://lab-neel.grenoble.cnrs.fr/.
A-site B-site
MA =1
2NgμBJBJ xA( ) =
Nm2
6kBTHA =
C
2THA
M B =1
2NgμBJBJ xA( ) =
Nm2
6kBTHB =
C
2THB
Therefore, total magnetisation is
M = MA + M B =C
2TH wMA w M B( ) + H w MA wMB( )[ ] =
C
2T2H + w + w ( )M[ ]
=M
H=
C
T +C
2w + w ( )
=C
T +
Néel temperature (TN)
Antiferromagnetism Paramagnetism
Magnetic Anisotropy
Magnetocrystalline anisotropy :
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Easy axis
Hard axis
Easy axis :
Magnetic anisotropy energy : minimum
Stable direction for spontaneous magnetisation
Hard axis :
Magnetic anisotropy energy : maximum
Unstable direction for spontaneous magnetisation
Magnetostriction
* http://www.gmmtech.co.jp/ ** http://www.ednjapan.com/
Electromagnet
Magnetostrictive material
Change
Flat panel speaker
Magnetic Domain Structures
Stable magnetic domain configuration is defined to minimize total energy :
U = Umag + Uex + Ua
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Umag : magnetostatic energy maximum when magnetic poles appear at the edge. minimum when no magnetic poles appear at the edge.
Uex : exchange energy maximum for antiparalell minimum for parallel
Ua : magnetic anisotropy energy maximum for hard axis minimum for easy axis
Magnetic Domain Walls
Bloch wall :
Néel wall :
Domain Wall Evolution with Film Thickness
* S. Chikazumi, Physics of Ferromagnetism (Oxford University Press, Oxford, 1997).
Magnetic domain walls change the configuration with film thickness :
Bloch wall
Néel wall
Cross-tie wall
Domain Wall Displacement in a M - H Curve
In a magnetisation process, domains are annihilated / nucleated by a field :
* J. Stoör and H. C. Siegmann, Magnetism (Springer, Berlin, 2006).
rotational domain motion
irreversible domain motion
Barkhausen jump