Galvanomagnetic Effects Sensors Based On Hall Effect

Jestr Journal of Engineering Science and Technology Review 2 (2009) 1-7

Lecture Note
JOURNAL OF
Engineering Science and
Technology Review

www.jestr.org

Galvanomagnetic Effects ¨Sensors based on Hall Effect¨
Vlassis N. Petousis* , Panagiotis D. Dimitropoulos

Department of Computer & Communication Engineering, University of Thessaly, Volos, Hellas
Received 18 April 2008; Accepted 16 February 2009

___________________________________________________________________________________________

Abstract

The Hall effect is the generation of a transverse electromotive force in a sample carrying an electric current and exposed
to perpendicular magnetic field. Depending on the sample geometry, this electromotive force may cause the appearance
of a transverse voltage across the sample, or a current deflection in the sample. The generation of this transverse voltage,
called Hall voltage, is the generally known way for the of the appearance of the Hall effect.
The resistance of this sample increasing under influence of the magnetic field, this called magnetoresistance
effect. Both the Hall effect and the magnetoresistance effect belong to the more general class of phenomena called
galvanomagnetic effects. Galvamomagnetic effects are the manifestations of charge transport phenomena in condensed
matter in the presence of a magnetic field.
The sensor applications of Hall effect became important only with the development of semiconductor
technology. For one thing, the Hall effect is only strong enough for this propose in some semiconductors. Therefore, the
first Hall effect magnetic sensor became commercially available in the mid 1950s, a few year after the discovery of high-
mobility compound semiconductors. Our goal in this paper is to understand the physically background of the Hall and
the magnetoresistance effects. We are going to discuss the effect of parameters in those phenomena and how we can
make better our technology to improve better efficiency.

Keywords: Hall Effect , Galvanomagnetic Effect, Hall Sensors, Wheel Hall Sensor, MR Effects.
___________________________________________________________________________________________

1.1 Approximate analysis of Hall effect In our analysis we use Smooth-drift approximation, this
approximation consist of assumption that charge carriers
In this section we will see the Hall effect. That give us an move uniformly as a result of an electric field or other
idea for the Hall effect and for magnetoconcetration effect driving forces, the velocity of the movement being the same
and help us to understand much more about solid state for all carriers and equal to the appropriate drift velocity. So
physics. we can replace the carrier velocity of each individual carrier,
To simplify the analysis, we also assume equilibrium in υ (Equation (1.1)), by the average drift velocity of all
carrier concentrations in samples, and thus neglect diffusion carriers. We neglect completely the thermal motion of the
currents. carriers and the energy dissipation effect of scattering by a
All galvanomagnetic effects come about as a smooth friction.
manifestation of the action of the Lorentz force on quasi-
free charge carriers in condensed matter. The Lorentz force
is the force acting on a charged particle in a electromagnetic 1.2 The (original) Hall effect
field. Moreover, the existence of the Lorentz force is
fundamental indication of a very presence of an electric and We shall first study the Hall effect in the form in which
/ or magnetic fields. It is give by Hall[1] discovered. Hall’s experimental device was a long
gold leaf. We shall also consider a long flat sample; but in
order to quickly come modern applications of Hall effect, we
F = eE + e(υ × B) (1.1)
shall study it in semiconductor samples.
Let us consider the transport of carriers in a long and
Here e denotes the particle charge (for electrons e=-q thin semiconductor strips, such as those shown in figures
and for holes e=q where q is he magnitude of an electron 1.1a and 1.1b.(Long strip means that the length of a strip l is
charge) E is the electric field, υ is the carrier velocity and B much larger than w).
the magnetic induction. The first term on the right hand side Suppose that in the strips, along the x direction, an
of (1.1) is often referred to as electrostatic force, and the external electric field Εe(Ex,0,0) is established, and the
second term as the Lorentz force. magnetic field is zero. Then in the Lorentz force (Equation
(1.1)) only the first term, witch is the electrical force exists.
______________
* E-mail address: petousis@inf.uth.gr As we know the electrical force causes the drift of charge
ISSN: 1791-2377 © 2008 Kavala Institute of Technology. All rights reserved. carriers. In the present case, charge carriers drift along the

V. N. Petousis , P. D. Dimitropoulos / Journal of Engineering Science and Technology Review 2 (2009) 1 -7

strip in two opposite directions. The drift velocities are given These forces push carriers towards the upper edges of the
by: strips. Consequently the carrier concentration at these upper
edges of the strips start to increase, while the carrier
concentration at the lower edges start to decrease.
υ dp = μ p Ee , υ dn = μ n Ee (1.2)

where μp and μn are the mobilities of holes and electrons, 1.4 Hall voltage
respectively. The associated current densities are given by :
A more tangible effect with associated with Hall field is the
J p = qμ p pEe , J n = qμ n nE e (1.3) appearance of transverse voltage between the edges of a
strip. This voltage is known as Hall voltage. Let us choose
two points M and N at the opposite edges of the strip, under
where p and n denote the hole and the electron density in p- the condition that both points lie in the same equipotential
type and n-type strip, respectively. plane when B=0(figure 1.1). Then the Hall voltage is given
by

1.3 Hall electric field N

Let us now expose the current – carrying strip to a magnetic
VH =
M
∫E H dz (1.9)
field. Let the magnetic induction B be collinear with the y-
axis(figure 1.1). Now on each charge carrier in the samples
and
both parts of Lorentz force (Equation((1.1)) act. Since we
assumed a uniform directional velocity of all carriers in the
sample, the magnetic forces acting on a carriers are give by: VHp = μ p Ex By w , VHn = μn Ex By w (1.10)

Fp = e[υ dp × B] , Fp = e[υ dp × B] (1.4) for our p-type and n-type strips, respectively. Here the
negative sign has been omitted and w denotes the width of
the strips. The generation of the transverse Hall electric field
and the associated Hall Voltage under the experimental
conditions similar to that shown in figure 1.1 are usually
referred to as the Hall effect. However, there is another,
more fundamental feature of the Hall effect: the Hall angle.

1.5 Hall angle

In the presence of a magnetic field, the total electric field in
the sample E=Ee+EH is not collinear with the external
electric field Ee. In the present case the current in the sample
is confined to the direction of the external electric field, and
the current density is collinear with the external electric field
(figure 1.1). Hence the total electric field is not collinear
with the current density either. Therefore, the Hall effect in a
long sample shows up through the tilting of the total electric
field relative to the external electric field and the current
density in the sample. With the aid of figure 1.2, the
Fig. 1.1: The Hall effect in a long plate of p-type (p) and n-type (n)
magnitude of the Hall angle is given by
material. Ee is the external electric field, B the magnetic induction, υd he ΕΗ
drift velocity of carriers, F the magnetic force, J the current density and tan ΘΗ = (1.11)
EH the Hall electric field. The magnetic forces press both positive and Εe
negative carriers towards the upper boundary of a strip. The Hall
voltage appears between the charged edges of the strip.

in the samples p and n respectively. These forces have the
same direction in both strips : since e=q and e =-q for hols
and electrons respectively, from equations(1.2) and (1.4) it
follows that:

Fp = qμ p [ Ee × B] , Fn = qμ n [Ee × B ] (1.5)

For our particular geometry the magnetic forces are
collinear with the z-axis:
Fig. 1.2 The vector diagrams of electric fields and current densities in
Fp = (0,0, qμ p E x B y ) , Fn = (0,0, qμ n E x B y ) (1.6)
long samples: (p), in a p-type material; (n), in n-type material J is the
current density, Ee the external electric field, EH the Hall electric field,
E the total electric field and ΘH the Hall angle.

2


induction perpendicular to the device plane. We omit the
For the reason which will become clear later, it is negative sign.
convenient to measure the Hall angle with respect to the Equation (1.16) shows why a plate is a preferential shape
direction of the total electric field, as shown in figure 1.2. of the Hall device with voltage output : for a given biasing
Then we may define the Hall angle as the angle of current and magnetic induction, the thinner the sample, the
inclination of the current density J with respect to the total higher the resulting Hall voltage.
electric field E. According to this definition and (1.11), we
obtain for the two particular cases from figures 1.1 and 1.2
1.7 The current deflaction effect
tan ΘΗp = μ p By , tan ΘΗn = − μn By (1.12)
Consider now the Hall effect in a very short Hall plate. A
sample for a Hall effect experiment is called short if its
For p-type and n-type semiconductor strip, respectively. dimension in the current direction is much smaller than that
The value of the Hall angle depends only on the applied in the direction of the magnetic force acting on carriers. As
magnetic induction and the mobility of the charge carriers. example of a short sample is the short strip shown in figure
The sign of the Hall angle coincides with the sign of the 1.3. the basic shape of this device, its position relative to the
charge carriers. coordinate system, and the notation are the same as in case
of the long strips in figure 1.3. Now, however, the strip is
short ,l<<w, and it is laterally sandwiched between two large
1.6 Hall coefficient current contacts. In such short samples the Hall effect takes a
form which is sometimes called current deflection effect.
From equations (1.3) and (1.8) we can find the relationship
between the current density and the Hall field:

1 1
EHp = − [ J × B] , EHn = [ J × B] (1.13)
qp qn

The equations can be rewritten as

EH = − RH [ J × B ] (1.14)

where RH is a parameter called the Hall coefficient. Equation
(1.14) is a modern representation of the conclusions made by
Hall after his experimental findings.
The Hall coefficient is a material parameter that
characterizes the intensity and the sign of the Hall effect in a
particular material. The unit of the Hall coefficient is VmA- Fig. 1.3 A short Hall plate. The distance between the current contacts
1 -1
T , which is sometimes expressed in a more compact form is smaller than the distance between the insulating boundaries (ib). (The
as ΩmT-1 or, equivalently, m-3C-1. picture was taken from P.S Popović “ Hall Effect Devices (Sensor
From a comparison of equations (1.13) and (1.14), we Series) Swiss Federal Institute of Technology Lausanne (E.P.F.L) 2nd
can find the Hall coefficients of strongly extrinsic Edition IoP 2004)
semiconductors
Let us now find a quantitative relationship between the
electric field and the current density in a short sample. We
1 1
RHp = , RHn = (1.15) shall again use the approximation which we introduced
qp qn above: we neglect the thermal motion of the carriers, and
assume that carriers move uniformly under the action of the
for p-type and n-type, respectively. The sign of the Hall Lorentz force.
coefficient coincides with the sign of the majority carriers,
and the magnitude of the Hall coefficient is inversely Fp = qE + q[υ p × B] , Fn = − qE − q[υn × B ] (1.17)
proportional to the majority carriers concentration. In the
practical applications of the Hall effect it is convenient to
for holes and electrons, respectively. The densities of the
operate with pure macroscopic and integral quantities
carriers are:
characterizing a Hall device. To this end, for a long Hall
device we obtain from equations (1.11) and (3.14) the Hall
voltage : J p ( B) = J p (0) + μ p [ J p ( B) × B]

IB⊥ J p ( B) = J p (0) + μ p [ J p ( B) × B] (1.18)
VH = RH (1.16)
t
Here Jp(B) and Jn(B) are the hole and electron current
densities respectively in the presence of the magnetic
Here t is the thickness of the strip, I is the device current
induction B.
given by I = Jwt and B⊥ the component of the magnetic

3


J p ( B) = qμ p pE p = μ p pFp = qpυ p

J n ( B ) = q μ n nEn = μ n nFn = − qnυn (1.19)

And Jp(0) and Jn(0) are the drift current densities due to
the electric field E when B=0:

J p (0) = qμ p pE
J n (0) = q μn nE (1.20)

Equations (1.17),(1.18) and (1.20) are graphically
represented in figure 1.4. In a short sample, a magnetic field Fig. 1.5 The current lines deflection effect in a short Hall plate, biased
deflects the current from its usual way along the electrical by a constant voltage (a) and a constant current (b). The broken lines are
field. Moreover using the diagrams in figure 1.4, we find the current lines at B=0, whereas the full lines are the current lines at
B ≠ 0.
tan Θ Hp = μ p B⊥ , tan Θ Hn = − μn B⊥ (1.21)
J p ( B) = σ pB E + σ pB μ p [ E × B ]
J n ( B) = σ nB E − σ nB μ n [ E × B ] (1.22)

where

σ p0 σ n0
σ pB = and σ nB = (1.23)
[1 + ( μ p B ) 2 ] [1 + ( μn B ) 2 ]

are the effective conductivities of p-type and n-type
materials in the presence of the magnetic induction B and,
σp0 and σn0 are the conductivities at B=0. Note that, at a given
electric field, the coefficients σΒ determine the longitudinal
current density in a infinitely short sample. Since a
materialization of the infinitely short sample is Corbino[2]
disc, we shall call the conductivity the coefficients σΒ the
Corbino conductivity.
Fig. 1.4 The graphical representation of the vector equations
(1.17),(1.18) and (1.20) that hold in a short Hall plates: (p) of a p-type The first term on the right-hand side of equation (1.22)
semiconductor ; (n) of a n-type semiconductor.(The picture was taken represent the lateral current densities. The integral of this
from P.S Popović “ Hall Effect Devices (Sensor Series) Swiss Federal current density over the contact surface is the device current.
Institute of Technology Lausanne (E.P.F.L) 2nd Edition IoP 2004) From equations (1.22) and (1.23) we deduce that the device
current decreases as the magnetic field, increases. Therefore,
if a sample exposed in magnetic field, everything happens as
2. The magnetoresistance effect if the conductivity of the sample material were decreased.
The corresponding relative increase in the effective material
By inspecting the diagrams in figure 1.4, we notice that the resistivity is given by
current density vector J(B) are smaller than J(0). This means
that in short Hall Plates a magnetic field also has additional
ρ pB − ρ p 0
effect: it causes a reduction in current density. = ( μ p B)2
This current deflection is illustrated in figure 1.5. As ρ p0
sown in figure 1.5(a), current density in presence of a
magnetic field is smaller than that with no magnetic field. ρ nB − ρ n 0
The attenuation in the current density is a consequence of
= ( μn B)2 (1.24)
ρn0
the current deflection effect: the deflected current lines
between the sample contacts are longer (see figure 1.5(a)).
The longer current path means also greater effective The increase in a material resistivity under the influence
resistance of the sample. of a magnetic field is called the magnetoresistance effect. In
To find how much the current density decrease due to a particular, the just described increase in resistivity due to the
magnetic induction, we have to solve the equation (1.18) current deflection effect in a short samples is called the
with respect to J(B). We can solve this equation and for geometrical magnetoresistance effect. The attribute “
geometrical “ reflects the fact that the effect is related to the
perpendicular magnetic field, such that E⋅B = 0, the geometry of the current lines, as illustrated in figure 1.5.
solutions of the equations (1.18) are

4


2.1 Galvanomagnetic current components 3. Geometrical factors

Up to now, in order to simplify our analysis, we have always Let us have a look at the distribution of the current density
assumed that our sample is plate-like, and that the magnetic and the electrical potential in a “realistic“ galvanomagnetic
field was perpendicular to the plate. Let us drop now this sample. Figure 1.7, shows such a two terminal rectangular
assumption and consider a general case. We only assume plate, which is somewhere between “ very long “ and “ very
that the vectors of electric field and magnetic induction have short “. By inspecting the current and the equipotential lines,
an angle and we seek the current density vector. we notice the following. Near the two isolating boundaries,
The basic equations relating the resulting current density current is forced to flow along the boundaries and the
J(E,B) with the electric field E and the magnetic induction B equipotential lines are strongly inclined: a Hall electric field
are given by is generated.
The device such us assumed have intermediate
geometry. Then a factor to related from geometry can be
J ( B ) = σ E + μ[ J ( B ) × B ] (1.25)
calculated. It is really that the Hall voltage of the Hall plate
with an arbitrary shape can de expressed as
where J ( E , B ) ≡ J ( B ) and σ is the material conductivity
at B=0 and μ is the mobility of the carriers. The total VH
galvanomagnetic current density (figure 1.6) generally
GH = (1.28)
VH ∞
consists of three components :
Where GH is a parameter called the geometrical
J = J1 + J 2 + J 3 (1.25a)
correction factor of a Hall voltage, and VH ∞ denotes the
Hall voltage in a corresponding long strip.
The Hall geometrical factor is a number limited by 0<
GH<1. For a long Hall device GH=1; for the very short Hall
device GH=0. Similarly, for the resistance of the device
exposed to a magnetic field we may write:

R( B)
GR = (1.29)
R∞ ( B)

Where GH is a parameter called the geometrical factor of
magnetoresistance, R(B) is the resistance of a considered
Fig. 1.6 The current density components arising in the presence of an
electric field E and the magnetic induction B. The coordinate system is magnetoresistor at an induction B, and R∞ ( B ) denotes the
so chosen so that E is collinear with the x axis and B lies in the xz resistance of a corresponding part of an infinitely long strip.
plane. The E and B vectors are not mutually orthogonal. The total
current density is the sum J=J1+ J2 + J3 . J1 is collinear with E, J2 is
collinear with [ExB] (witch is here collinear with the y axis), and J3 is
collinear with B. Note that we tacitly assume positive charge carriers. In 3.1 Common shapes of a Hall plates and magneto-
the case of negative charge carriers, the component J2 being resistors
proportional to the mobility, would have an inversed direction (along
the +y axis). (The picture was taken from P.S Popović “ Hall Effect
Devices (Sensor Series) Swiss Federal Institute of Technology Lausanne
Generally, any piece of electrically conductive material,
(E.P.F.L) 2nd Edition IoP 2004) fitted with four electrical contacts, could be used as a Hall
device. However, to be reasonably efficient and convenient
and specially we can write for a practical applications, a Hall device should preferably
be made in the form of a plate. The contacts should be
positioned at the boundary of the plate, far away from each
σ
J1 = ( )E other. The two contacts for sensing the Hall voltage should
1 + μ 2 B2 be placed so as to attain approximately equal potentials at a
zero magnetic field. Then the potential difference at the
σ
J2 = ( ) μ[ E × B ] sense contacts equals to Hall Voltage. The common
1 + μ 2 B2 potential at the sensing contacts should roughly equal the
mid-potential of the current contacts. Then the largest Hall
σ
J3 = ( ) μ 2 B[ E ⋅ B] (1.26) voltage can me measured. To simplify the design and
1+ μ B
2 2
fabrication of a Hall plate a highly symmetrical shape and a
uniform material and thickness of the plate are usual chosen.
And finally the total galvanomagnetic current density is: Some commonly used shapes of Hall plates are shown in
figure 1.7.

J σ E (1 − μ 2 B2 ) + μσ [ E × B] + μ 2σ B( E ⋅ B) (1.27)
3.2. Reducing offset voltage and 1/f noise

To removing ohmic offset we use, an elegant method that
exploits a property of the Hall-effect transducer to reduce
system offset. A four-way symmetric Hall-effect transducer
5


can be viewed as a Wheatstone bridge. Ohmic offsets can be model 1/f noise source by a fluctuating asymmetry
represented as a small ∆R, as shown in figure 1.8a. When resistance ΔR in figure 1.8. If the biasing current of a Hall
bias current is applied to the drive terminals, the output device spins fast enough so that the fluctuating voltage does
voltage appearing at the sense terminals is VH + VE, where not change essentially during one circle, then the system
VH is the Hall voltage and VE is the offset error voltage. shall not “see” the difference between the static and the
Now consider what happens if we take the transducer fluctuating offset and shall cancel both of them. In order to
and reconnect the bias and sense terminals, as shown in completely eliminate the 1/f noise of a Hall device, we have
figure 1.8b. All of the terminal functions have been rotated to set the spinning current frequency well above the 1/f noise
clockwise by 90°(“Spinning Current technique”). The sense corner frequency. For a very small device, this may be a
terminals are now connected to bias voltage, and the former very high frequency.
drive terminals are now used as outputs. Because the
transducer is symmetric with rotation, we should expect to
see, and do see, the same Hall output voltage. 4. Our suggestion for a novel shape Hall device : The 3D-
WHS

The lasts few months we design and we are ready to
developed a novel Hall sensor device which uses elaborate
spinning current technique. The novel 3D Hall device that
we call “3D Wheel Hall Senor” is presented in figures 1.9a
and 1.9b. The current enters the device, as presented in the
aforementioned figures, in two phases namely the even
phase (PHASE-P) and the odd phase (PHASE-N).

Fig. 1.7 Various shapes of Hall plates. Current contacts are denoted by
C, sense contacts by S, and C/S indicates that the current and sense are
interchangeable (a) (b)

The transducer, however, is not symmetric with respect Fig. 1.9 The novel 3D Hall device that we call “3D Wheel Hall Senor”
to the location of ∆R. In effect, this resistor has moved from or 3DWHS. (a) The even phase (PHASE-P) ; (b) the odd phase
the lower right leg of the Wheatstone bridge to the upper (PHASE-N).
right leg, resulting in a polarity inversion of the ohmic offset The device exploits the signals attributed to Hall voltage,
voltage. The total output voltage is now VH – VE. One way to Hall current, and geometric MR effect. As a result the device
visualize this effect is to see the Hall voltage as rotating in is equivalent to an “ideal” voltage or current Hall sensor
the same direction as the rotation in the bias current, while with geometrical factor of one (GH = 1). Moreover it
the ohmic offset rotates in the opposite direction. If one were provides for high-speed spinning, given that the voltage
to take these two measurements to obtain VH + VE and VH – distribution changes moderately between different phases.
VE, one can then simply average them to obtain the true This is equivalent to minimum charge injection that 0 in turn
value of VH. For this technique to work, the only requirement – allows spinning frequency increase. The device senses all
on the Hall-effect transducer is that it be symmetric with 3 filed dimensions, namely the flux-density of BZ is
respect to rotation. It is possible to build a circuit that is able proportional to the DC component of the output signal,
to perform this “plate-switching” function automatically. whereas the BX and BY components are proportional to the
first harmonic of the output signal. Finally the device can be
made in a way to reuse the current, if integrated in a
BiCMOS technology providing for matched JFETs.

Fig. 1.8 Effects of rotating bias and sense terminals on output. (The (a) (b)
picture was taken from Edward Ramsden “Hall Effect Sensors : Theory
and Application” 2006) Fig. 1.10 The novel 3D Hall device that we call “3D Wheel Hall
Senor” or 3DWHS. (a) The electric field and the equpotential lines in
the even phase (PHASE-P) ; (b) The electric field and the equipotential
Imagine that the 1/f noise (flicker) at the output terminals lines in the odd phase (PHASE-N).
of a Hall device as a fluctuating offset voltage. Then we can
6


We simulate in MatLab environment the electric field 5. Acknowlegements
and the equipotential lines inclination on the 3D-WHS in a
presence of magnetic induction to see the behavior of the The authors gratefully acknowledge the Pr. R P. Popovic
device in a real Hall effect conditions and the results (EPFL) and E. Ramsden, M. Shur and S.S.Sze for the useful
represented in figures 1.10a and 1.10b. bibliography and Pr. M. Vavalis for the useful discussion
and his scientific support.

______________________________
References

1. Hall. E. H 1879, On a new action of the magnet on electric 2. Corbino O. M. 1911, Electromagnetische Effekte,die von der
currents Am.J.Math. 1-p.287-292 Verzerrung herriihren, wehle ein Feld an der Bahn der
Ionen in Metallen hervobring. Phys. Z. 12. p.561

______________________________
Bibliography
1.P.S Popović “ Hall Effect Devices (Sensor Series) Swiss Federal 3.Michael Shur “Physics of Semiconductor Devices” (Prentice Hall
Institute of Technology Lausanne (E.P.F.L) 2nd Edition IoP 2004 Series in Solid State Physical electronics 1991)
2.Edward Ramsden “Hall Effect Sensors : Theory and Application” 4.Simon S. Sze “Physics of Semiconductor Devices” 3rd Edition
2006 2006

7

Galvanomagnetic Effects Sensors Based On Hall Effect

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Galvanomagnetic Effects Sensors Based On Hall Effect (20)

More from petousis (11)

Recently uploaded (20)

Galvanomagnetic Effects Sensors Based On Hall Effect