S.

Chandramouli: Hall Coefficient of Germanium 1

Hall Coefficient of Germanium
Sridhar Chandramouli
Physics Department, The College of Wooster, Wooster, Ohio 44691
April 29, 1999
This experiment experimentally measures the Hall coefficient of a
Germanium sample, and the number of carriers of electric current per unit
volume in that sample using formulae derived within the theory of the
paper. The Hall coefficient for the Germanium sample was found to be
-(1.907+0.071)*10-2 m3 /C, and the number of carriers was found to be
3.86*1020+0.14*1020 /m3 . The Hall coefficient, and the density of free
carriers for germanium has been previously found to be –8*10-2 m3 /C,4 and
1.0*1021 electrons/m3 respectively 6 . These results, in particular the sign of
the Hall coefficient show that conduction in Germanium is in fact
performed by electrons, and not holes as in many other semi-conductors.
The results of this experiment also verify previously published results, as
both sets of values are of the same order.

INTRODUCTION −ev d B = − eEH …2

In 1879, Hall observed that on placing a
current carrying conductor perpendicular to a where EH is the Hall Voltage, B is the magnetic
magnetic field, a voltage is observed field and e is the charge of an electron.
perpendicular to both the magnetic field and the
current. It was puzzling that the charge carriers, B(Magnetic Field)
which were assumed to be electrons, experienced
a sideways force opposite to what was expected.
This was later explained by the band theory of
solids.4 E[y]
The Hall Effect has been important in the
study of the mechanism of conduction in V(-)
semiconductors because both the mobility and θ
concentration of the charge carriers may be
measured, as opposed to only the conductivity I
E[x]
with conductivity experiments.
THEORY
Assuming that electrons, with charge e are
the current carriers, and they have a drift velocity,
vd in the direction of the applied electric field, V(+)
w t
they experience a force due to a
transverse magnetic field, B given by: I[t]=Iwt
F = −evd B …1 Figure 1: Voltage components for conduction by electrons
in a magnetic field, where w is the width of the conductor.
where the force, drift velocity, and the magnetic
field are perpendicular to each other, and the To gain a good understanding of the Hall
minus sign comes from the negative charge of the Effect, a classical particle view is beneficial. For
electron (figure 1). The sideways force, called the the specific case of Germanium or any other semi-
Lorentz Force, causes the electrons to deflect conductor, the charge carriers have a high
toward one face of the conductor, due to which probability of being at or around the point of
arises a transverse electric field, which is injection of impurities, giving the carriers a
measured as the Hall Voltage. At thermal definite location. Thus, using the classical
equilibrium, when the Lorentz Force exactly particle treatment, the drift velocity, vd , used in
matches the force due to the electric field (the equation 1, can be used to find the current density
Hall Voltage), we have: in a semi-conductor. Assuming that the collisions
that an electron undergoes are
 S. Chandramouli: Hall Coefficient of Germanium
completely random, the current density, j is: by moving into the hole. It appears like the hole
is moving in the opposite direction of the
j = −env d …3 electrons and thus, positive charges or holes are
where e is the charge of an electron, and n is the said to carry the current.
number of electrons per unit volume. The drift Holes may be introduced into a
velocity used in equation 1 is not a very germanium sample by careful injection of
convenient parameter. Using equations 2 and 3, impurities, the process called doping. This sheds
we can derive a convenient expression for the light on the possibility of positive current carriers,
Hall Voltage in terms of the current density. and thus some Hall Coefficient are positive,
because of conduction by positive holes in the
E V t valence bands. Equation 5 may be modified to
RH = H = ( H ) …4
jB I B include a positive charge instead of the –e in the
denominator. Thus, the sign of the Hall
where VH is the average Hall Voltage, or simply Coefficient may be used to discover the nature of
the Hall Voltage divided by the width across the charge carriers.
which the voltage was measured (the width of the
sample). The other parameters are given sense by Experimental Setup & Procedure
figure 1, and previous equations. To calculate the Hall Coefficient of
The applied electric field, and the Germanium, the power supply (Hewlett Packard
transverse electric field created by the magnetic 6216C) sends a current through the Ge sample,
field give a resultant electric field which is not while the permanent magnet provides a strong
parallel to the current. The angle θ, which magnetic field perpendicular to the direction of
separates the current from the resultant electric the current. A voltmeter (Keithley 197
field (figure 1), enables the derivation of a Autoranging Microvolt DMM) is connected
formula for the Hall Coefficient, already stated in perpendicular to both, the direction of flow of the
equation 4, in terms of the charge of the carrier, current and the magnetic field. This is used to
and the charge density. But in fact equation 4 is a make observations of the Hall Voltage, An
very simplified view of the system which involves ammeter (Tektronix CDM250 Digital Multimeter)
many simplifying assumptions regarding the is connected to the DC power supply to make
collisions of the electrons. In reality, the measurements of current flowing through the
probability of collision depends in a complicated germanium sample. A traveling Microscope
way on both the initial and final states.6 When (Gaertner Scientific Corporation), and a gauss
this is accounted for the equation reduces to: meter (Applied Magnetics Laboratories) complete
the equipment list.
3π 1 The DC power supply has control over the
RH = …5
8 n(−e) current passing through the sample. The current
may be varied by directly changing the current, or
Some Hall Coefficients are found to be by varying the voltage that drives the current.
positive, conflicting with the electron theory of The instruments are very sensitive to change in
conduction. The band theory of solids is essential either parameter, and extreme care must be taken
to explain the deflection of current carriers in to attain the desired level of current through the
semi-conductors like Germanium. When the Hall Ge sample. Once a desired current has been
Effect was first observed, the apparent opposite achieved, a transverse voltage reading is made at
deflection of the electrons, which were considered the voltmeter. The desired current was attained in
the only current carriers, was puzzling. This was different runs of the experiment, by alternately
resolved by the band theory of solids in the early keeping the current and the voltage constant, and
part of this century. Valence electrons occupy a varying the other parameter.
complete set/band of states that spans a finite First, the experiment was conducted
range of energies; therefore no electron may be without any magnetic field. Theoretically, there
moved to any other valence state without violating should be no transverse voltage as current carriers
Pauli’s Exclusion Principle, so valence electrons would not have deflected. Start the experiment at
are rigid and can carry no current. In metals, the 15 mA, and make readings of the transverse
bands overlap and so electrons move freely from voltage for successive decrements in the current,
the upper valence bands to the lower conduction down to 1mA, or even lower. Then slowly
bands and thus conduct. When an electron from a increase the current back to 15 mA once again and
filled valence band is removed, there is now a check for reproducibility of the data. The
hole in the valence band for other electrons to experiment is repeated under the influence of the
move into, and they may now conduct electricity
2
 S. Chandramouli: Hall Coefficient of Germanium
known magnetic field. The field is then inverted with the magnetic field are thus also
and the experiment redone yet again. systematically erred by the same amount.
The strength of the magnetic field and the Therefore, in graph 1, the voltage is non-zero
dimensions of the Ge sample are the other even for a zero magnetic field. In graph 1, the
measurements required. To measure the strength error bars are constant values. The negative B
of the magnetic field, the probe of the Gauss field merely signifies an opposite orientation of
meter is introduced exactly in the middle of the north and south poles.
space between the two poles, and a series of To correct for the error, the voltage
readings are made, because the field is very readings in the absence of any magnetic field are
variable. The average value of the readings is systematically forced to zero, and adjustments are
used in the determination of the Hall Coefficient. accordingly made to the Hall Voltage readings.
The dimensions of the Ge sample are measured The values of the zero magnetic field voltages are
using a traveling microscope. Only the thickness subtracted from both sets of Hall voltages. To
of the sample is required in the determination of calculate the Hall Coefficient of germanium, we
the Hall Coefficient. Nevertheless, measurements look to equation 4. One method that can be
were made of all the dimensions of the sample. employed to measure the Hall Coefficient of
Since the Ge sample is set within a frame and germanium is computing the Hall Coefficient for
should not be disturbed, its thickness is measured each of the values of current, and then using the
with a traveling microscope. The front and back mean and standard deviation of all those values.
end of the sample are focussed one after the other But there is a more elegant method. The factor
and the distance moved by the calipers recorded. VH/I is the slope of the plot of the current and the
Using simple trigonometric functions, the Hall Voltage. Therefore, a line can be fitted to the
thickness is easily estimated. values of the Hall Voltage at various values of
current, and the slope of this line may be used,
Data, Error, Error Propagation & Analysis
The magnetic field was measured as along with the error in fit (standard deviation), to
calculate the Hall Coefficient (graph 2). The
(171.7+1.3)*10-3 Tesla, which is the SI units for measurement of the magnetic field, and the
magnetic fields. A conversion factor of 10 4 was thickness of the germanium sample is conducted
used to convert between Gauss and Tesla. The as described in the
dimensions of the Ge sample were measured to be simply a standardprocedure, and their errors are
deviation of a series of
(4843.3+5.8)*10-6 m (length), (483.0+2.6)*10-5 m observations. These steps accounts for all the
(width), and (1815.0+6.5)*10 -6 m (thickness). random error involved in the observations.
0

-10 B=((171.7+1.3)e-3) Tesla 20

B=(171.7+1.3)e-3 Tesla
Voltage (milliVolts)

B=0 VH=((-3.1+1.0)e-1)-(1.81+0.01) I
Voltage (milliAmps)

-20
10

-30
B=0
0
-40 B=((-171.7+1.3)e-3) Tesla
-10
-50
VH=((-1.4+1.2)e-1)-((179.8+1.3)e-2) I
B=(-171.7+1.3)e-3 Tesla
-20
-60

2 4 6 8 10 12 14
Current (milliAmps) 2 4 6 8 10 12 14
Current (milliVolts)
Graph 1: Transverse (Hall) Voltage plotted against the
longitudinal current, uncorrected for systematic errors. The Graph 2: Current plotted against voltage, corrected for
opaque squares and circles are voltages for opposite systematic error, such that there is no voltage reading for
orientations of the magnetic field, while the crosses are B=0. Straight lines are fitted to the data. The markers are
voltages in the presence of no magnetic field. consistent with graph 1.

Systematic errors, probably due to faulty The slope of both the lines (current plotted against
calibration, causes the voltage recorded without Hall voltage) for the different orientations of the
the magnetic field to be consistently non-zero. magnetic field, are similar. Since the values are
The transverse voltage should be zero in the so consistent, the average value of the two slopes
absence of a magnetic field because none of the (VH/I) is used and the errors of the individual line
charge carriers are deflected, and hence no Hall fits also propagate as averages, since this error is
Voltage can be set up. The Hall Voltage readings
3
 S. Chandramouli: Hall Coefficient of Germanium
greater than simply the standard deviation of the References
1
two consistent slopes. The Hall Coefficient of Pugh, Emerson, M., Pugh, Emerson, W.,
germanium can thus be calculated: Principles of Electricity and Magnetism,
3 (Addison-Wesley Publishing Company, Inc.,
−2 −2 m
R H = −1.907*10 ± 0.071*10 …6 1960).
2
C Feynman, Richard, P., The Feynman Lectures on
On observing the sign of the Hall Physics, Volume III, (Addison-Wesley Publishing
Coefficient of germanium, it can be concluded 3 Company, Inc., 1965).
that the charge carriers in Ge are in fact electrons. Junior Independent Study Lab Manual, Spring,
The above value compares favorably with –8*10 -2 1999, (Unpublished).
4
m3 /C, reported by Lerner et.al.4 The negative sign Lerner, Rita, G., Trigg, George, L., Concise
may arise also imply due to the connection of the Encyclopedia of Solid-State Physics, (Addison-
voltmeter. Keeping track of the magnetic fields 5 Wesley Publishing Company, Inc., 1983).
therefore becomes very important. Seitz, Frederick, The Modern Theory of Solids,
Further, the density of the electrons is First Edition, (McGraw-Hill Book Company, Inc.,
calculated using equation 5. The Hall Coefficient 6 New York & London, 1940).
for germanium is known along with its associated Shockley, William, Electrons and Holes in
errors, Therefore, the density of conducting Semiconductors, (D. Van Nostrand Company,
electrons in Ge is: Inc.,1950.
n = 3.861*10 20 ± 0.14*10 20 electrons/m 3 …7
The density of the current carriers is also
consistent with values reported earlier. Shockely6
reported a value of 1.0*1021 electrons/m3 , and
Lerner et.al4 reported a value of the apparent
number of free electrons per atom as 1.7*10-9. To
convert to the units used in this paper,
(electrons/m3 ) we have to find the number of
atoms per unit volume for Ge. We find that there
are 4.4*10 28 atoms/m3 , and therefore, the density
of free electrons is: 4.4*1028 atoms/m3 * 1.7*10-9
electrons/atom = 7.48*1019 electrons/m3 .
Conclusion
The Hall Effect is important because it
enables us to make measures of mobility and
concentration, and gives insight into the
mechanism of conductivity in semi-conductors.
The Hall Coefficient and the density of carriers in
Ge are well known quantities, and have been
verified in this paper, given experimental
limitations, with values of the same order. Also,
it is established that the primary charge carriers
are electrons in our sample of germanium. In
some cases, conduction occurs in both the valence
bands and the conduction bands. Therefore,
electrons and positive holes simultaneously carry
current, though in the Ge sample used in the
experiment, the electrons predominate over the
positive holes.
If the experiment were conducted at low
temperatures, and a variable magnetic field used,
the Quantum Hall Effect could have been
observed. This is characterized by a step function
that arises when the resistivity of the sample is
plotted against the varying magnetic field.
_______________________________________
4