Experiment-1

Determination of Resistivity and Band Gap of

a Semiconductor by Four Probe Method

The most commonly used technique in the

semiconductor industry for measuring resistivity
is the four-point probe method. This method is
nondestructive and can be used to measure the
resistivity of ingots, as well as wafers.

We describe how a classic laboratory experiment

on electrostatics can be applied with some minor
modifications to study the four-point probe
measurement technique. The four- point probe
technique is one of the most common methods
for measuring resistivity. The classic
arrangement is to have four needle-like
electrodes in a linear arrangement with a current
injected into the material via the outer two elec-
trodes. The resultant electric potential
distribution is measured via the two inner
electrodes [Fig. 1]. By using separate electrodes
for the current injection and for the de-
termination of the electric potential, the contact
resistance between the metal electrodes and the
material will not show up in the measured
results. Because the contact resistance can be
large and can strongly depend on the condition
and materials of the electrodes, it is easier to
interpret the data measured by the four-point
probe technique than results gathered by two-
point probe techniques.

The four-point probe technique was originally

developed by Wenner in 1916 to measure the
earth’s resistivity. In geophysics it is referred to
as Wenner’s method. In 1954 Valdes adopted
the technique for semiconductor wafer resis-
tivity measurements. The technique has also
been applied to characterize electrolytes and to
analyze gases.

Today the four-point probe technique is widely

used in the semiconductor industry to monitor
the production process. Electrical measurements
are done on test structures to provide
information on the various process steps. For
example, resistivity measurements on doped
semiconductor structures provide information on
the active charge carrier concentration and the
mobility and are used as feedback to the doping
process. Resistivity measurements on aluminum
test structures provide information on the line-
width and thickness of the interconnect, which is
used to fine-tune the lithography process.
Four collinear metal probes with sharpened tips
are placed on the semiconductor. A constant
current (I) is passed through the two outer
probes, and the potential difference (V)
developed across the inner two probes is
measured using a high-input impedance
voltmeter. In this way the metal-semiconductor
contact resistance does not affect the results.
BASIC THEORY

Four sharp probes are placed on a flat surface

of the material to be measured (Fig.1). The
current is passed through the two outer
electrodes, and the floating potential is
measured across the inner pair. If the flat
surface on which the probes rest is adequately
large, it may be considered to be a semi-
infinite volume.
PROBES
I V I

SEMICONDUCTORS
1 2 3 4

S1 S2 S3

Fig. 1: Model for the four probe resistivity

measurement

The experimental circuit used for

measurement is illustrated schematically in
Fig. 2. A nominal value of probe spacing,
which has been found satisfactory, is an equal
distance of 2.0 mm between adjacent probes.
 GALVANOMETER
POTENTIOMETER
NANOAMMETER
MICROVOLTMETER
DIRECT
CURRENT V
SOURCE I
I

PROBES

Fig. 2: Circuit used for resistivity

measurement

In order to use the four-probe method, it is

assumed that:

1.The resistivity of the material is uniform in the

area of measurement.
2.If there is minority carrier injection into the
semiconductor by the current - carrying
electrodes, most of the carriers recombine
near the electrodes so that their effect on the
conductivity is negligible.
3.The surface on which the probes rest is flat
with no surface leakage.
4.The four probes used for resistivity
measurements are equally spaced and
collinear.
5.The diameter of the contact between the
metallic probes and the semiconductor
should be small compared to the distance
between probes.
6.The surfaces of the material may be either
conducting or non-conducting.

A conducting boundary (such as copper) is

one on which the sample is plated or placed.

A non-conducting boundary is produced

when the surface of the sample is in contact
with an insulator.
Case 1 - Resistivity Measurements on a Large
Sample
One added boundary condition is required to
treat this case namely, the probes are far from
any of the other surfaces of the sample and the
sample can thus be considered a semi- infinite
volume of uniform resistivity material.
Fig. 1 shows the geometry of this case. Four
probes are spaced S1, S2 and S3 apart. Current I
is passed through the outer probes (1 and 4) and
the floating potential V is measured across the
inner pair of probes 2 and 3.
The floating potential Vf a distance r from an
electrode carrying a current I in a material of
resistivity is given by
ρ I
V =
2πr

Calculation/Notes:
The electric potential around the electrodes can
be calculated by solving Laplace’s equation,
which is a combination of the equation of
continuity (Iin=Iout), Ohm’s law (V=IR), and the
E – V relation (E=-dV/dx):
 ∇ ∇
∇ = = = 0………….( )

We know that the electric field and electric

potential around and electric point charge are
given by

= … … … … … … … ( 1)

= … … … … … … … ( 2)
The current injected in the material will result
in a similar distribution of the electric field. If
we assume that the injected current I will be
distributed over a half sphere of radius r, the
electric field will be given by

= = = … … … … … (%)
!" # 2$
We then combine Eqs. (B) & (C) to obtain the
electric potential around a point source that
injects a current I in the material

=
2$
This equation can be used to calculate the
electric potential at the position of the sensing
probes.
_____________________________________

In the model shown in Fig. 1 there are two

current-carrying electrodes, numbered 1 and 4,
and the floating potential Vf, at any Y point in
the semiconductor is the difference between
the potential induced by each of the
electrodes, since they carry currents of equal
magnitude but in opposite directions Thus:
&' (
V = * − /…………… (1)
) +, +
.

where r1 = distance from probe number 1 and r4

= distance from probe number 4.
The floating potentials at probe 2 (Vf2), and at
probe 3 (Vf3) can be calculated from Eq.(1) by
substituting the proper distances as follows:

ρ I 1 1
V = 0 − 4
2π s s + s3

ρ I 1 1
V3 = 0 − 4
2π s + s s3
The potential difference V between probes 2 and 3
is then

V = V − V3
ρ I 1 1 1 1
= 0 + − − 4
2π s s3 s + s3 s + s

and the resistivity is computable as

2$
=
1 1 1 1
*s + s − s + s − s + s /
3 3

When the point spacings are equal, that is, S1

= S2 = S3 = S the above simplifies to:

5
ρ = 2πS……………………. (3)
(
 Case 2- Resistivity Measurements on a Thin
Slice-Conducting Bottom Surface

Two boundary conditions must be met in

this case; the top surface of the slice must be
a reflecting (non-conducting) surface and
the bottom surface must be an absorbing
(conducting) surface. Since the two
boundaries are parallel, a solution by the
method of images requires for each current
source an infinite series of images along a
line normal to the plane and passing through
the current source.
+I -I
n = +2

2W
-I +I
n = +1

2W
+I 1 2 3 4
-I TOP SURFACE
n = -0
(NON-CONDUCTING)
S S S W SLICE
BOTTOM SURFACE
-I +I (CONDUCTING)
W
n = -1
3S
2W
+I -I
n = -2

Fig. 3: Images for the case of the resistivity Probes

on a slice with conducting bottom surface
The model for this case is shown in Fig. 3.
The side surface of the slice is assumed to
be far from the area of measurement and,
therefore, only the effect of the bottom
surface needs to be considered. In this
analysis equal probe spacing S shall be
assumed. The width of the slice is W. The
array of images needed is indicated in Fig.
3, where the polarity and spacing of the first
few images are as shown.

The floating potential Vf2 at electrodes 2 is

= =
ρI 1 1
V = 7 8 (−1)9 − 8 (−1)9 @
2π :S + (2nW) :(2S) + (2nW)
9>?= 9>?=

Likewise, the floating potential at electrode (3)

can be obtained and
= =
ρI 1 4 4
V= 7 + 8 (−1)9 − 8 (−1)9 @
2π S :S + (2nW) :(2S) + (2nW)
9>?= 9>?=

…………………….. (5)


The resistivity then becomes

B'
=
CD (E/G)
……………… (6)
Where resistivity is computable from
E qs. (2 & 3) can be used if the point spacing
are different, but approximately equal. The
function G6 (W/S) is computed from
Q
G
HI (J/K) = 1 + 4E 8 (? )LM ? P
LR, :(G/E)# N( O)# :( G/E)# N( O)#

……………(7)

which is tabulated in Table I & plotted in Fig. 4.

TABLE - I
S. No. W/S G 6 (W/S) G7 (W/S)
1 0.100 0.0000019 13.863
2 0.141 0.00018 9.704
3 0.200 0.00342 6.931
4 0.33 0.0604 4.159
5 0.500 0.228 2.780
6 1.000 0.683 1.504
7 1.414 0.848 1.223
8 2.000 0.933 1.094
9 3.333 0.9838 1.0228
10 5.000 0.9948 1.0070
11 10.000 0.9993 1.00045
 1.0

0.7

0.5
0.4

0.3

0.2
S S S
G6(W/S)

0.1
W
0.07
CONDUCTING BOUNDARY
0.05
0.04

0.03

0.02

0.01
0.1 0.2 0.3 0.4 0.5 0.7 1.0 2 3 4 5 7 10
(W/S)

Fig. 4: G6 (W/S) for probes on a thin slice

with a conducting bottom surface.
Case 3 Resistivity Measurements on a Thin
Slice-Non-Conducting Bottom Surface.
The model for these measurements is like the
case 2, except that the bottom surface of the
slice is non-conducting. This means that all the
images of Fig. 3 have the same charge as the
current source. Thus all the images on a row
have equal charges and Eq. 5 describes the
potential difference across the inner pair of
probes, if (-1)n is removed from the equation.
Then,
B'
= ( )
CS E/G
……… …… (8)
where,
Q
G
HT (J/K) = 1 + 4E 8 (? )L M ? P
LR, :(G/E)# N(O)# :( G/E)# N( O)#

…………………..…(9)
This function G7(W/S) is tabulated in Table I.
and plotted in Fig. 5. For smaller values of W/S
the function G7 (W/S) approaches the case for an

2S
infinitely thin slice, or
GT (W/S) = log Y 2 … … … … . (10)
W
Thus for sample W/S < 0.25 or sample thickness
upto 0.5mm, the correction factor may be
obtained from equation (10) directly.
100.0

70.0

S S S
40.0

30.0

W
20.0

NON CONDUCTING
BOUNDARY

10.0

7.0

5.0
4.0

2.0

1.0
0.1 0.2 0.3 0.4 0.5 0.7 1.0 2.0 3.0 4.0 5.0 7.0 10.0

Fig. 5: G7 (W/S) for probes on a thin slice

with a non-conducting bottom surface.
Temperature dependence of resistivity of a
semiconductor:

Intrinsic semi-conduction: The process in which

thermally or optically excited electrons contribute
to the conduction is called intrinsic semi-
conduction. In the absence of photonic excitation,
intrinsic semiconduction takes place at
temperatures above 0 K as sufficient thermal
agitation is required to transfer electrons from the
valence band to the conduction band.

The total electrical conductivity is the sum of the

conductivities of the valence and conduction band
carriers, which are holes and electrons,
respectively. It can be expressed as

σ = e(neµe + nhµh) ……………….(11)

where ne , µe are the electron's concentration and

mobility, and nh , µh are the hole's concentration
and mobility, respectively.
Drift mobility determines the average drift
velocity in the presence of an applied external
field. It also depends on the temperature. The
mobility is a quantity that directly relates the drift
velocity vd of charge carriers to the applied
electric field E across the material, i.e.,
µ = vd/ E …….……….;…. (12)

In the intrinsic region the number of electrons is

equal to the number of holes, ne=nh=ni, so
equation (11) implies that,
σ = eZi (µe + µh)………………. (13)
The electron density (electrons/volume) in the
conduction band is expressed in the following
form
Zi = [ ]3/2 exp[-Eg/kBT]………………(14)
Where, N is a constant. Thus,

σ =e (µe + µh) []3/2 exp[-Eg/kBT]………. (15)

This shows that conductivity depends on

temperature it decreases exponentially with
decrease in temperature. Thus,
kl
= = exp j p…………………..(16)
i mn o

Where, = 1/ [e (µe + µh) []3/2] is assumed

constant.
 DIGITAL MICROVOLTMETER CONSTANT CURRENT SOURCE
Model : DMV-001 Model : CCS - 01

mA
1 mV
10 mV ZERO ADJ
RANGE 100 mV
CURRENT
1V
10 V
20mA 200mA
ON OUTPUT ON

OR

PID CONTROLLED OVEN

LOW CURRENT SOURCE
Model : PID-TZ Model : LCS - 02
2 µA
TEMP. CONTROLLER 20 µA
c 200 µA
CURRENT 2mA
RANGE
ADJ.

PV ON
OUTP UT
sv
SV2 AT OUT EV1

MD AT

Temp. Sensor

OVEN-600 OVEN-200

Oven
Selector
ON ON

MAINS

OVEN
ON

Fig. 6: Connection diagram for the set up

Fig. 7: Complete experimental set up

PROCEDURE

1. Measurement of resistivity of n-Ge:

1. Put the sample (n-Ge) on the insulating base plate of the four probe
arrangement. Unscrew the pipe holding the four probes and slowly let
the four probes rest at the middle of the sample. Apply a very gentle
pressure on the probes and tighten the pipe in this position.
2. Connect the outer pair of probes to the constant current source through
3-pin socket and the inner pair to the D.C. Microvoltmeter through an
BNC connector.
3. Place the Four Probe Arrangement in the oven.
4. Switch ‘ON’ D.C. Microvoltmeter and allow about 5 minutes time for
thermal stability. Adjust the zero of D.C. Microvoltmeter with the knob
provided for it.
5. Switch ‘ON’ the Constant Current Power Supply (for Ge) and check the
zero reading of D.C. Microvoltmeter at zero current reading again.
Adjust it, if required.
6. Increase the current gradually and note the corresponding voltage
readings.
7. Subtract the zero reading if any and check for Ohm’s Law (R=V/I). This
would ensure that the system is working properly and the systematic
readings could be taken.

2. Measurement of resistivity of n-Ge at various temperatures:

1. Put n-Ge sample in the four probe arrangement following the procedure
described in (I). The current will be fixed (say at 5mA) and temperature
is varied.
2. For different temperature settings, the PID Controlled Oven should be
used. It stabilizes the temperature at the desired value while recording
data.
3. Plot a graph for Log10 ~ (1/T) and calculate the enegy band gap.
OBSERVATIONS

Table 1: Data for resistivity of n-Ge

Thickness (W): ________ mm

S.No. Current (I) Voltage (volts) V/I

1
2
3
4
.
.
.
.
 Table 2: Data for resistivity -Temperature of n-Ge
Thickness (W): _________ mm

Probe current (I) =_-------------- mA (fixed)

S.No. Temp Voltage (volts) ρ T-1 x 10-3 Log10ρ

(T in K) (Ω.cm.) (K-1)
1
2
3
4
5
.
.
.
.
.

CALCULATIONS AND GRAPH

From Eq. (3)
5
ρ = 2πS
(

Correction Factor:
Since the thickness of the samples are small compared to the probe
distance a correction factor for it has to be applied. Further the bottom
surface is non-conducting in the present case, Eq. (9) will be applied.
=
HT (J/K)
The function G7 (W/S) may be obtained from Table-I or Fig. 5 for the
appropriate value of (W/S). For sample W/S<0.25, correction factor may
be obtained directly from Eqn. 10. Thus may be calculated for various
temperatures.

Energy Band Gap

z{x| &
We know, Ex = 2k y
/}
Where ~ is the Boltzamann's constant = 8.6 x 10-5 eV/deg & T is
temperature in kelvin. Use the graph to determine Eg.
 Determination of the band gap of a Ge Sample
1.300

1.250

1.200

1.150

1.100

1.050

1.000

0.950

0.900

0.850

0.800

0.750

0.700

0.650
log10ρ

0.600

0.550

0.500

0.450

0.400

0.350

0.300

0.250

0.200

0.150

0.100

0.050

0.000
1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40 2.60 2.80 3.00 3.20 3.40 3.60 3.80 4.00 4.20
-0.050

-0.100
T-1 X 10-3

Fig. 8: Sample Graph of Log10 • ~ (1/T) for n-Ge sample.