You are on page 1of 56

Techniques for Measuring the Electrical

Mary Anne Tupta

November 18, 2010

A GREATER MEASURE OF CONFIDENCE

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

• Methods for making resistivity measurements of conductors,

insulators, and semiconductors.

resistivity.

• Sources of measurement errors and ways to optimize resistivity

measurements

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

The electrical resistance of a material……

• is a basic material property
• defines how well the material will conduct an electric current
• is a common electrical measurement.

Ohm’s law relates the current (I) and the applied voltage (V) to the
material resistance (R) as follows:

V=IR
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Electrical Resistivity = longitudinal electrical resistance of a uniform

rod of unit length and unit cross-sectional area:

A
R
L
Current Source ρ = resistivity (Ω-cm)
R = resistance: V/I (Ω)
Voltmeter
V A = cross-sectional area of
L
sample (cm2)
L = distance between two
A
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Classification Type of Electrical Typical Resistivities

Conductor
Metals Good electrical 10-6 Ω-cm
conductors
Insulators Low electrical 109 to 1020 Ω-cm
conductivity
Semiconductors Intermediate levels of 10-3 to 107 Ω-cm
conductivity

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Conduction Band

Energy
Energy Gap
Energy Gap

Valence Band
Valence Band
Valence Band

Metal Insulator Semiconductor

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Measuring the Resistivity of

Conductors:
Metals and Superconductors (very low resistance)

Graphene and other Nanomaterials

(low voltage and low power - use same techniques)

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Basic Test Procedure: Voltmeter

V
1. Source current (I) through the
L
sample using one pair of leads.
w

across a second pair of leads a

known distance (L) apart.
Current Source
3. Calculate the resistivity (ρ) of
the sample using the cross-
V wt
sectional area (A=wt) and the    cm
distance between the voltmeter I L
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Sources of Error When Measuring Low Resistance

• Thermoelectric Voltages

• Johnson Noise

• Voltmeter Not Sensitive Enough

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Eliminate Lead Resistance by Using the 4-Wire Method

2-Wire Method 4-Wire Method
I I

V V

RSample
RSample

Measured Resistance: Measured Resistance:

VM/I = RSample + 2RLead VM/I = RSample
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Thermoelectric Voltages
Voltmeter Voltmeter
Metal A V
Metal A V

T1 Metal B T2

Thermoelectric voltages are generated when dissimilar metals (Metal A

and Metal B) in the circuit are at different temperatures (T1 and T2).

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

• Use an offset compensation method

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Use the Current Reversal Method

to Eliminate Voltage Offsets (VEMF)
Measurement with Positive Polarity Measurement with Negative Polarity

VEMF VEMF

I+ VM+ I- VM-
R R

VM+ = VEMF + IR VM- = VEMF - IR

Voltage Measurement:

VM   VM  VEMF  IR  (VEMF  IR )
VM    IR
2 2
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

and Noise

The Delta method consists of

alternating the current source
polarity and using a moving
calculate the resistance.
Averaging reduces the noise
bandwidth and therefore the noise.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

• External Noise Sources = interferences created by motors, computer

screens, or other electrical equipment

• Control these External Noise Sources by:

– Shielding and filtering
– Remove or turn-off the noise source
– When using DC instruments, integrate each measurement for an integer number of
power line cycles. The line cycle noise will “average out” when the integration time
is equal to an integration number of power line cycles.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Johnson Noise

measurements.

• In any resistance, thermal energy produces the motion of charged

particles. This charge movement results in Johnson noise.

• The formula for the voltage noise generated: V

rms  4kTRB
where k=Boltzmann’s constant, T= temp in K,
B=noise bandwidth in Hz, R=resistance of sample in ohms

• Reduce by:
1. reduce the measurement bandwidth –digital filtering (averaging
2. reduce temperature of the device
3. reduce the sample resistance (usually not practical)
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Use a Nanovoltmeter to Measure Voltage Drops

When measuring the resistances of conductors or other low power
materials, very small voltages are measured, typically in the microvolt and
nanovolt range.

To measure these very small voltage drops, use a sensitive voltmeter such
as a nanovoltmeter.
Verify the product specifications to
Model 2182A Nanovoltmeter make sure the measurement
resolution and accuracy will be
able to perform the sensitive

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Review of Measurement Considerations for Measuring

the Resistivity of Conductors Voltmeter

• Make a 4-wire measurement to eliminate the L

measurement accuracy Sample

Current Source

• Use an offset compensation technique to

eliminate voltage offsets and reduce noise:
Current Reversal Method or Delta Method

• Eliminate external noise sources and use line

cycle integration.

Continued……

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

• Reduce Johnson noise by reducing the

measurement bandwidth, averaging Vrms  4kTRB
of the sample

Model 2182A
• Use a sensitive voltmeter, such as a Nanovoltmeter
nanovoltmeter, to measure the low
voltages

• Use a low noise, bipolar current source

that can perform current reversals Model 6221 AC Model 6220 DC
+ DC Current Current Source
Source
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Measuring the Resistivity of Insulators

(high resistance)

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

The resistance of an insulator

is measured by: Sample
I Resistance

1. Applying a voltage to the

sample for a specified R
time period Voltage
Source
V A Ammeter

2. Measuring the resulting

current

V
R
3. Calculating the resistivity
using Ohm’s Law and
geometrical considerations
I
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Volume Resistivity of Insulator

LO
Volume Resistivity is a measure of the leakage
current directly through a material. A

HI
Test Procedure for Volume Resistivity:
1. Please sample between 2 electrodes of area (A). Electrode t

Sample
2. Apply potential difference (V) between the 2
HI
electrodes.
Voltage
Source
3. Wait specified time (60 seconds) and measure LO
current (I) using sensitive ammeter.

V A

4. Calculate resistivity based on the area of the
electrodes and thickness of the sample (t). Units
are Ω-cm. I t
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

I

Electrode
HI HI V w
 
L

Sample w A
LO
I L
LO

insulator.

Test Procedure for Surface Resistivity:

1. Place two electrodes a known distance (L) apart on sample.
2. Apply potential difference (V) between electrodes.
3. Wait specified time and measure current (I) with ammeter.
4. Calculate resistivity using width (w) of material and distance (L) between the
electrodes. The units are ohms or ohms per square.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

• Improper Measurement Instrumentation

• Electrification Time

• Test Voltage

• Background Currents

• Electrodes and Geometrical Considerations

• Electrostatic Interference

• Humidity
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Measuring the resistance of insulators usually involves measuring current in the

1E-9 and 1E-12 current range.

When measuring very small current, <100nA, it is important to use a sensitive

ammeter, such as an electrometer or picoammeter.

Voltage Source:

The 6517B has 1E-15 A

Sensitivity!

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Time Dependency of High Resistance Measurement

•Apply 250V
•Wait 60s 250V Step Response of Antistatic Bag
•Measure Current
•Calculate Resistivity

At 5 s, current=
1.5x10-10

At 60 s, current=
2.5x10-12

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Test Voltage

applied voltage.

• In general, as the voltage increases

V
the resistivity decreases 
• Sometimes the voltage may be varied intentionally to determine
the voltage dependence of an insulator.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Background Currents

When measuring very high resistances, background currents can cause

erroneous readings. Background may be due to:
•charge stored in the material
•static or triboelectric charge
•piezoelectric effects

Background currents can be equal or greater than the current stimulated by the
applied voltage.

If background current is same polarity as measured current, then resultant

current reading will be much higher than the true value.

If background current is the opposite polarity, these unwanted currents may

cause a reverse polarity current reading. This can cause the calculated
resistance to be negative!

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Alternating Polarity Method

Positive Voltage
Applied and
Exponential
Current Measured.

Negative Voltage
Applied and
Exponential
Current Measured.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Electrodes

Electrodes for use on an insulating material:

• Has good contact to the material: paint on electrodes or use flat metal
plates with conductive rubber

• Should be much lower resistance than the sample and should not
contaminate the sample

• Enables easy calculation of resistivity from the geometrical

considerations
Where:
Aelectrode

ρ=Volume Resistivity
R A=area of electrode on sample
t t=thickness of sample
R=measured resistance V/I
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Fixtures
Electrodes
Top Electrode

Test
Sample
HI
HI Voltage
Ammeter A Source
LO
LO 0V Ring
Guard

Volume resistivity is a measure of the leakage Center Electrode

current through the material, between the top
electrode and center bottom electrode. The Model 8009
outside ring electrode is guard.
Resistivity Test Chamber
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

The Surface Resistivity is

measured by placing two Electrodes
electrodes on the surface of the 0V Guard
test sample, applying a potential Test
difference between them, and Sample
measuring the resulting current. R
HI HI
Notice the surface resistivity (R) Ammeter A
is measured between the bottom
Center Electrode and the Ring LO LO
Electrode.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Model 65 High Resistivity Measurement Package

8009
Resistivity
Test Fixture

6524
Software

6517B Electrometer/
Voltage Source

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Shield

HI
HI
Ammeter Voltage
Sample Source
LO
LO

Connect Shield to LO terminal of Picoammeter or Electrometer

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Review of Measurement Considerations for Measuring

the Resistivity of Insulators

1. Use a picoammeter or electrometer for low current measurements

2. Use the same electrification time for each test to compare results

3. Use the same applied voltage for test comparisons Model 6517B

currents

7. Use an environmentally controlled room

Model 8009
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Measuring the Resistivity of

Semiconductors

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

The two most common methods for measuring the resistivity of

semiconductor materials are the

Both of these methods use a 4-wire method to eliminate both the

lead resistance and the contact resistance from affecting
measurement accuracy.

Photo courtesy of Lucas Signatone

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

4-Point Collinear Probe Method

Test Procedure:
HI LO
1. Place probes in center of wafer.
2. Source current from probes 1 to 4.
3. Measure voltage between probes 2 and 3.
4. Calculate resistivity: HI LO
V Measure

 V
Voltage
Source Current

Between 2
tk From 1 To 4 And 3
ln 2 I
4-Point
Where: Collinear
ρ= volume resistivity (ohm-cm) 1 2 3 4
Probe
V=voltage measured between 2 and 3
I=source current (A)
t = sample thickness (cm) Wafer
k=correction factor
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

2. Voltage offsets (use current reversal method to reduce)

3. Instrumentation

4. Issues with high resistance materials

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Test Set-Up Showing Circuit Resistances

Source Current
HI LO

HI LO
V RC=Contact Resistance
RL1 RL2 RL3 RL4 RS=Semiconductor Resistance

1 2 3 4
RC1 RC2 RC3 RC4 Only the voltage drop due to RS2 is
V
measured by the voltmeter.
RS1 RS2 RS3
Measure Voltage
Between 2 And 3
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Instrumentation for Mid-Range Resistances

(100mohm to <1Mohm)

Use a standard DC current source (bipolar to perform current reversals) to

force the current and a DMM to measure the voltage drop.

OR

Use a SourceMeter which can source the current and measure the voltage
drop.

ALSO

Photo courtesy of Lucas Signatone

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Photo courtesy of Lucas Signatone

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

• Use a 4-point collinear probe that has excellent isolation between

the probes to avoid leakage current errors
• Use a current source with high output impedance (1E14) to avoid
• Use a current source with a built-in guard to reduce the effects of
shunt capacitance
• Use voltmeters with high input impedance (1E14 ohms)
• Use shielding to avoid errors due to electrostatic interference
• Use differential electrometer method to avoid issues of common
mode current

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Common Mode Current Errors When Measuring High R

Current
Source
HI RC

AC

LO

Voltmeter Common
Mode
Current
V RV
i When the resistance of the
RIN
Contact AC sample, RS2, becomes on the
Resistance same order of magnitude as
HI LO
1 4
the isolation spec (input LO to
2 3
chassis) of the current source
RC1 RC2 RC3 RC4
and voltmeter, then common
RS1 RS2 RS3 mode current will flow
affecting the measurement
Sample Resistance accuracy.
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Making Differential Four-Point Probe Measurements

(For Very High Resistances >1Mohm)
HI LO

Voltmeter

V
HI LO

X1 X1
Buffer Buffer
HI LO HI LO

1 2 3 4
RC1 RC2 RC3 RC4

RS1 RS2 RS3

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

van der Pauw resistivity is a 4-probe

Force
technique that involves applying a current Current
and measuring a voltage using four small
contacts on a circumference of a flat,
1
arbitrarily shaped sample.
4 2
Test Procedure:
•Force Current (I) on adjacent terminals Measure
3 Voltage

•Repeat measurements around sample

•Calculate resistivity
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

van der Pauw Resistivity Method

• vdp configuration is
useful for measuring
Force
very small samples Current

1
B
• Easy to measure Hall
voltage using an 4 2
electromagnet to apply Measure
3 Voltage
the B field
V

• Force I and Measure V Hall Configuration

on opposite terminals
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

V5

1 2 1 2 1 2 1 2
V3 V7

4 3 4 3 4 3 4 3

V1

V6

1 2 1 2 1 2 1 2
V4 V8

4 3 4 3 4 3 4 3

V2

A series of 8 measurements are performed around the periphery of the sample

to compensate for offsets and are combined mathematically to compute the
resistivity.
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

1
Sample

Use a switch matrix to 4 2

automatically switch the current
3
source and voltmeter between all
four terminals of the sample.

1 2 3 4
Choose a switch system that will 1
Source

technique – current reversals – to

eliminate voltage offsets due to 3
the switch. V Voltmeter

Use same configuration to 4

Switch Matrix
measure the Hall voltage using
an electromagnet.
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Model 7065 Hall Effect

Card Configuration

Matrix Card that switches

current source and
voltmeter to the 4 terminals
of the sample.

Has unity gain buffers on

the card to avoid problems
with isolation or needing to
use an electrometer to
measure the voltage drops

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Using Four SMUs to Measure High R Samples

Model 4200-SCS with 4 SMUs and 4 preamps
– Input impedance >1016 
– Accurate low current sourcing, pA SMU1 i SMU2
I Source
– No leakage errors due to V Measure
mechanical switches 1 2
– Includes software to automate V Difference
4 3
measurements and calculate
resistivity SMU4 SMU3
Common V Measure

Model 4200-SCS Semiconductor

Characterization System

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Making Good Measurements on High Resistance

Samples
• Use electrostatic shielding to minimize electrical interference
– Shield the DUT and all sensitive circuitry
– Use shielded cabling
– Connect the shield to the low terminal of the system

• Use guarding to reduce the effects of leakage current in system

– Guarded current source
– Guarded voltmeters
– Use triax cable instead of coax cable

• Allow sufficient settling time

– Source I and measure V as a function of time to determine appropriate
settling time
– A diamond sample can take several minutes for settling

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Settling Time of a 1012Ω Resistance Sample

Wait at least
20 seconds
for a settled
measurement

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Summary

Use an appropriate method for measuring the

resistivity. The method will depend on if the material is
a conductor, insulator, or semiconductor.

Apply the proper measurement techniques to avoid

measurement errors.

A GREATER MEASURE OF CONFIDENCE | www.keIthley.com

Reference Materials
www.keithley.com

Edition

Application Notes: Hall effect, van der

Pauw method, four-point collinear probe
method, insulator resistivity, low voltage
measurements, low current
measurements, etc.

White Papers: Delta method, current

reversal techniques, etc.
A GREATER MEASURE OF CONFIDENCE | www.keIthley.com