Lake Shore Cryotronics Measure Ready M91 FastHall, Руководство пользователя

Страница: 19 / 195

9
|
www.lakeshore.com
1.5.5 Hall
Measurement
Optimization The measured components of mobility, Hall voltage, and resistance (for resistivity),
are captured in much the same way. An excitation current is passed through the
sample creating a voltage proportional to the desired quantity. In the case of Hall
voltage, magnetic field is also a necessary part of the process. In the ideal case, the
desired voltage is large compared to unwanted voltages, measurement errors, and
background noise. In reality, this is seldom the case. In many materials the unwanted
signals can be as large as, or larger than the desired voltage.
This section identifies several factors that must be considered when optimizing a Hall
measurement to separate the desired signals from the unwanted signals. It then
references various sections within the manual that explains these factors. Many of
the factors involve significant trade offs in sample preparation complexity,
equipment cost, and measurement time. This section also highlights many of the
features of the MeasureReady™ M91 FastHall™ measurement controller to help
optimize Hall measurements for a wide variety of samples, along with the dominant
error sources that make them necessary.
1.5.5.1 Sample Type and Structure
A simple, square van der Pauw sample structure is the most common structure, and it
is adequate for most materials. Section 1.5.6.1 describes how some more complex
van der Pauw structures can reduce uncertainty especially when it is difficult to
control the exact contact size and location. Similar information for Hall bar samples is
given in section 1.5.7.1. The M91 supports two of the most common sample types:
van der Pauw and Hall bar. There are significant advantages and disadvantages to
each type as described in section 1.5.6.4 and section 1.5.7.4.
1.5.5.2 Current Reversal
Hall voltage measurements suffer from all of the unwanted voltages present in
typical resistance measurements, in addition to those that arise in the presence of
magnetic field (section 1.5.5.4).
Unwanted contributions present in all measurements of this type include
thermoelectric voltage, offset voltage and offset current. Thermoelectric voltage (V
th
)
is generated any time there is a thermal gradient across dissimilar metals. It is almost
always present in the sample contacts, as well as system wiring and connections.
Offset voltage (V
os
) in the voltmeter is the voltage read on the meter when the inputs
are shorted (0V). One other unwanted measurement contribution is offset current
(I
os
) in the current source. This is the current produced when the source is set to zero
output. I
os
creates an unwanted voltage when it passes through the sample’s
resistance:
The desired voltage is proportional to the excitation current magnitude and polarity.
The unwanted signals mentioned above are not proportional to current. Current
reversal provides a straight forward method for separating the desired signals from
the undesired signals.
1.5.5.2.1 Nernst Effect Voltage (V
N
)
If a temperature gradient exists across the sample, then electrons tend to diffuse
from the hot end of the sample to the cold end of the sample. This diffusion current is
affected by magnetic field, producing a Hall voltage. The phenomenon is known as
the Nernst or Nernst-Ettingshausen effect. The resulting voltage is designated V
N
and
is proportional to magnetic field, but not to external current.