FILMETRICS F20 - The Baseline Measurement

The Baseline Measurement
The baseline measurement allows the FILMeasure software to take into account the response inher-
ent to the reflectance measurement hardware. It does this by measuring a reference sample and by
taking a “dark” reading. In any optical system there are many components whose characteristics
vary with wavelength (e.g., the output of the light source and the sensitivity of the spectrometer).
However, when reflectance measurements are made, only variations in reflectance vs. wavelength
due to the sample under test are of interest. Therefore, FILMeasure must perform a calibration to
determine the spectral response of the system. This is done by making a measurement of a "ref-
erence" sample that has known reflectance characteristics. Note that it is not necessary for the ref-
erence sample to be the same as the substrate upon which films to be measured reside. The only
purpose served by the reference sample is to permit calibration of the optical system. For example,
it is possible to use a Si wafer as the reference sample and then measure films on GaAs, InP, glass,
plastic, etc.
After the reference measurement is made a dark reading is taken. A non-zero dark level is due to
current leakage inherent to photodiodes, which causes each photodiode in the array slowly charges
up even when no light enters the spectrometer. Thus, in order to make an accurate measurement of
the light entering the spectrometer, it is necessary to subtract this “dark” current contribution. This
is the purpose of the “dark” reading, which measures the magnitude of the dark current for a given
integration time. When a “dark” measurement is made, a spectrum is measured that represents the
signal generated by the spectrometer when a sample of zero reflectance is measured. To simulate a
sample with zero reflectance during a “dark” measurement, a specularly reflecting sample can be
held at an angle with the light source turned on, or in many cases the light source may simply be
turned off momentarily during the dark measurement.
Due to drift in the light source and temperature of the spectrometer electronics, it is a good idea to
take a baseline periodically.