LIBS Analysis of Copper Sample
Goal:
To acquire elemental analysis data for a copper sample
Method:
LIBS
Hardware Used:
LIBS2000+ spectrometer
Big Sky Nd:YAG 50 mJ 1064 nm laser
LIBS Sample chamber with 125 mm focal length lens
OOILIBS software
Acquisition Parameters:
Data was acquired for different levels of surface cleaning (0 to 10 laser shots before acquiring data) and for samples where 10 laser shots were averaged. All of the spectra were collected with a 1.5 microsecond delay between when the laser fired and when data acquisition began.
Measurement Mode:
LIBS
Experimental Conditions:
The copper sample was placed in the LIBS sample chamber and the probe was focused until the maximum intensity was observed from the plasma. Three replicates at different locations on the sample were acquired using the following data acquisition parameters:
No surface cleaning (OOILIBS has a cleaning function that allows the user to fire several cleaning shots to clean contaminants from the surface of the sample before acquiring spectral data.)
10 cleaning shots
3 cleaning shots
10 laser shots averaged with no cleaning shots
Single laser shot resulted in surface pitting of the sample.
Results:
The average spectral data for the three replicates acquired using varying data acquisition parameters are shown in the figure below. The spectra were offset to facilitate comparison of spectral features.
For the bottom spectrum where no cleaning shots were fired, it appears that there is a significant amount of material on the surface of the sample as evidenced by the region between 500 and 600 nm. This region of the spectrum is filled with broad bands and elemental lines unique to the uncleaned surface. In the remaining spectra, when the samples are cleaned 3 or 10 times or 10 laser shots are averaged at the same location, the spectral data look similar. Even though the top spectrum where 10 laser shots at the same spot were averaged is similar to the condition where the surface is pre-cleaned with 10 shots before acquiring data, the top spectrum represents the average of 10 individual laser shots at the same location so it includes a contribution from the uncleaned spectrum shown at the bottom of the plot.
The elements identified in one of the replicate measurements for each set of data acquisition parameters are shown in the table. As anticipated based on the spectral data, the number of elements identified decreases with increasing cleaning shots. For example, numerous magnesium (Mg) and calcium (Ca) lines are observed when no cleaning shots are used to remove contaminants from the sample surface (first two columns in the table). With surface cleaning, the number of lines observed for these elements decreases. The data shown in the final two columns of the table for 10 shots averaged together without any surface cleaning still shows the Ca and Mg lines since the surface was not cleaned before the measurements.
Note that the library used to identify the spectral lines in the OOILIBS software is a partial library containing the persistent lines of the elements. With a more complete library, like the MIT Wavelength Tables containing thousands of spectral lines, additional elemental lines would be identified. As expected, several copper lines were observed for each measurement. In the case of the uncleaned sample, a larger number of lines was detected. It is important to note that these samples were measured in the presence of air. The sample chamber was not purged to remove the air. For this reason, the oxygen and nitrogen lines observed result from ambient air in the plasma in addition to any oxygen and nitrogen present in the sample.
