Five Verified Examples

PNNL references, single-block import, versioning and RefIMP retrieval.

1. Install the PNNL library and find methanol

  1. Download the NIST/PNNL example package containing three public-domain records from the NIST/PNNL Generator.
  2. Press Install library… and open the package. In Entity / material, choose methanol. Search for CAS 67-56-1 and select the matching spectrum.
  3. Check for 15,345 numeric points, wavenumber in cm⁻¹, spectrum type IR and k on the Y axis. Move the pointer over the plot to read the coordinate.
Methanol IR reference based on 2017 PNNL data.
Methanol IR reference based on 2017 PNNL data.

2. Compare ethanol and 2-propanol with separate Y scales

  1. Stay in the PNNL library, clear Entity / material and the search field, then select 2_propanol-k-pnnl-2017 and ethanol-k-pnnl-2017.
  2. Switch from overlay to top_bottom. Read Reference_A and Reference_B using each panel’s own Y scale.
  3. Try difference and read the difference over the common range. Restore the X view with Reset zoom.
The two PNNL references maintain independent Y scales.
The two PNNL references maintain independent Y scales.

3. Import the JCAMP k block into a personal library

  1. Extract ethanol-k-single.jdx from the ZIP into a folder. It is the second block of the 2017 PNNL JCAMP for ethanol, with the original samples intact. NIST source record.
  2. Select that folder in Import spectra… → Add directory…. Leave the JCAMP parsing fields blank and press Preview spectrum.
  3. Check for 15,347 samples and absorption_index. Select or create a user-owned destination. Import and open the record from that library.
Preview spectrum plots the k block using the coordinates from its headers.
Preview spectrum plots the k block using the coordinates from its headers.

4. Create library versions with an update and a removal

  1. Select the PNNL ethanol reference and open Update / remove reference…. Set a new ID and use Fork library.
  2. In the copy, select ethanol. Set a new ID and name, document the source and choose the CSV of PNNL data pairs extracted from the ZIP. Use CSV format, columns 1 and 2, header 1 and a comma delimiter. Run Update reference.
  3. In a third version, use Remove reference on the same reference. Check that the third library contains two spectra and the original PNNL library still contains three.
Version v3 contains the 2-propanol and methanol references. After selection, the record shows v3.
Version v3 contains the 2-propanol and methanol references. After selection, the record shows v3.

5. Find a RefIMP candidate and compare it with Session data

  1. Open the refimp_full_training_ir demonstration Session supplied with Generator 1.1.0. In the application, Import Session… accepts its Session ZIP. For the training corpus, use package tum.refimp.ftir.training.2022.gxsrl.zip. The example package is prepared with the supplied adapter for FTIR_MP_train_test_spectra.csv from TUM, which checks the source fingerprint. Open the Session and install the package through the GUI. The RefIMP Generator 1.1.0 (ZIP) and the 23 parts of the managed package with ID tum.refimp.ftir.training.2022, with the command to reassemble them, can be downloaded from the IR Spectroscopy manual.
  2. In IR candidates, choose ir.refimp.item13.ir.candidates, then select the first row. Verify refimp-train-00013 in tum.refimp.ftir.training.2022.
  3. Select ir.refimp.item13 in Session spectrum to compare the transmission measurement. Switch to the processed Dataset ir.refimp.item13.ir.processed and use top_bottom to read the two Y scales.

The measurements and corpus are from Weisser and Pohl (2022). The curves shown come from the RefIMP example and are licensed under CC BY-SA 4.0. The complete corpus is available from the source. RefIMP source · License.

The processed Dataset and the RefIMP reference appear on independent Y scales. Weisser and Pohl, CC BY-SA 4.0.
The processed Dataset and the RefIMP reference appear on independent Y scales. Weisser and Pohl, CC BY-SA 4.0.