GC-MS Surface Response Data for Instrument Optimization with Categorical Variables

The objective of this example case was to maximize peak height and minimize peak width of 11 iodinated standards analyzed using gas chromatography-mass spectrometry (GC-MS). Once the significant (or most significant) parameters have been determined in a prior screening experiment, another set of experiments must be designed in which a response surface objective experiment is performed to collect data for optimization. In this example case with mixed continuous and categorical variables, surface response experiments were performed using modified Box-Behnken design in which each statistically significant continuous parameter (e.g. split ratio, carrier gas flow rate, oven temperature ramp rate, and capillary column film thickness) was assigned a “low”, “middle”, and “high” value. These were measured with the experimental design for each type of the categorical parameter, inlet liner shape, which had three type options: straight, straight with wool, and Precision^®. The attached CSV files contain the modified Box-Behnken design (‘bbd2.csv’), the raw chromatographic data for the 75 distinct methods (‘Cat Vars Method #.csv’), and the retention times (‘bb2_rts.csv’) for each iodinated standard.