Last Update: 12/23/99

Analysis of Polymer Samples Using a Direct Insertion Probe and EI Ionization

By Steve M. Colby

Presented at ASMS, Palm Springs, CA., June 1997

INTRODUCTION

Quality control and process monitoring in the production of polymers requires a quick and simple method of analysis. We demonstrate the analysis of polymers using a direct insertion probe and the 5973 MSD. Emphasis is placed on the ability to identify impurities and the correct monomer composition. Average analysis time is under 5 minutes including sample preparation.

The manufacture of polymers requires careful monitoring and product characterization. This is often done via chromatography and may include the pyrolysis of the product material. These analyses are difficult to perform during production because of the time required for chromatography. Impurities or other problems in manufacture may therefore not be discovered until after significant material and time has been lost. Analysis via direct insertion probe and EI mass spectrometry can be performed within 5 minutes of sample collection from the process stream. This allows for relatively quick monitoring of manufacturing conditions. We describe the use of a new high temperature probe compatible with a bench top mass spectrometer and demonstrate the mass spectral information that can be obtained.

Experimental

A new microprocessor and computer controlled direct probe is used with an HP 5973 MSD mass spectrometer. The probe is capable of heating to 450 oC at rates of over 1000 degrees per minute. The probe control system is interfaced directly with the HP ChemStation software. This increases the reproducibility of data, since the probe temperature program can be coordinated with the mass spectrometer's data acquisition. It also allows the probe temperature program to be saved with the data files for improved QC and GLP procedures. In addition, the probe can be programmed to ramp at rates between 1 and 500 degrees per minute. Up to 14 ramp steps can be defined. The probe software incorporates a "Hold" feature that allows the user to stop a temperature program for a indefinite period. This unique feature allows the user to respond to the data as it is being collected and optimize acquisition conditions during the analysis process. Figure 1 shows the main screen for the probe control program. It includes a graphic illustrating the entered temperature profile.

PC Screen

Figure 1. Probe Control Screen

Ethylene/Vinyl Acetate Copolymer samples were purchased from Scientific Polymer Products, Inc. Copolymers with compositions of 9, 28, 50, and 70 percent Vinyl Acetate were dissolved in Toluene to form solutions of 1ug/uL. 1 uL aliquots were placed in probe vials for use in each analysis. The vials were tapped so that most of the sample was collected at the bottom of the vials. Toluene was evaporated by placing the vials in a heating block at 70 oC. It should be noted that the polymers could have been analyzed directly without the use of a solvent. This, however, would have reduced our ability to introduce small quantitative samples into the mass spectrometer. Samples were ballistically heated in the mass spectrometer to 450 oC. Ballistic heating resulted in a heating rate of approximately 1000 degrees per minute. Mass spectra were recorded for 3.53 minutes. Although, the data below shows that all useful information was obtained in less than 2 minutes.

Results

The total ion signal recorded for each of the four samples is shown in Figures 2A through 5A. Each sample produced a unique and reproducible profile. The reproducibility achieved is illustrated in Figure 6. This data includes 4 analyses of the 9% Vinyl Acetate sample. While there is variation in the total ion profile, it is always distinguishable from the signals obtained from the other samples. The variations may result from several factors.

The most critical aspect of reproducibility is the heating of the sample. The sample material may experience a range of temperatures as a function of its position within the sample vial. The top of the vial is influenced by the temperature of the mass spectrometer source and therefore can be at a different temperature than the bottom of the vial. If sample material is distributed at several location along the length of the vial, different parts of the analyte will experience different temperatures during the temperature program. This can even result in additional unwanted peaks. The heating of the probe tip is reproducible within 2 to 3 seconds, however, small variations in the thickness of the vial and the precise contact points between the glass and stainless steel of the probe can also effect the sample temperature significantly. We have found that using the same vial improves reproducibility. We are therefore investigating possible improvements in the vial manufacturing process.

Figures 2-5. A. Total ion profiles; B. Extracted Ion Chromatograms (figures 2=9%; 3=28%, 4=50%, 5=70%)

The total ion profiles shown in Figures 2A through 5A clearly indicate that a variety of poorly resolved species are detected when the probe is heated. These species include analyte components as well as products resulting from the thermal degradation and pyrolysis of the sample. The exact identity and ratio of the thermal degradation products is a function of the original polymer composition. Prediction of the signal intensity involves a statistical analysis of original monomer ratios as well as the expected fragmentation patterns. We have not yet attempted these calculations. Instead, our purpose is to show that the original polymer composition can have a dramatic effects on the data obtained and that this data can be used as a quality control tool.

Figure 6. Reproducibility of Total Ion Profile For 9% Sample

More information can be extracted from the total ion profiles using the HP ChemStation data manipulation software. "Extracted ion chromatograms" and background subtraction enables us to identify specific masses that correlate to most of the components that contribute to the total ion profiles. Extracted ion chromatograms are shown in Figures 2B through 5B. The masses selected are chosen to isolate individual thermal degradation products. As can be seen, the intensity of the detected species varies dramatically as a function of the polymer composition. Ions that correlate to all of the important features of the total ion profiles are easily identified. This greatly simplifies the interpretation of the data and distinguishes preferable regions of the spectrum for background subtraction. This is illustrated in Figure 7 wherein the small peak at 0.65 min (339 amu) from Figure 3B has been isolated and simplified through subtraction of the signal on either side of the peak. Such data processing techniques are necessary for the interpretation of multi-component electron impact mass spectrum.

Figure 7. Background Subtraction of a Small Component Identified Through Extraction Ion Signal. A. Raw Mass Spectral Data. B. Background Subtracted.

Conclusions

We have demonstrated that data indicating the relative composition of Ethylene/Vinyl Acetate copolymers can be obtained quickly using a direct probe and the HP 5973 MSD. The data manipulation capabilities of the HP ChemStation software are required to interpret multi-component electron impact mass spectra. The results indicate that simplified quality control procedures should be possible using the above techniques. Further work will be done to examine the sensitivity of this approach for the detection of impurities.

Name:		Organization/Country:
Email:		Phone/fax: