EXPERIMENT 2
GAS CHROMATOGRAPHY WITH A MASS SELECTIVE DETECTOR - (GCMSD)
PURPOSE
(1) To illustrate the power and limitations of GCMSD. In this experiment, a mass-selective detector (MSD) is used as the detector for a capillary gas chromatography column. This detector will provide the mass-to-charge ratio for compounds which elute from the column and which have been ionized prior to detection.
(2) Data from the GC-MSD experiment will be analyzed using a procedure known as cluster analysis. By examining the factor analysis results of the MSD data plotted in two dimensions (factor 1 versus factor 2) compounds with certain common characteristics appear in groups, or clusters, in the 2-D space. This idea can be used to do compound-class identification of unknown compounds. In addition, comparison of data obtained over a range of m/e values (intensity versus m/e value; called a mass spectrum) can be compared to a library of data. This can identify a compound unambiguously.
INTRODUCTION
Mass Spectrometry is an analytical technique that can reveal specific, characteristic, structurally related information about a compound. Most often, the questions that attract attention to mass spectrometry are qualitative in nature. For example, "Have I successfully synthesized a C19 sterol," or "What is the structure of the by-product from my synthesis reaction?" or "Is the compound that I isolated from plasma really norepinephrine?" Mass spectrometry can usually answer these questions.
The compound for mass spectroscopic analysis is injected into a high vacuum where the molecules can move freely in the evacuated space. Positive ions are commonly produced by electron ionization (EI), electron impact on the gaseous molecules of the sample. The most important ion sought in the mass spectrum is the molecular ion (the ionized, intact molecule) because it is a direct indicator of the molecule weight of the compound. Yet during the process of ionization considerable excess energy may be transferred to the molecular ion, which, depending on its stability (related to its structure), may decompose into various fragment ions. The resulting fragment ions define subsets of atoms which may be related to functional groups or structural components of the original molecule. The array of fragment ions represented by peaks in a mass spectrum is often called a fragmentation pattern. While an indication of molecular weight is usually sufficient to answer questions that will confirm the presence of a given compound, it is much better to compare the major peaks in the experimental spectrum to those in a reference spectrum of the authentic compound. Careful study of the fragmentation pattern is especially important when the mass spectrum is to be used to elucidate molecular structure or to distinguish between two or more structural isomers.
While some fragmentation of the sample molecules is necessary for structure evaluation, many compounds are so effectively fragmented by EI that no appreciable population of molecular ions survives to give an indication of the molecular weight in the mass spectrum. Chemical ionization (CI), involving ionization of the sample molecule by collisions with ions of a reagent gas (for example, the ions produced by electron ionization of methane), is a more "gentle" ionization technique which often produces abundant protonated molecular ions. The resultant spectrum provides a good indication of molecule weight but may have insufficient fragmentation for functional group and structural evaluation.
Of course, in order for mass spectrometry to be helpful the ions that are studied must originate from a pure source of the sample of interest, for to consider ions that are unrelated to the compound could be worse than having no mass spectral data at all. It is essential to insure that a sample contains no material that will produce unexpected ions which could interfere with interpretation. Thus, the combination of gas chromatography and mass spectrometry (GCMS) is often the method of choice. The gas chromatograph can separate the compound of interest from other potentially interfering compounds and present it to the mass spectrometer in purified form over a time interval sufficient to obtain the mass spectrum.
Per unit quantity of sample, mass spectrometry perhaps delivers more information concerning a molecule than does any other spectroscopic technique. Although mass spectrometry rarely solves a problem "single-handedly", it often provides the critical complimentary data that finally "crack" an identification problem. For a complete identification of a sample compound 0.1 to 1 micrograms of the sample may be required. If all that is needed is detection and verification of the presence of an already-identified compound the sample requirements are less severe (sometimes requiring only femtomole quantities).
Although mass spectrometry has been used successfully in the determination of many compounds (usually having a molecular weight of less than 1,000), one consistent obstacle is whether a compound of interest is amenable to vapor phase analysis. However, techniques such as field desorption (where a sample molecule is desorbed from the surface of a probe or electrode as an ion when it enters a high electric field), and chemical derivatization to increase thermal stability and volatility, extend the applicability of mass spectrometry to increasing numbers and types of compounds.
Finally it should be noted that there are three basic types of mass sensitive instruments encountered commonly in analytical chemistry laboratories. The first type is the magnetic sector mass spectrometer (see Figure 1).
Figure 1. Schematic of sector magnet mass spectrometer
The ion stream from the source is passed by a wedge-shaped magnet. The ions curve in response to the magnetic field and separate into several ions streams based on their individual trajectories in response to the magnetic field. At any given magnetic field strength (or, accelerating potential) only one stream has appropriate trajectory (based on it m/e value) to successfully survive the curved "track" to the detector. All other ions will collide with the walls of the track or with the detector slits and will not reach the detector to be recorded. It is common to sweep the magnetic field strength at constant accelerating potential to bring the different ion streams to the detector. The magnetic field sweep can be accomplished very quickly and thus an entire mass spectrum, covering a wide m/e range, can be recorded in a short time frame.
The second type is the quadrupole device which form a class of non-magnetic mass spectrometers and mass selective detectors (see Figure 2).
Figure 2. Schematic of quadrupole design
Our HP 5971A Mass Selective Detector is a quadrupole device. It employs a combination of dc and radiofrequency (rf) potentials as a mass "filter". The quadrupole typically consists of 4 cylindrical, parallel rods (10-25 cm in length) situated symmetrically in a square arrangement. Rods diagonally opposite each other are connected, in pairs, to dc and rf generators. Positive ions extracted from the ions source are accelerated into the quadrupole along the longitudinal axis of the the four rods. The ions are influenced by the combined dc and rf fields. In order for an ion to reach the detector it must traverse the quadrupole without colliding with any one of the metal rods. For any level of rf/dc voltage only ions of a specific m/e avoid collision and reach the detector. The entire mass spectrum is obtained as voltages are swept from a pre-established minimum to a maximum. In our MSD the rf/dc voltages are stepped in a manner that corresponds to 0.1 amu m/e jumps.
Lastly, there are time-of-flight devices which record the time it take for an accelerated ion to travel from the ion source to the detector (see Figure 3).
Figure 3. Schematic of time-of-flight mass spectrometer
Procedure:
This section includes a number of questions. Answers to these should be included in the report. These may be placed in either the experimental or discussion sections, as considered appropriate.
Supplies
1) Known Mixtures (in Pentane)
Alkanes Alkenes Alcohols Ketones Octane 1-Octene 2-Methyl-1-butanol MIBK Nonane 1-Nonene 1-Hexanol 2,4-Dimethyl-3-pentanone Decane 1-Decene Cyclohexanol 2-Heptanone Undecane 1-Undecene 1-Heptanol Cyclohexanone Dodecane 1-Dodecene Benzyl alcohol Acetophenone Tridecane 1-Tridecene 1-Octanol 1-Phenyl-1-propanone2) Unknown Mixture of 3 compounds
3) Test Solution (Decane in Pentane)
I. Instrument Start-up and Introduction
1) Start-up Instrument.
- Start an instrument log-book entry. Enter today's date, time, reason for using the instrument.
- Record the Carrier Gas (He) cylinder pressure and feed pressure (40-50 psig.).
- Record the Column Head Pressure (Detector B gauge) (80 kPa).
- Use the GC's built-in timer and the soap film flowmeter, to measure the Inlet Vent and Septum Purge flows. Adjust the flows using the TOTAL FLOW and SEPTUM PURGE valves as necessary The target values are
Inlet Vent = 55 ± 2 ml/min.
Septum Purge = 3 ± 0.5 ml/min.
- Record the actual values set in the log-book
NOTE: Please record anything else of significance that is done to the instrument (septum change, column switch, etc.) or any observations about instrument operation (peaks tailing, software errors, etc.).
2) Practice Injection Technique Load Method "ISOTEST.M" [M]ethods - [L]oad.. "isotest"
Run this method [M]ethod - [R]un..
- Enter a data file. Use "ISOTEST.D"
- Press the [Run Method] button. Overwrite the existing data file.
The software will give you a dialog box signaling that the instrument is ready.
- Rinse the syringe with 5 volumes of pentane. - Rinse the syringe with 5 volumes of the Decane in Pentane Test Solution - Load the syringe with 1 ul of the mixture - Inject the mixture.
The injection should be a smooth quick motion. The syringe should be removed quickly and smoothly after injection.
- Press [START] on the GC immediately after removing the syringe.
- Repeat this process until the retention times for two successive injection agree to within 2% of the average.
3) Examine the method for cluster analysis data collection - Load Method "GCMSEXPT.M" from the 5971 - Instrument1 Top Screen [M]ethods - [L]oad.. "GCMSEXPT"
- Edit the Method from the 5971 - Instrument1 Top Screen [M]ethods - [E]dit Entire Method.
Many dialog boxes will be presented. Most of these should be left unchanged (Use the "OK" button). The purpose here is to gain familiarity with the instrumentation. Please consult with the TA if any questions arise concerning any of the information. Use the software help feature to answer the questions concerning these parameters. These should be included in the report. Please make a note of the following critical parameters.
Solvent Delay = 1.50 min. - What is the purpose of this parameter?
MS Scan Parameters Mass Range: Low = 50 High = 250 Scans/sec = 3.3
- What does these vaules signify? How are they related?
Column Information
Note: this is found on the Inlet Pressure Programs dialog box, but there is no pressure program in use for this method. Column Length = 30 m Column Dia. (i.d.) = 0.25mm Flow (ml/min.) = 1.08
Using the inlet vent (split flow) value you measured when starting the instrument, calculate the split ratio.
- What does this mean?
Temperature Information Inj. A Setpoint = 250 C Det. B Setpoint = 280 C
Oven Program Init. Temp Setpoint = 80 C Init. Time = 1.00 min Final Temp. = 200 C Rate = 20_ C/min. Final Time = 1.00 min Next Run Time = 8.00 min
Select Reports LibSearch Report
Library Search Parameters Library Name: nbs75k.l
Library Search Report Options Spectrum to Use: Apex
What is the significance of this setting?
4) Enter the name of the first data file. [M]ethods - [Run]..
"C:\HPCHEM\1\DATAL\YYMMDD-N"
Where YY = the last 2 digits of the current year "94" for 1994.
MM = the numeric digits representing the current month "01" for January.
DD = the current day-of-the-month, must be two digits.
N = a sequential number for the data files, start with 1.
also enter operator name and the sample mixture name in the appropriate boxes.
II. Analyze the samples
1) Run the method [M]ethod - [Run].. - [Run Method] The software will give you a dialog box signaling that the instrument is ready.
- Rinse the syringe with 5 volumes of pentane.
- Rinse the syringe with 5 volumes of the mixture.
- Load the syringe with 1 ul of the mixture.
- Inject the mixture and press [START] on the GC.
The software will then present a dialog box concerning the solvent delay. DO NOT override the solvent delay. After clearing this dialog box, the chromatogram will begin to plot, as soon as the solvent delay time is over.
When the run is complete the software will process the chromatogram and the associated mass spectral data. It will then output a library search report.
- Perform one injection for each of the known mixtures and the unknown mixture. Be sure to increment the data file name for each of the injections. Change the Sample Name for each mixture as well.
You may notice that some of the known mixtures, and maybe the unknown mixture as well, show more peaks than expected.
- Why is this?
III. Analyze the Data
1) Load Data Analysis Module
[D]ata Analysis - [M]ain Panel
2) Start the MSCONV macro
- Type in: macro "MSCONV",go
This will execute a series of instructions that will allow the mass spectra for each of the compounds injected to be converted to a character form that is readable by Matlab. Matlab will be used to peform the cluster analysis.
3) Load the first data file
[F]ile - [L]oad
- Start with the first data file "YYMMDD-1.D"
It is critical that you process the files in order. This will setup a sequential numbering of the compounds from the earliest eluting alkane, Octane, through the latest eluting unknown compound. The sequential numbering of the compounds will be needed in order to interpret the cluster analysis
4) Convert the mass spectra for Matlab
- Type in: MSOUT
5) Repeat steps 3) and 4) for each data file.
6) Exit the Data Analysis Module
[F]ile - [E]xit - [Yes]
7) Start Matlab
The Icon for Matlab is in the Applications Program Group.
9) Start the GC/MS Cluster Analysis macro
- Type: gcms
10) Operate the Cluster Analysis macro This macro is interactive. Follow the instructions in it and read the narrative as they occur. At the end Matlab will print three Cluster plots and a list of the four eigenvectors used in these plots.
11) Exit the macro
- Type: Ctrl-C
12) Exit Matlab
- Type: quit
13) Print selected mass spectra
After completing the cluster analysis, begin reviewing the results. Consider the questions below to determine what additional information is needed. It will be useful to have some of the mass spectra for the complete analysis.
- To print mass spectra, return to the GC/MS Data Analysis module as performed previously. After loading a data file, mass spectra for individual compounds can be displayed and printed. To call up the mass spectrum for a compound, move the mouse cursor over the peak and then double-click the right mouse button. This will open the spectrum in a window below the chromatogram window.
- Experiment with this feature before any printing. It should be observed that the mass spectra will change depending upon where on the peak it is viewed.
- Why is this?
Print the spectra as close as possible to the top or "Apex" of the peaks. This is the spectra that was used for cluster analysis.
To print:
[F]ile - [P]rint - [T]IC & Spectra
Results:
- Identify the unknown compounds. Include no less than three independent pieces of evidence to substantiate the identification. Acceptable evidence includes: library search results, retention time match to known compound, cluster analysis results, interpretation of mass spectra.
Questions/Discussion:
1) Discuss the results of the cluster analysis, some important questions to consider are:
How many clusters can you identify in the plots?
By comparing the clustering to the mass spectra, is it possible to find justification for the patterns?
Why are some of the clusters very tight and others less so?
What real factors are effecting the cluster analysis?
2) Would it be possible to identify any of the unknown compounds using the GC alone? What does the mass spectrometer add to the analysis?
References:
1) Junk, G. A.; Richard, J. J., Anal. Chem., 1988, 60, 454.
2) Skoog, D. A.; West, D. M.; Holler, F. J. Fundementals of Analytical Chemistry, Sixth edition, Saunders College Publishing, 1992, Chaperts 26, 27, 29, 30.
3) Watson, J. T., Introduction to Mass Spectrometry: Biomedical, Environmental, and Forensic Applications, Raven Press, 1976.