Tobacco Documents Online

Authors: G. D. Byrd Date: September 11, 1989 J. N. Schumacher M. F. Borgerding ----- >K. W. Fowler M. E. Lovette Department: New Products Technology Notebook Pages: None Project No. and Title: 0218, Alpha R&DM, 1989, No. 251 Previous Reports: R&DM, 1987, No. 30 No. of Pages: 51 EBPANSION OF T8E TOBACCO/SMOKE MASS SPECTRAL LIBRARY (TOBACCO.L) OBJECTIVES: • To expand the number of compounds in the existing TOBACCO.L mass spectral library. • To create a spreadsheet for retention times of compounds in TOBACCO.L using standard conditions for both polar and nonpolar GC columns. i SUMMARY: The TOBACCO.L mass spectral library has been expanded to over 2,200 entries and a retention time spreadsheet has been created for many of the compounds. The spreadsheet includes retention times using standard polar and nonpolar GC conditions. The additional compounds were selected based on their occurrence in tobacco, tobacco smoke, and related samples. STATUS: As of September 5, 1989, there were 2,209 entries in the library and spreadsheet. Addition of new compounds and retention times will occur as time permits. Copies of the mass spectral library can be made available on CS80 tape, 3.5" floppy disk, or 20 megabyte Bernoulli cartridge.. Copies of the retention time spreadsheet can be made available on 3.5" or 5.25" floppy disks. NS KEYWORDS: Mass spectral library, mass spectrometry, gas chromatography, retention time spreadsheet, tobacco, smoke 1

This system has proven extremely useful in identifying chemical components in smoke and flavors from new products. We believe that this system could also be useful to others here at R. J. Reynolds R & D. EXPERIMENTAL Gas chromatography. Two columns, one polar and one nonpolar, were used to generate retention times. Both columns were commercial fused silica columns and each was operated under its own specific temperature programming parameters. The polar column was a 30 m x 0.25 mm fused silica capillary column with a 0.25 µm polyethylene glycol bonded film (DBWAX-30N, J&W Scientific, Folsom, CA). The GC oven was temperature programmed as follows: initial temperature initial time program rate final temperature = 37'C = 2 min = 2.5'C/min = 230'C The usual total run time was 95 min. The carrier gas was helium and the column head pressure was approximately 60 kPa using a direct interface to the mass spectrometer. Column back pressure was adjusted such that the retention times of pyrazine, nicotine, and glycerol were 13.2 ± 0.1 min, 44.3 ± 0.1 min, and 62.2 ± 0.1 min, respectively. The nonpolar column was a 60 m x 0.32 mm fused silica capillary column with a 5.0 µm methylsilicone bonded film (DB1- 60W, J&W Scientific, Folsom, CA). The GC oven was temperature programmed as follows: initial temperature = 35•C initial time = 10 min program.rate = 2.5•C/min final temperature = 275°C The usual total run time was 110 min. The carrier gas was helium and the column head pressure was approximately 90 kPa using a direct interface to the mass spectrometer. Column back pressure was adjusted such that the retention times of 2-propanol, toluene, octanal, and nicotine were 12.5 ± 0.1 min, 40.8 ± 0.1 min, 60.0 ± 0.1 min, and 86.8,t 0.2 min, respectively. Mass spectrometrv. The mass spectrometer used in these studies was a Hewlett-Packard HP 5970B "Mass Selective Detector"(Hewlett-Packard Company, Palo Alto, CA). This instrument is a quadrupole mass spectrometer with a scanning range of 10-800 u. Scanning range and rate were selected based on the sample or. standard analyzed.

Data system. The GC/MS system was controlled by a Hewlett- Packard HP 59970C ChemStation using revision 3.1.1 of the Pascal operating system. The mass spectral library editor (2] was used to enter spectra into the mass spectral library which was named "TOBACCO.L". For each spectrum entered the editor records the ten most significant peaks. The significance of a peak is its mass multiplied by its abundance. The library editor can also be used to search unknown compounds against the library. This editor uses a forward search algorithm based on the ten most significant peaks in the spectrum as defined above. Because entries cannot be deleted from the library once they are entered, multiple entries are included in the library for some compounds. However, these do not create problems during the search algorithm. Retention spreadsheet. A spreadsheet of retention times for comounds in the mass spectral library was maintained on an IBM PCt compatible computer using the Lotus i-2-3t'" spreadsheet program, Release 2 (Lotus Development Corporation, Cambridge, MA). Each entry included the compound number (which is the same as in TOBACCO.L), compound name, Chemical Abstract (CAS) Registry Number, molecular weight, and retention times on the DBWAX and DB1 columns. The spreadsheet file is named "TOBACCO.WK1". RESULTS AND DISCUSSION The compounds included in this library are generally tobacco and smoke related. Common solvents and impurities which might be expected in samples were also added. Table 1 summarizes the information in the spreadsheet as of September 5, 1989. The library contains over 2,200 entries but this includes duplicates and some compounds with only a partial identity. Duplicates account for 11% of the total entries and CAS registry numbers are given for 90% of the unique compounds. DBWAX retention times are available for 49% of the entries and DB1 retention times for 19% of the entries. A complete listing of TOBACCO.WK1 is given in Appendix 1 sorted alphabetically by compound name. Listing of compounds. Each entry has a unique compound number that is the same in TOBACCO.L and TOBACCO.WK1. Because of duplicate entries, multiple entry numbers exist for some compounds. It is possible that a compound may be entered twice and named in a different manner each time. In these cases the CAS number should be checked to verify the structure. Most compounds in TOBACCO.WK1 are named using a modified form of the IUPAC name. The modification was necessary to allow for logical alphabetizing of the names. For example, the IUPAC name for acetone is "2-PROPANONE" but is entered as "PROPANONE- 2". Substituents are listed by the primary root first followed by the position and name of substituents. An example of this is "PYRIDINE, 2,6-DIMETHYL-". When multiple functionalities occur 4

in a compound, their position is indicated by inserting the number just before the functionality as in "HEPTANE-2,6-DIONE". Some compounds are listed only by their common name such as "MYOSMINE", "MARIOLIDE", etc. When possible, both names are entered with the common name in parentheses following the IUPAC name. Most of the entries for which no CAS number is listed are only partially identified. In some cases the precise isomer (cis or trans) of the compound is not known. Certain codes that have been used historically at the R. J. Reynolds Tobacco Company have carried over into this library. These codes are listed in parentheses after the compound name. In some cases, only the code is listed for the compound and the exact structure is unknown. The most prevalent codes used are the "S-series" and refer to isolation work done on tobacco and smoke. These codes can be generally explained as follows for the example of entry #60, "SR43C". The "S" designates the division where the work was done. "R" is the code for the person who did the work and "43" is the compound number. "C" indicates that the compound is an isolate from tobacco. A "D" would have indicated that the compound was isolated from smoke. Use of mass spectral searches. The-results from a mass spectral library search should be used with basic skills in interpreting mass spectra to identify a compound. Each component must be reviewed individually to see if the spectrum is logical and consistent with the chromatography of the particular system. Practice has shown that the TOBACCO.L library is faster and usually more accurate in our work than the 10 peak NBS reference library (NBS REVE.L) offered by Hewlett-Packard. Consequently, we tend to search TOBACCO.L first. However, the use of other libraries and other searching algorithms can assist in solving identities when TOBACCO.L cannot. A macro for searching three different libraries and displaying the results was written and is listed in Appendix 2. In addition to the 10 peak searches for TOBACCO.L and NBS REVE.L, this macro offers access to a probability-based matching algorithm [3] that can be used with Hewlett-Packard's NBS REVF.L library. Use of retention times. The retention times listed in TOBACCO.WK1 were produced on systems that were calibrated as described in the experimental section. Practice has shown that retention times will shift in actual samples by as much as a minute. Factors affecting this are solvent, manner of injection (split or splitless), and instrumental variability. This should be kept in mind when matching retention times for unknowns. The primary purpose of the retention times here is to provide more confidence in the assignment of a particular structure. Homogolous series. In cases where a good mass spectral match is obtained but no retention time is available, it is useful to guess what type of retention behavior might be 5

expected. Several homologous series of selected compounds have been extracted from TOBACCO.WK1 and are plotted as retention time versus carbon number. Table 2 lists these compounds and the corresponding plots are shown in Figures 1-8. G. D. rd F. B M"v Submitted: September 11, 1989 Approved: ac.. q/IZ~SS Accepted : ~ O~j DISTRIBUTION Dr. M. F. Borgerding Dr. G. D. Byrd Dr. H. L. Chung Dr. E. L. White Dr. B. M. Lawrence Dr. M. W. Ogden Mr. R. W. Williams a. °Yt . S tm"& J. . Schumacher K ler • n; •; • Dr. G. R. DiMarco Dr. W. M. Hildebolt Dr. R. A. Lloyd Dr. M. E. Stowe Dr. A. W. Hayes Mr. L. A. Lyerly Ms. B. T. Hodge Mr. K. L. Rush Dr. B. M. Gordon Dr. C. R. Ashcraft 6

REFERENCES 1. Schumacher, J. N. and Murphy, J. J., "Tobacco/Smoke Mass Spectral Library", RDM, 1987, No. 30. 2. HP 59970C ChemStation Operator's Manual, Hewlett-Packard Company, Palo Alto, CA, Publication No. 59970-90016, March 1986. 3. MacLafferty, F. W., InterRretation of Mass Svectra, 3rd Edition, University Science Books, Mill Valley, CA, pp.. 232- 233 (1980). 7

TABLE 1: Information Available in TOBACCO.WK1 Spreadsheet as of September 5, 1989. Total Number of Entries = 2,209 Number of Unique Entries = 1,975 Number of Compounds with CAS No. = 1,772 Number of Compounds with DBWAX R. T. = 975 Number of Compounds with DB1 R. T. = 377 8

TABL E 2: Homogolous Series from TOBACCO.WK1 Retention Times°. ALKA NES ALCOHOLS C# COMPOUND DBWAX DB1 C# --- COMPOUND ------------- DBWAX ------ DB1 ------ 1 Methane 1 Methanol-1 3.2 6.3 2 Ethane 2 Ethanol-1 3.9 11.4 3 Propane 3 Propanol-1 6.8 17.6 4 Butane 4 Butanol-1 10.9 -28.3 5 Pentane 5 Pentanol-1 15.5 38.8 6 Hexane 1.5 23.7 6 Hexanol-1 20.4 48.4 7 Heptane 1.7 34.3 7 Heptanol-1 25.8 57.2 8 Octane 2.2 44.3 8 Octanol-1 30.8 65.8 9 Nonane 3.4 53.3 9 Nonanol-1 35.5 73.2 10 Decane 5.6 61.6 10 Decanol-1 41.0 80.0 11 Undecane 9.4 69.3 il Undecanol-1 45.4 86.5 12 Dodecane 13.8 76.6 12 Dodecanol-1 49.6 13 Tridecane 18.9 83.1 13 Tridecanol-1 53.7 14 Tetradecane 22.5 89.2 14 Tetradecanol-1 57.7 15 Pentadecane 29.5 15 Pentadecanol-1 61.4 16 Hexadecane 33.1 100.3 16 Hexadecanol-1 65.2 17 Heptadecane 39.1 17 Heptadecanol-1 18 Octadecane 42.9 18 Octadecanol-1 71.5 19 Nonadecane 47.4 19 Nonadecanol-1 20 Eicosane 51.6 20 Eicosanol-l 78.4 21 Heneicosane 55.3 21 Heneicosanol-1 22 Docosane 59.3 22 Docosanol-1 85.4 23 Tricosane 62.7 24 Tetracosane 66.2 25 Pentacosane 69.7 26 Hexacosane 72.3 27 Heptacosane 28 Octacosane 78.5 29 Nonacosane 30 Tricontane 86.2 31 Hentriacontane 32 Docotricontane 96.2 °Plots of Carbon Number versus Retention Time are shown in Figures 1-8. 9

TABLE 2, continued ALDEHYDES KETONES-2 C# COMPOUND DBWAX DB1 C# COMPOUND DBWAX DB1 --- ------------- ------ ------ --- ------------- ------ ------ 1 Methanal 1 Propanone-2 2.4 11.4 2 Ethanal 1.8 6.2 2 Butanone-2 3.2 20.6 3 Propanal 2.1 11.7 3 Pentanone-2 4.8 30.4 4 Butanal 2.9 21.0 4 Hexanone-2 8.6 40.9 5 Pentanal 6.2 31.6 5 Heptanone-2 11.9 50.2 6 Hexanal 8.9 42.0 6 Octanone-2 17.1 59.1 7 Heptanal 12.0 51.5 7 Nonanone-2 22.3 67.1 8 Octanal 16.7 60.0 8 Decanone-2 27.4 74.4 9 Nonanal 22.0 67.9 10 Decanal 27.6 75.3 11 Undecanal 32.3 82.0 12 Dodecanal 37.6 88.3 13 Tridecanal 42.7 94.4 14 Tetradecanal 47.7 100.3 ACID S METHYL ESTERS C# COMPOUND DBWAX DB1 C# COMPOUND DBWAX DB1 ------ ------ --- ------------- ------ ------ 1 Formic 1 Formate 2 Acetic 24.6 21.0 2 Acetate 2.4 15.1 3 Propanoic 29.1 30.9 3 Propionate 4 Butanoic 4 Butanoate 5 Pentanoic 5 Pentanoate 6 Hexanoic 6 Hexanoate 7 Heptanoic 48.4 7 Heptanoate 8 Octanoic 8 Octanoate 9 Nonanoic 9 Nonanoate 27.4 10 Decanoic 60.1 83.0 10 Decanoate 33.1 11 Undecanoic 64.0 11 Undecanoate 12 Dodecanoic 67.5 12 Dodecanoate 42.7 13 Tridecanoate 14 Tetradecanoate 51.5 NITR ILES ALKENES-1 C# --- COMPOUND ------------- DBWAX ------ DB1 C# --- COMPOUND ------------- DBWAX ------ DB1 ------ 2 Ethane 5.1 10.3 2 Ethene 3 Propane 5.9 17.4 3 Propene 4 Butane 8.6 28.2 4 Butene 5 Pentane 5 Pentene-1 1.4 12.6 6 Hexane 6 Hexene-1 1..6 22.5 7 Heptane 7 Heptene-1 8 Octane 28.4 65.9 8 Octene-1 2.5 43.1 9 Nonane 9 Nonene-1 3.9 52.3 10 Decane 10 Decene-1 6.5 10

L80 I I I 1 I 1 0 0 0 0 0 0 + XVM80 O 2138Wf1N NOB~"J 0 OZ 9L 0 0 0 0 0 0 0 0 ,. + + S3NM)I'ly-u L 3z~nou 0 tL 0 0 + + 0 + + + 8 + + + 0 'r- OZ ~-0* f-09 'r- 09 'r- OL 51016 9793 A t