What types of new compounds were added and where did the new data come from?
What are NIST's quality assurance criteria?
How do NIST98 and NIST02 compare from a statistical perspective?
What other MS techniques can the NIST02 EI spectra be used with?
Are there any journal articles describing NIST's spectrum evaluation process?
Are there any journal articles relating to search algorithms used in the NIST Search Program?
Is there a demo program available?
How can the NIST Search Program v2.0 be integrated with Agilent's ChemStation(tm) software?
These EPA Superfund Statements of Work (SOWs) cite the NIST/EPA/NIH Mass Spectral Library as a suitable for the requirements of the method.
Low Concentration Organic Analysis Analytical Method (olc32d.pdf)
Analytical Method for the Analysis of
Trace Concentrations of Volatile Organic Compounds (som10d-trace.pdf)
Analytical Method for the Analysis of
Low/Medium Concentrations of Volatile Organic Compounds (som10d-lmvoa.pdf)
Analytical Method for the Analysis of
Semivolatile Organic Compounds (som10d-svoa.pdf)
These EPA RCRA methods cite the NIST/EPA/NIH Mass Spectral Library as a suitable for the requirements of the method.
Semivoltatile Organic Compounds by GC-MS (8270c.pdf)
Voltatile Organic Compounds by GC-MS(8260b.pdf)
Q:What types of new compounds were added and where did the new data come from?
New spectra were added for:
-new commercially available compounds including drugs and intermediates
-replicates for important compounds
-derivatives (mostly silyl and acyl)
-chemical weapons related compounds (precursors and decomposition products)
These are some of the sources of new spectra. They were all checked by NIST.
| 14,000 | Japan AIST/NIMC Collection | Commercially available common organic compounds |
| 6,976 | Russian Academy of Sciences
Institute of Petrochemical Synthesis |
Mostly Derivatives (silyl, acyl) |
| 7,182 | NIH | Synthetic analogues of natural compounds, drugs, drug metabolites, and their intermediates |
| 2,400 | Asinex | Synthetic compounds |
| 1,746 | Berlin University (Loewis) | Compounds with simple functional groups and their derivatives |
| 1,735 | NIST | Commercially available common compounds, pesticides, drugs |
| 1,200 | Institute of Elementorganic Chemistry | Synthetic compounds, some N-containing cycles |
| 1,022 | Eastman Chemical Company | Commercial and synthetic compounds and silyl derivatives |
| 749 | Spectra from the source "RARE" | Common organic compounds |
| 692 | CA Lab Services | Pesticides & Herbicide |
| 549 | University of Missouri | Common organic compounds |
| 406 | Verifinn (Finland) | Chemical weapons and precursors |
| 348 | HD-Science (UK) | Silyl derivatives of drugs |
| 204 | Susan Richardson | Halogenated organic compounds |
| 145 | Australia (Antolasic) | Common compounds, drugs |
| 138 | Military Institute of Chemistry and Radiometry (Poland) | Warfare agents |
| 19 | Koch Institute, Berlin (Melchert) | TMS Derivatives of "Vitamin E" family compounds. |
Q: What are NIST's quality assurance criteria?
A: New spectra must be complete in every sense. Chemical structures and other chemical information are added, the evaluation process is documented and the data are collected on calibrated instruments.
This is a rough outline of the criteria and process of spectrum evaulation at NIST
Related Chemical Information
-formula
-chemical name and synonym
-chemical structure
-CAS number
-MW consistency
General Rules
-confirm molecular ion and important fragment ions
-origin of [M-X] ions and the most intense peaks
-low mass region
-deduce origin of errors
Compare with Other Spectra
-replicates and isomers
-homologous series
-derivatives
-spectra in other collections
Spectrum Selection
-completeness
-no spurious peaks
-no saturation
Q: How do NIST98 and NIST02 compare from a statistical perspective?
A: The differences are clear. There are many new spectra covering
larger molecules along with better chemical name and CAS number data. Many
spectra were rerun in both the main and replicates libraries so the quality
of spectra has also increased substantially.
| NIST98 | NIST02 | ||
| total spectra | 129,136 | 175,214 | +35.7% |
| spectra with a CAS number | 90,311 | 134,949 | +49.4% |
| spectra with a unique CAS number | 69,031 | 107,105 | +55.2% |
| chemical structures | 107,829 | 147,350 | +36.6% |
| chemical names and synonyms | 255,234 | 440,764 | +72.7% |
| median peak count per spectrum | 79 | 99 | +25.3% |
| average peak count per spectrum | 96 | 111 | +15.6% |
| spectra with less than 20 peaks | 12% | 5% | |
| spectra with less than 10 peaks | 2% | 0.5% |
The NIST02 main library increased from 107,886 compounds to 147,370
compounds.
91,586 spectra from the NIST98 main library
1,331 spectra from the NIST98 replicates library
54,183 new spectra
The NIST02 replicates library increased from 21,250 compounds to 27,844
compounds.
14,050 spectra from the NIST98 replicates
library
7,378 spectra from the NIST98 main library
6,416 new spectra
Excluded from NIST02 (nearly all excluded spectra were rerun):
8,652 spectra from the NIST98 main library
5,869 spectra from the NIST98 replicates library
Q: What other MS techniques can the NIST02 EI spectra be used with?
A: Most data come from quadropole mass spectromteters.
1) Spectra measured on magnetic sector and TOF instruments can be searched against NIST02 with good results.
2) Searching with ion trap spectra can be productive but the results need to be considered thoughtfully.
NIST compiled this table searching ion trap spectra of pesticides. You
can see the results are compare well.
| quadropole | ion trap | |
| anthracene-d10- | 87 | 87 |
| chlordane | 78 | 87 |
| chlorothalonil | 95 | 92 |
| chloropropham | 95 | 94 |
| chrysene-d12 | 87 | 79 |
| dimethoate | 83 | 72 |
| diphenylamine | 99 | 93 |
| ethion | 76 | 79 |
| methidathion | 95 | 82 |
| methoxychlor | 86 | 61, noisy data |
| permethrin | 91 | 75 |
| o-phenylphenol | 100 | 96 |
| phosmet | 84 | co-elutes with trifluralin |
| pirimiphos methyl | 98 | 90 |
| terbufos | 98 | 85 |
| trifluralin | 94 | 90 |
Ion trap spectra of small molecules with oxygen, nitrogen, etc., require attention. Small heteratom molecules are known to produce mass spectra that differ between quadropole and ion trap instruments. The differences relate to ion intensities and product ions.
Library searching is difficult when the ion trap spectrum is not unique or the molecular ion is small or nonexistent..
Ion trap spectra of small molecules (MW 50-225) more often than not
were found in NIST98.
| acetic acid, 1 methylethyl ester |
| acetic acid, butyl ester |
| butanoic acid, 3-oxo, methyl ester |
| 1-butanol |
| 1,4-cyclohexanedimethanol |
| 1-butanol |
| 1,4-cyclohexanedimethanol |
| 1,4-dioxane |
| ethanol, 2-mercapto- |
| ethanol, 2-phenoxy- |
| ethyl acetate |
| glycerin |
| hexanal, 2-ethyl- |
| 2-hexanone, 5-methyl- |
| methylene chloride |
| N-methyl-N(2hydroxyethyl)carbamic acid, methyl ester |
| 2-propenoic acid, ethyl ester |
| propananoic acid, 3-ethoxy, ethyl ester |
| triethylamine |
| p-xylene |
| 1-butanol |
| 2-butenal |
| ethanol, 2-butoxy- |
| hexane |
| methane, diethoxy- |
| pentane |
| 1-propanol |
| propanpoic acid, 2-methyl, 2-methylpropyl ester |
| 1,4-cyclohexanedimethanol |
| ethanol, 2-ethoxy- |
| ethyl ether |
| hexane |
| octane |
| 1-propanamine |
NIST has or shall proceed with the following steps to better accomodate
ion trap data.
1) Include ion trap spectra of small molecules in the NIST MS library
2) Modify the library search algorithm with respect to ion trap data.
3) Assess variations in spectra from ion trap instruments.
Q: Are there any journal articles describing NIST's spectrum evaluation process?
A: The Critical Evaluation of a Comprehensive Mass Spectral Library, J Am Soc Mass Spec, 1999, 10, 287-289
Q: Are there any journal articles relating to search algorithms used in the NIST Search Program?
A:
Optimization and Testing of Mass Spectral Library Search Algorithms for Compound Indentification, J Am Soc Mass Spec, 1994, 5, 859-866
Chemical Substructure Indentification by Mass Spectral Library Searching, J Am Soc Mass Spec, 1995, 6, 644-655
Estimating Probablilities of Correct Identification From Results of Mass Spectral Library Searches, J Am Soc Mass Spec, 1994, 5, 316-323
Q: Is there a demo program available?
A: Yes. The demonstration software download is about 9 MB. Click here to download.
Q: How can the NIST Search Program v2.0 be integrated with Agilent's ChemStation(tm) software?
A: The distribution CD has a folder Support.cs with a file menu_mod.wri
which describes the several macros used to run the NIST Search Program
v2.0 from the ChemStation menus. The macros are also in the folder.
Click to return to the FDM NIST02 page
Click to return to the FDM home page
ChemStation is a trademark of Agilent Technologies.