NMAM Methods Update

A laboratory response has answered concerns about technologically outdated and problematic methods.

• By Stephanie M. Pendergrass, Dr. Donald D. Dollberg
• Oct 01, 2005

THE National Institute for Occupational Safety and Health publishes the NIOSH Manual of Analytical Methods (NMAM), a primary reference that has been a reliable source of analytical methods used in occupational health laboratories throughout the world for more than 25 years. The NMAM, despite several revisions and supplements over the years, contains many methods that are technologically outdated. In addition, various NMAM users have noted that some of the methods are problematic or totally nonfunctional.

In an effort to address this situation, the NIOSH NMAM editor conducted a survey of the various NMAM clients (Schlecht and Cassinelli, 1997). These survey results identified a number of problematic methods requiring either re-evaluation or new method development. The resulting laboratory research followed the NIOSH "Guidelines for Air Sampling and Analytical Method Development" (Kennedy et al, 1995). Although the NMAM survey identified methods based upon various chromatography techniques, the majority of the methods targeted for evaluation were developed using gas chromatography (GC).

At the start of this project, each GC method was prioritized based upon client need using the results of the NMAM survey. The methods, identified by their priority ranking, are listed in Table 1. In addition to correcting any problems associated with the individual methods, a number of related goals were incorporated into the method development research. For example, those methods that still specified packed columns were revised (through laboratory evaluation) to use the appropriate capillary column. Capillary column chromatography allowed the limit of detection (LOD) and limit of quantitation (LOQ) to be lowered for each analyte evaluated.

For some analytes, new or additional desorption efficiency (DE) studies were required to meet the current method development criteria (Kennedy et al, 1995). Others required DE studies at levels substantially lower (5x LOQ to approximately 0.1x Recommended Exposure Level or Permissible Exposure Level) than in the original method in order to more realistically evaluate current occupational exposure levels. Finally, after a review of all of the methods, it was determined that most lacked sufficient storage stability data as required by current method development guidelines (Kennedy et al, 1995). Thus, this article will endeavor to provide a general comparison of the new capillary column-based methods versus the original, outdated, and/or problematic packed column methods.

Materials and Methods
Analytical Standards
All standards used in these studies were obtained from Aldrich Chemical Company (Milwaukee, Wis.). All analytes were 99 percent pure with the exception of: 1,2-dichloroethylene (98 percent), 1,1,2-trichloroethylene (97 percent), p-tert-butyltoluene (95 percent), o-xylene (98 percent), 2-heptanone (98 percent), 3-heptanone (98 percent), mesityl oxide (98 percent), 5-methyl-3-heptanone (97 percent), and isophorone (97 percent).

Desorption Solvents
Solvents for this study include: carbon disulfide (low benzene grade, 99.9+ percent, Aldrich Chemical Co., Milwaukee, Wis.); methylene chloride (99.9+ percent, Burdick and Jackson, Muskegon, Mich.): isopropanol (HPLC/pesticide grade, Burdick and Jackson, Muskegon, Mich.); and methanol (pesticide grade, Fisher Scientific Co., Fair Lawn, N.J.).

Solid Sorbent Tubes
Solid sorbent tubes used in this research include: Anasorb CSC (400/200 mg, # 226-09), Silica Gel (150/75 mg, # 226-10), Anasorb CSC (100/50 mg, # 226-01), Anasorb 747 (140/70 mg, # 226-81A), XAD-2 (400/200 mg, # 226-30-06), Anasorb CMS (150/75 mg, # 226-121, SKC, Inc., Eighty-four, Pa.) and ORBO-92 (carbon molecular sieve, 160/80 mg, # 2-0362, Supelco, Inc., Bellefonte, Pa.).

Analytical Parameters
The gas chromatographic analyses for each method evaluation was performed using a Hewlett-Packard 6890 GC, equipped with a flame ionization detector (FID), and a HP6890 series autosampler (Hewlett-Packard Co., Avondale, Pa.). The following Restek7 fused silica capillary columns were used in place of the outdated packed columns: 30-m Rtx7-1 (0.32 mm ID, 1.00 micrometer), 30-m Rtx7-35 (0.53 mm ID, 3.00 micrometer), 30-m Stabilwax7 (0.32 mm ID, 1.00 micrometer), 30-m Stabilwax7-DA (0.32 mm ID, 1.00 micrometer), 30-m Rtx7-200 (0.32 mm ID, 1.00 micrometer), and a 30-m Rtx7 502.2 (0.32 mm ID, 1.80 micrometer) (Restek7 Corp., Bellefonte, Pa.). In all cases, the carrier gas used was helium (2.8 to 5.0 mL/min, with flow rate dependent on the capillary column employed and/or the requirements of the method development) and the injection volume was 1 mL. Both split and splitless injections were employed, depending on chromatographic need.

Results and Discussion
The major purposes of re-evaluating the selected methods were: (1) to update the old packed column approach in favor of the current capillary column technology; (2) improve, when possible, the limits of detection (LOD) and limits of quantitation (LOQ); (3) improve desorption efficiency (DE) through minor solvent changes; (4) reassess method recovery/storage stability data by conducting such studies at more representative concentrations found in current occupational environments, and, when appropriate, collect storage stability data for a 30-day time period.

Additionally, during each initial packed column method development, the DEs were determined separately for each analyte. Because most of the DEs do not represent "real world" occupational exposure scenarios where a number of different analytes may be collected on a single solid sorbent tube, DE studies were conducted with each sorbent tube containing multiple analytes. To achieve these objectives, some of the methods required an alternative solid sorbent media and/or a different desorption solvent choice. Completely new method development was required for a few of the outdated methods to satisfy the current method development criteria. Also, during these evaluations it was possible to combine some single analyte methods into methods containing an analogous series of analytes (i.e., methyl ethyl ketone was added to the Ketones I method [NMAM 2555], and trichloroethylene was added to the Halogenated hydrocarbons method [NMAM 1003]).

Priority 1 Methods
Three of the four priority 1 methods identified by the NMAM survey have been completed. The first method, NMAM 1606 (acetonitrile), represents an entirely new method development, with the parameters of interest shown in Table 2.

The second method development of this group to be completed was NMAM 2005 (Nitroaromatic compounds). The analytes in this method included nitrobenzene, o-nitrotoluene, m-nitrotoluene, p-nitrotoluene, and 1-chloro-4-nitrobenzene. A 30-m Rtx7-5 Amine fused silica capillary column allowed for the resolution and subsequent analysis of the three nitrotoluene isomers, previously not possible with the packed column used in the original method. Both the original packed column method and the new method used silica gel sorbent tubes (150/75 mg) and methanol as the desorption solvent.

The final priority 1 method development completed in the initial phase of the NMAM update project was NMAM 1453 (Vinyl acetate) (Data Chem Laboratories, 2003).

Priority 2 Methods
As was the case with the priority 1 methods, three methods were evaluated in this second group. The first method, NMAM 1501 (Aromatic hydrocarbons), included the following analytes: benzene, toluene, ethylbenzene, o-xylene, m-xylene, p-xylene, cumene, p-tert-butyltoluene, alpha-methylstyrene, beta-methylstyrene, styrene, and naphthalene. Vinyl toluene, an analyte included in the original method, was dropped from the method because a standard could not be obtained. Beta-methylstyrene and the xylene isomers (not resolved in the original method) represent new additions to the method. In order to accommodate all the analytes in this method, two columns were utilized: Stabilwax7 and an micrograms Rtx7-35 fused silica capillary columns--both resolved and separated all analytes. The analytes were divided into two groups (A and B) based upon the column used for the analysis. While the actual DE results for each analyte were slightly improved using Anasorb CSC (Anasorb 747 was also evaluated and performed equally well) as the solid sorbent media using 1 mL of carbon disulfide as the desorption solvent, when compared to the results in the original method, it should be noted this improvement was achieved at significantly lower analyte concentrations

For example, benzene, which had a DE recovery of 96.4 percent at the 88 microgram level, and alpha-methylstyrene, which had a DE recovery of 91 percent at the 687 microgram level in the original method development, had recoveries of 98 percent at the 4 microgram level and 100 percent at the 36 microgram level, respectively, in the updated method. It should be noted that both styrene and naphthalene failed the current laboratory method development guidelines criteria² at the levels evaluated in this method development (5x LOQ to 0.1x REL/PEL). Styrene, which passed the original DE criteria at levels 0.5x, 1.0x, and 2.0x the REL/PEL, was kept in the new method with these results noted. For naphthalene, both the original and current DE data failed to meet the method criteria; hence, the compound was omitted from the revised method.

Significant improvement in the individual LODs was achieved for all analytes in the revised NMAM 1501, ranging from 0.4 microgram/sample for benzene, toluene, ethyl benzene, and styrene to 2.0 micrograms/sample for naphthalene; whereas, in the original packed column method, the LOD was listed as 10 micrograms/sample for all analytes. A 30-day storage stability study, absent in the original method, was completed for each analyte. All analytes with the exception of naphthalene were stable for 30 days, with recoveries ranging from 97 percent for alpha-methylstyrene to 104 percent for toluene.

The second method evaluated was NMAM 2555 (Ketones I). This method includes the following analytes: acetone, methyl ethyl ketone, 2-pentanone, methyl isobutyl ketone, 2-hexanone, diisobutyl ketone, and cyclohexanone. Methyl ethyl ketone (MEK), previously a single analyte method (NMAM 2500), was a new addition to this method. This evaluation focused on an alternative solid sorbent as well as desorption solvent. With Anasorb CMS (150/75 mg), carbon disulfide (1 mL) as the desorption solvent, and using an Rtx-35 fused silica capillary column, MEK can be included in the new Ketones I method. Previous attempts to sample MEK with coconut shell charcoal resulted in reduced recoveries (50-70 percent) because of the interaction of the MEK with the coconut shell charcoal media.

The third method evaluated in this group was NMAM 1403 (Alcohols IV), which included the following analytes: 2-methoxyethanol, 2-ethoxyethanol, and 2-butoxyethanol. All of the analytes were resolved using a Stabilwax7-DA fused silica capillary column in place of the original packed column. Analytes were sampled on Anasorb CSC (100/50 mg) and 1 mL of methylene chloride/methanol (95:5) was used as the desorption solvent. While the actual DE recoveries achieved in this method development were not significantly different from those reported in the original packed column method, it should be noted that the results for the current method development were achieved at significantly lower analyte concentrations. Table 6 highlights these differences.

The fourth and last method evaluated in the priority 2 series was NMAM 1003 (Halogenated hydrocarbons). The analytes included in this method development were as follows: benzyl chloride, bromochloromethane, 1,1-dichloroethane, 1,1,1-trichloroethane, trichloroethylene, chloroform, 1,2-dichloroethylene, tetrachloroethylene (perchloroethylene), carbon tetrachloride, o-dichlorobenzene, p-dichlorobenzene, 1,2-dichloroethane, 1,1,2-trichloroethane, chlorobenzene, 1,2,3-trichloropropane, hexachloroethane, and bromoform. Trichloroethylene, previously a single analyte method (NMAM 1022), was incorporated into the new method because it belongs with this series of analogous halogenated hydrocarbons.

Because of the large number of analytes in this method, the use of both Rtx7-502.2 and Rtx7-35 fused silica capillary columns was required for the resolution and separation of the various analytes present. Two columns were required because no single column evaluated could achieve baseline separation of all analytes, even when separated into two groups of eight. The analytes were divided into two groups (A and B), based upon which column was used for the analysis. An additional reason for the required separation of the analytes into two groups was that previous experience had shown that when more than six to eight analytes are present on a single sorbent tube, the individual DE results for each analyte were reduced. The reduction in DE is independent of the combination of the analytes present.

Analytes were sampled on Anasorb CSC (100/50 mg) and 1 mL carbon disulfide was used as the desorption solvent. Overall, the actual DE results were slightly improved to quantitative levels (95-100 percent recovery) for each analyte studied when compared to the original packed column method. As with previous methods, the results are more significant when it is realized that they were achieved at substantially lower concentrations. Some examples of this improvement in DE recovery at lower analyte concentrations include tetrachloroethylene and o-dichlorobenzene, which had a DE recovery of 96 percent at the 2100 microgram level and 86 percent at the 500 micrograms/sample level in the original packed column method, respectively. Contrast with the new method, where tetrachloroethylene had a DE of 97 percent at the much lower concentration level of 8 micrograms/sample and o-dichlorobenzene had a DE of 98 percent at the low concentration of 19 micrograms/sample. Similar improvements in the LODs for each analyte were achieved in the new method, where they ranged from 0.6 microgram/sample for benzyl chloride to 4.0 micrograms/sample for carbon tetrachloride, as compared to a reported LOD of 10 micrograms/sample for all analytes in the original packed column method.

A 30-day storage stability study (not performed initially) was conducted as part of this method update. All of the analytes with the exception of bromoform were stable for 30 days, with recoveries ranging from 91 percent for bromochloromethane to 103 percent for 1,1-dichloroethane. Bromoform was stable for 14 days (104 percent), but the recovery dropped off significantly to 81 percent after 30 days.

Priority 3 Methods
After completion of the relevant priority 2 methods, subsequent research was directed to the priority 3 methods, which were selected from the methods of a lesser priority in the NMAM survey.¹ Currently, three methods have been completed. The first method evaluated was NMAM 2552 (Methyl acrylate). Table 8 compares the original method with the updated method.

The second method evaluated was NMAM 2537 (Methacrylates), which included the analytes methyl and ethyl methacrylate. Because it was part of an analogous series of methacrylates and analysis has been previously requested on occasion, ethyl methacrylate, not previously included in the original method, was added to this method. A comparison of the original and revised method is shown in Table 8.

The final method evaluated in this group was NMAM 2553 (Ketones II). The analytes included in this method development included 2-heptanone, 3-heptanone, 5-methyl-3-heptanone, mesityl oxide, and camphor.

Priority 4 Methods
The Priority 4 methods included NMAM 2556 (Isophorone), NMAM 1460 (Isopropyl acetate), and NMAM 1618 (Isopropyl ether). The first method evaluated in this group was NMAM 2556 (Isophorone).

In the original method, isophorone was collected on petroleum charcoal sorbent tubes (100/50 mg), desorbed with 1 mL carbon disulfide, and analyzed using a packed column. Whereas in the new method, isophorone was collected on XAD-4 sorbent tubes (80/40 mg), desorbed with 1 mL of diethyl ether, and analyzed on a 30-m Stabilwax-DA fused silica capillary column. With the capillary column, the LOD was improved to 1 microgram/sample in the new method versus 20 micrograms/sample in the original packed column method. A significant improvement in the DE recovery was achieved with the new method, 94 percent at 55 micrograms versus only 86 percent at 849 micrograms in the original packed column method. A 30-day storage stability study was completed for the new method (89 percent at 275 micrograms), while no storage stability study was conducted in the original method.

The second method evaluated was NMAM 1460 (Isopropyl acetate). In both the new and original methods, isopropyl acetate was collected on coconut shell charcoal sorbent tubes (100/50 mg). However, in the new method, when the desorption solvent was changed to carbon disulfide/methanol (99:1), a significant improvement in the DE recovery was achieved, 99 percent at 17 micrograms versus 86 percent at 3750 micrograms in the original packed column method. The use of an Rtx-1 fused silica capillary column allowed the LOD to be lowered to 0.2 microgram/sample in the new method versus 10 micrograms/sample in the original packed column method. In the new method, a 30-day storage stability study was completed (100 percent at 100 micrograms), while in the original method no storage stability study was conducted.

The final method evaluated was NMAM 1618 (Isopropyl ether). In both the new and original methods, isopropyl ether was collected on coconut shell charcoal (100/50 mg) and desorbed with 1 mL of carbon disulfide. Using an Rtx-1 fused silica capillary column, the LOD was improved to 0.2 microgram/sample versus 10 micrograms/sample in the packed column method. An improvement in the DE recovery at much lower levels was achieved (101 percent at 11 micrograms versus 94 percent at 3160 micrograms). Again, a 30-day storage stability study was completed in the new method (94 percent at 615 micrograms), while no results were available in the original method.

Conclusion
The methods identified and evaluated in this project have incorporated the most recently available fused silica capillary column technology. By selection of alternative desorption solvents and improved or new solid sorbent media, combined with the improved resolution afforded by capillary column chromatography, the desorption efficiency (DE) was improved for each analyte evaluated at substantially lower concentrations (approximately 5-10x LOQ to 0.1x REL/PEL). Capillary columns, coupled with more sensitive gas chromatographs and data collection systems, resulted in lower LOD values by a factor of ten- to twenty-fold for most analytes evaluated. Thirty-day storage stability studies, previously lacking in the original packed column methods, were successfully completed to meet current method development criteria (Kennedy et al, 1995).

Finally, additional benefits included the combination of several single analyte methods into related multi-analyte methods. For example, trichloroethylene (previously NMAM 1022) was incorporated into the halogenated hydrocarbons method (NMAM 1003), and methyl ethyl ketone (previously NMAM 2500) was incorporated into the ketones I method (NMAM 2555). Several analyte isomers were added to some methods. Some examples include the addition of o, m, and p-xylene and beta-methylstyrene to the aromatic hydrocarbons method (NMAM 1501). Also, it should be noted that naphthalene was removed from the aromatic hydrocarbons methods because it failed to meet the acceptance criteria for DE and storage stability during method development.² Thus, based upon these results, it can be concluded that the methods discussed in this manuscript have successfully addressed the problems and concerns identified in the NMAM client survey.

The methods described in this text will equip occupational and environmental hygienists with new and/or improved procedures for the sampling of workplace hazards. Analysis of field samples in the laboratory will incorporate the most technologically modern gas chromatography techniques, such as lower detection limits and improved sample recoveries. These new or revised methods are also flexible enough to be applied to the sampling and analysis of any of these chemicals if they were ever employed in a chemical terrorism incident.

This article appeared in the October 2005 issue of Occupational Health & Safety.

References
1. Data Chem Laboratories, Inc. (2003), Vinyl Acetate: NMAM 1453, under the auspices of NIOSH contract # 200-95-2955.

2. Kennedy, E.R., Fischbach, T.J, Song, P., Eller, P.E., and Shulman, S.A. (1995). Guidelines for Air Sampling and Analytical Method Development and Evaluation, Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication 95-117.

3. Schlecht, P.C. and Cassinelli, M.E. (1997). Survey of Analytical Methods: Parts I and II. The Synergist; 26: 8-9.

The authors note that mention of company names and/or products does not constitute endorsement by the Centers for Disease Control and Prevention (CDC). They would like to express their appreciation to Data Chem Laboratories, Inc. for its efforts in developing an improved method for vinyl acetate (NMAM 1453) as part of this project.

This article originally appeared in the October 2005 issue of Occupational Health & Safety.