NY.gov Portal State Agency Listing Search all of NY.gov
D E C banner
D E C banner

Disclaimer

The New York State Department of Environmental Conservation has added a link to a translation service developed by Microsoft Inc., entitled Bing Translator, as a convenience to visitors to the DEC website who speak languages other than English.

Additional information can be found at DEC's Language Assistance Page.

Appendix C - Volatile Organics Monitoring Methods Comparison

VOC Data Comparison Study Between Collocated Ambient Air Samples Collected In Summa Canisters And On Adsorption Tubes

New York State
Department of Environmental Conservation
Division Of Air Resources
Bureau Of Air Quality Surveillance
Special Sampling Section

March, 2000

1. Introduction

The New York State Department of Environmental Conservation (DEC), Division of Air Resources (DAR), Bureau of Air Quality Surveillance (BAQS), analyzes ambient air from various sites around New York State for volatile organic compounds (VOCs). Several different ongoing projects employ different air sampling techniques depending on the individual project goals. This presents BAQS with an opportunity to compare collocated Toxics data from air samples that were collected using different sampling techniques. This paper will focus on two, namely, air samples collected on Carbotrap 300 6628 adsorbent tubes and air samples collected in SUMMA canisters. The GC/MS instrumentation and analytical methods that were employed for the analyses of the adsorption tubes and SUMMA can samples were different, however, they are not considered to have had any confounding influence on the results. Any observed variability between collocated samples collected on adsorbent tubes and in evacuated six liter canisters are assumed to be due to the differences in the air sampling techniques.

Data collected at the end of calendar year 1997 and the first half of calendar year 1998 was used in this comparison study.

2. Sampling Sites

Three sites were selected for the co-located sampler comparison study. These sites were chosen because they are (1) within urban areas with moderate to heavy traffic flow; (2) they are within areas routinely accessible to DEC field personnel; and (3) because VOC levels have historically fallen in the single to tens of ppbV range. The first site was on the rooftop of the Atrium Mall in Troy, New York. This site is within the geographic region of the Hudson River Valley just north and east of the city of Albany and has a population of 55,000 people. The other two sites are located downstate within the boundaries of the Freshkills landfill in Staten Island, New York. These two sites are the District 2 Garage, which is located in the northeast corner of the landfill, and the Richmond Avenue site which is located in the southeast corner of the landfill. Both sites are just downwind of the landfill proper.

The Freshkills landfill was opened on April 16th, 1948 and is the worlds largest dump. It is the only remaining landfill in the city of New York and receives the residential trash of all 5 boroughs of NYC at a rate of approximately 14,000 tons per day. The landfill is a 3000-acre site with 175 foot mounds. New York State Governor George Pataki and New York City Mayor Rudolph Giuliani agreed to set December 31, 2001 as the closing date for the Fresh Kills landfill.

3. Data Acquisition Of Samples Collected In Canister

SUMMA canister data was acquired using an Entech 1-805-527-5939. 7000 pre-concentrator and a Finnigan 408-433-4800. Magnum GC/MS ion trap run in scan mode. Liquid nitrogen was used as cryogen for the pre-concentrator and the sub-ambient GC method. A SGE BP-1 capillary column (50 meter, 0.22 mm id, 1.0 micron film thickness) was used for compound separation.

4. Data Acquisition Of Samples Collected On Adsorption Tubes

Adsorption tube data was acquired using a Perkin-Elmer ATD-400 pre-concentrator with a Hewlett-Packard 201-599-5737. 6890 GC and 5972 mass spectrometer run in SIM mode. Similar to the canister method, a SGE BP-1 capillary column (50 meter, 0.22 mm id, 1.0 micron film thickness) was used for compound separation.

5. Data Comparison

VOC data were compared using a Microsoft Excel macro written by Avi Teitz of EPA's Region 2. The grades assigned to the data comparisons, using the GC/MS data as the reference, included: Excellent, Good, Fair, Poor, and Fail. Grading criteria were based on percent differences for concentrations ranging between 1.0 to > 5.0 ppbC and the absolute differences for concentrations ranging from 0.1 to 1.0 ppbC. Comparisons for compounds falling into the lower concentration ranges were given more latitude since it is more difficult to measure their concentrations reproducibly.

VOC comparison criteria used in Excel macro

Table 1. VOC comparison criteria used in Excel macro.

The minimum detection limits of the Toxics compounds using, both, tube and canister analytical methods are listed in Table 2 below. The 5972 MDLs were calculated using 37 calibration standard runs from actual analytical batches over the 1997 and 1998 calendar years while the Trap MDLs were calculated using 5 calibration standards from actual analytical batches. It should be noted that the EPA's method for determining detection limits is based on the precision of repeated calibration standard injections.

Table 2
Compound H.P. 5972 Magnum Ion Trap
Vinylchloride 0.24 0.33
Dichloromethane 0.62 0.25
Chloroform 0.23 0.33
1,2-dichloroethane 0.10 0.21
1,1,1-trichloroethane 0.28 0.19
Benzene 0.35 0.20
Carbontetrachloride 0.09 0.21
1,2-dichloropropane 0.09 0.18
Trichloroethene 0.14 0.17
1,1,2-trichloroethane 0.14 0.16
Toluene 0.27 0.11
Tetrachloroethene 0.32 0.18
Chlorobenzene 0.16 0.20
Ethylbenzene 0.13 0.14
m/p-Xylene 0.29 0.18
o-Xylene 0.18 0.08
1,3-dichlorobenzene 0.13 0.31
1,4-dichlorobenzene 0.11 0.33
1,2-dichlorobenzene 0.13 0.30

Toxics MDLs in ppbV. HP 5972 numbers reflect 37 standard runs over 6 to 7 months of sampling. Magnum numbers reflect 5 consecutive standard runs.

6. Summary For Troy

Troy Summary
Rating > 5 ppbC 3-5 ppbC 1-3 ppbC 0.5-1 ppbC 0-0.5 ppbC
Excellent 0 0 2 8 259
Good 0 1 2 6 6
Fair 0 0 3 7 3
Poor 0 1 3 2 0
Fail 0 0 0 0 1

frequency distribution by concentration range for Troy site

7. Summary For District Garage

District Garage Summary
Rating > 5 ppbC 3-5 ppbC 1-3 ppbC 0.5-1 ppbC 0-0.5 ppbC
Excellent 0 1 3 18 226
Good 0 3 6 5 5
Fair 0 0 3 5 1
Poor 0 1 7 1 0
Fail 0 0 0 0 0

District Garage summary

8. Summary For Richmond Avenue

Richmond Ave. Summary
Rating > 5 ppbC 3-5 ppbC 1-3 ppbC 0.5-1 ppbC 0-0.5 ppbC
Excellent 0 0 1 6 251
Good 0 1 5 12 9
Fair 0 0 6 6 1
Poor 0 0 4 2 0
Fail 0 0 0 0 0

frequency distribution by concentration range for Richmond Avenue site

9. Discussion

Overall, the comparisons between canister samples and tube samples were quite good with most falling into the excellent and good categories. Only one comparison failed. This failure was at the Troy site on 01/23/98 where toluene was found at 0.61 ppbv (can sample) and 1.42 (tube sample) ppbV. These results differ by a factor of 2.3. The failure criterion for comparing two results at that level is >0.8 ppbV absolute.

The Excel macro output shows that most analytes were detected at levels below 0.5 ppbV. However, there were some exceptions such as benzene, toluene, ethylbenzene and ortho, meta and para xylenes (BTEX). Toluene was the only compound found within the 3.0 to 5.0 ppbV range by either method. None was found greater than 5.0 ppbV.

In most cases the adsorption tube results showed lower levels of VOCs than canister results. This bias could be due to problems such as tube breakthrough, calibration bias or passive loses by diffusion. It is the author's opinion that any losses are most likely to be the result of passive diffusion. It is not uncommon for sample tubes to sit for several days in the field sampler before field personnel retrieve and properly cap both ends. Cans samples do not suffer passive diffusion since whole air trapping is not an adsorption mechanism.

This study and the resulting comparisons give the monitoring staff a very confident outlook on the change from the adsorbent based air sampling method to the canister based air sampling method. No significant problems with the resulting data will be observed as a result of the change and, in fact, the move will save significant manpower and reduce the error bars around future VOC monitoring data.

References

  1. SUPELCO INC., Supelco Park, Bellefonte, PA, 16823-0048, USA, Phone 1-800-247-6628.
  2. Scientific Instrumentation Specialists, Inc. P.O. Box 8941, Moscow, Idaho, 83843.
  3. Entech Instruments Inc., 2265 Ward Ave, Unit A, Simi Valley, Ca., 93065, Phone 1-805-527-5939.
  4. ThermoQuest/Finnigan, 355 River Oaks Parkway,San Jose, Ca., 95134, Phone 1-408-433-4800.
  5. SGE Inc., 2007 Kramer Lane, Austin TX.,78758, Phone 1-512-837-7190.
  6. Perkin-Elmer, 761 Main Ave., Norwalk, Ct., 06859-0241, Phone 1-800-762-8288.
  7. Hewlett Packard Company, West 120 Century Rd., Paramus, N.J. 07652, Phone 1-201-599-5737.

Acknowledgments

Brian Aho for Magnum Ion Trap operation and review of the report.

John Miller for the HP 5970 operation, the "Excel" macro execution and the author of the comparison report.