EPA/600/A-94/045
Field Comparison of Portable Gas Chromatographs with Method TO-14
Richard E. Berkley and Maribel Colon
US Environmental Protection Agency, Research Triangle Park, NC 27711
Jesus Gonzalez and Israel Droz
University of Puerto Rico School of Public Health, San Juan, PR 00936
Jeffrey Adams, Christopher Fortune, and Karen Oliver
ManTech Environmental Services, Research Triangle Park, NC 27709
Daniel R. Coleman, CMS Research Corporation, Birmingham, AL 35244
Clayton Wood, HNU Systems, Incorporated, Newton, MA 02161

-------
         Field Comparison of Portable Gas Chromatographs with Method TO-14

                         Richard E. Berkley and Maribel Colon
          US Environmental Protection Agency, Research Triangle Park, NC 27711
                            Jesus Gonzalez and Israel Droz
           University of Puerto Rico School of Public Health, San Juan, PR 00936
                   Jeffrey Adams, Christopher Fortune, and Karen Oliver
            ManTech Environmental Services, Research Triangle Park, NC 27709
                                  Daniel R. Coleman
                    CMS Research Corporation, Birmingham, AL 35244
                                    Clayton Wood
                     HNU Systems, Incorporated, Newton, MA 02161

ABSTRACT

       A field-deployable prototype fast gas chromatograph (FGC) and two

commercially-available portable gas chromatographs (PGC) were evaluated by measuring organic

vapors in ambient air at a field monitoring site in metropolitan San Juan, Puerto Rico.  The data

were compared with simultaneous grab samples which were collected in six-liter

Summa-polishedฎ canisters and analyzed by method TO-14. Because of fluctuating retention

times, the FGC produced no useable data. High humidity levels may have adversely affected its

performance.  Both commercially-available PGCs performed successfully, and data from twenty

analyses were compared with the reference method.

       This paper has been reviewed in accordance with US Environmental Protection Agency's

peer and administrative review policies and approved for presentation and publication.  Mention

of trade names or commercial products does not constitute  endorsement or recommendation for

use.

INTRODUCTION

       During June 1993 a field study, organized under auspices of US Environmental Protection

-------
Agency, Atmospheric Research and Exposure Assessment Laboratory, Research Triangle Park,




NC, was conducted near San Juan, Puerto Rico to compare data produced by three gas




chromatographs with reference data from samples which were collected simultaneously in six-liter




Summa-polishedฎ canisters and subjected to laboratory analysis by Method TO-14 [1]. The




purpose of this study was to assess the performance of the field-deployable chromatographs.




They were (1) a prototype high-speed (fast) chromatograph (FGC)  developed by Professor




Steven Levine at the University of Michigan School of Public Health in cooperation with HNU




Systems under Environmental Protection Agency Cooperative Agreement No. CR 817123 with




the University of Michigan, (2) the CMS MINICAMSฎ, and (3) the HNU Model 311ฎ. Field-




deployable (portable) gas chromatographs (PGC) offer the advantages that they produce data




immediately, and they can produce more data at lower cost than laboratory-based methods. In




addition, samples analyzed on-site do not undergo lengthy storage pending analysis, as samples




analyzed by laboratory-based methods usually must.




       There are some disadvantages to use of PGCs, the principal one being generally lower




quality performance. PGCs often incorporate less-sophisticated components than are used in




laboratory instruments in order to reduce cost and size. Many operate isothermally, and most use




short columns to decrease analysis times, so they are not able to analyze as many compounds per




run as laboratory instruments. PGCs can be difficult to operate properly because of inferior




components and because they experience ambient temperature fluctuations and other




consequences of deficient shelter which  are not normally encountered by laboratory instruments.




       In previous field studies of this kind, EPA purchased PGCs and evaluated them, or




employed a contractor to do so [2-6]. More recently, an EPA contractor borrowed several




instruments and evaluated them together during a field study [7]. In the multi-instrument  field

-------
study, it was noticed that the number of PGCs which can be evaluated is limited by the number of




competent operators available, which is limited in turn by available space inside shelter at the




sampling site. The predicament of an operator can become intolerable when confronted by




several different instruments of variable and unknown temperament, functioning under the adverse




conditions typical of a field study and without immediate access to technical support.  In fact,




extensive direct assistance by factory technical representatives proved crucial to success when




several different PGCs were evaluated simultaneously.  Operators were overworked, and it was




apparent that one technically-trained person is needed for each kind of instrument. In this study,




each instrument was operated by a factory representative.




EXPERIMENTAL




       The rationale for the study was direct comparison of data produced by different methods,




using TO-14 analysis of canister grab samples as the reference method. Periodically, all




instruments collected samples simultaneously through collocated inlets as a canister was filled




nearby (within 20 cm).  Ambient concentrations were lower than expected. In order to obtain




some data at higher concentrations, samples were collected inside a mobile laboratory, where




concentrations were higher than outside.  Because of this, and because site selection was




constrained by requirements for shelter and line power, the data are not necessarily representative




of conditions in Catano. The purpose of this study was to evaluate the instruments under field




conditions, not the air in Catano.




Site Selection




       Site selection was determined primarily by the current location of the least-transportable




participating instrument; the FGC was located at the University of Puerto Rico School of Public




Health. Secondary consideration in site selection was given to using a site where there was

-------
perception of an air pollution problem.  The most convenient and practical of several sites




considered was a monitoring station operated by the Junta de Calidad Ambiental (Environmental




Quality Board) at the jail in Bayamon, which was within the San Juan Metropolitan Area and




generally downwind of the Catano industrial area.  JCA had two small shelters on site with air




conditioning and electric power. The PGC operators collaborated to rent a large recreational




vehicle to which electric power from a JCA building was connected so it could be used as a




mobile laboratory during the study. A schematic (not to scale) of the area around the site is




shown in Figure 1.




Audit Standard




       Occasionally, each instrument analyzed an audit standard which had been prepared in the




EPA/AREAL laboratory in Research Triangle Park, North Carolina. The audit standard was




certified before the study by analysis with a gas chromatograph equipped with dual flame




ionization and electron capture detectors (GC/FID+ECD). After the study it was analyzed again




by a gas chromatograph with a mass selective detector (GC/MSD). Results of these analyses are




compared in Figures 2 and 3, which include all 41 compounds on the TO-14 target list.  There




was little change in the audit standard between the two analyses. Canister samples of ambient air




collected during the study were analyzed by GC/MSD.




Canister Sampling




       Canisters were filled by holding the valve within 20 cm of the group of PGC inlets and




opening it just far enough to fill the canister while the PGCs were sampling. This procedure could




not ensure that all methods analyzed exactly the same air, but it should produce quite similar




results during calm weather if air is well-mixed with vapors from remote sources.




High-Speed Chromatograph

-------
       The prototype FGC was developed by Levine. Details of its unique features have been




presented elsewhere [8-12].  A small volume of air was isolated in a sample loop, then passed




through a Monelฎ capillary cooled to -120ฐC by vapor which was boiled out of a reservoir of




liquid nitrogen at a controlled rate.  Sample loops with volumes from 0.25 to 5 ml could be used.




After the air sample was flushed into the Monel capillary by a helium stream, the capillary was




placed in line with the column by operating a multi-port valve, and the Monel capillary was




rapidly heated by capacitive discharge.  The 0.25 mm id column was operated isothermally




between 40-200ฐC.  Carrier flow velocity was within the range of 60 to 175  cm/sec.  The injection




bandwidth was 10 msec.  Dual ECD and 10.2 eV PID used an electrometer which was specially




designed to cope with extremely narrow peaks. During preliminary laboratory evaluation with a




flame ionization detector (FID), which has essentially no volume, analyses were completed in less




than 30 s, and excellent separation was demonstrated. Detection limits for TO-14 target list




compounds varied between 1-300 ppb, depending on the sensitivity of the FID to the individual




compound [10].  Using an ECD, separation was slightly less, and it took nearly one minute to




complete a chromatogram, but detection limits for highly halogenated compounds appeared to be




below one ppb [13]. During the present study, the length of the 0.25 mm id column was




increased from 10 m to 30 m in an attempt  to elute benzene after the initial baseline excursion.




Commercially-Available Portable Gas Chromatographs




       Two commercially-available portable gas chromatographs were evaluated in the study,




each operated by a factory representative.  Operators were responsible for calibration,




maintenance, and troubleshooting their own equipment.




       CMS Minicams  The Series 2000 Field MINICAMS was a field-deployable gas




chromatograph which operated on  110 V AC line power.  It measured 25.4 X 30.5 X 21.6 cm,

-------
weighed 8.6 kg, and had an on-board microcomputer which used a custom software package to




control operation and communicate data to an external computer.  The unit used in this study was




equipped with a FED and a solid sorbent preconcentrator.  Samples were collected on a




proprietary carbon-based sorbent bed from an air stream which was drawn through it at 200




ml/min for 260 s. The sample was injected by operating a 3-way valve to place the sorbent bed in




the carrier stream flowing in the opposite direction to the sample stream.  Then it was rapidly




heated to 350ฐC for 40 s.  The 15 m X 0.32 mm fused silica column was coated with 5 ^m DB-1.




During analysis, column temperature was increased ballistically from 40ฐ to 180ฐC at a rate of




329ฐC/min.




       HNU Model 311.  The HNU 311 is a field-deployable gas chromatograph which operates




on 110 V AC line power.  It measures 38.1 X 56.4 X 24.6 cm and weighs 30 kg. An on-board




microprocessor can deliver data directly to an on-board printer or pass it to an external computer




for processing or storage on disk.  The unit used in this study was equipped with both ECD and




PID, but the compounds for which it was calibrated were detected by PID. A continuously-




pumped air stream was sampled periodically for 20 s by diverting the contents of a sample loop




with a multi-port valve. The 25 m X 0.53 mm fused silica column was coated with 1.0 ^m of




NB-30.  Nitrogen carrier flow rate was 15 mL/min and column temperature was 80ฐC.




RESULTS AND DISCUSSION




       The study was carried out during June 22-25, 1993.  The first day was entirely spent in




setting up equipment, connecting utility supplies, and startup. Routine periodic autosampling by




the two PGCs began on June 23 and continued until the study ended on June 25. Periodic




manually-initiated sampling by the FGC was done during daylight hours.




High-Speed Gas Chromatograph

-------
       The FGC was plagued by significant shifts in retention times throughout the study, and its




sensitivity was diminished. This problem may have resulted from water in the samples, since




relative humidity was much greater than had been encountered during laboratory evaluation.




Ambient temperature at this field site was about 35ฐC and relative humidity was near 100%.




Another possibility is that some instrument failure may have led to part or all of the problem. The




cause is under investigation.  If necessary, moisture could be removed from samples before




analyzing them by this method. No useful data were produced by the FGC during this study.




Portable Gas Chromatographs




       Both PGCs required several hours for setup, warm-up, and tune-up.  Once operational,




both performed  effectively for the duration of the study.




       Data from Audit Standard Analyses. Each PGC performed several analyses of the




audit standard mixture of Method TO-14 target list compounds. Data from several analyses of




the audit standard by the CMS MINICAMS are shown in Figure 4,  and HNU 311 audit data are




shown in Figure 5. MINICAMS results for o-xylene were high. Otherwise, field audit data are in




good agreement with laboratory analyses for both PGCs.




       Comparison of PGC and TO-14 Data.  Canister samples were analyzed for all 41




compounds on the TO-14 target list, while the PGCs were calibrated only for benzene, toluene,




m,p-xylene, and o-xylene. Table 1 shows data from simultaneous canister and PGC sampling of




outdoor air, and Table 2 shows similar data for air sampled inside the mobile laboratory. Indoor




air samples were taken because outdoor analyte levels were lower than expected, as seen




throughout Table  1.  Because of its fuel tank, engine, generator, refrigeration equipment,




upholstery, and limited ventilation, air inside the mobile laboratory contained much higher levels




of TO-14 target compounds.   Inspection of the data in Tables  1 and 2 shows that agreement

-------
between methods, though not exact, was close.




       An understanding of how well the methods agree can be obtained by taking the average of




the absolute values of the differences between PGC and TO-14 results for each compound for all




comparison runs.  Average absolute differences are represented as red horizontal lines in Figure 6.




The standard deviation of the absolute values of the differences is shown by the length of the bar




above and below the red line. This figure shows that the differences are finite, real, and relatively




small, except at the lowest concentrations. The averages are not much larger than their standard




deviations, which suggests that the differences are essentially random and do not contain a large




component of systematic error.  Averaging the differences without taking absolute values, so  that




positive and negative values tend to cancel each other, produces the results shown in Figure 7. A




negative bias of insignificant  magnitude is apparent.




       Retention times are contrasted with standard deviations of retention times in Figure 8.




The width of the red boxes at the ends of the bars represents standard deviation, which was small




enough to suggest that chromatographic conditions were stable during the study, and that the




peak identification algorithms were finding no more than one actual compound for each analyte.




CONCLUSIONS




       The causes of the retention time instability and loss of sensitivity experienced by the Fligh-




Speed Gas Chromatograph remain to be determined.  If high moisture levels were responsible,




then a dry purge preconcentrator injection system will be incorporated. Other indicated repairs or




modifications will also be done, after which the FGC will be re-evaluated at a field study in some




hot and humid environment.  Both PGCs performed as claimed by their manufacturers and are




clearly suitable  for use as field deployable instruments for analyzing volatile organic compounds in




ambient air, provided that adequate shelter and utility support  are available.

-------
REFERENCES

1. Compendium of Methods Tor the Determination or Toxic Organic Compounds in
Ambient Air.  Environmental Protection Agency, Atmospheric Research and Exposure
Assessment Laboratory, Research Triangle Park, NC 27711. EPA-600/4-84-017. June 1988.

2. Richard E. Berkley, Jerry L. Yarns, William A. McClenny, James Fulcher.  "Field Evaluation
of Photovac  10S70 Portable Gas Chromatograph."  Proc. 1989 EPA/AWMA International
symposium on Measurement of Toxic and Related Air Pollutants, AWMA, Pittsburgh, PA, 1989,
pp. 19-26.

3. Richard E. Berkley, Karen D. Oliver, Keith Kronmiller.  "Performance Optimization of
Photovac 10S70 Portable Gas Chromatograph." Proc. 1990 EPA/AWMA International
Symposium on Measurement of Toxic and Related Air Pollutants, AWMA, Pittsburgh, PA, 1990,
pp 849-854.

4. Richard E. Berkley. "Evaluation of Emission Sources and Hazardous Waste Sites Using
Portable Chromatographs." Proc. Second International Symposium on Field Screening
Methods for Hazardous Wastes and Toxic Chemicals, ICAIR Life Systems, Inc.,  Cleveland, OH,
1991, Paper 003.

5. Jerry L. Yarns, R. E. Berkley.  "Measurement of Volatile Organic Compounds in Two
Autoex-Soviet Field Studies Using Portable Gas Chromatography." Proc. 1991
EPA/AWMA International symposium on Measurement of Toxic and Related Air Pollutants,
AWMA, Pittsburgh, PA, 1991.

6. Richard E. Berkley, Jerry  L. Yarns, Joachim Pleil. "Comparison of Portable Gas
Chromatographs and Passivated Canisters for Field Sampling Airborne Toxic Organic
Vapors  in  the United States and the USSR." Environ. Sci. Technol., Vol. 25, No. 8, 1991,
1439.

7. R. E.  Berkley, M. Miller, J. C. Chang, K. Oliver, C.  Fortune.  "Evaluation of
Commercially-Available Portable Gas Chromatographs." Proc. 1992 EPA/AWMA
International symposium on  Measurement of Toxic and Related Air Pollutants, AWMA,
Pittsburgh, PA, 1992, pp 413-418.

8. Robert F. Mouradian, Steven P. Levine, Richard D.  Sacks. "Evaluation of a  Nitrogen-
Cooled, Electrically-Heated Cold Trap Inlet for High-Speed Gas Chromatography." J.
Chrom.  Sci. 28, 1990, 643.

9.  S. P. Levine, H.  Q. Ke, R. F. Mouradian, R. Berkley, J. Marshall. "High Speed
Chromatography for Air Monitoring." Proc. Second International  Symposium on Field
Screening Methods for Hazardous Wastes and Toxic Chemicals, ICAIR Life Systems, Inc.,
Cleveland, OH, 1991, Paper 007.

-------
10. Huiqiong Ke, Steven P. Levine, Robert F. Mouradian, Richard E. Berkley.  "High Speed
Chromatography for Air Monitoring." Proc. 1991 EPA/AWMA International symposium on
Measurement of Toxic and Related Air Pollutants, AWMA, Pittsburgh, PA, 1991.

11. H. Q. Ke, S. P. Levine, R. F. Mouradian, R. E. Berkley.  "Fast Gas Chromatography for
Air Monitoring: Limits  of Detection and Quantitation". Amer. Ind. Hyg. Assoc. J., 53,
1992, 130.

12. H. Q. Ke, S. P. Levine, R. Berkley. "Analysis of Complex Mixtures of Vapors in Air by
Fast-Gas Chromatography."  Air Waste Manag Assoc. J., 42, 1992, 1446.

13. H. Q. Ke, R. E. Berkley. Unpublished data.

-------
Table 1. Canister Grab Sample and Portable Gas  Chromatograph  Data  ฃrom Outdoor Air at Bayamon C^rcel

1,1,1-Tri-
Dichloro- chloro-
me thane ethane
June 23, 1993
12:23
TO-14 0.26
CMS
HNU
12:46
TO-14 0.80
CMS
HNU
13:00
TO-14 ND
CHS
HNU
13:23
TO-14 0.24
CMS
HNU
June 24, 1993
10:23
TO-14 ND
CMS
HNU
11:26
TO-14 0.24
TO-14 0.00
CMS
HNU
11:39
TO-14 ND
CMS
HNU
11:53
TO-14 4.95
CMS
HNU
12:08
TO-14 4.08
CMS
HNU
June 24, 1993
12:21
TO-14 7.67
CMS
HNU
12:28
TO-14 0.26
CMS
HNU
15:23
TO-14 0.88
CMS
HNU
June 25, 1993
09:24
TO-14 0.82
CMS
HNU
09:31
TO-14 0.61
CMS
HNU
09:58
TO-14 1.03
CMS
HNU


ND



ND



ND



ND




ND



ND
ND



ND



ND



ND




0.45
0.00
0.00

ND



ND




0.23
0.00
0.00

0.27
0.00
0.00

ND


Ethyl -
Benzene Toluene benzene


0.
0.
0.

1.
0.
0.

0.
0.
0.

0,
0,
0.


2.
2.
0

2
2
4
0

2
2
0

0
4
0

2
1
0


2
3
0

1
2
0

0
1
0


1
0
0

0
2
0

1
2
0


91
40
00

,77
80
00

.65
.60
.00

,77
,00
.00


.05
.20
,00

.57
.42
.10
.00

.00
.20
.00

.73
.00
.00

.14
.50
.00


.29
.30
.00

.54
.70
.00

.73
.30
.00


.15
.00
.00

.88
.20
.00

.09
.80
.00


2
6
3

5
5
10

1
2
2

5
a
10


15
20
12

7
8
10
12

6
5
7

0
a
5

4
2
a


5
7
6

4
5
3

2
1
0


2
4
0

2
4
0

1
3
0


.20
.50
.20

.38
.00
.30

.37
.00
.20

.38
.80
.70


.79
.10
.80

.95
.38
.60
.20

.82
.40
.10

.44
.30
.90

.75
.90
.10


.54
.00
.30

.50
.70
.90

.23
.90
.00


.50
.70
.00

.57
.20
.00

.95
.20
.00






0.












0.



0
0



0







0




0



0








0



0








ND



39



(TO



ND




.64



.93
.00



.49



ND



.30




.18



.13



ND




.19



.19



ND


1,3,5- 1,2,4-
m,p- 4-Ethyl- Trimethyl Trimethyl
Xylene o-Xylene toluene benzene benzene


0.00
1.10
0.00

1.51
2.40
1.40

0.00
0.70
0.30

0.00
0.00
0.00


2.23
3.60
1.10

3.18
3.05
0.00
3.00

1.74
2.00
1.80

0.00
2.60
0.90

1.02
0.90
1.50


1.58
2.10
1.10

1.07
1.70
0.00

0.14
0.70
0.00


0.84
0.00
0.00

0.50
0.00
0.50

0.56
1.70
0.00


0.
3.
0.

0,
0.
1,

0.
0.
2

0,
0,
1,


0,
2 ,
2 .

0
0
4
2

0
0.
3

0
1
2

0
0
0


0
0
2

0
0
0

0
0
0


0
0
1

0
0
0

0
0
0


24
00
80

,65
.90
,90

.00
.00
.70

.00
.00
.20


.83
,00
,60

.73
.78
.10
.90

.58
.80
.10

.00
.00
.70

.39
.50
.00


.69
.90
.70

.33
.90
.89

.11
.70
.00


.37
.00
.00

.25
.00
.00

.00
.70
.00


ND ND 0.11
0.00
0.00

0.16 ND 0.54
0.00
0.00

ND ND ND



ND ND ND




0.58 0.45 1.78



0.26 0.24 0.73
0.00 0.00 0.00



0.37 0.13 0.82



ND ND ND



ND ND 0.54




0.26 0.11 0.45



0.36 0.09 0.69



ND ND 0.20




ND ND 0.37



ND ND 0.36



0.09 ND 0.28
0.00
0.00

-------
Table 2.  Canister Grab Sample and Portable Gas Chromatograph Data from Air Inside Temporary Mobile Laboratory


June
15:36
TO-14
CMS
HNU
June
15:44
TO-14
TO-14
CMS
HNU
15:56
TO-14
CMS
HNU
June
09:38
TO-14
CMS
HNU
09:44
TO-14
CMS
HNU
1,1,1-Tri-
Dichloro- chloro-
Freon 11 methane ethane
23, 1993

0.93 19.67 1.74


24, 1993
(Replicate)
0.95 16.68 1.36
1.14 17.01 1.36



1.32 20.44 1.88


25, 1993

0.80 14.26 1.14



1.26 15.96 1.24


Benzene


3.
0.
6.


5.
6.
9.
6.

7.
10.
8.


5.
9.
4.

6.
11.
3.


92
50
40


79
01
10
90

38
10
30


63
40
30

33
60
60
Toluene


12
15
22


14
14
16
21

16
17
23


15
17
13

16
19
14


.70
.80
.80


.61
.30
.60
.90

.07
.20
.00


.28
.50
.90

.20
.50
.70
Ethyl- m,p-
benzene Xylene Styrene


0.53 2
4
4


0.70 3
0.59 2
4
2

0.71 3
4
3


0.93 3
0
0

0.98 3
5
1


.23 1.89
.40
.70


.28 1.78
.99 1.30
.30
.80

.74 2.28
.30
.30


.54 1.20
.00
.60

.55 1.67
.00
.60
1,3.5-
4-Ethyl- Trimethyl
o-Xylene toluene benzene


1
4
4


1
1
4
4

1
5
3


1
3
2

2
4
0


.89 0.32 0.13
.10
.90


.83 0.22 0.26
.72 0.12 0.21
.80
.50

.99 0.21 0.27
.30
.30


.62 0.25 0.27
.60
.10

.01 0.26 0.24
.20
.00
1,2,4-
Trimethyl-
benzene


1.08




1.06
1.01



1.17




1.39



1.11



-------
Table 3. Summary of Differences Between TO-14 Data
        and Portable Gas Chromatograph Data

Benzene
CMS Minicams
Maximum Absolute Difference
Mean Absolute Difference
Minimum Absolute Difference
Std Dev of Abs Difference
Bias (Mean Difference)
HNU Model 311
Maximum Absolute Difference
Mean Absolute Difference
Minimum Absolute Difference
Std Dev of Abs Difference
Bias (Mean Difference)

5
1
0
1
-0

5
1
0
1
1

.27
.67
.05
.41
.92

.27
.56
.03
.06
.09
Toluene m,p-Xylene

7
2
0
1
-1

10
3
0
2
-1

.86
.33
.33
.70
.93

.10
.69
.28
.66
.80

3.
1.
0.
0.
-0.

2.
0.
0.
0.
0.

54
16
00
95
35

94
69
00
78
20
o-Xylene

3.
1.
0.
1.
-1.

3.
1.
0.
1.
-1.

34
21
00
16
15

43
48
00
02
24

-------
                                   LEGENDS FOR FIGURES

Figure 1. Schematic of the field study area (not to scale) showing the location of the sampling site on the
grounds of the Bayamon Carcel and the nearest industrial facility upwind in Catano.

Figure 2. Consecutive laboratory analyses of audit standard using GC/FID+ECD before the study (red),
followed by GC/MSD after the study (green), showing the first 21 (of 41) compounds to elute.

Figure 3. Consecutive laboratory analyses of audit standard using GC/FID+ECD before the study (red),
followed by GC/MSD after the study (green), showing the last 20 (of 41) compounds to elute.

Figure 4. Replicate field analyses of the audit standard by CMS MINICAMS, showing data for each of the
four compounds for which it was calibrated during the study and contrasted with results of laboratory
analyses of the audit standard by GC/FID+ECD before the study (red) and GC/MSD after the study (green).

Figure 5. Replicate field analyses of the audit standard by HNU Model 311, showing data for each of the four
compounds for which it was calibrated during the study and contrasted with results of laboratory analyses of
the audit standard by GC/FID+ECD before the study (red) and GC/MSD after the study (green).

Figure 6. Averages of the absolute differences between TO-14 reference results and PGC field results (red
lines). Centered upon the averages are bars, the lengths of which represent the standard deviations of the
absolute differences.

Figure 7. Averages of the differences between TO-14 reference results and PGC field results. The lengths of
the bars represent the average of the differences.  Negative differences tend to cancel positive differences,
revealing a slight negative bias in five of eight  cases and a slight positive bias in the other three.

Figure 8. Average retention times and their standard deviations. The length of each bar represents average
retention time. The widths of the red boxes at the ends of the bars are  proportional to the standard deviations
of the retention times.

-------
:"-"''''":
                                                        
-------



.--.. . 8ซซ*wiy
•• ... . 	 ;.•-•
. . atuฐ^.
.'. fttcfKXykme ......

. .. o-Xtfซปv.

'••'' '• :•. ' •„:'.•'• .
.-.,.
i • • +
-•• " -r . v...{

•....{ "". .:v.. :;v.".' 3
.... '" . *. .'..0.'" '• ' * '.V
.•:-'-.•. '..' 	 \'"\$
--. ,,l-:rr1
v i:':J\"'3'r:l
l,,..l,i.l...ll^.l':M.!,'i!!i.ii!!!.!!!;!i'1
•;. 	 ; . • •..•:.-.-J
• •• tt.. ...-...•..•; *q:
' .' ".-'
. . . .. •
• 	 	 i • . . .
f A ....
."'.'•* . . . . • •.

l...:.;;:i:::::i :?.• .•• i
|:I:'Y:V:-:.:;{:- y;:vv..;v
." ; :.;Z".. i: .:-.-.' .-.J
p-_f:'' I'...'..---.. yV
J-.7V ••.'.5 : • .• " :
TTT1- •••:.. J ••.".: : .•ซ

ri^cwso 	 nu



'..". ••"
......
	 :.
. .-;;;,',' ';,
... ' ..."
"!^:-- •'••-.•
. .''.: .' '.•'••':
•• • ••• •' . ."."
•••" •:• ••:.':.-:!



'"' . .

".;
."••'
/>."••;


•5 ' :.
;. s,

-------
Field Comparison of Portable Gas Chromatographs with Method TO-14
Richard E. Berkley and Maribel Colon
US Environmental Protection Agency, Research Triangle Park, NC 27711
Jesus Gonzalez and Israel Droz
University of Puerto Rico School of Public Health, San Juan, PR 00936
Jeffrey Adams, Christopher Fortune, and Karen Oliver
ManTech Environmental Services, Research Triangle Park, NC 27709
Daniel R. Coleman, CMS Research Corporation, Birmingham, AL 35244
Clayton Wood, HNU Systems, Incorporated, Newton, MA 02161

-------
                                  TECHNICAL REPORT DATA
                           (Hast fttdInttrutHont on iHt rtvtnt btfort
 . REPORT NO.
  EPA/60O/A-94/045
I. TITLE ANDSUSTITLl
 Field Comparison of  Portable Gas Chromatographt
 with Method TO-14
                                                           I. REPORT BAT!
                                                           I. PERFORMING ORGANIZATION CODE
 AUTHORIS)
          R E Berkley,  M Colon, J Gonzalez, I Droz,
  J Adams, C Fortune,  K Oliver, D R  Coleman, C Wood
I. PERFORMING ORGANIZATION REPORT NO.
 PERFORMING ORGANIZATION NAME AND ADORES*

  US EPA/AREAL/MRDD/AMRB/RTP, NC
                                                           10. PROGRAM ELEMENT
                                                              Y105C  A04  04
                                                           11. CONTRACT/BRANT NO.
                                                              CR - 817123-02-0
12. SPONSORING AOINCT NAME AND AOOREM
                                                           1J.TVPE OF REPORT AND PERIOD COVERED
                                                           14. SPONSORING AGENCY CODE
                                                                EPA/600/09
It. SUPPLEMENTARY MOTES
It. ABSTRACT

'-"A  field-deployable prototype fast gas  chromatograph  (FGC)  and two cotmnercially-
'"' Available portable gas  chromatographs  (PGC) were evaluated by measuring organic
  vlpors in ambient afr  at a field monitoring site in  metropolitan San Juan,  Puerto
  Rico.  The data were  compared with  simultaneous grab samples which were
  collected in six-liter  Summa-polished  canisters and  analyzed by method TO-14.
  Because of fluctuating retention  times, the FGC produced no useable data    High
  humidity levels may  have adversely  affected its performance.  Both commercially
  available PGCs performed successfully, and data from twenty analyses were compared
  with the reference method. '  .
IT.
                                KIT WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTlPIERS/DPEN ENOEO TERMS
                                                                            COSATi Field/Croup
ซ•. DISTRIBUTION STATEMENT
  RELEASE TO PUBLIC
                                               tt. SECURITY CLASS
                                                UNCLASSIFIED
               "•"0-
-------