The Unregulated Contaminant Monitoring Program (UCMR3) recently announced a new rule that requires public drinking water facilities to collect data on roughly 30 contaminants from January 2013 through December 2015. For the VOC portion of this rule, seven (down from the original nine) were selected for analysis by US EPA Method 524.3, using selective ion monitoring (SIM) to reach MRLs as low as 0.03 µg/L.
To evaulate this low-level method, we applied our standard 524.3 purge and trap parameters (524.3 application note). Table one displays adequate calibration curves across the low part-per-trillion level. The real issues started when we ran MDLs that did not return consistent results. Since it was the first run of this application, we did not know if the purge and trap parameters needed to be adjusted or if the GC/MS needed optimized.
Table 1: First GC/MS with Standard 524.3 Conditions
|
Analyte |
Q Ion |
Confirming Ion |
Quadratic Regression Factor (r2) |
MDL (ppt) |
%RSD of MDLS |
|
Chlorodifluoromethane |
51 |
67 |
0.9951 |
5.49 |
27.9 |
|
Chloromethane |
50 |
52 |
0.9952 |
2.72 |
20.8 |
|
1,3-Butadiene |
54 |
53 |
0.9964 |
3.13 |
13.2 |
|
Bromomethane |
94 |
96 |
0.9973 |
3.17 |
10.8 |
|
1,1-Dichloromethane |
63 |
65 |
0.9980 |
3.71 |
15.7 |
|
Bromochloromethane |
128 |
130 |
0.9989 |
3.77 |
13.8 |
|
n-Propylbenzene |
91 |
120 |
0.9955 |
3.27 |
11.6 |
|
1,2,3-Trichloropropane |
110 |
112, 75 |
0.9994 |
1.37 |
4.22 |
|
sec-Butylbenzene |
105 |
134 |
0.9963 |
2.13 |
7.44 |
In parallel to the above test, we were also evaluating an application for 1,2,3-trichloropropane using SIM scan, at the ppt level, to meet California regulations. This system was optimized for low level analysis and could be used for an UCMR3 application as well. We once again used 524.3 purge and trap parameters and got very reproducible results. The results are displayed in Table 2 and in the UCMR3 application note (UCMR3 application note).
Many of the problems we encountered running the first GC/MS system can be attributed to water. However, there may be some optimizations required to discover the right split ratio and mass spec settings (voltage, gain, etc.) for this application.
Table 2: Second GC/MS – Optimized to run low level samples
|
Analyte |
Q Ion |
Confirming Ion |
Quadratic Regression Factor (r2) |
MDL (ppt) |
%RSD of MDLS |
|
Chlorodifluoromethane |
51 |
67 |
0.9998 |
1.26 |
5.2 |
|
Chloromethane |
50 |
52 |
0.9996 |
0.92 |
2.9 |
|
1,3-Butadiene |
54 |
53 |
1.0000 |
2.16 |
9.4 |
|
Bromomethane |
94 |
96 |
0.9998 |
1.09 |
4.0 |
|
1,1-Dichloromethane |
63 |
65 |
0.9998 |
1.08 |
3.4 |
|
Bromochloromethane |
128 |
130 |
0.9996 |
1.69 |
5.5 |
|
n-Propylbenzene |
91 |
120 |
0.9997 |
0.87 |
3.3 |
|
1,2,3-Trichloropropane |
110 |
112, 75 |
0.9994 |
0.86 |
2.7 |
|
sec-Butylbenzene |
105 |
134 |
1.0000 |
0.59 |
2.1 |
We can apply the lessons learned from running these evaluations to optimize all our GC systems to run this application. Even though we specialize in sample prep instrumentation, we also devote many hours to method development and optimizing the conditions of the analyzer. Our analysts have a wealth of experience in method development for GC/MS and we are happy to share this knowledge with our customers when they need it.