Sample Contamination
Tim Loftus
Currently there is an effort by many regulatory agencies to lower environmental pollutant limits. Most of us are seeing this directly with our new NPDES permits. Copper and lead limits, for example, are commonly in the part per billion ranges. Since some of these limits are so low, even small amounts of contamination can lead to highly inaccurate results. It’s important to realize that good sample results don’t start with calibrated instruments or excellent laboratory technique (although these are important). Good sample results start with good sampling, and one of the major components of good sampling is clean equipment.
Parts per million (ppm) and parts per billion (ppb) are common units of measurement in the wastewater field. Now with more accurate analyzers we are starting to hear about parts per trillion (ppt), especially with organics and mercury analyses. To put the ppm, ppb, and ppt into perspective, think of these in terms of time. A million seconds is about eleven days. To accurately measure wastewater with a copper value of 1 ppm is like accurately picking one specific second out of eleven days. A billion seconds is equal to about 32 years. Our NPDES permit requires that we discharge less than 12 ppb copper. Again, putting this in terms of time, we must accurately measure better than a 12-second block of time within a span of 32 years. For those of you lucky enough to deal with environmental mercury samples in the ppt range, it takes about 32,000 years to get one trillion seconds. The point is if you’re sampling for parameters at such low levels, especially in the ppb and ppt range, it is extremely easy to contaminate a sample to the level where you may violate your NPDES permit (or other) limit.
Space constraints do not permit details of every situation, but in general clean all sample tubes, collection containers, and collection equipment (funnels, filtering apparatus, etc.) according to the recommendations outlined in the test procedures. Most cleaning procedures are test specific and involve washing with soap and water, followed by an acid rinse, then with deionized water. Sometimes an organic solvent is used as a rinse depending on the test. Cleaning procedures can be found in Standard Methods for the Examination of Water and Wastewater, EPA Methods for Chemical Analysis of Water and Wastes (EPA-600 /4-79-020), and the test guidelines and methods outlined in 40 CFR 136. These cleaning recommendations are adequate for most sampling events. There is, however, an additional level of “clean sampling” not mentioned in these publications. It involves body suits and a number of special procedures and is used mostly for sampling ultra low levels of mercury. These techniques are not covered here since few of us need to sample at this level - yet.
Oftentimes, too, preservatives can add contamination to a sample. Acids used for metal preservation can have trace levels of the analyte in question. Always use the highest quality preservatives when sampling for trace analytes.
Sample contamination can also be avoided using just common sense. For example, don’t sample for VOCs when the grass is being cut nearby, or sample for metals using a Mason jar with a metal cap.
With environmental limits becoming more stringent, it’s important to keep samples free from contamination so you don’t mistakenly report a false violation. Violations, as we all know, creates more work for us. Like the ppb/seconds analogy, clean sample containers, clean equipment, contamination-free preservatives, and a little common sense will save you time - in more ways than one.
As usual, check your federal, state and local regulations. You may have additional sampling regulations that you must meet.
This article was written under the auspices of the New England Water Environment Association (a chapter of the Water Environment Federation) Laboratory Practices Committee. Please visit the NEWEA website at www.newea.org for membership information and other opportunities.