Sound Environmental Sampling Techniques Translate into Better Lab Data: Five Sampling Pitfalls to Avoid

As an environmental testing lab, we see many environmental samples a day – some for ongoing monitoring purposes, some strictly for compliance reasons, and some for investigative work.  We also appreciate that the cost associated with collecting these samples often goes far beyond the cost of the analytical tests themselves when you factor in expenses associated with trained field staff, sampling equipment, transportation to/from remote site locations etc. Therefore, it concerns us when we receive samples at our various labs that contain inconsistencies and non-conformances, particularly knowing that some of these pitfalls will undoubtedly affect the quality of the resulting lab data.

 

If big decisions rest on the numbers coming from your environmental lab, then please read on to learn how the top five sampling pitfalls can impact your results.

 

Pitfall #1:  Improper use of sampling containers

 

One of the first pitfalls to avoid concerns the proper use of your sampling containers.  This may seem basic enough, but it is surprising how many times we in the lab industry see sample submissions that fall drastically short in this respect.  When you set out to sample, you’ve undoubtedly discussed your sampling program with your analytical lab ahead of time to ensure you have the right bottle sets for the job.  It may appear excessive to you when you flip open the lid of your sampling cooler and find a myriad of glass and plastic bottles sent from the lab for what you feel is a pretty straightforward sample collection plan, but the choice of sampling containers has been well-considered by your lab and should not be second-guessed. For example, organics love to cling to plastics, which is why glass bottles with Teflon-lined septa are generally used for samples that will be analyzed for organic compounds.  Light is also a destructive force for some analytes which is why amber glass is provided, or bottle forms are used to protect the sample from the effects of light degradation for certain analytical tests.

 

As for the plastics, a closer look will reveal they aren’t all the same.  Testmark provides PET plastics for general chemistry parameters, higher-grade HDPE bottles for nutrient analysis, and acid-rinsed HDPE bottles for metals analysis after extensive in-house testing confirmed that trace amounts of silicon, aluminum, titanium, magnesium, chromium, zinc and iron can leach out of HDPE plastic and into your sample if not acid-rinsed first.

 

Then there is the issue of volume.  If the lab provides a set of bottles, please use them – all of them.  The ability of your lab to achieve a low method detection level (MDL) often is as much a function of the methodology and analytical equipment used as it is the volume of the sample.  Bottles returned only half-filled may run the risk of yielding a higher MDL as MDL is inversely proportional to sample volume.

 

Finally, it is important to consider the proper use of preservatives.  Again, this is something you can request from the lab when placing your bottle order (pre-charged sample bottles) which can drastically prolong the sample hold time as well as improve the quality of the data itself.  Preservatives, similar to bottles, are analyte-specific.  In general, preservatives serve the purpose of stabilizing the chemistry by either reducing the biological activity in the sample after it is collected, neutralizing other chemicals such as chlorine, or in some cases trapping the analyte of interest to prevent loss to head-space.  Pre-charged bottles have specific quantities of preservative (generally an acid or a base) in them and are fit for purpose.  Field errors such as overfilling a pre-charged bottle essentially dilute the effect of the preservative.  Another common error is submitting preserved samples for lab filtration and analysis as it is often flawed to filter samples after preservation.

 

Pitfall #2:  Improper sampling technique for the analysis of volatiles

 

The amount of headspace in water and soil samples can often greatly affect the quality of your lab data.  For example, water samples collected for the analysis of volatile organics, THMs, dissolved oxygen and suphide should not contain headspace.  The goal is to fill the bottle just enough to achieve a positive meniscus and then seal it with the lid.  A quick field check means flipping your bottle over and seeing if there are any visible air bubbles in your sample.  An air bubble that covers the entire surface of the bottom of a 40mL vial (used for the analysis of BTEX/VOC, F1 and THMs) is too much air.  Volatile compounds, by their very nature basically want to ‘be air’ which means you may be losing some of your target compounds to headspace.

 

Particular attention has been given to the case of volatile organics in soil in light of the amended Brownfields Regulation.  Best practice involves the use of either hermetically sealed sampling kits, or the more convenient methanol pre-charged soil sampling kits.  Testmark promotes the use of the latter and provides Terracore® samplers to our clients which allows them to extract two soil plugs and insert them in pre-charged methanol vials.  The methanol serves to “trap” the volatile compounds in solution until they can be sparged at the lab.

 

 

Pitfall #3:  Excessive particulate matter in water samples

 

This is a tricky one to avoid given some of the realities that exist when drawing from silt-laden or under-developed groundwater monitoring wells.  We often hear from clients about the logistical challenges of filtering samples in the field when there are time constraints or they are working in remote locations.  At times, the only viable solution may be to collect the sample as is (no filtration) and get it to the lab as soon as possible to be filtered there prior to analysis.

 

It is challenging to think of any analytical test that can’t be affected by the presence of sediment in the sample.  The presence of suspended particulate can greatly affect almost all microbiological, nutrient or organic tests results.  Take the example of polyaromatic hydrocarbons (PAH) which tend to adsorb to particulate to the extent that samples with visible sediment may be biased high if the intent was to know the content in the water phase. The revised Record of Site Condition provides some accommodation for this in the form of qualifiers for filtered and unfiltered PAH samples and also underscores that the best practice for metals analysis is to field filter.

 

The case of metals is also greatly affected by the presence of sediment and here we get into the realm of colloidal chemistry, which can be seen almost as a science unto itself.  Colloidal and particulate metal may be found in a number of complexes including hydroxides, oxides, silicates and sulfides.  Furthermore, metals in water are continually adsorbing and desorbing to and from sediment; they are in a constant state of flux dependent largely in part on the water chemistry.   Adsorption removes the metal from water and stores it in the substrate.  Desorption returns the metal to the water.  Metals may be desorbed from the sediment from such things as increases in salinity, decreases in redox potential or decreases in pH.  Dissolved metals are generally in low concentrations in natural water bodies.  The bottom line is that particulate matter in your sample can give fodder to all these biases.

 

Pitfall #4:  Failure to incorporate quality control measures in your sampling plan

 

It’s often been said that North Americans have become the most over-insured society on the planet.  We have been known to insure our health, lives, cars, homes, vacations, kids, pets, even body parts – pretty much everything.  But surprisingly, when we function at our jobs, it is remarkable just how insurance-adverse we are.  A properly executed sampling plan should include some form of quality assurance and quality control.  Think of it this way – for the inconvenience of a few extra bottle sets, you have some insurance that your sampling design, your field work and your choice of lab is validated.

 

The US Environmental Protection Agency (EPA) has a good number of guidance documents that address how to incorporate critical QA/QC measures into your sampling plan.  For example, they recommend one trip blank per cooler.  A trip blank is a clean matrix sample (lab-grade water) taken from the lab to the sampling site and transported back to the lab without having been exposed to sampling procedures.  Field blanks are analyte-free water (lab-grade water) that is poured and preserved in the field exactly as if it were an actual field sample.  Where trip blanks are meant to provide QA for the quality of the bottles and denote any lab-based contamination, field blanks go one step further by also providing a QA check on the physical sampling technique used in the field.  The EPA recommends the use of 1 field blank/day/matrix or 1 blank/20 samples/matrix (whichever is more frequent).

 

Other insurance you should consider building into your sampling plan includes taking field duplicates or splitting samples between labs.  If you opt to include these elements, just be sure your sampling technique is valid to ensure you aren’t introducing sample bias by, for example, taking consecutive samples as opposed to splitting a homogenized sample.

 

Finally, a major QA/QC consideration when dealing with environmental chemistry is to ensure you are transporting samples with due regard to temperature and hold-time requirements.  Your lab should provide you with guidance in regards to sample hold-times as they do vary according to test method.  In general, protocols favour that samples be maintained between 2-8°C during transport.

 

Pitfall #5:  Poorly developed sampling plans

 

Sampling is an arena that Testmark doesn’t enter into firsthand.  We consider our core business to be the provision of quality analytical testing services and we stick to that.  However, we are extremely aware of the impact sloppy sampling regimes will have on the interpretation of lab data.  Once again, the US EPA is a great resource to turn to as they have developed extensive guidance documents in regards to how to carry out water, soil, sediment, biota and air sampling programs.  All factors must be considered, from the influence of stratified chemical and thermal layers occurring in natural water bodies, to best practices in sinking and developing groundwater wells.  Statistical representation in terms of number of samples and spatial distribution is also critical.

 

From a lab perspective, we experience a similar situation when we perform the preparation of soil samples for testing.   Extractable organics typically require 20 grams of sample to be analyzed from a much larger one submitted to the lab.  The key then becomes ensuring that the 20 grams isolated from the container is extracted representatively to ensure that the sample actually used for analysis is representative of the entire sample submitted.  This involves sound adherence to sample preparation protocols which include mixing, disaggregation, sieving and drying.  We spend a great deal of effort ensuring the subsamples taken from your submitted sample are representative and this only accentuates your responsibility in obtaining a sample that will allow meaningful data from a project standpoint.

 

The end goal is that the sample is meant to represent the whole.  A sound sampling plan should consider the impact of all variables including ones such as statistics, time, temperature, seasonality, numeracy and location.

 

Undoubtedly, there are other pitfalls beyond the ones mentioned above that can impact the quality of your lab data.  However, the main message here is to take the time beforehand to properly prepare and execute your sampling plan and to learn as much as you can in regards to best practices for sampling.  Time spent upfront at the planning stage, as well as care taken in the field, can translate into much more meaningful results when the data comes back from the lab.  Consider your lab your partner throughout the process.  Much of the technical direction concerning the proper use of laboratory bottles, hold times and preservatives is information that they will readily provide.