On-Demand MDLs for the Environmental Laboratory
Kenneth E. Osborn*
Laboratory Services Division
East Bay Municipal Utility District
Oakland, California
kosborn@ebmud.com
Francois Rodigari
East Bay Municipal Utility District
ABSTRACT
The traditional approach to the determination of detection limits (MDL) represents a
challenge: MDLs do not reflect day-to-day analytical variation, they are an inadequate
metric for monitoring on-going performance, and they require a non-routine commitment
of analytical resources. The EBMUD Laboratory has adopted a QC procedure requiring
the analysis of a False Negative Quality Sample (FNQS) included with every analytical
batch. Recovery of the FNQS is used to monitor batch accuracy in the range near
detection. A web-based MDL calculator, designed by EBMUD staff, is used to generate
EPA compliant detection limits using the FNQS database. Experience at our laboratory has
demonstrated that this approach reduces resource requirements for maintaining up-to-date
detection limits; provides data end-users with a measure of accuracy in the region of
detection; and provides a more reliable set of detection limits. This presentation will cover
the concepts, details of the calculation tool, and some potential pitfalls.
KEYWORDS: MDL, FNQS, false-negative, detection limits
INTRODUCTION
The current EPA procedure for the determination of the method detection limit (MDL), as
specified in 40CFR Part 136 - Appendix B, can be difficult and expensive to carry out.
This is particularly true if the optional iterative procedure is also performed as part of an
MDL study. While some analytical methods specify a frequency for performing MDL
studies, most do not. When making decisions on how to manage scheduling and
administrating MDL studies, environmental laboratories must take into account the
practices recommended by their certifying authority, method requirements, and operational
factors.
An MDL study is typically performed when preparing the initial demonstration of
labora