Page 31 - Geochemistry of Oil Field Waters

20 ANALYSIS OF OILFIELD WATERS

documented. The value of research investigations which use oilfield brine
analyses depends upon the validity of the laboratory results.
A program to insure the reliability of analytical data is mandatory because
of the importance of the laboratory results and the actions that they
produce. An established routine control program applied to analytical tests is
necessary to assure the precision and accuracy of the final results. The use of
spiked samples can measure quality, while the use of analytical-grade
reagents is a control measure. Quality control varies with the type of as*-
sis. For example, the frequent standardization of the titrant used in a titra-
tion is an element of quality control, while instrument calibration in an
instrumental method is also a quality control function.
The specific methodology employed should be carefully documented
regardless of the method used; thus the data user or reviewer can apply the
associated precision and accuracy when interpreting the laboratory data.

Choosing an analytical method

Widespread use of an analytical method indicates that it probably is
reliable and will produce valid results. Use of a little-known procedure forces
the data user to accept the judgment of the analyst.
The following criteria are useful in selecting analytical methods:
(1) The desired constituent should be measured with sufficient precision
and accuracy in the presence of the interferences normally found in oilfield
waters.
(2) The method must utilize the skills and equipment available in the
oilfield water laboratory.
(3) The method should be sufficiently tested and used by several labora-
tories to establish its validity.

Precision

Precision is the reproducibility among replicate observations, and in
quality control it is determined on actual water samples containing inter-
fering constituents. Several methods to determine precision are available and
the following is representative:
(1) Study four separate concentration levels, including a low concentra-
tion near the sensitivity level of the method, two intermediate concentra-
tions, and a concentration near the upper limit of application of the method.
(2) Make seven replicate determinations at each of the concentrations
tested.
(3) To allow for changes in conditions, the precision study should use at
least 2 hours of normal laboratory operation.
(4) To permit the maximum interferences in sequential operation, the
samples should be run in the following order: high, low, intermediate, inter-
mediate. Repeat this series seven times to obtain the desired replication.