Instant Notes: Analytical Chemistry

the detector response and the mass or concentration of the analyte more

complex curvilinearor logarithmic regressioncalculations are required.

For any analytical procedure, it is important to establish the smallest amount of

an analyte that can be detected and/or measured quantitatively. In statistical

terms, and for instrumental data, this is defined as the smallest amount of an

analyte giving a detector response significantly different from a blank or back-

ground response (i.e. the response from standards containing the same reagents

and having the same overall composition (matrix) as the samples, where this is

known, but containing no analyte). Detection limits are usually based on esti-

mates of the standard deviation of replicate measurements of prepared blanks.

A detection limit of two or three times the estimated standard deviation of the

blanks above their mean, x

_

B, is often quoted, where as many blanks as possible

(at least 5 to 10) have been prepared and measured.

This is somewhat arbitrary, and it is perfectly acceptable to define alternatives

provided that the basis is clear and comparisons are made at the same proba-

bility level.

Where components of a sample other than the analyte(s) (the matrix) interfere

with the instrument response for the analyte, the use of a calibration curve

based on standards of pure analyte may lead to erroneous results. Such matrix

interference effectscan be largely if not entirely avoided by preparing calibra-

tion standards where known amounts of pure analyte are added to a series of

equal sized portions of the sample, a procedure known as spiking. In addition,

one portion of sample is not spiked with analyte. (Note: if spiking sample solu-

tions with analyte changes the volume significantly, volume corrections must be

applied.)

The effects of the matrix on measurements of the analyte in both the spiked

and unspiked samples should be identical. The instrument responses are then

used to construct a calibration graph where the x-axis values are the added

amounts of analyte and the response for the unspiked sample is at x=0 (i.e., the

curve does NOT pass through the origin). The regression line is calculated and

extrapolated back to give a negative intercept on the x-axis at y=0, which corre-

sponds to the amount of analyte in the sample (Fig. 4).

The less reliable procedure of extrapolation rather than interpolation is

outweighed by the advantage of eliminating or minimizing matrix interference.

The method of standard addition is widely used, particularly when the

composition of the sample matrix is variable or unknown so that the response of

a reagent/matrix blank would be unreliable. At least three and preferably more

spiked samples should be prepared, but if the amount of sample is limited, as

few as one sample must suffice. It is especially useful with such analytical tech-

niques as flame and plasma emission spectrometryand potentiometry(Topics

E4, E5 and C8).

Example

The calcium level in a clinical sample was determined by flame emission

spectrometry using a standard addition method, which gave the following data:

Spiked calcium (ppm) 0 10 20 30 40 50

Emission intensity 0.257 0.314 0.364 0.413 0.468 0.528

at 423 nm

Standard
addition

Limit of
detection

46 Section B – Assessment of data