Standard Addition

(c) David L. Zellmer, Ph.D.

Department of Chemistry
California State University, Fresno
September 2, 1998

Standard addition must be used whenever the matrix of a sample changes the analytical
sensitivity of the method. In other words, the slope of the working curve for standards made with
distilled water is different from the same working curve made up in swamp water (or whatever
your real world matrix happens to be).

To do constant volume standard addition we need the following solutions and glassware.

A constant volume (Vunk) of the unknown solution is added to each of four volumetric flasks of
volume Vflask. Then a series of increasing volumes of stock solution are added (Vstd). Finally,
each flask is made up to the mark with solvent and mixed well. The concentration and volume of
the stock solution added should be chosen to increase the concentration of the unknown by about
30% in each succeeding flask.

The actual concentration of the analyte in any given flask will be given by:

The instrumental response to the analyte will be R = (K)(concentration), so

Now set Csa = CstdVstd/Vflask.

After measuring the response for a series of standard additions, we plot the results, then
extrapolate to y = 0 to get the value of that Csa would need to be to equal the response of the
unknown. We use this negative value to compute Co.

Here is an example of this calculation carried out on a spreadsheet using simulated data.
Remember that the Parameters section sets up the simulation, while the Calculations section
carries out the analysis in much the same way you would do it.

The data in the blue box are plotted in the graph. The figures shown in red in the Calculated
section would not be known to the analyst. All the analyst knows is the concentration due to the
additions of the standard, so please don't try to create the figures in red when working up your
own data! The point here is that the instrument is responding to the sum of the unknown and
standard addition analyte.

A simple Linear Least Squares (LLS) analysis is made using the SLOPE(y-array, x-array) and
INTERCEPT(y-array, x-array) functions of Microsoft Excel. From y=mx+b we calculate the
value of x at y=0, which is -30.60, Co = -(-30.60)*50mL/10mL or 153 µg/mL. This agrees with
the value of Co of 153 µg/mL set in the Parameters section of the simulation.

A more advanced simulation adds the effect of instrumental uncertainty. The analyst must
calculate both the value of Co and its associated error. See Gran's Plotting for pH Titrations for
full details on computing the error of an X-axis intercept using Excel. A quick example will be
given here.

Our instrumental error was set at 0.2 out of a maximum response of about 10, or a 2% error. The
effect of this error on the final result is quite a bit larger than 2%. The 95% Confidence Limit of
Co is 32.6 for a value of 139.3, or a 23% error! Some analysts prefer to use the "1-sigma" error,
which is much smaller. Here that would be "s sub c, y=0" or a 2.047 error in the Csa value of
27.85 or 7.4% error.

With the errors this large, the analyst would have to report the result as 140 +or- 30 µg/mL. The
extra significant figures would not be justified.

If you think these computed errors seem excessively large, consider the animation below where
the simulated analysis is repeated over and over. Note how the X-intercept bounces around.
 The Method of Standard Additions

In many cases the intensity of the signal of the analyte is affected by the composition of the
matrix, by the temperature and other factors.

One of the methods to overcome these problems is the method of standard additions. Two
conditions have to be fulfilled for successful application of the method:

(a) the calibration graph must be linear,

(b) the calibration curve of the analyte passes through the origin.

The signal intensity of the sample solution is measured and then portions of a solution of the
element at a known concentration are added and the signal intensity is measured after each
addition. The optimal size of each addition is that which gives a signal 1.5 to 3 times that of the
sample.

For each addition:

where Cx and (Cx + Cs) are the concentration of the analyte without and with the standard
addition, respectively,

signalx and signalx+s are the signal intensities of the solutions containing Cx and (Cx + Cs).

A plot of the signal intensities of the solutions vs. the added concentrations yields a straight line.
The concentration of the analyte is determined from the point at which the extrapolated line
crosses the concentration axis at zero signal.
An example is given in Table 1 and Fig.1.

Table 1. Data for the determination of Cu2+ by the standard-addition method.

Sample volume - 20.00 ml; concentration of the Cu2+ solution used for standard
additions - 0.5 mM.
Solution No. Volume of 0.5 mM Concentration of the Signal
standard addition in the (absorbance)
Cu2+ added, cell,
1 0 0 0.100

2 100 249 0.152

3 200 495 0.199

4 300 739 0.248

Two modes of
preparing the
solutions are
used, depending
on the type of cell
used in a specific
analysis:

(a) The
analyte
and
mixtures
of analyte
with the
respective standard additions are prepared in separate volumetric flasks.

(b) A known amount of analyte is introduced into a cell (with or without additions of known
volumes of other constituents). The volume of the standard additions is small so as not to
change the matrix composition of the sample. The analytical signal is recorded. Standard
 additions are made in succession. The changes in the concentration of the analyte as a
result of the additions must be calculated.

The main advantage of the method is that the matrix remains unchanged. The method is suitable
for cases in which the matrix is complex or is difficult to reproduce.

The method of sample bracketing

If the calibration graph is curved, the standards are chosen so that they are very close to the
concentration of the unknown. After the initial run of the unknown, two standards are chosen -
one with a concentration lower than the unknown and one higher. The segment between the two
standards can be considered linear and an accurate determination can be realized.

The method is suitable for the case of non-linear calibration curves and when reasonable
matching of the matrix can be achieved.

Tutorial: Standard Additions

Quantitative Analysis with Multiple Standard Additions

The general procedure for standard additions is to obtain measurements after adding increasing
amount of analyte to portions of the sample. There are actually two related ways of doing this:
one in which the sample portions are all made up to the same volume after the standard has been
added, and one in which the volumes of the sample solutions are different.

Example: Multiple Standard Addition with dilution to

constant volume

A 3.12g stick of chewing gum is ashed at high temperature, dissolved in 5mL of concentrated
nitric acid and then diluted to the mark in a 100mL volumetric flask. To 25 mL volumetric flasks
are added a 5.00 mL aliquot of the dissolved chewing gum sample and an addition of 300.0 ppm
aluminum. After dilution to volume, the solution is analyzed for aluminm by graphite furnace
atomic absorption at 309 nm. The data are shown in the following table. Report the aluminum
content of the gum as a confidence interval.

vol added standard,

Signal, A-s
mL

0 0.214

0.1 0.386

0.2 0.554
 0.3 0.728

Answer: 48.1 +/- 2.8 ppm (95% CI)

Sample dilution is not always desirable, since it degrades analytical sensitivity. The following
example shows how the standard addition method may be used without dilution.

Example: Multiple Standard Addition without additional

dilution

Anodic stripping voltammetry can be used to measure the leachable lead content of pottery and
crystalware. A sample of pottery being considered for import is leached with 4% acetic acid for
24 hr. A 50.00 mL aliquot is transferred to an electrolysis cell, and after controlled-potential
electrolysis at -1.0V (vs SCE) for 2 minutes, an anodic scan is recorded with the differential
pulse method. The procedure is repeated with new 50.00 mL aliquots to which 25, 50, and 75
microliter spikes of standard containing 1000 ppm lead have been added.

vol added standard,

limiting current, uA
uL

0 2.5

25 3.7

50 4.9

75 6.0

What was the concentration of lead in the original acetic acid leach? Report your answer as a
confidence interval.

Answer: 1.07 +/- 0.16 ppm (95% CI)