Lecture Feb. 7th 2007.

Answers to quiz. There are 1.00 x 10^6^ microgram (µg) in 1.00 g.

The percentage by mass of O in HNO_3_ is 100 x (3 x 16)/[(3 x 16) + 14 + 1] = 4800/63 = 76 (2 digits).

Added 20 found 15 (45 - 30), so recovery is 15/20 = 0.75 = 75%.

50% (see Fig 5.2 on page 97)

The method of standard additions.

Calibration is needed at the time of measurement. Instruments may have to be recalibrated during a working period to correct for drift. Interferences are always a potential problem. They can be overcome by creating a matrix matched set of calibration standards. If the matrix is not known, then a series of calibration solutions can be produced by adding increasing concentrations of standard to the same amount of sample and making up to same final volume.

Then plot instrument response versus concentration added and extrapolate back to find the concentration in the sample. Correct by the dilution factor to find the concentration in the sample solution.

Check for interferences (or loss of analyte) by spiking a known amount or concentration into the sample at some point in the procedure and then compare result with expected based on analysis without spike plus known spike concentration (see quiz question above). For trace analyses (sub ppm concentrations) 85 - 115% is OK.

To compensate for changes in instrument response during analysis, add an internal standard to samples and standards and plot ratio of responses versus ratio of concentrations. The internal standard must be absent in the sample and must behave the same as the analyte when the measurement conditions change.

For instruments in which a discrete volume is introduced (such as in HPLC or GC), a prior experiment is needed to establish the response factor for the analyte relative to the internal standard. The response factor is the ratio of calibration slopes for the two species (analyte and IS). As the slope of a calibration based on a single standard (and the assumption of zero response for zero analyte) is the ratio of response to concentration, a response factor is a ratio of ratios! If this is not making sense, work through the example on page 102. Try to deduce the concentration of X without looking at Dan Harris's worked solution.