YSI 9600 - Manual (Page 74)

8.9 TEMPERATURE EFFECTS
The effects of temperature on the performance of the 9600 fall into two categories. First, the output of the
LED (the light source in the absorbance measurement) is temperature dependent. Second, the rate of the
chemical reactions is highly dependent on temperature.
Absorbance measurements are based on the amount of light entering (P
o
) and exiting (P) your sample flow
cell. See Beer’s law above. The 9600 uses a green LED as the light source and a light-to-frequency
converter (a photodiode with a current-to-frequency converter) to measure P
o
and P in units of Hz. The
instrument corrects for temperature effects on the LED by use of an additional (reference) photodiode
which takes a reading independent of the sample/blank/standard in the detector flow cell. This reference
reading will display a temperature dependence identical to the temperature dependence of the
sample/blank/standard photodiode readings. By simply ratioing all sample/blank/standard photodiode
readings to a simultaneous reference reading, most of the effects of temperature are removed as shown
below. Without ratioing, there could be a 19% difference in the photodiode reading across the temperature
range; with ratioing this difference is reduced to less than 4% across the entire range and less than 2% from
10-40 ˚C. This effect is documented in the figure below.
DETECTOR READINGS NORMALIZED TO THE 20 C READING; WITH
AND WITHOUT TEMPERATURE CORRECTION
0.8
0.85
0.9
0.95
1
1.05
1.1
1.15
1.2
0 5 10 15 20 25 30 35 40 45
TEMPERATURE, C
NO
RM
ALIZED DETECTO
R
READIN
G
Raw
Corrected
It is important to note that the self calibration of the instrument (running of blanks and standards at least
twice a day) will remove these residual temperature effects as shown below.
The effects of temperature on the chemical reaction rates can be ignored if the reaction is allowed to reach
equilibrium. If this is the case, then the remaining temperature effects are thermodynamic, i.e. position of
equilibrium may be temperature dependent. The figure below demonstrates empirically that temperature
effects on the chemistry are minimal. Note that a coefficient of variation of only 1.16 % was seen over a
wide temperature range (1 to 40 C) for analysis of a solution which was approximately 2 mg/L in nitrate-N.
YSI Environmental 74 Model 9600 Nitrate Monitor