CATION CONDUCTIVITY TEMPERATURE COMPENSATION

DAVID M. GRAY, Thornton Associates, Inc., Waltham, MA

ANTHONY C. BEVILACQUA, Ph.D., Thornton Associates, Inc., Waltham, MA

Presented at the International Water Conference, Pittsburgh, PA, November 1997. (IWC-97-48)

KEYWORDS: cation conductivity, temperature compensation, power plant cycle chemistry

SUMMARY: Cation conductivity is one of the most important on-line cycle chemistry measurements in
today's power plant. It is relatively simple, sensitive, low cost and reliable. But is it accurate? Often taken
for granted, cation conductivity accuracy depends heavily on the sample temperature variability and the
temperature compensation algorithm used. The cation conductivity temperature compensation
performance of current instrumentation has been examined over a wide range of conductivity and
temperature and compared with new developments. Surprisingly large errors have been found.

INTRODUCTION More than a decade later, little of available high

purity conductivity instrumentation has changed
The importance of temperature compensation of despite improvements in microprocessor
high purity water conductivity measurements memory and computational methods. There is
has been widely recognized in the power also greater reliance on cation conductivity as
industry. The development of microprocessor- the key water chemistry parameter, particularly
based conductivity instrumentation in the 1980s with oxygenated treatment. Presented here is a
brought with it the first commercially available background of conductivity temperature
equipment to provide reasonable temperature compensation evolution in the power industry, a
compensation accuracy. It also initiated concern recent evaluation of cation compensation
about the algorithms used. At the same time, the methods in use and a comparison with new
first EPRI (Electric Power Research Institute) developments.
guidelines for cycle water chemistry were
published, setting specific and cation BACKGROUND
conductivity limits for various points in the
cycle. Maintaining absolute values rather than Conductivity is primarily used as a measure of
just monitoring trends for upsets was seen as ionic concentration. However, conductivity is
vital to long term corrosion and deposition also sensitive to temperature. Nearly all
reduction. The high and variable conductivity conductivity measurements and specifications
temperature coefficients, the wide variation of are therefore referenced or compensated to a
many sample temperature controls and the fixed temperature, usually 25°C, in order to
vulnerability of sample temperature to ambient refer to a specific level of contamination. That
changes downstream of temperature controls is, the conductivity is readout as if the
justify real concern over the type of temperature of the sample were at 25°C,
compensation provided. regardless of the actual

1
measurement temperature. The methods to algorithms as well as precise non-linear
provide this temperature compensation vary conductivity to concentration conversion for
widely in the level of complexity and accuracy, specific acids and caustic. These extensive
as this work will demonstrate. algorithms can be activated by simple menu
selection.
Close temperature control of samples in the
sample line is a good, though expensive, Other instruments require the user to enter a
practice. It minimizes secondary effects of matrix of conductivity to concentration data plus
temperature such as high temperature reactions, an appropriate temperature coefficient in order
varying attraction of trace ions to built-up solids to achieve direct concentration readout. Users
on the sample line wall and ambiguities of should be aware of this requirement since the
compensation when the composition of the data is often not available and typically requires
sample differs significantly from that for which extensive laboratory work to obtain. The
temperature compensation algorithms are resulting temperature compensation may have
intended. However, temperature control is not a limited range or limited accuracy since many
complete substitute for temperature percent concentration samples have non-linear
compensation since controls do fail and develop temperature characteristics.
wide tolerances and samples are influenced by
ambient conditions in the lines between the LOW CONDUCTIVITY TEMPERATURE
cooler and the sensors. COMPENSATION

HIGH CONDUCTIVITY TEMPERATURE Pure waters present a different challenge to

COMPENSATION appropriate temperature compensation. With
most materials, the ion concentration remains
The temperature coefficients of most neutral salt relatively constant with temperature. The
ionic contaminants in water are approximately temperature effect is largely due to the improved
linear, near 2% of conductivity value per °C. ionic mobility at higher temperature, which in
This has allowed instrumentation for general turn is due to the decreasing viscosity of water.
purpose conductivity measurements to use very For this reason, most salts in water have about
simple temperature compensation methods. It the same temperature coefficient, near 2%/°C.
should be noted that most portable and
laboratory conductivity instrumentation still In pure waters, the hydrogen and hydroxide ions
uses this limited type of compensation. from the dissociation of the water itself
dominate the conductivity. As the temperature
Acids and bases have somewhat lower of water increases, the degree of dissociation
temperature coefficients, on the order of 1.6- rises dramatically. The hydrogen and hydroxide
1.9% per °C. This is because of the higher but ion concentrations increase nearly 22-fold over
less temperature-sensitive conductivity of the range of 0 to 100°C. This causes a much
hydrogen (hydronium) and hydroxide ions. higher and more non-linear conductivity vs.
Some conductivity instrumentation has the temperature relationship. At 25°C the slope or
ability to set in a linear temperature coefficient sensitivity is approximately 5% per ºC for pure
but depends on the user to determine the water and at low temperatures is over 7% per
appropriate value to enter. For acid/base °C. See Figure 1. The properties of pure water
measurements in the percent concentration conductivity with temperature have been
range (such as used for demineralizer repeatedly examined, resulting in improved
regeneration) some instruments have carefully accuracy.1-6
matched non-linear temperature compensation

2
The compensation for these properties has also water conductivity/temperature characteristics
evolved and improved. High purity water and by closer mathematical curve matching to
compensation has traditionally been recognized those characteristics over wider temperature
to consist of two separate components: the ranges. The present state of the art for neutral
properties of the solvent, pure water, and the salt compensation is the G.E.-type equation
properties of the solute, the salts in the water. using the most recent Thornton/Light
This is appropriate for neutral salt impurities expressions for Cw and replacing the 0.02 linear
since the ionization of water is influenced only coefficient with a more accurate non-linear
by temperature, not by trace concentrations of expression.6 However, some high purity
salts. Similarly, the mobility and conductivity of instruments have not kept up with these more
trace salt ions varies with temperature, recent developments.
independent of the ionization of water.
All of the above developments based on
The combination of these independent
separate solvent and solute effects in G.E.-type
temperature effects was codified decades ago by
equations are appropriate only for BWR samples
General Electric for use with measurements on
and polished makeup water, where neutral salts
boiling water reactor (BWR) samples. This was
are typically the major contaminants. Most
the so-called G.E. equation, Eq. (1). Note
other specific conductivity samples will have
specific parts of the equation: Cw and 0.0545
ammonia or amines dominating the
describe the properties of pure water and 0.02
conductivity. Cation conductivity samples have
represents the typical 2%/C temperature
had their cations replaced by hydrogen ion and
coefficient of neutral salts. There is a direct
thus have become acidic solutions. Since these
correlation between water's physical properties
samples are bases or acids rather than neutral
and the parts of the equation describing the
salts, a different situation exists. The hydroxide
compensation.
ion of the base and the hydrogen ion of the acid
have a direct impact on the ionization of water
CT − Cw which also produces these ions. The additional
C = + 0.0545 (1)
25o C 1 + 0.02 ( T − 25) hydroxide or hydrogen ions tend to suppress the
ionization of the water itself.
CT = total conductivity at temperature
(raw or uncompensated) This interaction and its variation with
Cw = conductivity of pure water at temperature require an entirely different
approach to temperature compensation since the
Temperature
solute and water effects cannot be separated. It
T = sample temperature (°C)
is not correct as sometimes proposed that the
C25°C = conductivity compensated to 25°C
G.E.-type equation can be modified by putting
in a different linear coefficient for ammonia or
The original expression for Cw in the GE cation conductivity samples. Such an approach
equation was an empirical curve match, Eq. (2), does not take into account that the ionization of
which could be calculated fairly easily with a water has been greatly influenced by the
slide rule.7 presence of the bases or acids in these samples
and therefore the Cw term is no longer
Cw = (0.0545) (0.55 e 0.0363 (T) - 0.356) (2) applicable. This fact was pointed out some years
ago but as not heeded in the development of
The performance of compensation has been some instrumentation.8
improved both by more accurate knowledge of

3
It is fairly straightforward to calculate the The best available data were used for the ion
conductivity of a solution of known acid or base product of water, the equivalent ionic
concentration at various temperatures (see conductances of hydrogen, hydroxide and
appendix). However, it is not straightforward to chloride ions, plus the density of water, all
compensate those values back to the reference varying over temperature.6 These were used in
conductivity at 25°C: the difficult job a standard physical chemistry equations (see
conductivity instrument must accomplish. For appendix) to obtain the reference values of
this reason, advanced temperature compensation conductivity vs. temperature at various
algorithms have been kept proprietary, buried in concentrations of hydrochloric acid which
microprocessor software. Their performance, represented particular conductivity values at
however, can be readily assessed. 25°C.

TEMPERATURE COMPENSATION Simulated uncompensated conductivity signals

EVALUATION (resistances) for the full range of temperatures
were determined from the reference values
Standard high purity neutral salt compensation above and connected to the input of several
methods using various modified G.E.-type manufacturers' instruments. Resistances were
equations give fairly comparable results. Only made compatible with the appropriate cell
the most demanding semiconductor ultrapure constant for each instruments' range.
water applications require the slightly higher Instruments were made to display precise
accuracy provided by the most recent work in corresponding temperatures by adjusting the
that area.6 However, with cation compensation, temperature input resistance as needed. The
it was suspected that accuracy with various resulting readout of compensated conductivity
instruments would vary widely because of the was tabulated over the range of hydrochloric
prevalence of false assumptions, limited acid concentrations and over a wide temperature
mathematical capabilities and perhaps some range. These results were then compared with
complacency. It was decided to evaluate the the reference values at 25°C for the
cation temperature compensation performance corresponding HCl concentration to identify the
while eliminating other possible sources of error errors. Results are shown in Figures 2-7 and
such as cell constant, temperature measurement, Tables 2 and 3.
cable and cell capacitance, sample
contamination, etc., since these are often Readily apparent is the variation in accuracy
installation-dependent. A sampling of among instruments due to widely varying
instruments that provide cation compensation approaches to the difficult challenge of cation
for power plant applications was evaluated. temperature compensation. Instrument A's
They represent the variety of equipment method apparently attempted to match most
installed in U.S. power plants today. closely at 0, 25 and 50 °C points as shown in
Figure 2. Instrument B gave a fairly good match
As a basis for the evaluation, a table of reference from 0-40°C but deviated widely above that, in
values was established, based on various low Figure 3. Instrument C did not have a true cation
concentrations of hydrochloric acid in pure compensation algorithm but did allow changing
water. Hydrochloric and sulfuric acids have the linear temperature coefficient in its G.E.-
been the most commonly used equivalent for type equation. No instructions were provided for
cation conductivity.9 Their conductivity vs. an appropriate coefficient and attempts at using
temperature properties are similar to each other estimated values gave large errors, examples of
and to many other acids that could be present. which are shown in Figure 4. Some
improvement might be possible with more

4
information. Instrument D, Figure 5, appeared conductivity measurement and 3) the effective
to match the acid characteristics at high compensation accuracy of the instrumentation
concentration and might be suitable for makeup over the sample conditions. Careful questioning
water treatment cation exchange monitoring. about the instrumentation and actual evaluation
However, for cycle chemistry measurement it is may be necessary since virtually all published
clearly unacceptable. accuracy specifications are of the direct
conductance measurement capability. Cation
A more recent unique three-dimensional curve compensation performance over temperature
fitting matrix algorithm was developed by and conductivity ranges is never specified.
Thornton Associates for its instrumentation. Its
performance in instrument E is illustrated in This work identifies significant errors in widely
Figure 6. A comparison with the other figures accepted equipment for this measurement. There
makes it clear there has been significant is clearly more work needed to evaluate all
progress toward more accurate cation cation and specific conductivity temperature
conductivity temperature compensation. Further compensation methods, preferably in an
comparison is provided with part of the data independent investigation. At the same time
(simulated pure water and 0.1 µS/cm cation there is the need for some instrument
conductivity) listed in Tables 2 and 3 and manufacturers to develop major improvements
plotted in Figures 7 and 8. This latest method to their algorithms.
improves the accuracy of cation compensation
by well over an order of magnitude. ACKNOWLEDGMENTS

CONCLUSION Appreciation is expressed to Robert A. Venditti

for his software wizardry in implementing high
Wherever cation conductivity measurement is accuracy compensation methods and to Mark J.
needed, accuracy requirements should consider: Cruickshank and Nicholas Morash for their
1) the realistic anticipated temperature and assistance in obtaining much of the data
conductivity ranges of the sample, 2) the presented here.
acceptable magnitude of error in cation

REFERENCES

1. Otten, G., "Measuring Water Purity by Specific Resistance," American Laboratory, July 1972, pp.
41-46.
2. Light, T. S., "Temperature Dependence and Measurement of Resistivity of Pure Water," Analytical
Chemistry, vol 56, June 1984, pp. 1138-1142.
3. Hunt, R. C., "Measurement of Conductivity in High Purity Water," Ultrapure Water Journal,
July/August 1985, pp. 26-29.
4. Light, T. S., and S. L. Licht, "Conductivity and Resistivity of Water from the Melting to Critical
Points," Analytical Chemistry, vol. 59, Oct. 1987, pp. 2327-2330.
5. Thornton, R.D., and T.S. Light, “A New Approach to Accurate Resistivity Measurement of High
Purity Water,” Ultrapure Water, July/August 1989, pp. 14-26.
6. Light, T. S., R. D. Thornton, K. R. Morash, C. H. Saunders and A. C. Bevilacqua, "Measurement of
the Resistivity of High-Purity Water at Elevated Temperatures," Ultrapure Water Journal, Dec.
1994, pp. 18-26.

5
7. Laurent, M. S., Unpublished G.E. Pure Water Equation, General Electric Vallecitos, CA Nuclear
Center.
8. Gray, D. M. and A. S. Tenney, "Improved Conductivity/Resistivity Temperature Compensation for
High Purity Water," Ultrapure Water Journal, Jul/Aug. 1986, pp. 27-30.
9. Test Method for On-Line Determination of Anions and Carbon Dioxide in High Purity Water by
Cation Exchange and Degassed Cation Conductivity, D4519, American Society for Testing and
Materials, Conshohocken, PA, 1994.

APPENDIX

Standard values for comparison of conductivity were calculated from the following equations across the
temperature range:

C = { λH+ [H+] + λOH- [OH-] + λCl- [Cl-] } d

[H+] = [OH-] + [Cl-]
Kw = [H+] [OH-]
where
C = total conductivity at temperature
λi = respective equivalent ionic conductance at temperature
[ ] = respective ion concentration
d = density of water at temperature
Kw = ion product of water at temperature
Table 1
Instrument Temperature Compensation Methods
Temp. Compensation Method Availability Applications
Fixed 2%/C coefficient Common with general purpose Cooling towers, wastewater,
analog and portable meters. manufacturing rinse waters, etc…
where modest accuracy is
acceptable
Adjustable linear coefficient Many on-line microprocessor- Processes where temperature
based instruments coefficients are known and
temperature range is limited
Preprogrammed direct % Many on-line microprocessor- Deionizer regeneration acids and
concentration readout with non- based instruments bases
linear compensation
Matrix % concentration readout Some on-line microprocessor- Process solutions where extensive
with linear coefficient based instruments concentration vs. conductivity and
temperature data are available
High purity water with neutral Some on-line microprocessor- Pure makeup water, BWR cycle
salts based instruments samples
Cation conductivity/ammonia/ Some on-line microprocessor- Cation and degassed cation
ETA (ethanolamine) based instruments conductivity samples, ammonia
and ETA treated samples

6
 Table 2
Instruments A,B,C, D, & E Cation Temperature Compensation Data for Pure Water
Instrument A Instrument B Instrument C Instrument D Instrument E
Comp Comp Comp Comp Comp
Temp Cond Error Cond Error Cond Error Cond Error Cond Error
(°C) (µS/cm) (%) (µS/cm) (%) (µS/cm) (%) (µS/cm) (%) (µS/cm) (%)
0 0.053 -3.7 0.055 -0.1 0.062 13 0.05514 0.2
5 0.043 -22 0.055 -0.1 0.063 14 0.05514 0.2
10 0.043 -22 0.054 -1.9 0.08 45 0.065 18 0.05511 0.1
15 0.045 -18 0.055 -0.1 0.05 -9.2 0.068 24 0.05513 0.1
20 0.049 -11 0.055 -0.1 0.06 9.0 0.076 38 0.05509 0.1
25 0.055 -0.1 0.055 -0.1 0.064 16 0.080 45 0.05517 0.2
30 0.049 -11 0.055 -0.1 0.05 -9.2 0.089 62 0.05507 0.0
35 0.046 -16 0.056 1.7 0.06 9.0 0.099 80 0.05505 0.0
40 0.046 -16 0.056 1.7 0.06 9.0 0.110 100 0.05513 0.1
45 0.052 -5.5 0.057 3.5 0.07 27 0.121 120 0.05500 -0.1
50 0.059 7.2 0.058 5.3 0.08 45 0.138 151 0.05514 0.2
55 0.050 -9.2 0.059 7.2 0.08 45 0.156 184 0.05496 -0.2
60 0.048 -13 0.060 9.0 0.09 64 0.175 218 0.05483 -0.4
65 0.049 -11 0.062 13 0.10 82 0.200 263 0.05486 -0.4
70 0.060 9.0 0.062 13 0.11 100 0.220 300 0.05535 0.5
75 0.069 25 0.064 16 0.12 118 0.253 360 0.05531 0.5
80 0.059 7.2 0.066 20 0.280 409 0.05515 0.2
85 0.052 -5.5 0.067 22 0.308 459 0.05534 0.5
90 0.054 -1.9 0.069 25 0.342 520 0.05509 0.1
95 0.045 -18 0.071 29 0.377 585 0.05538 0.6
100 0.085 54 0.076 38 0.419 661 0.05509 0.1

7
 Table 3
Instruments A,B,C, D, & E Cation Temperature Compensation Data at 0.100 µS/cm
Instrument A Instrument B Instrument C Instrument D Instrument E
Comp Comp Comp Comp Comp
Temp Cond Error Cond Error Cond Error Cond Error Cond Error
(°C) (µS/cm) (%) (µS/cm) (%) (µS/cm) (%) (µS/cm) (%) (µS/cm) (%)
0 0.099 -1.0 0.100 0.0 0.098 -2.0 0.0998 -0.2
5 0.096 -4.0 0.101 1.0 0.100 0.0 0.0998 -0.2
10 0.094 -6.0 0.101 1.0 0.200 100 0.101 1.0 0.0999 -0.1
15 0.094 -6.0 0.101 1.0 0.140 40 0.108 8.0 0.0999 -0.1
20 0.096 -4.0 0.101 1.0 0.110 10 0.110 10 0.0999 -0.1
25 0.101 1.0 0.100 0.0 0.101 1.2 0.118 18 0.0998 -0.2
30 0.094 -6.0 0.100 0.0 0.090 -10 0.121 21 0.0997 -0.3
35 0.090 -10 0.099 -1.0 0.090 -10 0.129 29 0.0998 -0.2
40 0.090 -10 0.099 -1.0 0.090 -10 0.139 39 0.0997 -0.3
45 0.094 -6.0 0.099 -1.0 0.090 -10 0.155 55 0.0998 -0.2
50 0.102 2.0 0.105 5.0 0.100 0.0 0.166 66 0.0998 -0.2
55 0.094 -6.0 0.101 1.0 0.100 0.0 0.183 83 0.0996 -0.4
60 0.092 -8.0 0.103 3.0 0.110 10 0.201 101 0.0995 -0.5
65 0.095 -5.0 0.103 3.0 0.110 10 0.220 120 0.1002 0.2
70 0.103 3.0 0.105 5.0 0.130 30 0.245 145 0.1002 0.2
75 0.114 14 0.108 8.0 0.140 40 0.272 172 0.1002 0.2
80 0.106 6.0 0.109 9.0 0.299 199 0.1001 0.1
85 0.102 2.0 0.111 11 0.333 233 0.0997 -0.3
90 0.105 5.0 0.114 14 0.367 267 0.1000 0.0
95 0.131 31 0.115 15 0.402 302 0.0996 -0.4
100 0.128 28 0.122 22 0.438 338 0.1003 0.3

8
 7
0.0551 uS/cm - 0 ppb HCl
Sensitivity (% Conductivity change/°C)

0.0700 uS/cm - 2.70 ppb HCl

6
0.100 uS/cm - 6.42 ppb HCl
0.250 uS/cm - 20.6 ppb HCl
5 0.500 uS/cm - 42.4 ppb HCl
1.000 uS/cm - 85.4 ppb HCl
4 2.50 uS/cm - 214 ppb HCl
5.00 uS/cm - 428 ppb HCl
10.00 uS/cm - 856 ppb HCl
3

0
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 1
Sensitivity of Cation Conductivity to Temperature

25
0.0551 uS/cm - 0 ppb HCl 0.0700 uS/cm - 2.70 ppb HCl
Compensated Conductivity Error (%)

20 0.100 uS/cm - 6.42 ppb HCl 0.250 uS/cm - 20.6 ppb HCl
15 0.500 uS/cm - 42.4 ppb HCl 1.000 uS/cm - 85.4 ppb HCl
2.50 uS/cm - 214 ppb HCl 5.00 uS/cm - 428 ppb HCl
10 10.00 uS/cm - 856 ppb HCl
5
0
-5
-10
-15
-20
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 2
Instrument A Cation Conductivity Temperature Compensation Performance

9
 25
Compensated Conductivity Error (%)

20
15
10
5
0
-5
0.0551 uS/cm - 0 ppb HCl 0.0700 uS/cm - 2.70 ppb HCl
-10 0.100 uS/cm - 6.42 ppb HCl 0.250 uS/cm - 20.6 ppb HCl
0.500 uS/cm - 42.4 ppb HCl 1.000 uS/cm - 85.4 ppb HCl
-15
2.50 uS/cm - 214 ppb HCl 5.00 uS/cm - 428 ppb HCl
-20 10.00 uS/cm - 856 ppb HCl
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 3
Instrument B Cation Conductivity Temperature Compensation Performance

25
Compensated Conductivity Error (%)

20
15
10
0.0551 uS/cm - 0 ppb HCl
5 0.0700 uS/cm - 2.70 ppb HCl
0 0.100 uS/cm - 6.42 ppb HCl
0.250 uS/cm - 20.6 ppb HCl
-5
0.500 uS/cm - 42.4 ppb HCl
-10 1.000 uS/cm - 85.4 ppb HCl
-15 2.50 uS/cm - 214 ppb HCl
5.00 uS/cm - 428 ppb HCl
-20
10.00 uS/cm - 856 ppb HCl
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 4
Instrument C Cation Conductivity Temperature Compensation Performance

10
 25
Compensated Conductivity Error (%)

20
15
10
0.0551 uS/cm - 0 ppb HCl
5
0.0700 uS/cm - 2.70 ppb HCl
0 0.100 uS/cm - 6.42 ppb HCl
0.250 uS/cm - 20.6 ppb HCl
-5
0.500 uS/cm - 42.4 ppb HCl
-10 1.000 uS/cm - 85.4 ppb HCl
-15 2.50 uS/cm - 214 ppb HCl
5.00 uS/cm - 428 ppb HCl
-20
10.00 uS/cm - 856 ppb HCl
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 5
Instrument D Cation Conductivity Temperature Compensation Performance

25
Compensated Conductivity Error (%)

20 0.0551 uS/cm - 0 ppb HCl 0.0700 uS/cm - 2.70 ppb HCl

0.100 uS/cm - 6.42 ppb HCl 0.250 uS/cm - 20.6 ppb HCl
15 0.500 uS/cm - 42.4 ppb HCl 1.000 uS/cm - 85.4 ppb HCl
10 2.50 uS/cm - 214 ppb HCl 5.00 uS/cm - 428 ppb HCl
10.00 uS/cm - 856 ppb HCl
5
0
-5
-10
-15
-20
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 6
Instrument E Cation Conductivity Temperature Compensation Performance

11
 25
Compensated Conductivity Error (%)

20
15
10
5
0
-5
-10 Instrument A
Instrument B
-15 Instrument C
Instrument D
-20
Instrument E
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 7
Cation Conductivity Temperature Compensation Performance for Pure Water

25
Compensated Conductivity Error (%)

20
15
10
5
0
-5
-10 Instrument A
Instrument B
-15 Instrument C
-20 Instrument D
Instrument E
-25
0 10 20 30 40 50 60 70 80 90 100
Temperature (°C)
Figure 8
Cation Conductivity Temperature Compensation Performance at 0.1 µS/cm

12