Impact of Film-Forming Amines on the Reliability of Online ... 2012. 14(9), 560-567.pdf · Impact of Film-Forming Amines on the Reliability of Online Analytical ... Three commonly

560 PowerPlant Chemistry 2012, 14(9)

PPCHEM

INTRODUCTION

Organic additives like oxygen scavengers or film-formingamines (FFA) are widely used in small industrial water-steam cycles, and their use in larger installations is sug-gested more and more frequently. Hardly any reportsdescribe their influence on the online monitoring system,and their impact on specific parameters has never beenreviewed properly [1].

A field test in a waste incineration plant [2] showed signifi-cant coating effects on conductivity and oxygen probes.With the test methods available in the field, no negativeeffects on pH and sodium probes could be measured. Butto draw a final conclusion, the recorded data showed toomany pattern noises coming from the operation mode ofthe plant. Furthermore, in this field test only one FFA prod-uct was involved.

This situation was the main motivation to repeat the test ina controlled laboratory-like installation with as few disturb-ing effects as possible. Additionally, the tests were per-formed with three different brands of film-forming aminesto give more relevant results.

ONLINE TEST SYSTEM

The flow scheme is shown in Figure 1. The raw water waspurified by reverse osmosis, mixed-bed ion exchange andUV disinfection to obtain ultrapure water (UPW). Table 1lists the average specifications of the UPW.

Oxygen Removal

The oxygen level could be reduced from saturation(8 mg · L–1) to below 10 µg · L–1 with three Liqui-Cel®Membrane Contactors connected in series. For a reduc-tion to 10 µg · L–1 O2 the contactors were evacuated with

Impact of Film-Forming Amines on the Reliability of Online Analytical Instruments

Impact of Film-Forming Amines on the Reliability ofOnline Analytical Instruments

© 2012 by Waesseri GmbH. All rights reserved.

ABSTRACT

There are very few reports describing the effects of the dosing of film-forming amines (FFAs) on the online monitoringequipment. This paper describes controlled, laboratory-like tests performed to ascertain the impact of three differentbrands of film-forming amines on system parameters. Specific conductivity, pH drift, cation resin retention, pH stabil-ity, sodium step response, sodium calibration, oxygen sensor response and ORP probe response were considered.While with some measurements no negative influence could be observed, with other equipment there was a loss ofsensitivity and speed of response time due to coating effects. There were also differences in some results dependingon the FFA used.

Marco Lendi and Peter Wuhrmann

NaCl/NH

addition3 FFA

addition

Input

UPW

Oxygen

removal

Exposed

instruments

Reference

instruments

Figure 1:

Flow scheme of the test system.

Parameter Range

Sodium 0.01–0.02 µg · L–1

Silica 0.1 µg · L–1

Conductivity 0.055–0.057 µS · cm–1

Oxygen Oxygen saturated (8–9 µg · L–1)

Table 1:

Average specifications of the input ultrapure water.

Heft2012-09:Innenseiten 25.10.12 17:53 Seite 560

561PowerPlant Chemistry 2012, 14(9)

PPCHEMImpact of Film-Forming Amines on the Reliability of Online Analytical Instruments

high vacuum and purged with nitrogen. In this way it waspossible to keep a stable oxygen level below 10 µg · L–1 forhours. The oxygen was only reduced for a short period totest the oxidation-reduction potential (ORP) and the oxy-gen probe's response while the other data was recordedwith saturated oxygen levels.

Dosing Points

Dosing was designed for a flow-dependent addition suchthat a constant concentration of a given substance couldbe maintained over several days. The first dosing pointwas used to dose sodium chloride solution and/or ammo-nia. With the second dosing point, the film-forming amines(FFA) were dosed. The reference instruments weremounted upstream of the second dosing point, so thatthese instruments were never exposed to film-formingamines.

Online Instruments

The complete list of the tested instruments is given inTable 2.

Test Procedures and Materials

Phase 1: Recording of the base lineDuring this phase, no FFA was dosed. Tests likesodium step responses or oxygen reductionwere performed to see the instruments reactionunder laboratory conditions.

Phase 2: Dosing of a small amount of FFAA target concentration of 0.5 mg · L–1 as FFAwas dosed. The goal of this phase was toobserve possible effects with low FFA concen-tration.

Phase 3: Verification stageThe dosing of FFA was interrupted, perfor-mance tests were continued. This phase gavean impression of the irreversibility of the coatingeffects.

Phase 4: High concentration FFA dosingTarget concentrations up to 3 mg · L–1 as FFAwere dosed. Depending on brand and applica-tion, these are the recommended concentra-tions of the manufacturers.

Three commonly used film-forming amine brands weretested. In the following report, the names as well as themanufacturers have been made anonymous. The shownFFA concentrations were calculated based on the flow-dependent addition and were frequently measured withthe available test methods of the used products.

TEST RESULTS

Conductivity Probe

The installed conductivity probes on the AMI Deltacon DGare titanium 2-electrode types with a cell constant of0.04 cm–1 and an integrated PT1000 temperature sensor.

Instruments Parameter Sample

Specific conductivityAcid conductivity

AMI Deltacon DGDegassed acid conductivity Reference instrumentsCalculated pH

AMI Sodium P Sodium

Specific conductivityAcid conductivity

AMI Deltacon DGDegassed acid conductivityCalculated pH

AMI pH Ion selective pH Exposed instruments

AMI ORP Oxidation-reduction potential

AMI Soditrace Trace sodium

AMI Oxytrace QED Trace oxygen with online verification

AMI Sodium A Sodium

Table 2:

Online instruments and parameters.



PPCHEM

Points of interest were:

(1) Coating effects on specific conductivity (SC) probe

(2) Drift between pH calculation and ion selective meas-urement

(3) Retention capacity of the cation exchange resin

A summary of the impact of the FFAs tested on specificconductivity, pH drift, and cation resin retention is given inTable 3.

Coating Effects on Specific Conductivity ProbeThis sensor is in direct contact with the FFA. The acid con-ductivity and the degassed acid conductivity sensors areprotected by the cation exchange resin. Of the threetested brands, only one substance did not decrease thespecific conductivity compared with acid and degassedacid conductivity.

Two criteria were established to detect a coating effect:

• Decrease in specific conductivity(SC) compared with degassed acidconductivity (DC) during FFA dos-ing;

• Lower specific conductivity valuesof the exposed instrument com-pared with the reference after alonger period of FFA dosing.

Figure 2 shows the behavior of the SCwith FFA no. 3. The FFA concentrationof substance no. 3 was 1 mg · L–1

during the whole measuring period. Atthe beginning, the SC reading was17.38 µS · cm–1, then it dropped to14.58 µS · cm–1. The decrease of2.06 µS · cm–1 within three days isclearly an indication of coating on theconductivity probe. This is supportedby the observation that during thesame period the degassed acid con-ductivity (DC) is more or less constantbetween 0.143 and 0.164 µS · cm–1.

Figure 3 shows the coating effect of FFA no. 2. After athree-day exposure time at 1 mg · L–1 FFA, a stepresponse with a target concentration of 100 µg · L–1

sodium was performed. The readings of the non-exposedreference conductivity probe and the exposed sensorwere compared.

At the beginning, the reading of the exposed sensor is0.489 µS · cm–1, which is considerably lower than the SCof the reference instrument (0.513 µS · cm–1). Figure 3 alsoshows that the coating effect disappeared about half anhour after the FFA dosing was stopped.

Difference between pH Calculation and Ion SelectiveMeasurement According to VGB-450L [3], the pH ofa sample can be calculated with the specific and acidconductivity. As shown previously, the specific conductiv-ity sensors are coated by two of the three film-formingamines. As a result, the calculated pH should alsodecrease compared to the ion selective measurement witha glass electrode.

FFA no. 1 FFA no. 2 FFA no. 3

Coating on SC probe No Yes Yes

pH drift No No Yes

Retention capacity Poor Poor Poor

Table 3:

Summary of results on specific conductivity, pH drift, and resin retention.

50

40

Conductivity

[µS

cm

]·

–1

Time [day]

0 1 2 3

30

20

100.5

0.4

0.3

0.2

0.1

Specific conductivity (SC)

Degassed acid conductivity (DC)

Figure 2:

Coating of specific conductivity (SC) probe during an exposure to 1 mg · L–1 of FFAno. 3.




PPCHEM

In Figure 4 the deviation between thecalculated and measured pH duringFFA no. 3 dosing is shown as ��pH.Over a three-day exposure time, thedeviation was –0.10 pH. Anotherpiece of evidence for the 'coating the-ory' is the fact that in the shown timethe ion selective measured pH (sepa-rated measuring and reference elec-trode) is not drifting.

A coating-like behavior of FFA no. 2could be demonstrated with a stepresponse, but the measured deviationwas much lower than the one fromFFA no. 3. The influence on pH calcu-lation was too small to demonstrate asignificant drift.

Retention Capacity of the CationExchange ResinThe quality of the cation exchangeresin is a key factor in measuring areliable and stable acid and degassedacid conductivity. The resin utilizedwas a strongly acidic cationexchanger in H+ form. Under normaloperation, the tested resin will holdback ammonia-treated water forapproximately 1 month before theresin is exhausted.

The dosed FFA substances containnot only the film-forming amines.According to the Material Safety DataSheets of the tested substances, themixtures contain a high load of othercationic molecules. Figure 5 showsthe beginning of a dosing (phase 2)with a low concentration of FFA no. 1.After three days, it seems that thecation exchange resin cannot holdback all cationic substances becausethe degassed acid conductivity (DC)of the exposed instrument is begin-ning to rise. Figure 6 shows the con-tinued test with FFA no. 1 after 44days of exposure. After the dosing ofFFA no. 1 was stopped, the DC of theexposed instrument remained on ahigh level. It seems that the cationexchange resin is continuing to release ionic substances.The DC reading went back to normal after a new bottle ofresin was installed. This is the proof that the source of theionic substances responsible for the higher DC readingswas the cation exchange resin.

FFAs no. 2 and no. 3 gave the same picture. The only dif-ference between the substances was the time until ionicsubstances broke through.

Conductivity

[µS

cm

]·

–1

SC reference instrument

SC exposed instrument

Dosing of NaCl

Dosing of FFA

0.7

0.6

0.5

0.4

0.3

0.2

0.1

Na

Dosin

g[m

gkg

]·

–1

0.5

0.4

0.3

0.2

0.1

0

2.5

2.0

1.5

1.0

0.5

0

00:00 00:30 01:00 01:30

Time [hh:minmin]

FFA

Dosin

g[m

gkg

]·

–1

Figure 3:

Step response with a sodium chloride solution after a FFA dosing period (substanceno. 2) to compare the specific conductivity (SC) reading of the exposed and referenceinstrument.

�pH

�pH

Control line

Calculated pH

Ion selective measured pH

0.8

0.6

0.4

0.2

0

–0.2

–0.4

pH

11

10

9

8

7

0

Time [day]

0.5 1.0 1.5 2.0 2.5 3.0

Figure 4:

Drift between the calculated and ion selective measured pH (1 mg · L–1 FFA no. 3).




PPCHEM

pH/Sodium (Ion Selective GlassElectrode)

The installed sodium reference instru-ment was an AMI Sodium P and theexposed sodium probes were on anAMI Soditrace. On both instruments,the same types of measuring and ref-erence electrodes were used. The ref-erence electrode was a calomel(Hg/HgCl) electrode with a KCl liquidjunction.

The installed pH electrode used a sep-arated reference electrode systembased on Ag/AgCl with a KCl liquidjunction.


(1) pH: stability during long-termexposure

(2) Sodium: reaction time during stepresponse

(3) Sodium: calibration data

Table 4 shows a summary of the testresults.

pH Stability Figure 7 shows a 10-day exposure period with FFA no. 1.The stability during that test wasgood, and the response to changes inFFA concentrations was fast andreversible. Therefore, no negativeinfluence could be noticed.

Sodium Step Response The nor-mal sodium level during the test wasbetween 0.02 µg · L–1 and 0.10 µg · L–1

sodium. The sensors were regener-ated weekly with a sodium-free etch-ing solution. With flow-dependentinjection of a sodium chloride solution,a 'contamination' of 200 µg · L–1 Na

DC of reference instrument

DC of exposed instrument

FFA dosing

1.0

0.5

2.0

1.5

1.0

0.5

0

0

Time [day]

2 4 6 8

0.1

0.05

Conductivity

[µS

cm

]·

–1

FFA

Dosin

g[m

gkg

]·

–1

Figure 5:

Retention capacity of cation exchange resin with respect to the degassed acidconductivity (DC) during a dosing period of FFA no. 1.

DC of reference instrument

DC of exposed instrument

FFA dosing

1.0

0.5

5.0

4.0

3.0

2.0

1.0

0

44

Time [day]

46 48 50 54

0.1

0.05

Conductivity

[µS

cm

]·

–1

FFA

Dosin

g[m

gkg

]·

–1

52

New cationexchange resin

Figure 6:

Retention capacity of cation exchange resin with respect to the degassed acidconductivity (DC) after a dosing period of FFA no. 1.


pH stability No influence No influence No influence

Sodium step response No influence No influence No influence

Sodium calibration No influence –– –

Table 4:

Summary of results on pH stability, sodium step response, and sodium calibration.



was simulated. The most importantparameter to evaluate the perfor-mance of an ion selective sodium sen-sor is the response time to concentra-tion changes. Figure 8 summarizesfive step responses during long-termexposure to different concentrationsof FFA no. 1. No significant loss of thesodium probe's response time can beobserved.

Oxygen and ORP Sensor

For direct oxygen measurement aClark oxygen sensor with a gold cath-ode and silver anode was used.Additionally, a Faraday electrode foronline verification was built in.

The ORP was measured with a sepa-rated platinum (Pt) measuring elec-trode and reference electrodes with aKCl liquid junction were used.


(1) Clark probe: response time duringoxygen concentration reduction

(2) ORP probe: response time duringoxygen concentration reduction

Response Time of Oxygen and ORPProbes to Oxygen ConcentrationChanges The long-termstability tests were performed withoxygen saturated ultrapure water(8 mg · L–1O2). To test the oxygen andORP probes' performance, oxygenwas removed (see Figure 1) from thesample water. The oxygen concentra-tion of the water during removal wasthe same. Table 5 shows a summary ofthe test results.


PPCHEM


Oxygen response time Good Good Good

ORP response time Bad Bad Bad

Table 5:Summary of results on oxygen sensor response and ORP response.

pH

12

11

10

9

8

7

6

10

8

6

4

2

0

0

Time [day]

2 4 6 8 10 12

FFA

Dosin

g[m

gkg

]·

–1

pH

FFA dosing

Figure 7:

Long-term stability of ion selective pH probe during dosing of FFA no. 1.

300

250

200

150

100

50

0

0.00

Time [h]

0.01 0.02 0.03 0.04

3 days (0.5 mg kg FFA)

12 days (1 mg kg FFA)




·

·

·

·

·

–1

–1

–1

–1

–1

Na

[µg

kg

]·

–1

Figure 8:

Recorded step responses from an AMI Soditrace during long-term exposure to FFAno. 1.




PPCHEM

Figure 9 shows three examples of ORPresponses during different FFA dosingphases. The ORP readings in Figure 9are standardized because the FFAconcentration itself also influences theORP value. The red line (0.5 mg · L–1)shows an ORP response after a 5-dayperiod of FFA dosing while the greenline (1 mg · L–1) was recorded after a13-day dosing period.

No alteration of the Clark sensor char-acteristics was observed under any ofthe conditions tested in these experi-ments (Figure 10).

CONCLUSION

The impact of three different condition-ing reagents containing film-formingamines on online instrumentation wastested in a laboratory-like installation.On pH, ion selective sodium measure-ment as well as on a Clark-type oxy-gen probe, no negative influence couldbe observed. On oxygen reductionpotential (ORP) measurement, alltested FFA substances result in a lossof sensitivity and speed of responsetime due to coating effects. The impacton conductivity measurement differedbetween the tested substances. Withtwo tested substances, coating effectson the conductivity probe could beobserved which resulted in a drift ofthe specific conductivity reading aswell as in a drift of the calculated pH.With one film-forming amine product,such effects could not be observed.

OR

P[m

V]

20

0

–20

–40

–60

–80

–100

–120

–140

–160

–180

8 000

6 000

4 000

2 000

0

00:00

Time [day]

00:20 00:40 01:00 01:20 01:40

O[µ

gkg

]2

–1

·

0 mg kg FFA

0.5 mg kg FFA

1 mg kg FFA

Oxygen level

·

·

·

–1

–1

–1

Figure 9:

Reaction of an ORP probe during oxygen removal with different FFA concentrations ofFFA no. 1.

200

150

100

50

0

1.2

1.0

0.8

0.6

0.4

0.2

0

0

Time [day]

2 4 6 8 10 12

FFA

Dosin

g[m

gkg

]·

–1

FFA concentration

O level2

O[µ

gkg

]2

–1

·

14 16

Figure 10:

Response of a Clark oxygen probe during FFA dosing (FFA no. 1).



PowerPlant Chemistry 2012, 14(9)

REFERENCES

[1] Verheyden, K. S., Ertryckx, R. A. M., De Wispelaere,M., Poelemans, N., PowerPlant Chemistry 2003, 5(9),516.

[2] Lendi, M., Wuhrmann, P., EPRI Major ComponentReliability Meeting, 2011 (Barcelona, Spain). Paper#16.

[3] VGB Guideline for Boiler Feedwater, Boiler Water andSteam of Steam Generators with a PermissibleOperating Pressure of > 68 bar, 1988. VGB PowerTech e.V., Essen, Germany, VGB-R 450-Le.

THE AUTHORS

Marco Lendi (B.S., Chemistry, University of AppliedScience, Zurich, Switzerland) joined Swan AnalyticalInstruments in 2009, working first as a quality managerand from 2010 in SWAN's customer support group. In2012 he was promoted to the R&D group.

Peter Wuhrmann (Ph.D., Analytical Chemistry, SwissFederal Institute of Technology (ETHZ), Zurich, Switzer -land) conducted several years of research work with ionsensitive microelectrodes for intracellular ion measure-ments at the Institute of Cell Biology at the Swiss FederalInstitute of Technology (ETHZ). He was a member of thefounding group of Swan Analytical Instruments, where hehas been responsible for R&D since 1991.

CONTACT

Marco LendiSwan Analytical Instruments AGP.O. Box 3988340 HinwillSwitzerland

E-mail: [email protected]

Dr. Peter Wuhrmann

[email protected]

Impact of Film-Forming Amines on the Reliability

Degassed Cation ConductivityAnalyzer AMI Deltacon DG

ANALYTICAL INSTRUMENTS

SWAN ANALYTISCHE INSTRUMENTE AG · CH-8340 HINWILwww.swan.ch · [email protected] · Phone +4144 9436300

� Automatic and continuous measurement of total, cation anddegassed cation conductivity.

� Re-boiler according to Larson-Lane (ASTM D4519-94).

� Tightly controlled degassing temperature.

� Calculation of sample pH andammonia concentration.

Ask for technical documentation or check our homepage.

www.swan.ch


Documents

Impact of Film-Forming Amines on the Reliability of Online ... 2012. 14(9), 560-567.pdf · Impact of Film-Forming Amines on the Reliability of Online Analytical ... Three commonly