Karl Fischer Titration Has Long Been the Method of Choice for Water Content Determination, Mainly Due To

AGCKarl Book Fischer 20% Titration Cyan for Water Content Determination METTLER TOLEDO Methods Compact Coulometric Application brochure Compact Volumetric Titration Excellence 38 METTLER TOLEDO

EDITORIAL

Dear Reader

For a big variety of products in a multitude of industries, knowing the water content is crucial with respect to product quality and properties. If the water content is not correct

drugs may perish too early (Pharmaceutical Industry) lubricants may not protect against wear anymore (Oil Industry) food may not taste as expected (Food Industry) and many more

Karl Fischer Titration has long been the method of choice for water content determination, mainly due to

short analysis time high accuracy and repeatability high selectivity to water directness (no additional reference method necessary)

With the new generation of METTLER TOLEDO Karl Fischer Titrators, water determination has never been more simple and secure. All models feature a brilliant touch screen, One ClickTM operation, and an individual home screen with user specific shortcuts. The titrators can be operated in many different languages, stand-alone and / or using the instrument control software LabX® titration. The new line consists of

Compact volumetric KF titrators (V20 / V30) Compact coulometric KF titrators (C20 / C30) Modular and flexible Excellence titrator models for general titration and volumetric KF water determination (T70 / T90)

In the first part, this brochure lists the methods of choice for your specific samples. The results, the methods used and more background information is given in a second part for a big variety of samples in numerous industry segments. Thirdly, an overview of Karl Fischer reagents is given to help finding the ones best suited for your sample. Last but not least, METTLER TOLEDO methods, available ready-to-use from our instruments, are described in detail.

This brochure, in combination with the Good Titration PracticeTM Brochure (GTP) for Karl Fischer Titration, provides you with all necessary information for simple and secure water content determination of your sample. We wish trouble-free analyses!

Mike Treyer Rolf Rohner Market Support Manager Marketing Manager BA Titration BA Titration

METTLER TOLEDO

Contents

Method Title

Volumetric Karl Fischer Titration

M300 Concentration and Content Determination with Water Standard 10.0 mg/g

M301 Water Content Determination in Di-Sodium Tartrate Dihydrate 15.66%

M302 Water Determination In Solvents: Toluene

M303 Water Determination in Ketones: Acetone

M304 Water Determination with Internal Extraction (Homogenizer): Milk Powder

M305 Water Determination with External Extraction: Tobacco

M306 Water Determination with Manual KF Drying Oven: Corn Starch

M307 Water Determination in Gases: Air

M311 Water Content Determination in Water Standard Tablet 10 mg

M312 KF-Oven Check with 5.55% Solid Water Standard

M313 Automated Water Determination at Different Temperatures

Coulometric Karl Fischer Titration

M314 Coulometer Check with Water Standard 1.00 mg/g

M315 Coulometric KF-Oven Check with 1% Solid Water Standard

M391 Coulometric Water Determination In Solvents: Toluene Dry

M392 Coulometric Water Determination in Gases: Nitrogen

M393 Coulometric Water Determination in Ketones: Acetone Dry

M394 Coulometric Water Determination with External Extraction: Sugar

M395 Coulometric Water Determination with Manual Drying Oven: PET granules

M396 Automated Coulometric KF Analysis at Different Temperatures: Polystyrene

METTLER TOLEDO - I - Karl Fischer Titration Applications METTLER TOLEDO

Karl Fischer Titration Applications

1 Karl Fischer Titration: The Method at a Glance...... 1 1.1 Solid samples...... 1 1.2 Liquid samples...... 4 1.3 Titration Methods ...... 6 1.4 Sample preparation and input...... 9

2 Food...... 11 2.1 Applications: Sugar and Sugar Products...... 11 2.2 Applications: Fats, Oils and Dairy Products ...... 14 2.3 Grains and Starchy Food...... 18 2.4 Applications: Spices...... 21 2.5 Applications: Various Foodstuffs ...... 25 2.6 Applications: Beverages and Drink Concentrates ...... 28

3 Cosmetics ...... 30 3.1 Coulometric determinations ...... 30 3.2 Volumetric determinations ...... 31

4 Pharmaceuticals ...... 33 4.1 General remarks ...... 33 4.2 Coulometric determinations ...... 34 4.3 Volumetric determinations ...... 35

5 Inorganic raw materials ...... 38 5.1 General remarks ...... 38 5.2 Coulometric determinations ...... 39 5.3 Volumetric determinations ...... 40

6 Organic raw materials ...... 42 6.1 General remarks ...... 42 6.2 Coulometric determinations ...... 43 6.3 Volumetric determinations ...... 45

7 Solvents...... 47 7.1 General remarks ...... 47 7.2 Coulometric determinations ...... 48 7.3 Volumetric determinations ...... 49

8 Petroleum and mineral oil products ...... 51 8.1 General remarks ...... 51 8.2 Coulometric determinations ...... 52 8.3 Volumetric determinations ...... 54

METTLER TOLEDO - II - Karl Fischer Titration Applications METTLER TOLEDO

9 Plastics, polymers and adhesives ...... 56 9.1 General remarks ...... 56 9.2 Coulometric determination...... 57 9.3 Volumetric determinations ...... 58

10 Dyes and agrochemicals ...... 60 10.1 General comments...... 60 10.2 Coulometric determinations ...... 60 10.3 Volumetric determinations ...... 60

11 Detergents and Surfactants...... 63 11.1 General remarks ...... 63 11.2 Coulometric determinations ...... 63 11.3 Volumetric determinations ...... 63

12 Silk, wool, cellulose, paper and wood ...... 65 12.1 General remarks ...... 65 12.2 Coulometric determinations ...... 65 12.3 Volumetric determinations ...... 66

13 Building materials...... 67 13.1 General remarks ...... 67 13.2 Coulometric determinations ...... 67 13.3 Volumetric determinations ...... 68

14 References ...... 69 14.1 Additional literature ...... 70 14.2 Official Standards...... 70

15 Appendix ...... 71 15.1 Formula for the external extraction...... 71 15.2 Formula for the external dissolution...... 72

16 Karl Fischer Reagents...... 73 16.1 A choice of manufacturers ...... 73 16.2 Coulometry: A choice of reagents and solvents ...... 73 16.3 Volumetric analysis: titrants and solvents...... 77 16.4 Water Standards for Karl Fischer volumetric titration...... 79

17 Hazards and waste disposal tips ...... 80 17.1 One-component reagents ...... 80 17.2 Two-component reagents: ...... 80 17.3 Reagents for coulometry:...... 80 17.4 Safety data for the KF-components and auxiliary solvents: ...... 80

METTLER TOLEDO - III - Karl Fischer Titration Applications METTLER TOLEDO Application M300-09 Concentration and Content Determination with Water Standard 10.0 mg/g

The concentration and content determination with liquid water standard 10.0 mg/g are performed by volumetric Karl Fischer titration.

Sample 1.5-2 g Preparation and Procedures Water Standard 10.0 mg/g (HYDRANAL® Water Standard 10.0) - Drawn out approximately 1 mL of water standard Certified value: 10.01 ± 0.04 mg/g 10.0 into a 10 mL syringe with needle to rinse it before sample analysis. Compound Water, H2O M = 18.01 g/mol - Discard the 1 mL standard portion and drawn the rest of the standard completely into the syringe. - The water standard is then injected into the KF Chemicals 50 mL methanol as a solvent titration vessel in aliquots of 1-1.5 mL. - The weight is determined by back-weighing technique.

Titrant HYDRANAL® -Composite 5 (5 mg H2O/mL), or CombiTitrant 5 apura®

(5 mg H2O/mL) Standard Water Standard 10.0 mg/g (HYDRANAL® Water Standard 10.0)

Indication DM143-SC Electrode Remarks - First, enter the certified value of the standard in CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 the setup of the titrator. → (RNH)•SO4CH3 + 2 (RNH)I - Before starting analysis, the concentration determination is performed by pressing “Start conc.” (Water Standard 10.0 mg/g): - The concentration determination is calculated Calculation Water content in mg/g: according to the formula: CONC = CONT*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000) ( VEQ*CONC-(TIME*DRIFT/1000) ) * C ------CONT = Conc. of liquid water standard (mg/g) m CONC(alt) = CONC(alt) refers to the Setup value

current at the time of calculation. VEQ: Titrant consumption to EP TIME: Total time since sample - Subsequently, the sample analysis is started by request pressing “Start sample” on the display. C: 1 (constant for mg/g calc.)

Waste Organic solvents disposal

Author, Cosimo De Caro, MSG Version July 2009

METTLER TOLEDO Page 1 of 4 Titration Application M300-09 Instruments - KF Compact Volumeters V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1) - XP205 Balance

Accessories - LabX Titration Software - 10 mL syringe with needle

Results Standard: 10.01 ± 0.04 mg/g

No. Comment / ID Start time Sample size and results 1/7 -- 15.07.2009 17:18:52 1.20280 g R1 = 9.98 mg/g Content 2/7 -- 15.07.2009 17:23:28 1.72360 g R1 = 10.02 mg/g Content 3/7 -- 15.07.2009 17:27:36 1.44504 g R1 = 10.03 mg/g Content 4/7 -- 15.07.2009 17:31:29 1.49153 g R1 = 9.97 mg/g Content 5/7 -- 15.07.2009 17:35:34 1.75782 g R1 = 9.99 mg/g Content 6/7 -- 15.07.2009 17:39:34 1.72488 g R1 = 9.98 mg/g Content 7/7 -- 15.07.2009 17:43:20 2.71281 g R1 = 10.00 mg/g Content

Statistics Rx Name n Mean value Unit s srel[%] R1 Content 7 10.00 mg/g 0.02 0.223

Additional results Reagent n Content[mg/g] srel[%] Comments

CombiTitrant 5 apura® 5 10.02 ± 0.02 mg/g same parameters as for HYDRANAL® Composite 5 reagent HYDRANAL® Titrant 5 6 10.03 ± 0.03 mg/g 2-components control parameters, Other parameters, see “Method” HYDRANAL® Titrant 5 E 6 10.00 ± 0.01 mg/g Other parameters, see “Method”

Titration curve

E – t curve V – t curve

Sample 4/7

METTLER TOLEDO Page 2 of 4 Titration Application M300-09 Table of measured values

Volume Meas. value H2O Drift Time mL mV mg µg/min s 0.00000 515.1 0.0000 2.3 0 0.00000 515.1 0.0000 2.3 0 0.00000 515.4 0.0000 0.0 1 0.00125 515.4 0.0059 175.8 2 0.00625 515.1 0.0293 586.3 3 0.01525 515.2 0.0715 1073.1 4 0.02675 515.2 0.1255 1506.0 5 0.04350 515.1 0.2041 2040.8 6 0.06425 515.0 0.3014 2583.7 7 0.08875 514.5 0.4164 3122.8 8 0.11750 515.4 0.5513 3675.0 9 0.15000 515.0 0.7037 4222.4 10 0.18675 515.1 0.8761 5204.8 11 0.26100 511.9 1.2245 7239.4 12 0.33425 513.5 1.5681 9141.5 13 ... … … … … 3.17075 87.7 14.8757 55.7 148 3.17075 87.5 14.8757 55.7 149 3.17100 90.1 14.8769 48.8 150 3.17125 89.1 14.8780 48.8 151 3.17125 81.9 14.8780 48.8 152 3.17125 74.4 14.8780 48.8 153 3.17125 75.3 14.8780 48.8 154 3.17125 74.3 14.8780 48.8 155 3.17125 88.5 14.8780 41.8 156 3.17125 76.0 14.8780 13.9 157 3.17125 76.0 14.8780 13.9 157 Sample 4/7

Comments - This application represents a general method for determination in organic solvents, e.g. methanol. - To achieve precise results it is necessary to weigh the sample accurately, and in particular, to first clean the syringe with a few mL of liquid standard which are then discarded. - To ensure a more efficient cleaning, the syringe is gently shaken to allow the standard to absorb the moisture on the inner wall of the syringe. Subsequently, the whole liquid standard is aspired into the syringe. - When using two-component reagents, the corresponding titrant has to be selected in the titration function, e.g. select “KF 2-comp 5”.

METTLER TOLEDO Page 3 of 4 Titration Application M300-09 Method V2.0 001 Title Extra statistical functions No Type Karl Fischer titration Compatible with T70/T90/V20/V30 007 Record ID M300 Summary Per sample Title Water Standard 10.0 mg/g Results No Author decaro Raw results No Date/Time 15.07.2009 14:58:15 Table of measured values No Modified at 15.07.2009 15:02:35 Sample data No Modified by decaro Resource data No Protect No E-V curve No SOP None E-t curve No V-t curve No

002 Sample H2O-t No Sample Drift-t No

Number of IDs 1 H2O-t & Drift-t curves No ID 1 -- V-t & drift-t No Entry type Weight Method No Lower limit 0.0 g Series data No Upper limit 5.0 g Density 1.0 g/mL 008 End of sample Correction factor 1.0 Open series Yes Weight per piece 1.0 g Temperature 25.0°C ------Autostart Yes Control parameters for the two-components reagents: Entry After addition HYDRANAL® Titrant 5 Concentration Titrant KF 1-comp 5 ... Nominal concentration 5 mg/mL 005 Titration (KF Vol) [1] Standard Water-Standard 10.0 Titrant Entry type Weight Titrant KF 2-comp 5E Lower limit 0.0 g Nominal concentration 5 mg/mL Upper limit 2.0 g Reagent type 2-comp Temperature 25.0°C Sensor Mix time 10 s Type Polarized Autostart Yes Sensor DM143-SC Entry After addition Unit mV Concentration limits Yes Indication Voltametric Lower limit 4.5 mg/mL Ipol 24.0 µA Upper limit 5.6 mg/mL Stir Speed 35 % 003 Titration stand (KF stand) Predispense Type KF stand Mode None Titration stand KF stand Wait time 0 s Source for drift Online Control Max. start drift 25 µg/min End point 100.0 mV Control band 400.0 mV 004 Mix time Dosing rate(max) 3.0 mL/min Duration 15 s Dosing rate(min) 100.0 µL/min Start Normal 005 Titration (KF Vol) [1] Termination Titrant Type Drift stop relative Titrant KF 1-comp 5 Drift 15.0 µg/min Nominal concentration 5 mg/mL At VMax 10.0 mL Reagent type 1-comp Min. time 0 s Sensor Max. time infinity Type Polarized ------Sensor DM143-SC Control parameters for the two-components ethanolic reagents: Unit mV HYDRANAL® Titrant 5 E Indication Voltametric Ipol 24.0 µA ... Stir 005 Titration (KF Vol) [1] Speed 35 % Titrant Predispense Titrant KF 2-comp 5E Mode None Nominal concentration 5 mg/mL Wait time 0 s Reagent type 2-comp Control Sensor End point 100.0 mV Type Polarized Control band 400.0 mV Sensor DM143-SC Dosing rate(max) 5.0 mL/min Unit mV Dosing rate(min) 80.0 µL/min Indication Voltametric Start Normal Ipol 15.0 µA Termination Stir Type Drift stop relative Speed 35 % Drift 15.0 µg/min Predispense At VMax 10.0 mL Mode None Min. time 0 s Wait time 0 s Max. time infinity Control End point 100.0 mV 006 Calculation R1 Control band 400.0 mV Result tape Predefined Dosing rate(max) 3.0 mL/min Result Content Dosing rate(min) 250.0 µL/min Result unit mg/g Start Normal Formula R1=(VEQ*CONC- Termination TIME*DRIFT/1000)*C/m Type Drift stop relative Constant C= 1 Drift 15.0 µg/min Decimal places 2 At VMax 10.0 mL Result limits No Min. time 0 s Record statistics Yes Max. time infinity

METTLER TOLEDO Page 4 of 4 Titration Application M300-09 METTLER TOLEDO Application M301-09 Water Content Determination in Di-Sodium Tartrate Dihydrate

The concentration and content determination with the solid water standard di-sodium tartrate dihydrate

(15.66% H2O) are performed by volumetric Karl Fischer titration.

Sample 50 mg sodium tartrate dihydrate Preparation and Procedures (HYDRANAL® Standard sodium tartrate dihydrate) - A mixing time of 180 s is required to ensure that the finely ground di-sodium tartrate is completely dissolved. Sodium tartrate, which is not finely Compound Water, H2O ground, needs a significantly longer mixing time. M = 18.01 g/mol - The solid standard must be completely dissolved into methanol to achieve correct results. Chemicals 60 mL methanol as a solvent - A turbid solvent solution indicates that the solid (max. 3 samples) standard can not be dissolved anymore. In this or case, the solvent must be replaced. 120 mL methanol (6 samples) - The solvent should always be replaced after three samples, i.e. after a total amount of max. Titrant HYDRANAL® -Composite 5 150 mg sodium tartrate dihydrate. (5 mg H2O/mL), - When adding the standard with the weighing or boat, care must be taken that no standard is ® CombiTitrant 5 apura adhering to the beaker wall or the electrode. (5 mg H2O/mL) Here, a weighing boat with rubber flexible tube can be very useful. Standard Water Standard 10.0 mg/g ® (HYDRANAL Water Standard 10.0) - The weight is determined by back-weighing technique. Indication DM143-SC Electrode Remarks - Before analysis, the concentration determination CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 is performed by pressing “Start conc.” (Water → Standard 10.0 mg/g). (RNH)•SO4CH3 + 2 (RNH)I - The concentration is calculated according to the formula:

CONC = CONT*10*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000)

Calculation Water content in %: CONT = Conc. of liquid water standard (%) ( VEQ*CONC-(TIME*DRIFT/1000) ) * C CONC(alt) = CONC(alt) refers to the setup value at the time ------of calculation. m - Subsequently, the sample analysis is started by VEQ: Titrant consumption to EP pressing “Start sample” on the display. TIME: Total time since sample request C: 0.1 (constant for % calc.)

Waste Organic solvents disposal

Author, Cosimo De Caro, MSG Version July 2009

METTLER TOLEDO Page 1 of 4 Titration Application M301-09 Instruments - KF Compact Volumeters V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1) - XP205 Balance

Accessories - LabX Titration Software - Weighing boat, ME-23952

Results Standard: 15.66 ± 0.05 % Titrant: HYDRANAL®-Composite 5 No. Comment / ID Start time Sample size and results 1/6 -- 22.07.2009 08:56:10 0.04763 g R1 = 15.70 % Content 2/6 -- 22.07.2009 09:08:29 0.04843 g R1 = 15.62 % Content 3/6 -- 22.07.2009 09:18:25 0.04105 g R1 = 15.82 % Content 4/6 -- 22.07.2009 09:27:37 0.03881 g R1 = 15.55 % Content 5/6 -- 22.07.2009 09:39:46 0.05013 g R1 = 15.54 % Content 6/6 -- 22.07.2009 09:49:22 0.05044 g R1 = 15.58 % Content Statistics Rx Name n Mean value Unit s srel[%] R1 Content 6 15.64 % 0.11 0.689

Titration curve

Sample 2/6

METTLER TOLEDO Page 2 of 4 Titration Application M301-09 Table of measured values

Volume Meas. value H2O Drift Time mL mV mg µg/min s 0.00000 680.4 0.0000 0.0 0 0.00000 680.5 0.0000 0.0 1 0.00100 680.7 0.0046 138.6 2 0.00575 678.7 0.0266 531.8 3 0.01450 674.1 0.0671 1005.8 4 0.02700 671.6 0.1249 1498.2 5 0.04375 665.1 0.2023 2023.1 6 0.06400 662.4 0.2960 2536.7 7 0.08175 657.9 0.3780 2835.2 8 0.09925 658.4 0.4590 3059.7 9 0.11550 667.8 0.5341 3204.6 10 0.12500 667.8 0.5780 3433.8 11 0.16450 643.4 0.7607 4518.9 12 0.22900 570.3 1.0589 6263.3 13 0.23325 648.7 1.0786 6249.6 14 …….. 1.64425 80.0 7.6034 171.7 163 1.64425 63.0 7.6034 137.4 164 1.64425 82.4 7.6034 96.1 165 1.64425 75.9 7.6034 54.9 166 1.64425 67.9 7.6034 27.5 167 1.64425 101.8 7.6034 27.5 168 1.64425 77.8 7.6034 27.5 169 1.64450 117.1 7.6045 27.5 170 1.64450 81.3 7.6045 27.5 171 1.64475 121.6 7.6057 13.7 172 1.64525 94.8 7.6080 13.7 172 Sample 2/6 Size: 0.04843 g

Comments - Sodium tartrate dihydrate is slowly and poorly dissolving in methanol. Therefore, a stirring time of the least 3-5 minutes is needed to completely dissolve it. - The solubility of sodium tartrate dehydrate is limited in methanol: in general, it is not possible to perform more than 3 determinations of approx. 50 mg aliquot within 60 mL of methanol. - After three determinations, methanol has to be exchanged since it becomes turbid an the subsequent measurements clearly deviate from the average value of the first three samples. - On the other hand, if one would like to run a sample series of 6 samples, then at least 120 mL methanol is added into the titration vessel. - To improve solubility, only use finely ground sodium tartrate dihydrate. - Air humidity has a strong influence on the results. When the moisture in the air is too high, then the results are higher, and their deviation is also pronounced. It is recommended to work under controlled conditions in order to reduce air humidity. - To get accurate and precise results, it is necessary to perform the measurements in e.g. a controlled environment to reduce the interference of air humidity.

METTLER TOLEDO Page 3 of 4 Titration Application M301-09 Method V2.0 001 Title 007 Record Type Karl Fischer titration Summary Per sample Compatible with T70/T90/V20/V30 Results No ID M301 Raw results No Title Di-Sodium-Tartrate-15.66% Table of measured values No Author METTLER TOLEDO Sample data No Date/Time 20.07.2009 08:32:46 Resource data No Modified at 21.07.1009 07:31:23 E-V curve No Modified by METTLER TOLEDO E-t curve No Protect No V-t curve No

SOP None H2O-t No Drift-t No 002 Sample H2O-t & Drift-t curves No Number of IDs 1 V-t & drift-t No ID 1 -- Method No Entry type Weight Series data No Lower limit 0.04 g Upper limit 0.08 g 008 End of sample Density 1.0 g/mL Open series Yes Correction factor 1.0 Temperature 25.0°C Autostart No Entry After addition Concentration Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Standard Water-Standard 10.0 Entry type Weight Lower limit 0.0 g Upper limit 2.0 g Temperature 25.0°C Mix time 10 s Autostart Yes Entry After addition Concentration limits Yes Lower limit 4.5 mg/mL Upper limit 5.6 mg/mL

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 25 µg/min

004 Mix time Duration 180 s

005 Titration (KF Vol) [1] Titrant Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Reagent type 1-comp Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol 24.0 µA Stir Speed 35 % Predispense Mode None Wait time 0 s Control End point 100.0 mV Control band 400.0 mV Dosing rate(max) 5 mL/min Dosing rate(min) 80 µL/min Start Normal Termination Type Drift stop relative Drift 15.0 µg/min At VMax 10.0 mL Min. time 0 s Max. time infinity

006 Calculation R1 Result tape Predefined Result Content Result unit % Formula R1=(VEQ*CONC- TIME*DRIFT/1000)*C/m Constant C= 0.1 Decimal places 2 Result limits No Record statistics Yes Extra statistical functions No

METTLER TOLEDO Page 4 of 4 Titration Application M301-09 METTLER TOLEDO Application M302-09 Water Determination in Solvents

The water content in dry toluene is determined by volumetric Karl Fischer titration.

Sample Toluene, dry Preparation and Procedures 3-4 g - Drawn out approximately 2-3 mL of sample into a 10 mL syringe with needle to rinse it before sample analysis. Compound Water, H2O M = 18.01 g/mol - Discard the sample into disposal container and fill the syringe again with the sample. - The sample is then injected into the KF titration Chemicals 50 mL methanol as a solvent vessel in aliquots of 2.5-3 mL. ® e.g. CombiMethanol apura - The weight is determined by back-weighing technique. - If the amount of water is very small, the titration can also be started with the parameter “Cautious” ® Titrant HYDRANAL -Composite 2 (see Control Parameters). (2 mg H2O/mL), or ® CombiTitrant 2 apura Note: (2 mg H2O/mL) The sample can be prepared by extracting water Standard Water Standard 10.0 mg/g from toluene using molecular sieves. For this, (HYDRANAL® Water Standard 10.0) toluene is added into a flask containing molecular sieve on the bottom. The flask can be closed with a septum. Indication DM143-SC Electrode Remarks - First, the concentration determination is CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 performed by pressing “Start conc.” (Water → Standard 10.0 mg/g, approx. 0.8 g). (RNH)•SO4CH3 + 2 (RNH)I - The concentration determination is calculated according to the formula:

CONC = CONT*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000) Calculation Water content in ppm: CONT = Conc. of liquid water standard (mg/g) CONC(alt) = CONC(alt) refers to the Setup value ( VEQ*CONC-(TIME*DRIFT/1000) ) * C current at the time of calculation. ------m - Subsequently, the sample analysis is started by pressing “Start sample” on the display. VEQ: Titrant consumption to EP TIME: Total time since sample request C: 1000 (constant for ppm calc.)

Waste Organic solvents disposal

Author, Melanie Nijman, MSG Version July 2009

METTLER TOLEDO Page 1 of 4 Titration Application M302-09 Instruments - KF Compact Line Volumeters V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1.) - XP205 Balance

Accessories - LabX Titration Software - 10 mL syringe with needle

Results HYDRANAL® -Composite 2 No. Comment / ID Start time Rx Result Unit Name

1/8 -- 16.07.2009 10:45:18 R1 = 254.8 ppm Content 2/8 -- 16.07.2009 10:48:48 R1 = 254.4 ppm Content 3/8 -- 16.07.2009 10:52:09 R1 = 252.9 ppm Content 4/8 -- 16.07.2009 11:03:13 R1 = 253.7 ppm Content 5/8 -- 16.07.2009 11:06:32 R1 = 252.9 ppm Content 6/8 -- 16.07.2009 11:11:09 R1 = 254.7 ppm Content 7/8 -- 16.07.2009 11:19:48 R1 = 256.6 (2) ppm Content 8/8 -- 16.07.2009 11:24:52 R1 = 256.7 (2) ppm Content

Statistics Rx Name n Mean value Unit s srel[%] R1 Content 6 253.9 ppm 0.864870 0.341

Excluded results: 7/8, 8/8

Additional results R1 Content (apura® CombiTitrant 2) 6 (7) 97.4 ppm 0.671565 0.690 R1 Content (apura® Titrant 2) 6 97.1 ppm 0.511534 0.527 R1 Content (HYDRANAL® Titrant 2) 6 105.2 ppm 0.4 0.393

Titer determination Titer (HYDRANAL® Composite 2) 6 2.243095 mg/mL 0.003866 0.172 Titer (HYDRANAL® Titrant 2) 7 1.967454 mg/mL 0.005653 0.287 Titer (apura® CombiTitrant 2) 6 2.034590 mg/mL 0.003792 0.186 Titer (apura® Titrant 2) 6 2.105130 mg/mL 0.006554 0.311

Titration curve

Sample 1/8

METTLER TOLEDO Page 2 of 4 Titration Application M302-09 Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s

0.00000 NaN 0.0000 0.0 690.0 NaN 0 0.00000 0.00000 0.0000 0.0 690.1 0.1 0 0.00000 0.00000 0.0000 0.0 690.3 0.2 1 0.00050 0.00050 0.0011 33.6 690.4 0.1 2 0.00100 0.00050 0.0022 44.8 690.4 0.0 3 0.00150 0.00050 0.0034 50.4 690.8 0.4 4 0.00225 0.00075 0.0050 60.5 690.6 -0.2 5 0.00325 0.00100 0.0073 72.9 689.7 -0.9 6 0.00450 0.00125 0.0101 86.5 689.0 -0.7 7 0.00600 0.00150 0.0135 100.9 688.1 -0.9 8 0.00775 0.00175 0.0174 115.9 687.3 -0.8 9 0.01000 0.00225 0.0224 133.2 685.7 -1.6 10 ...... 0.35125 0.00025 0.7879 89.9 140.5 15.7 121 0.35175 0.00050 0.7890 73.3 129.1 -11.4 122 0.35250 0.00075 0.7907 70.0 112.7 -16.4 123 0.35275 0.00025 0.7913 63.3 120.7 8.0 124 0.35325 0.00050 0.7924 63.3 113.2 -7.5 125 0.35375 0.00050 0.7935 63.3 98.9 -14.3 126 0.35375 0.00000 0.7935 56.6 97.2 -1.7 127 0.35450 0.00075 0.7952 60.0 97.4 0.2 128 0.35475 0.00025 0.7957 60.0 101.4 4.0 129 0.35500 0.00025 0.7963 53.3 95.1 -6.3 130 0.35500 0.00000 0.7963 50.0 93.5 -1.6 132 0.35550 0.00050 0.7974 50.0 83.1 -10.4 132 0.35550 0.00000 0.7974 40.0 98.9 15.8 134 0.35575 0.00025 0.7980 40.0 97.1 -1.8 134 0.35600 0.00025 0.7985 36.7 91.9 -5.2 136 0.35625 0.00025 0.7991 33.3 89.0 -2.9 136 0.35625 0.00000 0.7991 33.3 88.7 -0.3 138 0.35625 0.00000 0.7991 23.3 87.2 -1.5 139 0.35625 0.00000 0.7991 20.0 93.2 6.0 139 Sample 1/8

Comments - This application represents a general method for determination in organic solvents, e.g. methanol, which do not lead to side reactions with the Karl Fischer reagents. - To achieve precise results it is necessary to weigh the sample accurately, and in particular, to first clean the syringe with a few mL sample which is then discarded. - To ensure a more efficient cleaning, the syringe is gently shaken to allow for absorption of the moisture on the inner wall of the syringe. Subsequently, the syringe is completely filled. - Care has to be taken to start analysis at approx. the same potential value when titrating samples of low water content. In order to get accurate and precise results. - When titrating several samples be sure to replace the KF solvent in the cell when the sample quantity is reaching not more than 50% of the total amount.

METTLER TOLEDO Page 3 of 4 Titration Application M302-09 Method (V2.0) 001 Title Type Karl Fischer titration 007 Record Compatible with T70 / T90 / V20 / V30 Summary Per sample ID M302 Results No Title Toluene Raw results No Author METTLER TOLEDO Table of measured values No Date/Time 16.07.2009 08:40:11 Sample data No Modified at 16.07.2009 08:40:11 Resource data No Modified by METTLER TOLEDO E-V curve No Protect No E-t curve No SOP None V-t curve No

H2O-t No 002 Sample Drift-t No

Number of IDs 1 H2O-t & Drift-t curves No ID 1 -- V-t & drift-t No Entry type Weight Method No Lower limit 3.0 g Series data No Upper limit 8.0 g Density 1.0 g/mL 008 End of sample Correction factor 1.0 Open series Yes Weight per piece 1.0 g Temperature 25.0°C Autostart No Entry After addition Concentration Titrant KF 1-comp 2 Nominal concentration 2 mg/mL Standard Water-Standard 10.0 Entry type Weight Different reagents: Lower limit 0.0 g Upper limit 2.0 g apura® CombiTitrant 2 Temperature 25.0°C … Mix time 10 s Termination Autostart Yes Type Drift stop relative Entry After addition Drift 15.0 µg/min Concentration limits Yes At VMax 10.0 mL Lower limit 1.7 mg/mL Min. time 10 s Upper limit 2.3 mg/mL Max. time infinity … 003 Titration stand (KF stand) Type KF stand Titration stand KF stand HYDRANAL® Titrant 2 Source for drift Determination … Max. start drift 25 µg/min Control End point 100.0 mV 004 Mix time Control band 400.0 mV Duration 15 s Dosing rate(max) 3 mL/min Dosing rate(min) 100 µL/min 005 Titration (KF Vol) [1] Start Cautious Titrant Termination Titrant KF 1-comp 2 Type Drift stop relative Nominal concentration 2 mg/mL Drift 15.0 µg/min Reagent type 1-comp At VMax 10.0 mL Sensor Min. time 10 s Type Polarized Max. time infinity Sensor DM143-SC … Unit mV Indication Voltametric Ipol 24.0 µA apura® Titrant 2 Stir … Speed 35 % Control Predispense End point 100.0 mV Mode None Control band 400.0 mV Wait time 0 s Dosing rate(max) 3 mL/min Control Dosing rate(min) 100 µL/min End point 100.0 mV Start Cautious Control band 400.0 mV Termination Dosing rate(max) 2 mL/min Type Drift stop relative Dosing rate(min) 40.0 µL/min Drift 15.0 µg/min Start Cautious At VMax 10.0 mL Termination Min. time 10 s Type Drift stop relative Max. time infinity Drift 15.0 µg/min … At VMax 10.0 mL Min. time 0 s Max. time infinity

006 Calculation R1 Result tape Predefined Result Content Result unit ppm Formula R1=(VEQ*CONC- TIME*DRIFT/1000)*C/m Constant C= 1000 Decimal places 1 Result limits No Record statistics Yes Extra statistical functions No

METTLER TOLEDO Page 4 of 4 Titration Application M302-09 METTLER TOLEDO Application M303-09 Water Determination in Ketones

The water content in acetone (ketone) is determined by volumetric Karl Fischer titration.

Sample Acetone Preparation and Procedures 1.5-2 g - Drawn out approximately 2-3 mL of sample into a 10 mL syringe with needle to rinse it before sample analysis. Compound Water, H2O M = 18.01 g/mol - Discard the sample into disposal container and fill the syringe again with the sample. - The sample is then injected into the KF titration Chemicals 60 mL solvent vessel in aliquots of approx. 2-3 mL. ® HYDRANAL KetoSolver - The weight is determined by back-weighing or technique. apura® CombiSolvent 5 Keto - If the amount of water is very small, the titration can also be started with the parameter “Cautious” ® Titrant HYDRANAL Composite 5 K (see Control Parameters). (5 mg H2O/mL) or ® apura CombiTitrant 5 Keto

Standard Water Standard 10.0 mg/g (HYDRANAL® Water Standard 10.0)

Indication DM143-SC Electrode Remarks - First, the concentration determination is CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 performed by pressing “Start conc.” (Water → Standard 10.0 mg/g, approx. 0.8 g). (RNH)•SO4CH3 + 2 (RNH)I - The concentration determination is calculated according to the formula:

Waste Organic solvents disposal

Author, Market Support Group Anachem Version Maria-José Schmid, August 2009

METTLER TOLEDO Page 1 of 4 Titration Application M303-09 Instruments - KF Compact Line Volumeters V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1) - XP205 Balance

Accessories - LabX Titration Software - 10 mL syringe with needle

Results HYDRANAL® Composite 5 K / KetoSolver No. Comment / ID Start time Sample size and results 1/6 -- 20.08.2009 09:53:25 R1 = 0.209 % Content 2/6 -- 20.08.2009 11:06:25 R1 = 0.216 % Content 3/6 -- 20.08.2009 11:48:21 R1 = 0.218 % Content 4/6 -- 20.08.2009 12:02:55 R1 = 0.205 % Content 5/6 -- 20.08.2009 12:09:48 R1 = 0.209 % Content 6/6 -- 20.08.2009 12:19:20 R1 = 0.214 % Content

Statistics Rx Name n Mean value Unit s srel[%]

R1 Content 6 0.212 % 0.005 2.340

------

apura® CombiTitrant 5 Keto/CombiSolvent 5 Keto

Statistics Rx Name n Mean value Unit s srel[%] R1 Content 6 0.209 % 0.005 2.182

Titration curve

Sample 1/6 20.08.2009 09:53:25

METTLER TOLEDO Page 2 of 4 Titration Application M303-09 Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s 0.00000 NaN 0.0000 0.0 671.1 NaN 0 0.00000 0.00000 0.0000 0.0 671.1 0.0 0 0.00000 0.00000 0.0000 0.0 671.7 0.6 1 0.00100 0.00100 0.0049 153.2 672.7 1.0 2 0.00550 0.00450 0.0267 552.6 672.2 -0.5 3 0.01400 0.00850 0.0680 1045.9 659.5 -12.7 4 0.02675 0.01275 0.1299 1590.5 617.2 -42.3 5 0.02700 0.00025 0.1311 1333.2 612.0 -5.2 6 0.02750 0.00050 0.1335 1161.1 630.3 18.3 7 0.03075 0.00325 0.1493 1134.0 634.6 4.3 8 0.03800 0.00725 0.1845 1243.9 621.2 -13.4 9 0.04950 0.01150 0.2404 1456.7 579.7 -41.5 10 … … … … … … … 0.97900 0.00025 4.7538 50.5 147.5 -0.4 232 0.97900 0.00000 4.7538 43.3 145.6 -1.9 234 0.97900 0.00000 4.7538 28.8 149.2 3.6 234 0.97900 0.00000 4.7538 21.6 147.0 -2.2 236 0.97925 0.00025 4.7550 28.8 147.5 0.5 236 0.97950 0.00025 4.7562 36.1 146.9 -0.6 238 0.97950 0.00000 4.7562 36.1 147.8 0.9 238 0.97975 0.00025 4.7575 36.1 146.7 -1.1 240 0.97975 0.00000 4.7575 36.1 147.8 1.1 241 0.97975 0.00000 4.7575 28.8 147.9 0.1 242 0.98000 0.00025 4.7587 28.8 147.7 -0.2 243 0.98050 0.00050 4.7611 43.3 146.1 -1.6 244 0.98050 0.00000 4.7611 43.3 149.1 3.0 245 0.98050 0.00000 4.7611 43.3 146.3 -2.8 246 0.98075 0.00025 4.7623 43.3 149.4 3.1 246 Sample 1/6

Comments - Ketones such as acetone react with methanol forming ketals. In this reaction water is released:

R-CO-R + 2 CH3OH → R-C(OCH3)2-R + H2O

Therefore, methanol-free solvent and titrant are used to avoid this side-reaction. For this, special reagents (the so-called K-reagents) are commercially available. - However, side-reactions can not be completely suppressed. The methanol-free reagents also react slowly with acetone. This leads to an increased drift value at the end of each sample titration. Thus, it is necessary to wait for a constant drift value (e.g. approx. 5 minutes) after each sample determination for the side reaction to subside. - The drift value is also increasing with increasing number of titrated samples. Thus, the solvent must be replaced after 2-3 samples, depending on the sample size. This can be done by defining in the setup the solvent exchange after 2-3 titrated samples (Setup > Global > Solvent control > Monitoring no. of samples). - To achieve precise results it is necessary to weigh the sample accurately, and in particular, to first clean the syringe with a few mL of sample which is then discarded. - To ensure a more efficient cleaning, the syringe is gently shaken to allow for absorption of the moisture on the inner wall of the syringe. Subsequently, the syringe is completely filled.

METTLER TOLEDO Page 3 of 4 Titration Application M303-09 Method (V2.0) 001 Title 006 Calculation R1 Type Karl Fischer titration Result tape Predefined Compatible with T70/T90/V20/V30 Result Content ID M303 Result unit % Title Acetone Dry Formula R1=(VEQ*CONC- Author METTLER TOLEDO TIME*DRIFT/1000)*C/m Date/Time 18.08.2009 11:49:10 Constant C= 0.1 Modified at 18.08.2009 11:49:10 Decimal places 3 Modified by METTLER TOLEDO Result limits No Protect No Record statistics Yes SOP None Extra statistical functions No

002 Sample 007 Record Number of IDs 1 Summary Per sample ID 1 -- Results No Entry type Weight Raw results No Lower limit 3.0 g Table of measured values No Upper limit 8.0 g Sample data No Density 1.0 g/mL Resource data No Correction factor 1.0 E-V curve No Weight per piece 1.0 g E-t curve No Temperature 25.0°C V-t curve No

Autostart No H2O-t No Entry After addition Drift-t No

Concentration H2O-t & Drift-t curves No Titrant KF 1-comp 5K V-t & drift-t No Nominal concentration 5 mg/mL Method No Standard Water-Standard 10.0 Series data No Entry type Weight Lower limit 0.6 g 008 End of sample Upper limit 1.8 g Open series Yes Temperature 25.0°C Mix time 10 s Autostart Yes Entry After addition Concentration limits Yes Lower limit 4.5 mg/mL Upper limit 5.8 mg/mL

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Determination Max. start drift 25 µg/min

004 Mix time Duration 5 s

005 Titration (KF Vol) [1] Titrant Titrant KF 1-comp 5K Nominal concentration 5 mg/mL Reagent type 1-comp Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol 10.0 µA Stir Speed 35 % Predispense Mode None Wait time 0 s Control End point 150 mV Control band 300.0 mV Dosing rate(max) 8 mL/min Dosing rate(min) 4 µL/min Start Normal Termination Type Delay time Delay time 10 At VMax 10.0 mL Min. time 0 s Max. time infinity

METTLER TOLEDO Page 4 of 4 Titration Application M303-09 METTLER TOLEDO Application M304-09A Water Determination in Milk Powder (Internal Extraction with Homogenizer)

The water determination of milk powder is performed by volumetric Karl Fischer titration with a homogenizer.

Sample 0.1 – 0.2 g milk powder Preparation and Procedures - Before analysis, the concentration determination is performed by running method M300 (Water Standard 10.0 mg/g). Compound Water, H2O - The titration vessel needs to be filled with more M = 18.01 g/mol solvent i.e. 120 mL. This is due to the fact that the dispersing aggregate of the homogenizer has to be immersed into the solvent for at least 2 cm. 120 mL methanol as a solvent Chemicals - Add the milk powder with a glass weighing boat.

Titrant HYDRANAL® -Composite 5 (5 mg H2O/mL), or CombiTitrant 5 apura®

(5 mg H2O/mL) Standard Water Standard 10.0 mg/g (HYDRANAL® Water Standard 10.0)

Indication DM143-SC Electrode Remarks - The homogenizer time should be 60-90 s. Chemistry CH3OH + SO2 + 3 RN + I2 + H2O → - Use an additional mix time of 15 s after (RNH)•SO4CH3 + 2 (RNH)I homogenizing. - Do not use the parameter “Autostart”! - The concentration is calculated according to the formula: Calculation Water content in %: CONC = CONT*10*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000) VEQ*CONC-(TIME*DRIFT/1000)*C/m CONT = Conc. of liquid water standard (%) VEQ: Titrant consumption to EP CONC(alt) = CONC(alt) refers to the setup value at the time TIME: Total time since sample of calculation. request C 0.1 Reference: For more detailed information on this application and on how to connect the homogenizer, please Organic solvents Waste refer to: disposal METTLER TOLEDO Applications Brochure 27, “Karl Fischer Titration with An Homogenizer: An Introduction with Practical Examples” Market Support Group Anachem, Author, 08/2000, ME-51 725 053. Version Maria José Schmid, August 2009

METTLER TOLEDO Page 1 of 4 Titration Application M304-09A Instruments - KF Compact Volumeter V30 (V2.0), Titration Excellence T70/T90 (V3.1.1) - Adapter Set 12, ME-51 107 534 (For Kinematica Polytron PT1200) Note: The homogenizer must be purchased from the producer/Not a MT product - TBox DR42 Accessories - XP205 Balance - LabX Titration Software - Glass weighing boat ME-23952

Results Sample: Milk powder Titrant: HYDRANAL®-Composite 5

No. Comment / ID Start time Sample size and results 1/6 -- 21.08.2009 10:02:31 0.1114 g R1 = 5.59 % Content 2/6 -- 21.08.2009 10:15:00 0.1002 g R1 = 5.61 % Content 3/6 -- 21.08.2009 10:32:54 0.1146 g R1 = 5.57 % Content 4/6 -- 21.08.2009 11:23:04 0.1078 g R1 = 5.45 % Content 5/6 -- 21.08.2009 11:41:45 0.1036 g R1 = 5.53 % Content 6/6 -- 21.08.2009 11:52:32 0.1053 g R1 = 5.69 % Content

Statistics Rx Name n Mean value Unit s srel[%] R1 Content 6 5.57 % 0.08 1.443

Additional results R1 Content 4 5.40 % 0.12 2.141

Titration curve

Sample 1/6

METTLER TOLEDO Page 2 of 4 Titration Application M304-09A Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s 0.00000 NaN 0.0000 0.0 794.0 NaN 0 0.00000 0.00000 0.0000 0.0 793.8 -0.2 0 0.00000 0.00000 0.0000 0.0 794.0 0.2 1 0.00000 0.00000 0.0000 0.0 794.0 0.0 2 0.00150 0.00150 0.0075 144.5 792.9 -1.1 3 0.01425 0.01275 0.0710 1064.7 788.9 -4.0 4 0.02700 0.01275 0.1345 1646.8 784.0 -4.9 5 0.04350 0.01650 0.2167 2241.5 773.8 -10.2 6 0.05925 0.01575 0.2951 1940.4 777.6 3.8 7 0.05975 0.00050 0.2976 2330.0 788.0 10.4 8 0.07100 0.01125 0.3537 2100.9 788.6 0.6 9 0.08300 0.01200 0.4134 2257.4 787.8 -0.8 10

… … … … … … …

1.26050 0.00025 6.2788 118.4 0.3 0.3 203 1.26050 0.00000 6.2788 118.4 -1.4 -1.4 204 1.26125 0.00075 6.2825 103.6 -0.8 -0.8 205 1.26125 0.00000 6.2825 103.6 -3.1 -3.1 206 1.26125 0.00000 6.2825 59.2 2.0 2.0 207 1.26125 0.00000 6.2825 59.2 -0.5 -0.5 208 1.26125 0.00000 6.2825 59.2 3.7 3.7 209 1.26125 0.00000 6.2825 59.2 -3.1 -3.1 210 1.26125 0.00000 6.2825 59.2 1.3 1.3 211 1.26125 0.00000 6.2825 59.2 0.8 0.8 212 1.26125 0.00000 6.2825 29.6 -0.5 -0.5 213 1.26125 0.00000 6.2825 22.2 -0.4 -0.4 213 Sample 1/6

Comments - The homogenizer is controlled by TBox and is mounted in the titration stand. - An additional power cable for the connection from the TBox DR42 to the power supply unit of the homogenizer is needed. This cable is commercially available in any electronics and electrical components shop. - For this, a special cover is needed for the titration vessel as well as an adapter. These parts are contained in the Adapter Set 12 (ME-51 107 534). This adapter set can be used with the Polytron 1200 Homogenizer and the PT-DA 1212/2EC of Kinematica. The dispersing aggregate has a diameter of 12 mm. - There is also the possibility of using a homogenizer with a larger diameter (18 mm). This homogenizer can be purchased from IKA. For this homogenizer it is necessary to use the Adapter Set 18 (ME-51 107 409). - The use of the homogenizer in combination with a KF Volumetric Instrument is described in detail in Titration Applications Brochure No. 27 (ME-51 725 053) - The homogenizers are not Mettler-Toledo products and therefore they cannot be purchased from Mettler-Toledo. The homogenizer has to be purchased from the producers: 12 mm Homogenizer: Kinematica www.kinematica.ch 18 mm Homogenizer: IKA www.ika.de

METTLER TOLEDO Page 3 of 4 Titration Application M304-09A Method (V2.0) 001 Title Type Karl Fischer titration 007 Calculation R1 Compatible with T70/T90/V30 Result tape Predefined ID M304 Result Content Title Milk powder (homogenizer) Result unit % Author METTLER TOLEDO Formula R1=(VEQ*CONC- Date/Time 02.08.2006 15:00:00 TIME*DRIFT/1000)*C/m Modified at 02.08.2006 15:00:00 Constant C= 0.1 Modified by METTLER TOLEDO Decimal places 2 Protect No Result limits No SOP None Record statistics Yes Extra statistical functions No 002 Sample Number of IDs 1 007 Record ID 1 -- Summary Per sample Entry type Weight Results No Lower limit 0.1 g Raw results No Upper limit 1.0 g Table of measured values No Density 1.0 g/mL Sample data No Correction factor 1.0 Resource data No Weight per piece 1.0 g E-V curve No Temperature 25.0°C E-t curve No Autostart No V-t curve No

Entry After addition H2O-t No Concentration Drift-t No

Titrant KF 1-comp 5 H2O-t & Drift-t curves No Nominal concentration 5 mg/mL V-t & drift-t No Standard Water-Standard 10.0 Method No Entry type Weight Series data No Lower limit 0.0 g Upper limit 2.0 g 008 End of sample Temperature 25.0°C Open series Yes Mix time 10 s Autostart Yes Entry After addition Concentration limits Yes Lower limit 4.5 mg/mL Upper limit 5.6 mg/mL

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 25 µg/min

004 Homogenizer Name Homogenizer TTL Duration 60 s

005 Mix time Duration 15 s

006 Titration (KF Vol) [1] Titrant Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Reagent type 1-comp Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol 24.0 µA Stir Speed 35 % Predispense Mode None Wait time 0 s Control End point 100.0 mV Control band 400.0 mV Dosing rate(max) 5 mL/min Dosing rate(min) 80 µL/min Start Normal Termination Type Drift stop relative Drift 15.0 µg/min At VMax 10.0 mL Min. time 0 s Max. time infinity

METTLER TOLEDO Page 4 of 4 Titration Application M304-09A METTLER TOLEDO Application M305-09 Determination of the water content in tobacco by external extraction

Determination of the water content of tobacco with external extraction by volumetric KF titration.

Sample Tobacco Preparation and Procedures 1) The water content of the solvent (methanol) is determined first and is automatically stored in the blank value B[Blank KF]. Compound Water, H2O 2) Afterwards, approximately 2-5 g tobacco is M = 18.01 g/mol added to the solvent (approx. 60g), put for 15 min in an ultrasonic bath and left stand for another 15 min in order to allow for water Chemicals 60 mL apura® CombiMethanol extraction. Total extraction time is 30 min. or 3) The clear solution (0.5-1g) is then injected into HYDRANAL® Methanol dry the KF titration vessel using a syringe with a as a solvent for the titration cell needle and the weight determined by back weighing. Titrant apura® CombiTitrant 5 (5 mg H2O/mL) or HYDRANAL® Composite 5 (5 mg H2O/mL)

Standard Water Standard 10.0 mg/g (HYDRANAL® Water Standard 10.0)

Indication DM143-SC Remarks - Before analysis, the concentration determination CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 is performed by pressing “Start conc.” (Water → (RNH)•SO CH + 2 (RNH)I 4 3 Standard 10.0 mg/g). - The concentration determination is calculated according to the formula:

CONC = CONT*10*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000)

Calculation External extraction (B in %) %: CONT = Conc. of liquid water standard (%) R1=(100/(100-C)*(C*msol/mext-B*msol/mext) CONC(alt) = CONC(alt) refers to the setup value at the time C=(VEQ*CONC-TIME*DRIFT/1000)*0.1/m of calculation.

VEQ: Titrant consumption to EP - Subsequently, the sample analysis is started by TIME: Total time since sample request CONC: Concentration of titrant pressing “Start sample” on the display. msol: Weight of solvent mext: Weight of extracted sample m: Sample weight Literature: • ISO 6488-1; 2004: Tobacco and tobacco Waste Karl Fischer waste should be products -- Determination of water content -- disposal disposed of as halogenated Karl Fischer method organic solvents. • Schweizerisches Lebensmittelbuch 2005: Tabak und Tabakerzeugnisse, 1.3.3. Author, Maria-José Schmid „Wasserbestimmung nach Karl Fischer“ www.slmb.bag.admin.ch/slmb/index.html Version Market Support Group, July 09

METTLER TOLEDO Page 1 of 4 Titration Application M305-09 Instruments - KF Compact Line Volumeters V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1) - XP205 Balance

Accessories - LabX titration pro software - 10 mL syringe with needle

Results No. Comment / ID Start time Sample size and results 1/6 -- 17.08.2009 14:34:44 1.0188 g R1 = 8.27 % External extraction (B in %) 2/6 -- 17.08.2009 14:43:15 1.0318 g R1 = 8.44 % External extraction (B in %) 3/6 -- 17.08.2009 14:49:49 1.1922 g R1 = 8.34 % External extraction (B in %) 4/6 -- 17.08.2009 15:01:25 0.7626 g R1 = 8.31 % External extraction (B in %) 5/6 -- 17.08.2009 15:04:49 0.9726 g R1 = 8.25 % External extraction (B in %) 6/6 -- 17.08.2009 15:08:24 1.0212 g R1 = 8.31 % External extraction (B in %) Statistics Rx Name n Mean value Unit s srel[%] R1 External extraction 6 8.32 % 0.07 0.804 (B in %)

Blank: R1 Blank 3 0.014374 % 0.000228 1.588

Additional results: R1 Ext. extraction 6 8.25 % 0.08 0.985 R1 Ext. extraction 5 9.06 % 0.04 0.504 R1 Ext. extraction 4 9.07 % 0.11 1.260

Titration curve

Sample 1/6

METTLER TOLEDO Page 2 of 4 Titration Application M305-09 Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s 0.00000 NaN 0.0000 9.6 680.8 NaN 0 0.00000 0.00000 0.0000 9.6 680.8 0.0 0 0.00000 0.00000 0.0000 0.0 680.9 0.1 1 0.00125 0.00125 0.0062 186.6 680.9 0.0 2 0.00625 0.00500 0.0311 622.6 676.3 -4.6 3 0.01525 0.00900 0.0760 1139.4 667.8 -8.5 4 0.02825 0.01300 0.1407 1688.6 642.2 -25.6 5 0.04025 0.01200 0.2005 2004.9 652.0 9.8 6 0.04125 0.00100 0.2055 1761.2 672.0 20.0 7 0.04650 0.00525 0.2316 1737.2 671.2 -0.8 8 0.05525 0.00875 0.2752 1834.7 647.8 -23.4 9 0.06850 0.01325 0.3412 2047.3 649.6 1.8 10 0.07025 0.00175 0.3499 2078.8 655.3 5.7 11

0.60550 0.00000 3.0161 81.3 82.7 3.1 117 0.60550 0.00000 3.0161 66.6 83.1 0.4 118 0.60550 0.00000 3.0161 66.6 89.2 6.1 119 0.60550 0.00000 3.0161 66.6 86.8 -2.4 120 0.60550 0.00000 3.0161 66.6 82.8 -4.0 121 0.60550 0.00000 3.0161 66.6 87.7 4.9 122 0.60550 0.00000 3.0161 66.6 94.0 6.3 123 0.60550 0.00000 3.0161 59.2 92.4 -1.6 124 0.60550 0.00000 3.0161 22.2 92.3 -0.1 125 0.60550 0.00000 3.0161 0.0 87.1 -5.2 125

Sample 1/6

Comments - To achieve precise results it is necessary to weigh the sample accurately, and in particular, to first clean the syringe with a few mL of sample which are then discarded. - To ensure a more efficient cleaning, the syringe is gently shaken to allow for absorption of the moisture on the inner wall of the syringe. Subsequently, the syringe is completely filled.

METTLER TOLEDO Page 3 of 4 Titration Application M305-09 Method (V2.0)

001 Title Type Volumetric external Termination extraction Type Drift stop relative Compatible with T70 / T90 / V30 Drift 15.0 µg/min ID M305 At VMax 10.0 mL Title Tobacco (ext. extraction) Min. time 0 s Author METTLER TOLEDO Max. time infinity Date/Time 02.08.2009 15:00:00 Modified at 02.08.2009 15:00:00 006 Calculation R1 Modified by METTLER TOLEDO Result tape Predefined Protect No Result External extraction SOP None (B in %) Result unit % 002 Sample Formula R= 100/(100-C)*(C*msol/mext-B*msol/mext) Number of IDs 1 ID 1 -- Constant C= (VEQ*CONC-Time*Drift/1000)*0.1/m Entry type Weight Decimal places 2 Lower limit 0.0 g Result limits No Upper limit 5.0 g Record statistics Yes Density 1.0 g/mL Extra statistical functions No Solvent weight [g] 0.0 Extraction weight [g] 0.0 007 Record Correction factor 1.0 Summary Per sample Temperature [°C] 25.0 Results No Autostart Yes Raw results No Entry After addition Table of measured values No Concentration Sample data No Titrant KF 1-comp 5 Resource data No Nominal concentration 5 mg/mL E-V curve No Standard Water-Standard 10.0 E-t curve No Entry type Weight V-t curve No

Lower limit 0.0 g H2O-t No Upper limit 2.0 g Drift-t No Temperature 25.0°C H2O-t & Drift-t curves No Mix time 10 s V-t & drift-t No Autostart Yes Method No Entry After addition Series data No Concentration limits Yes Lower limit 4.5 mg/mL 008 End of sample Upper limit 5.6 mg/mL Open series Yes Blank Source Setup Blank Blank KF Unit % Entry type Weight Lower limit 0.0 g Upper limit 5.0 g Temperature [°C] 25.0 Mix time 15 s Autostart No Entry After addition Limits No

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 25 µg/min

004 Mix time Duration 15 s

005 Titration (KF Vol) [1] Titrant Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Reagent type 1-comp Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol 15.0 µA Stir Speed 45 % Predispense Mode None Wait time 0 s Control End point 100.0 mV Control band 400.0 mV Dosing rate(max) 5 mL/min Dosing rate(min) 80 µL/min Start Normal

METTLER TOLEDO Page 4 of 4 Titration Application M305-09 METTLER TOLEDO Application M306-09A Water Determination in Corn Starch with Drying Oven DO308

The water content in corn starch is determined by volumetric Karl Fischer titration using a drying oven.

Sample Corn starch, 0.1 g Preparation and Procedures Preparation DO308/air pump: - The oven must first be heated out (incl. aluminum boat) at 300°C for approx. 20 min until the drift is less

Compound Water, H2O than 10 µg/min before starting measurement (see M = 18.01 g/mol operating instruction for DO308), the air gas flow rate is adjusted to 200 mL/min. - Set the temperature for the sample measurement to 180°C. ® Chemicals 60 mL HYDRANAL -Methanol Dry or Sample measurement: 60 mL CombiMethanol apura® - The titrator is in Stand-by.

- Weigh the sample in a glass weighing boat and tare it. - Open the valve of the DO308 oven in order to stop the ® Titrant HYDRANAL -Composite 5 air flowing into the titration cell. (5 mg H2O/mL), - The sample is transferred into the aluminum boat of or the DO308 through the sample opening on the oven ® CombiTitrant 5 apura tube. (5 mg H O/mL) 2 - Press start sample and close the valve to allow for air flowing into the titration cell. Standard Water Standard 10.0 mg/g (Merck® Water Standard 1.0%) - Push the oven boat into the heating zone of the oven. - Back weighing the empty glass boat and enter the sample size. Indication DM143-SC Electrode Remarks - The concentration of the titrant should be CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 determined first with method M300 and approx. 1 → (RNH)SO4CH3 + 2 (RNH)I g of 10.o mg H2O /g water standard is used: Titer (n=3): (4.901135 ± 0.011893) mg/mL (srel: 0.243%) - The transfer of the sample into the DO308 has to Calculation Water content in ppm: be done very carefully to reduce the error of the sample size. An alternative sample transfer is (VEQ*CONC-(TIME*DRIFT/1000)) * C ------also show (see “Comments”). m VEQ: Titrant consumption to EP TIME: Duration of a sample analysis from the end of Standby until the end of the method function Titration (KF Vol) C: 1000 (constant for ppm calc.) Waste Organic solvents disposal

Author, Cosimo De Caro, MSG Version July 2009

METTLER TOLEDO Page 1 of 5 Titration Application M306-09A Instruments - KF Compact Line Volumeter V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1) - KF drying Oven DO308 with drying unit (silica gel, molecular sieve) and air pump - XP205 Balance Accessories - LabX Titration Software - Glass weighing boat ME-23952 - 10 mL syringe with needle

Results

No. Comment / ID Start time Sample size and results 1/6 Corn starch 28.07.2009 09:50:36 0.0969 g R1 = 13.29 % Content 2/6 Corn starch 28.07.2009 10:15:08 0.1308 g R1 = 13.21 % Content 3/6 Corn starch 28.07.2009 10:32:54 0.1327 g R1 = 13.30 % Content 4/6 Corn starch 28.07.2009 10:54:45 0.1075 g R1 = 13.25 % Content 5/6 Corn starch 28.07.2009 11:12:33 0.0997 g R1 = 13.27 % Content 6/6 Corn starch 28.07.2009 11:30:22 0.1342 g R1 = 13.39 % Content

Statistics Rx Name n Mean value Unit s srel[%] R1 Content 6 13.28 % 0.06 0.456

Titration curve

sample 1/6

METTLER TOLEDO Page 2 of 5 Titration Application M306-09A Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s 0.00000 NaN 0.0000 0.0 594.3 NaN 0 0.00000 0.00000 0.0000 0.0 594.3 0.0 0 0.00000 0.00000 0.0000 0.0 594.4 0.1 1 0.00100 0.00100 0.0049 146.9 594.4 0.0 2 0.00575 0.00475 0.0282 563.6 593.6 -0.8 3 0.01425 0.00850 0.0698 1047.6 589.0 -4.6 4 0.02700 0.01275 0.1323 1587.9 587.5 -1.5 5 0.04375 0.01675 0.2144 2144.2 579.7 -7.8 6 0.06450 0.02075 0.3161 2709.6 570.8 -8.9 7 0.08450 0.02000 0.4141 3106.1 562.6 -8.2 8 0.10125 0.01675 0.4962 3308.3 567.0 4.4 9 0.12225 0.02100 0.5992 3595.0 567.3 0.3 10 0.14150 0.01925 0.6935 4119.9 570.0 2.7 11 0.17350 0.03200 0.8503 5022.5 494.1 -75.9 12 0.25125 0.07775 1.2314 7147.9 63.5 -430.6 13 0.25125 0.00000 1.2314 6900.4 485.8 422.3 14 0.25425 0.00300 1.2461 6616.6 569.7 83.9 15 0.28150 0.02725 1.3797 6922.3 509.3 -60.4 16

2.62700 0.00000 12.8753 21.8 88.1 9.3 895 2.62700 0.00000 12.8753 14.6 84.1 -4.0 896 2.62700 0.00000 12.8753 14.6 86.3 2.2 897 2.62700 0.00000 12.8753 14.6 83.4 -2.9 898 2.62700 0.00000 12.8753 7.3 87.5 4.1 899 2.62700 0.00000 12.8753 0.0 96.1 8.6 900 Sample 1/6

Principle DO308 KF Oven: Mechanism of the sample transport with magnet

Guide rod with magnet

Slider Glass sample boat

METTLER TOLEDO Page 3 of 5 Titration Application M306-09A

Comments - The drying oven can be used for different samples; it is especially used for samples which would give side reactions with the Karl Fischer reagent by direct measurement. - It is important to know about the thermal decomposition of the measured sample at a specific

temperature. In fact, decomposition product such as for instance CO2, reacts with iodine leading to a higher water content. In such case, the external extraction with e.g. chloroform would be the most suitable method for sample prearation. Example:

Sample Method H2O content (%)

Washing powder DO308, 150°C, 900s, 200 mL/min N2 10.6 ± 0.4 (Migros) External extraction with CHCl3 for 2h 1.0 ± 0.03

Alternative solution for sample transfer and weighing: - Weigh a small funnel with a 100 mL glass beaker (W1). - Weigh the sample in a metal weighing boat (ME-4507) and tare it (see photograph below). - Press start sample, transfer it with the funnel in the aluminum boat of the DO308 and push it into the heating zone of the oven. - Back weighing the empty metal boat (S1). - Weigh the funnel with the glass beaker (W2).

Calculation of the sample weight: Sample weight = [S1] - ([W2] – [W1])

Sample preparation: Sample in metal weighing boat

Transfer of the sample into drying oven

METTLER TOLEDO Page 4 of 5 Titration Application M306-09A Method (V2.0)

001 Title Type Karl Fischer titration 007 Record Compatible with T70/T90/V20/V30 Summary Per sample ID M306 Results No Title Corn starch (manual oven) Raw results No Author METTLER TOLEDO Table of measured values No Date/Time 27.07.2009 14:59:07 Sample data No Modified at 28.07.2009 14:05:33 Resource data No Modified by METTLER TOLEDO E-V curve No Protect No E-t curve No SOP None V-t curve No

H2O-t No 002 Sample Drift-t No Number of IDs 1 H2O-t & Drift-t curves No ID 1 Corn starch V-t & Drift-t No Entry type Weight Method No Lower limit 0.0 g Series data No Upper limit 0.15 g Density 1.0 g/mL 008 End of sample Correction factor 1.0 Open series Yes Weight per piece 1.0 g Temperature 25.0°C Autostart No Entry Arbitrary Concentration Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Standard Water-Standard 10.0 Entry type Weight Lower limit 0.0 g Upper limit 2.0 g Temperature 25.0°C Mix time 10 s Autostart Yes Entry After addition Concentration limits Yes Lower limit 4.5 mg/mL Upper limit 5.6 mg/mL

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 25 µg/min

004 Mix time Duration 60 s

005 Titration (KF Vol) [1] Titrant Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Reagent type 1-comp Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol 24.0 µA Stir Speed 35 % Predispense Mode None Wait time 0 s Control End point 100.0 mV Control band 350.0 mV Dosing rate(max) 5 mL/min Dosing rate(min) 80 µL Start Normal Termination Type Delay time Drift 10 s At VMax 10.0 mL Min. time 900 s Max. time 900 s

METTLER TOLEDO Page 5 of 5 Titration Application M306-09A METTLER TOLEDO Application M307-09 Water Determination in Gases by Volumetric Karl Fischer Titration

The moisture content in air is determined by volumetric Karl Fischer titration.

Sample Air Preparation and Procedures - To determine the water content of gas (air), the air flow must be directed into the titration vessel during a specific period of time. Compound Water, H2O M = 18.01 g/mol Three-way valve Flow meter

50 mL methanol Air (gas) Chemicals sample

- Adjust the air stream to a constant flow rate. In this application, the rate was set to 200 mL/min. ® Titrant HYDRANAL -Composite 2 - Purge the system prior the determination, and (2 mg H2O/mL), close the three-way valve. or CombiTitrant 2 apura® - Start the pre-titration, and wait in standby until

(2 mg H2O/mL) the drift has decreased to a constant value. Standard Water Standard 10.0 mg/g - Open the three-way valve, and let the air flow (HYDRANAL® Water Standard 10.0) and stop it after 5 min. - Calculate the gas volume from the time and flow Indication DM143-SC Electrode rate. In this case, 200 * 5 = 1000 mL = 1 L. Remarks - The air (gas) is introduced into the titration vessel CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 via an inlet tube. → (RNH)•SO4CH3 + 2 (RNH)I - The density of the air under test conditions is entered to allow for calculation of the appm content: Density = 1.141 g/L at 20°C and 95.7 kPa

Calculation Water content in %: - Since the density is given in g/L, then also the air volume must be entered in g/L. ( VEQ*CONC-(TIME*DRIFT/1000) ) * C ------m

VEQ: Titrant consumption to EP TIME: Total time since sample request C: 0.1 (constant for % calc.)

Waste Organic solvents disposal

Author, Melanie Nijman, MSG Version July 2009

METTLER TOLEDO Page 1 of 4 Titration Application M307-09 Instruments - KF Compact Volumeter V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1) - Gas inlet tube ME-105091

Accessories - LabX Titration Software - XP205 Balance - 10 mL syringe

Results HYDRANAL® Composite 5

No. Comment / ID Start time Sample size and results ______1/6 -- 12.08.2009 13:34:35 400 mL R1 = 9.905 µg/mL Content R2 = 8680.608 ppm Content 2/6 -- 12.08.2009 13:38:56 400 mL R1 = 9.998 µg/mL Content R2 = 8762.593 ppm Content 3/6 -- 12.08.2009 13:43:03 400 mL R1 = 9.974 µg/mL Content R2 = 8741.834 ppm Content 4/6 -- 12.08.2009 13:47:06 400 mL R1 = 10.028 µg/mL Content R2 = 8788.532 ppm Content 5/6 -- 12.08.2009 13:57:18 400 mL R1 = 9.807 µg/mL Content R2 = 8595.518 ppm Content 6/6 -- 12.08.2009 14:01:25 400 mL R1 = 9.783 µg/mL Content R2 = 8573.726 ppm Content Statistics Rx Name n Mean value Unit s srel[%] ______R1 Content 6 9.916 µg/mL 0.102 1.032 R2 Content 6 8690.468 ppm 89.664 1.032

apura® CombiTitrant 2 ______R1 Content 6 9.505 µg/mL 0.045 0.469 R2 Content 6 8330.515 ppm 89.664 0.469

Titration curve

Sample 1/6

METTLER TOLEDO Page 2 of 4 Titration Application M307-09 Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s 0.00000 NaN 0.0000 0.0 718.3 NaN 0 0.00000 0.00000 0.0000 0.0 718.2 -0.1 0 0.00000 0.00000 0.0000 0.0 718.2 0.0 1 0.00175 0.00175 0.0033 99.5 718.3 0.1 2 0.01225 0.01050 0.0232 480.2 717.7 -0.6 3 0.03275 0.02050 0.0620 979.6 716.7 -1.0 4 0.06325 0.03050 0.1198 1529.7 715.8 -0.9 5 0.10400 0.04075 0.1970 2111.0 714.2 -1.6 6 0.15475 0.05075 0.2932 2706.2 713.3 -0.9 7 0.21550 0.06075 0.4083 3266.1 711.1 -2.2 8 0.27900 0.06350 0.5286 3775.4 708.0 -3.1 9 0.35650 0.07750 0.6754 4357.3 707.0 -1.0 10

….. ….. …… ….. ….. …… ……

2.09125 0.00100 3.9618 126.6 94.5 -0.6 87 2.09125 0.00000 3.9618 106.9 90.8 -3.7 88 2.09125 0.00000 3.9618 98.5 90.2 -0.6 89 2.09125 0.00000 3.9618 81.6 91.0 0.8 90 2.09125 0.00000 3.9618 70.3 94.2 3.2 91 2.09125 0.00000 3.9618 56.3 93.1 -1.1 92 2.09125 0.00000 3.9618 45.0 91.0 -2.1 93 2.09125 0.00000 3.9618 45.0 90.7 -0.3 94 2.09125 0.00000 3.9618 22.5 92.8 2.1 95 2.09125 0.00000 3.9618 11.2 91.8 -1.0 96 Sample 1/6

Comments - Sampling/sample addition for gases: If possible, you should titrate the gas sample directly from the source. If not, you must fill the gas either into special gas sample tubes or into small steel cylinders. 1. You must purge the sample vessels and the tubing thoroughly beforehand with the gas. 2. If you use sample vessels, you can determine the amount of dispensed gas by differential weighing. 3. If you are able to liquefy the gas, you should also inject it into the titration vessel in liquid form.

Note: 1. Pre-dispense a sufficient amount of buffer solution into the titration vessel to determine the water content of acid gases, e.g. hydrochloric acid. 2. If you titrate large quantities of gas in the same solvent, you should replace the evaporated methanol that has been lost in the gas stream depending on titration time and number of determinations.

3. CO2 gas can not be titrated into the titration vessel directly, because iodine reacts with it. CO2 gas must be directed through a water-dissolving, water-free absorption liquid in which CO2 itself does not dissolve. The water contained in the gas is then absorbed by the liquid and can be determined by means of Karl Fischer titration in a process similar to external extraction.

METTLER TOLEDO Page 3 of 4 Titration Application M307-09 Method (V2.0) 001 Title Type Karl Fischer titration 007 Calculation R2 Compatible with T70 / T90 / V30 Result tape User defined ID M307 Result Content Title Air Result unit ppm Author METTLER TOLEDO Formula R2= Date/Time 12.08.2009 13:18:00 (VEQ*CONC- Modified at 12.08.2009 13:32:30 TIME*DRIFT/1000)* Modified by METTLER TOLEDO *C/(m*d*f) Protect No Constant C= 1000 SOP None Decimal places 3 Result limits No 002 Sample Record statistics Yes Number of IDs 1 Extra statistical func. No ID 1 -- Entry type Fixed volume 008 Record Volume 400 mL Summary Per sample Density 0.001 g/mL Results No Correction factor 1.141 Raw results No Weight per piece 1.0 g Table of measured values No Temperature 25.0°C Sample data No Autostart No Resource data No Entry Before E-V curve No Concentration E-t curve No Titrant KF 1-comp 2 V-t curve No

Nominal concentration 2 mg/mL H2O-t No Standard Water-Standard 10.0 Drift-t No Entry type Weight H2O-t & Drift-t curves No Lower limit 0.0 g V-t & drift-t No Upper limit 2.0 g Method No Temperature 25.0°C Series data No Mix time 10 s Condition No Autostart Yes Entry After addition 009 End of sample Concentration limits Yes Open series Yes Lower limit 1.7 mg/mL Upper limit 2.3 mg/mL

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Note: Max. start drift 25 µg/min A) The density of air 1.141 g/L has to be entered using the factor and the density parameters. 004 Mix time Duration 120 s B) The measuring unit for the density in the titration method is given in g/mL. Therefore, a factor of 1000 must be taken 005 Titration (KF Vol) [1] into account when entering the density value. Titrant Titrant KF 1-comp 2 In our case: Nominal concentration 2 mg/mL d = 0.001 Reagnt type 1-comp f = 1.141 Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol 24.0 µA Stir Speed 35 % Predispense Mode None Wait time 0 s Control End point 100.0 mV Control band 400.0 mV Dosing rate(max) 5 µL/mL Dosing rate(min) 80.0 µL Start Normal Termination Type Drift stop relative Drift 15.0 µg/min At VMax 10.0 mL Min. time 0 s Max. time 180

006 Calculation R1 Result tape User defined Result Content Result unit µg/mL Formula R1=(VEQ*CONC- TIME*DRIFT/1000)*C/m Constant C= 1000 Decimal places 3 Result limits No Record statistics Yes Extra statistical func. No

METTLER TOLEDO Page 4 of 4 Titration Application M307-09 METTLER TOLEDO Application M311-09A Water Content Determination in Water Standard Tablet 10 mg

The concentration and content determination with water standard tablet (10 mg H2O) are performed by volumetric Karl Fischer titration.

Sample Water standard tablet 10 mg Preparation and Procedures (HYDRANAL® Fastrate™) - The suction tube of the Solvent Manager which is mounted into the titration cell has to be avoided due to the following reasons: Compound Water, H2O M = 18.01 g/mol 1. the tablet can be clogged between tube and glass wall. 2. the dissolved powder of the tablet can not be efficiently Chemicals 80 mL methanol as a solvent - A mixing time of at least 60 s is required in order to completely dissolve the standard tablet and the water is released. - The solvent should be replaced after 6 tablets. Titrant HYDRANAL® -Composite 5 - When adding you should not touch the tablets (5 mg H2O/mL), with the fingers. or CombiTitrant 5 apura®

(5 mg H2O/mL) Standard Water Standard 10.0 mg/g (HYDRANAL® Water Standard 10.0)

Indication DM143-SC Electrode Remarks - First, the concentration determination is CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 performed by pressing “Start conc.” (Water → Standard 10.0 mg/g, approx. 1 g). (RNH)•SO4CH3 + 2 (RNH)I - The concentration determination is calculated according to the formula:

CONC = CONT*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000) Calculation Water content in mg: CONT = Conc. of liquid water standard (mg/g) CONC(alt) = CONC(alt) refers to the Setup value VEQ*CONC-(TIME*DRIFT/1000) current at the time of calculation. - Subsequently, the sample analysis is started by VEQ: Titrant consumption to EP TIME: Total time since sample pressing “Start sample” on the display. request - After two samples, the titration solvent shows a relatively high drift. Thus, it may take longer to achieve the starting drift. If this takes too long, Waste Organic solvents the starting drift parameter in the method should disposal be increased.

Author, Market Support Anachem Version Melanie Nijman, July 2009

METTLER TOLEDO Page 1 of 4 Titration Application M311-09A Instruments - KF Compact Volumeter V20/V30 (V2.0) - Titration Excellence T70/T90 (V3.1.1)

Accessories - LabX Titration Software - XP205 Balance - 10 mL syringe

Results Water Standard tablet: 10.0 ± 0.3 mg Titrant: HYDRANAL®-Composite 5

No. Comment / ID Start time Sample size and results 1/6 -- 14.08.2009 11:26:56 1 pcs R1 = 10.298 mg Content 2/6 -- 14.08.2009 11:32:52 1 pcs R1 = 10.283 mg Content 3/6 -- 14.08.2009 11:39:51 1 pcs R1 = 10.185 mg Content 4/6 -- 14.08.2009 11:47:55 1 pcs R1 = 10.275 mg Content 5/6 -- 14.08.2009 11:54:12 1 pcs R1 = 10.411 mg Content 6/6 -- 14.08.2009 12:03:07 1 pcs R1 = 10.232 mg Content

Statistics Rx Name n Mean value Unit s srel[%]

R1 Content 6 10.281 mg 0.076 0.739

Titrant: apura®-CombiTitrant 5 Statistics Rx Name n Mean value Unit s srel[%]

R1 Content 6 10.192 mg 0.085 0.832

Titration curve

Sample 1/6

METTLER TOLEDO Page 2 of 4 Titration Application M311-09A Table of measured values

Volume Meas. value H2O Drift Time mL mV mg µg/min s 0.00000 718.3 0.0000 0.0 0 0.00000 718.2 0.0000 0.0 0 0.00000 718.2 0.0000 0.0 1 0.00175 718.3 0.0033 99.5 2 0.01225 717.7 0.0232 480.2 3 0.03275 716.7 0.0620 979.6 4 0.06325 715.8 0.1198 1529.7 5 0.10400 714.2 0.1970 2111.0 6 0.15475 713.3 0.2932 2706.2 7 0.21550 711.1 0.4083 3266.1 8 0.27900 708.0 0.5286 3775.4 9 0.35650 707.0 0.6754 4357.3 10

…. …. …. …. ….

2.09125 94.5 3.9618 126.6 87 2.09125 90.8 3.9618 106.9 88 2.09125 90.2 3.9618 98.5 89 2.09125 91.0 3.9618 81.6 90 2.09125 94.2 3.9618 70.3 91 2.09125 93.1 3.9618 56.3 92 2.09125 91.0 3.9618 45.0 93 2.09125 90.7 3.9618 45.0 94 2.09125 92.8 3.9618 22.5 95

2.09125 91.8 3.9618 11.2 96 Sample 1/6

Comments --

METTLER TOLEDO Page 3 of 4 Titration Application M311-09A Method 001 Title Type Karl Fischer titration 007 Record Compatible with T70 / T90 / V20 / V30 Summary Per sample ID M311 Results No Title Water standard tablet 10 Raw results No mg Table of measured values No Author METTLER TOLEDO Sample data No Date/Time 12.08.2009 16:35:32 Resource data No Modified at 12.08.2009 16:39:15 E-V curve No Modified by METTLER TOLEDO E-t curve No Protect No V-t curve No

SOP None H2O-t No Drift-t No

002 Sample H2O-t & Drift-t curves No Number of IDs 1 V-t & drift-t No ID 1 -- Method No Entry type Fixed pieces Series data No Pieces 1 pcs Weight per piece 1.0 g 008 End of sample Correction factor 1.0 Open series Yes Weight per piece 1.0 g Temperature 25.0°C Autostart No Entry After addition Concentration Titrant KF 1-comp 5 Nominal concentration 5 mg/mL Standard Water-Standard 10.0 Entry type Weight Lower limit 0.0 g Upper limit 2.0 g Temperature 25.0°C Mix time 10 s Autostart Yes Entry After addition Concentration limits Yes Lower limit 4.5 mg/mL Upper limit 5.6 mg/mL

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 25 µg/min

004 Mix time Duration 60 s

006 Calculation R1 Result tape Predefined Result Content Result unit mg Formula R1=(VEQ*CONC- TIME*DRIFT/1000)*C Constant C= 1 Decimal places 3 Result limits No Record statistics Yes Extra statistical functions No

METTLER TOLEDO Page 4 of 4 Titration Application M311-09A METTLER TOLEDO Application M312-09 Automated Volumetric KF Analysis: Check with 5.55% Oven Water Standard

The system Volumeter - KF Oven sample changer is checked by using a solid oven water standard. The water content of the standard has a certified value of 5.55 ± 0.05 %. Water is evaporated in the KF oven sample changer and transferred into the titration cell where it is determined by volumetric KF analysis.

Sample HYDRANAL® water standard Preparation and Procedures KF-oven 5.55 ± 0.05 % Sample preparation: approx. 0.2 g - Place a STROMBOLI glass vial onto the balance and tare the balance. Weigh approx. 0.1 g of the Compound Water, H2O sample into the vial. M = 18.01 g/mol - Close the vial with the aluminum foil, fix the blue rubber and place it on the sample rack.

Filling the STROMBOLI sample rack: 50 mL dry methanol Chemicals - Place 4 empty vials for the blank determination at the positions 1–4 of the sample rack. - Place 8 sample vials for the sample determination at the positions 5–12 of the sample rack. ® Titrant HYDRANAL Composite 5 Remark: A maximum total number of 14 blank and sample vials can be measured.

Performing the analyses: - Create a two-loop sample series with 4 blanks and 8 samples. ® Standard HYDRANAL Water Standard - Start the pre-titration, and wait in standby until 10.0 the drift has decreased to a constant value. - When the online drift is e.g. in the order of 10 Indication DM143-SC Electrode µg/min, start the determinations. Remarks - First, the concentration determination is CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 performed by pressing “Start conc.” (Water → Standard 10.0 mg/g, approx. 0.8 g). (RNH)•SO4 CH3 + 2 (RNH)I - The concentration determination is calculated according to the formula:

CONC = CONT*m/(VEQ-(DRIFT/CONC(alt))*TIME/1000) Calculation Water content in %: CONT = Conc. of liquid water standard (mg/g) R1 = (VEQ*CONC – B[Blank]/1000 – CONC(alt) = CONC(alt) refers to the Setup value TIME*DRIFT/1000) * C / m current at the time of calculation.

VEQ: Consumption at EP [mL] - Subsequently, the sample analysis is started by CONC: Titrant conc [mg/mL] pressing “Start sample” on the display Blank: Blank value [μg] TIME: Total time since sample request [min] - Two sample loops are used in this method: the DRIFT: Drift [μg/min] first loop allows for the blank value determination, C: 0.1 (constant for % calc.) whereas the second one allows for the Waste Organic solvents determination of the water content in the disposal standard. - All sample loops are terminated at fixed time, e.g. t(min) = t(max). Optimum results are obtained with short mix time (e.g. 60 s), and long titration Author, Martin Lanz, R&D time (termination at fixed time, e.g. 900 s). Version July 2009

METTLER TOLEDO Page 1 of 4 Titration Application M312-09 Instruments - Compact Volumeter V30 (V2.0), T70/T90 Titration Excellence (V3.1.1) - STROMBOLI KF Oven Sample Changer - Drying unit with air pump (molecular sieve, silica gel).

Accessories - XS205 Balance - LabX Titration Software

Results All results

1/4 (1/2) 14.07.2009 13:37:02 R1 = 599.2 µg Stromboli blank value 2/4 (1/2) 14.07.2009 13:53:47 R1 = 680.4 µg Stromboli blank value 3/4 (1/2) 14.07.2009 14:10:28 R1 = 587.2 µg Stromboli blank value 4/4 (1/2) 14.07.2009 14:27:09 R1 = 573.1 µg Stromboli blank value

1/8 (2/2) 14.07.2009 14:43:53 R2 = 5.524 % Content blank comp. (B in µg) 2/8 (2/2) 14.07.2009 15:00:53 R2 = 5.505 % Content blank comp. (B in µg) 3/8 (2/2) 14.07.2009 15:17:51 R2 = 5.502 % Content blank comp. (B in µg) 4/8 (2/2) 14.07.2009 15:34:51 R2 = 5.517 % Content blank comp. (B in µg) 5/8 (2/2) 14.07.2009 15:51:48 R2 = 5.512 % Content blank comp. (B in µg) 6/8 (2/2) 14.07.2009 16:08:50 R2 = 5.501 % Content blank comp. (B in µg) 7/8 (2/2) 14.07.2009 16:25:46 R2 = 5.505 % Content blank comp. (B in µg) 8/8 (2/2) 14.07.2009 16:42:46 R2 = 5.489 % Content blank comp. (B in µg)

Statistics

Rx Name n Mean Unit s srel [%] R1 Stromboli blank value 4 610.1 µg 48.1 7.9 R2 Content blank comp. (B in µg) 8 5.507 % 0.011 0.2

Titration curve

METTLER TOLEDO Page 2 of 4 Titration Application M312-09 Table of measured values

Volume Increment H2O Online-Drift Signal Change Time mL mL mg µg/min mV mV s

0.00000 NaN 0.0000 0.0 591.9 NaN 0 0.00000 0.00000 0.0000 0.0 591.9 0.0 0 0.00000 0.00000 0.0000 0.0 592.3 0.4 1 0.00125 0.00125 0.0063 189.3 591.6 -0.7 2 0.00625 0.00500 0.0316 631.4 575.0 -16.6 3 0.01050 0.00425 0.0530 795.7 581.5 6.5 4 0.01400 0.00350 0.0707 848.7 549.7 -31.8 5 0.01975 0.00575 0.0998 997.8 544.0 -5.7 6 0.02200 0.00225 0.1111 952.6 558.0 14.0 7 0.02375 0.00175 0.1200 899.8 554.0 -4.0 8 0.02725 0.00350 0.1377 917.7 556.3 2.3 9 0.02950 0.00225 0.1490 894.2 556.0 -0.3 10 0.06550 0.03600 0.3309 975.3 89.9 -466.1 11 0.06600 0.00050 0.3334 1928.2 164.8 74.9 12 0.06875 0.00275 0.3473 1793.2 176.8 12.0 13 0.07300 0.00425 0.3688 1748.1 183.0 6.2 14 0.08375 0.01075 0.4231 1770.7 354.7 171.7 15 0.10250 0.01875 0.5178 1920.7 61.2 -293.5 16 0.10250 0.00000 0.5178 2415.9 102.0 40.8 17 0.10325 0.00075 0.5216 2385.9 181.0 79.0 18 0.11575 0.01250 0.5848 2656.0 140.9 -40.1 19 0.11975 0.00400 0.6050 2708.5 122.7 -18.2 20 0.12225 0.00250 0.6176 2693.6 246.9 124.2 21 ….. ….. ….. ….. ….. ….. ….. 2.41225 0.00000 12.1865 0.0 86.2 0.8 896 2.41225 0.00000 12.1865 0.0 89.6 3.4 897 2.41225 0.00000 12.1865 0.0 87.5 -2.1 898 2.41225 0.00000 12.1865 0.0 89.7 2.2 899 2.41225 0.00000 12.1865 0.0 96.0 6.3 900 2.41225 0.00000 12.1865 0.0 96.0 0.0 900

Comments

There is no standby at the end of the series, i.e. the Standby method function is not defined: - The titrator and STROMBOLI are stopped. - STROMBOLI stops heating, and the air pump is turned off. - The sample rack turns back to the drift position. - The drift beaker is kept out of the oven. This is done for security reason. - To improve accuracy and reproducibility, the glass vials can be first dried in a drying oven at 130°C over night. The heated vials are cooled to room temperature in a desiccator before usage. - A heating temperature of 230°C has shown to be optimum for complete water evaporation of this standard sample.

METTLER TOLEDO Page 3 of 4 Titration Application M312-09 Method (V2.0)

001 Title 008 End of sample Type Stromboli Compatible with T70/T90/V30 009 Blank ID M312 Name Blank Stromboli Title Oven stand. 5.55% (Stromboli) Value B= Mean[R1] Author METTLER TOLEDO Unit µg Date/Time 14.07.2009 12:00:00 Limits No Modified at 14.07.2009 12:00:00 Modified by METTLER TOLEDO 010 Sample Protect No Number of IDs 1 SOP None ID 1 KF oven standard 5.55% Entry type Weight 002 Drift determination Lower limit [g] 0.1 Wait time [s] 60 Upper limit [g] 0.3 Duration [min] 5 Density [g/mL] 1.0 Correction factor 1.0 003 Sample Temperature [°C] 25.0 Sample Entry Arbitrary Number of IDs 1 ID 1 Blank 011 Titration stand (Stromboli TTL) Entry type Fixed pieces Type Stromboli TTL Pieces [pcs.] 1 Titration stand Stromboli TTL 1 Weight per piece [g] 1.0 Oven temperature [°C] 230 Correction factor 1.0 Source for drift Determination Temperature [°C] 25.0 Max. Start drift [µg/min] 25 Analysis start Manual Concentration 012 Mix time Titrant KF 1-comp 5 Duration [s] 60 Nominal conc. [mg/mL] 5 Standard Water-Standard 10.0 013 Titration (KF Vol) [2] Entry type Weight Titrant Lower limit [g] 0.0 Titrant KF 1-comp 5 Upper limit [g] 2.0 Nominal conc. [mg/mL] 5 Temperature [°C] 25.0 Reagent type 1-comp Mix time [s] 10 Sensor Autostart Yes Type Polarized Entry After addition Sensor DM143-SC Concentration limits Yes Unit mV Lower limit [mg/mL] 4.5 Indication Voltametric Upper limit [mg/mL] 5.6 Ipol [µA] 24.0 Stir 004 Titration stand (Stromboli TTL) Speed [%] 35 Type Stromboli TTL Control Titration stand Stromboli TTL 1 End point [mV] 100.0 Oven temperature [°C] 230 Control band [mV] 400.0 Source for drift Determination Dosing rate(max)[mL/min] 5 Max. Start drift [µg/min] 25 Dosing rate(min)[μL/min] 80 Start Normal 005 Mix time Termination Duration [s] 60 Type Delay time Delay time [s] 10 006 Titration (KF Vol) [1] At Vmax [mL] 10.0 Titrant Min. time [s] 900 Titrant KF 1-comp 5 Max. time [s] 900 Nominal conc. [mg/mL] 5 Reagent type 1-comp 014 Calculation R2 Sensor Result Content blank compensated Type Polarized Result unit % Sensor DM143-SC Formula R1=(VEQ[2]*CONC-B[Blank Unit mV Stromboli]/1000- Indication Voltametric TIME[2]*DRIFT/1000)*C/m Ipol [µA] 24.0 Constant C= 0.1 Stir Decimal places 3 Speed [%] 35 Result limits No Control Record statistics Yes End point [mV] 100.0 Extra statistical func. No Control band [mV] 400.0 Dosing rate(max)[mL/min] 5 015 Record Dosing rate(min)[μL/min] 80 Summary Per sample Start Cautious Results No Termination Raw results No Type Delay time Table of measured values No Delay time [s] 10 Sample data No At Vmax [mL] 10.0 Resource data No Min. time [s] 900 E-V curve No Max. time [s] 900 E-t curve No V-t curve No

007 Calculation R1 H2O-t No Result Stromboli blank value Drift-t No Result unit µg H2O-t & Drift-t curves No Formula R1=(VEQ*CONC- V-t & drift-t No TIME*DRIFT/1000)*C Method No Constant C= 1000 Series data No Decimal places 1 Result limits No 016 End of sample Record statistics Yes Extra statistical func. No

METTLER TOLEDO Page 4 of 4 Titration Application M312-09 METTLER TOLEDO Application M313-09A

Automated Volumetric KF Analysis at Different Temperatures: Copper Sulfate Pentahydrate

Copper sulfate pentahydrate is heated at different temperatures to find out the optimum heating temperature for volumetric Karl Fischer titration. This is achieved by means of a KF Oven sample Changer in combination with a volumetric KF instrument.

Sample Copper sulfate pentahydrate Preparation and Procedures 0.035-0.050 g - The titer determination is performed by running water content: 36.07% method M300 using a 10.0 mg/g water standard. - Place a STROMBOLI glass vial onto the balance. Compound Copper sulfate pentahydrate, - Tare the balance. CuSO4•5H2O, M=249.68 g/mol - Weigh 0.05-0.06 g CuSO4•5H2O into the vial. (purity: 99.0-100.5%) - Close the vial with the aluminum foil, fix the blue rubber and place it on the sample rack. Chemicals 50 mL methanol dry - Start the pre-titration, and wait in standby until the drift has decreased to a constant value. - When the online drift is e.g. in the order of 10 µg/min, confirm the sample determination. Note: Before each sample place an empty vial for the blank Titrant One component titrant determination at the specific temperature. HYDRANAL® Composite 5 Sample series with different samples: - Always place an additional empty glass vial between the last measured sample vial and the subsequent blank vial at higher temperature when running a series with different samples at different increasing Standard 1.5-2 g temperatures. Water Standard 10.0 mg/g - In fact, the subsequent temperature is first achieved in (HYDRANAL® Water Standard 10.0) the last sample vial before moving the next one into the oven. This can lead to decomposition, combustion Indication DM143-SC Electrode or explosion if the last sample is thermally not stable. Remarks - Max. 7 samples can be analyzed if for each CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 sample a blank vial is used. Thus, up to 7 → temperatures can be defined (one for each (RNH)•SO4CH3 + 2 (RNH)I sample loop). - Following temperatures have been selected: 180, 200, 240, 260, 280 and 300°C.

Calculation Water content in %: - CuSO4•5H2O decomposes before melting, losing four water molecules at 110 °C and all five at 150 (VEQ*CONC-B /1000-TIME*DRIFT/1000)*C °C. At 650 °C, CuSO4•5H2O decomposes into ------m copper oxide (CuO) and sulfur trioxide (SO3). Its blue color is due to water of hydration. When VEQ: Titrant consumption at EQP heated in an open flame the crystals are TIME: Total time since sample request B = B[Blank Stromboli] dehydrated and turn grayish-white. C: 0.1 (constant for %) - All sample loops are terminated at fixed time, e.g. t(min) = t(max). Optimum results are Waste Organic solvents obtained with short mix time (e.g. 60 s), and long disposal titration time (termination at fixed time). - The water contents are plotted as a function of the heating temperature to select the optimum Cosimo De Caro, Author, heating temperature. Version MSG AnaChem, July 2009

METTLER TOLEDO Page 1 of 5 Titration Application M313-09A Instruments - KF Compact Line V30 Volumeter (V2.0) - T70/T90 Titration Excellence (V3.1.1) - STROMBOLI KF Oven Sample Changer - Drying unit with air pump (molecular sieve, silica gel). Accessories - XS205 Balance - LabX Titration Software

Results

Series start time 27.07.2009 13:33:30

No. Comment / ID Start time Rx Result Unit Name

1/1 (1/12) Blank 180°C 27.07.2009 14:01:17 R1 =517.4 µg Stromboli blank value 1/1 (2/12) Copper sulfate 180°C 27.07.2009 14:17:19 R2 =29.03 % Content blank comp. (B in µg) 1/1 (3/12) Blank 200°C 27.07.2009 14:37:21 R3 =546.9 µg Stromboli blank value 1/1 (4/12) Copper sulfate 200°C 27.07.2009 14:53:14 R4 =32.45 % Content blank comp. (B in µg) 1/1 (5/12) Blank 240°C 27.07.2009 15:11:56 R5 =531.3 µg Stromboli blank value 1/1 (6/12) Copper sulfate 240°C 27.07.2009 15:27:50 R6 =35.14 % Content blank comp. (B in µg) 1/1 (7/12) Blank 260°C 19.01.2009 16:24:12 R7 =88.3 µg Stromboli blank value 1/1 (8/12) Copper sulfate 260°C 27.07.2009 16:01:38 R8 =35.70 % Content blank comp. (B in µg) 1/1 (9/12) Blank 280°C 27.07.2009 16:19:20 R9 =467.3 µg Stromboli blank value 1/1 (10/12) Copper sulfate 280°C 27.07.2009 16:35:13 R10 =35.82 % Content blank comp. (B in µg) 1/1 (11/12) Blank 300°C 27.07.2009 16:52:59 R11 =385.2 µg Stromboli blank value 1/1 (12/12) Copper sulfate 300°C 27.07.2009 17:08:51 R12 =35.72 % Content blank comp. (B in µg)

Titration curve

Sample 1/1 (12/12)

METTLER TOLEDO Page 2 of 5 Titration Application M313-09A Table of measured values

Volume Increment H2O Online drift Signal Change Time mL mL mg µg/min mV mV s 0.00000 NaN 0.0000 0.0 264.9 NaN 0 0.00000 0.00000 0.0000 0.0 268.8 3.9 0 0.00000 0.00000 0.0000 0.0 293.6 24.8 1 0.00100 0.00100 0.0051 151.4 318.3 24.7 2 0.00375 0.00275 0.0189 392.0 337.9 19.6 3 0.01050 0.00675 0.0531 837.7 333.8 -4.1 4 0.01850 0.00800 0.0935 1193.3 325.7 -8.1 5 0.02025 0.00175 0.1023 1096.3 327.7 2.0 6 0.02600 0.00575 0.1314 1212.7 315.1 -12.6 7 0.02975 0.00375 0.1503 1218.9 304.8 -10.3 8 0.03125 0.00150 0.1579 1141.5 323.0 18.2 9 0.03425 0.00300 0.1731 1128.7 336.4 13.4 10 0.04075 0.00650 0.2059 1223.2 321.3 -15.1 11 0.05450 0.01375 0.2754 1636.0 270.1 -51.2 12 0.06125 0.00675 0.3095 1808.6 265.8 -4.3 13 0.06650 0.00525 0.3360 1883.6 241.9 -23.9 14 0.08025 0.01375 0.4055 2093.7 210.4 -31.5 15 0.08125 0.00100 0.4106 1883.6 220.6 10.2 16 0.08700 0.00575 0.4396 2003.7 238.8 18.2 17 0.09600 0.00900 0.4851 2101.3 205.5 -33.3 18

...... … ..… .....

2.60750 0.00000 13.1757 0.0 95.8 1.4 893 2.60750 0.00000 13.1757 0.0 96.1 0.3 894 2.60750 0.00000 13.1757 0.0 94.8 -1.3 895 2.60750 0.00000 13.1757 0.0 98.2 3.4 896 2.60750 0.00000 13.1757 0.0 96.1 -2.1 897 2.60850 0.00100 13.1807 30.0 93.8 -2.3 898 2.60875 0.00025 13.1820 37.5 94.8 1.0 899 2.60875 0.00000 13.1820 37.5 96.0 1.2 900 Sample 1/1 (12/12)

Comments

Water content in copper sulfate pentahydrate (t = 900 s)

30 Water content / %

25 160 180 200 220 240 260 280 300 320 Heating temperature / °C

- In the above diagram, the water content has been plotted as a function of the heating temperature for an extraction time of 900 s. More results from several sample series have been added into it. The diagram shows that the water content remains approximately the same for a temperature of above

240°C (tmin = tmax = 900 s).

METTLER TOLEDO Page 3 of 5 Titration Application M313-09A - Method M313 consists of 12 sample loops. Thus, 6 different temperatures have been tested, since for each temperature one blank vial and one sample vial has to be titrated. As a consequence, not all possible sample loops have been programmed (total number of loops with Stromboli: 14). - The blank value is dependent on temperature, and therefore it must be determined at each temperature (see also M396):

Blank value of glass vials for STROMBOLI KF Oven Sample Changer

700.0 600.0 500.0

g water 400.0 μ 300.0 200.0 100.0 Blank / 0.0 0 50 100 150 200 250 300 350 Temperature / °C

Termination: fixed time, tmin = tma x= 600 s

- With the KF oven it is recommended to heat the sample using a fixed titration time parameter tmin = tmax, i.e. water is evaporated and transferred in the cell during this time, and immediately titrated until the analysis is stopped at a fixed time tmin = tmax defined in the method. In this way, evaporation of water is avoided which lead to lower water content. - The heating (or extraction) time is also a relevant parameter when working with the KF drying oven. As for temperature, the optimum titration time has to be selected to get accurate and precise results. The appropriate time can be found by keeping the heating temperature constant throughout the

method, while sample loops have different fixed titration time, e.g. tmin = tmax = 300 s, 600 s, 900 s, and 1200 s. An example is given in application M396.

- IMPORTANT: There is no stand-by at the end of the series, i.e. the stand-by method function is not defined: - The titrator and STROMBOLI are stopped - Stromboli stops heating, and the air pump is turned off. - The sample rack turns back to the drift position. - The drift beaker is kept out of the oven. This is done for security reason.

METTLER TOLEDO Page 4 of 5 Titration Application M313-09A Method (V2.0) 001 Title ... Type Stromboli Compatible with T70 / T90 / V30 008 Record Method ID M313 Summary Per sample Title Temperature ramp (Stromboli) Results No Author METTLER TOLEDO Raw results No Date/Time 23.07.2009 10:41:59 …… … Modified at 30.07.2009 10:15:58 Modified by METTLER TOLEDO 009 End of sample Protect No SOP None 010 Blank Name Blank Stromboli 002 Drift determination Value B= Mean[R1] Wait time [s] 900 Unit µg Duration [min] 5 Limits No

003 Sample 011 Sample Sample Number of IDs 1 Number of IDs 1 ID 1 Copper sulfate 180°C ID 1 Blank 180°C Entry type Weight Entry type Fixed pieces Lower limit 0.03 g Pieces 1 pcs. Upper limit 0.04 g Weight per piece 1.0 g Density 1.0 g/mL Correction factor 1.0 Correction factor 1.0 Temperature 25.0°C Temperature 25.0°C Start analysis Start automatically Entry Arbitrary Drift stability 25 µg/min Dt 60 s 012 Titration stand (Stromboli TTL) Entry Arbitrary Type Stromboli TTL Concentration Titration stand Stromboli TTL 1 Titrant KF 1-comp 5 Oven temperature 180°C Nominal concentration 5 mg/mL Source for drift Determination Standard Water-Standard 10.0 Max. Start drift 25 µg/min Entry type Weight Lower limit 0.0 g 013 Mix time Upper limit 2.0 g Duration 15 s Temperature 25.0°C Mix time 10 s 014 Titration (KF Vol) [2] Autostart Yes Titrant Entry After addition Titrant KF 1-comp 5 Concentration limits Yes Nominal concentration 5 mg/mL Lower limit 4.5 mg/mL Reagent type 1-comp Upper limit 5.6 mg/mL Sensor Type Polarized 004 Titration stand (Stromboli TTL) Sensor DM143-SC Type Stromboli TTL Unit mV Titration stand Stromboli TTL 1 Indication Voltametric Oven temperature 180°C Ipol 24.0 µA Source for drift Determination Stir Max. start drift 25 µg/min Speed 45 % Control 005 Mix time … Normal Duration 15 s Termination … 006 Titration (KF Vol) [1] Titrant 015 Calculation R2 Titrant KF 1-comp 5 Result Content blank comp. (B in µg) Nominal concentration 5 mg/mL Result unit % Reagent type 1-comp Formula R2=(VEQ[2]*CONC - B[Blank Sensor Stromboli]/1000 - Type Polarized TIME[2]*DRIFT/1000)*C/m Sensor DM143-SC Constant C= 0.1 Unit mV Decimal places 2 Indication Voltametric … Ipol 24.0 µA Stir 016 Record Speed 45 % ... Control 017 End of sample End point 100.0 mV ... Control band 400.0 mV The first two loops are subsequently repeated 5 times (5 Dosing rate(max) 5 mL/min different temperatures): Dosing rate(min) 80 µL/min Start Normal Blank (loop3) + Sample (loop4) 200°C Termination Blank (loop5) + Sample (loop6) 240°C Type Delay time Blank (loop7) + Sample (loop8) 260°C Delay time 10 s Blank (loop9) + Sample (loop10) 280°C At V(max) 10.0 mL Blank (loop11) + Sample (loop12) 300°C Min. time 900 s ... Max. time 900 s 092 End of sample

007 Calculation R1 Result Stromboli blank value Result unit µg

Formula R1=(VEQ*CONC - TIME*DRIFT/1000)*C Constant C= 1000 Decimal places 1

METTLER TOLEDO Page 5 of 5 Titration Application M313-09A METTLER TOLEDO Application M314-08 Water Content Determination in Water Standard 1.0 mg/g

Determination of the water content of 1.0 mg/g water standard.

Sample 1 g Preparation and Procedures Water Standard 1.0 mg/g (HYDRANAL® Water Standard 1.0) - Drawn out approximately 0.5 mL of water standard 1.0 into a 5 mL syringe with needle to rinse it before sample analysis. Compound Water, H2O M = 18.01 g/mol - Discard the 0.5 mL standard portion and drawn the rest of the standard completely into the syringe. Chemicals HYDRANAL®-Coulomat AG as - The water standard is then injected into the KF anolyte. titration vessel in aliquots of 1 mL. ® HYDRANAL -Coulomat CG as - The weight is determined by back-weighing catholyte. technique.

Titrant --

Standard --

Indication DM143-SC Electrode Remarks -- Chemistry CH3OH + SO2 + 3 RN + I2 + H2O → (RNH)•SO4CH3 + 2 (RNH)I

Calculation Water content (mg/g)

(ICEQ/10.712-(TIME*DRIFT) ) ------C*m

ICEQ: Current cons. to EP TIME: Total time since sample request C: 1000 (constant for mg/g calc.) Waste Karl Fischer waste should be disposal disposed of as halogenated organic solvents.

Author, Melanie Nijman (MSG), Version December 2008

METTLER TOLEDO Page 1 of 4 Titration Application M314-08 Instruments - C20D/C30D Compact Line Karl Fischer Titrator with generator electrode with diaphragm - XS205 Balance

Accessories - LabX Titration Software - Solvent Manager - 5 mL syringe with needle

Results

Sample ID Date Sample size Result 01/10 Water standard 1.0 20.11.2008 1.0469 g R1 = 0.998 mg/g Content 02/10 Water standard 1.0 1.0179 g R1 = 0.999 mg/g Content 03/10 Water standard 1.0 1.0011 g R1 = 0.999 mg/g Content 04/10 Water standard 1.0 0.9132 g R1 = 1.002 mg/g Content 05/10 Water standard 1.0 1.0195 g R1 = 1.002 mg/g Content 06/10 Water standard 1.0 0.9585 g R1 = 1.002 mg/g Content 07/10 Water standard 1.0 1.0301 g R1 = 0.998 mg/g Content 08/10 Water standard 1.0 1.1264 g R1 = 0.997 mg/g Content 09/10 Water standard 1.0 1.0412 g R1 = 1.000 mg/g Content 10/10 Water standard 1.0 0.9693 g R1 = 0.999 mg/g Content

Rx Name n Mean value Unit s srel [%] R1 Content 10 1.000 mg/g 0.002 0.184

Titration curve

METTLER TOLEDO Page 2 of 4 Titration Application M314-08 Table of measured values

METTLER TOLEDO Page 3 of 4 Titration Application M314-08 Method

001 Title Type Karl Fischer titration Coul. Compatible with C20 / C30 Method ID M314 Title Water standard 1.0 mg/mL Author METTLER TOLEDO Date/Time 25.08.2008 15:00:00 Modified on -- Modified by METTLER TOLEDO Protect No SOP None

002 Sample Sample Number of IDs 1 ID 1 -- Entry type Weight Lower limit [g] 0.0 Upper limit [g] 2.0 Density [g/mL] 1.0 Correction factor 1.0 Temperature [°C] 25.0 Autostart No Entry After addition

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift [µg/min] 25

004 Mix time Duration [s] 15

005 Titration (KF Coul) [1] Sensor Type Polarized Name DM143-SC Unit mV Indication Voltametric Ipol [µA] 5.0 Stir Speed [%] 45 Control End point [mV] 100 Control band [mV] 250 Rate Normal Generator current Automatic Termination Type Drift stop relative Drift [µg/min] 3.0 Min. time [s] 90 Max. time [s] 3600

006 Calculation R1 Result type Predefined Result Content Result unit mg/g Formula R1=(ICEQ/10.712- TIME*DRIFT)/(C*m) Constant 1000 Decimal places 3 Result limits No Record statistics Yes Extra statistical functions No

007 Record Summary Per sample Results No Raw results No Table of meas. values No Sample data No Resource data No E – C No E – t No C – t No H2O – t No Drift – t No H2O – t & Drift – t No C – t & Drift – t No Method No Series data No

008 End of sample Open series Yes

METTLER TOLEDO Page 4 of 4 Titration Application M314-08 METTLER TOLEDO Application M315-08 Automated Coulometric KF Analysis: Check with 1% Oven Water Standard

The system Coulometer-KF Oven sample Changer is checked by using a solid oven water standard. The water content of the standard has a certified value of 1.00 ± 0.03 % . Water is evaporated in the KF oven sample changer and transferred into the titration cell where it is determined by coulometric KF analysis.

Sample apura® Oven water standard Preparation and Procedures 0.08-0.1 g - Place a STROMBOLI glass vial onto the balance. - Tare the balance.

Compound Water, H2O - Weigh 0.08-0.1 g KF oven water standard into M = 18.01 g/mol the vial. - Close the vial with the aluminum foil. - Fix the blue rubber onto the vial. Chemicals -- - Place the sample vial onto the STROMBOLI sample rack. - Start the pre-titration, and wait in standby until the drift has decreased to a constant value. ® Titrant 100 mL HYDRANAL Coulomat - When the online drift is e.g. in the order of 10 AG Oven µg/min, confirm the sample determination.

Note: Standard -- An empty vial for blank determination is placed before a series of sample vials.

Indication DM143-SC Electrode Remarks - Two sample loops are used in this method: the CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 first loop allows for the blank value determination, → whereas the second one allows for the (RNH)•SO4 CH3 + 2 (RNH)I determination of the water content in the standard. - Each loop can consists of more than one sample vials. Calculation Water content in %: - A heating temperature of 180°C has proven to be (ICEQ/10.712-TIME*DRIFT-B[Blank]) optimum for complete water evaporation. ------(C*m) - All sample loops are terminated at fixed time, e.g.

ICEQ: Current at EP t(min) = t(max). Optimum results are obtained TIME: Total time since sample request with short mix time (e.g. 60 s), and long titration Blank: Blank value time ( t(min) = t(max)=900 s ). C: 10000 (constant for % calc.) - To improve accuracy and reproducibility, the Waste Organic solvents glass vials can be first dried in a drying oven at disposal 130°C over night. The heated vials are cooled to room temperature in a desiccator previous use.

Author, MSG AnaChem, Version Cosimo De Caro, Sept 2008

METTLER TOLEDO Page 1 of 4 Titration Application M315-08 Instruments - KF Compact Coulometer C30X - STROMBOLI KF Oven Sample Changer - Drying unit with air pump (molecular sieve, silica gel).

Accessories - XS205 Balance - LabX Titration Software

Results Results

R1 (1/1) (1/2) 508.2 µg R2 (1/3) (2/2) 1.01 % R2 (2/3) (2/2) 1.01 % R2 (1/3) (2/2) 1.04 %

Statistics

Rx Name n Mean value Unit s srel[%] ______

R1 Stromboli blank 1 547.1 µg NaN NaN value R2 Content blank 3 1.02 % 0.02 1.447 compensated

Additional series: R2 Content blank 3 0.98 % 0.04 4.506 compensated R2 Content blank 3 0.96 % 0.05 5.331 compensated R2 Content blank 3 1.03 % 0.02 1.490 compensated R2 Content blank 7 1.09 % 0.04 3.288 compensated R2 Content blank 3 1.03 % 0.03 2.926 compensated R2 Content blank 3 1.00 % 0.01 1.228 compensated

Titration curve

METTLER TOLEDO Page 2 of 4 Titration Application M315-08 Table of measured values

Consumpn Increment Signal Change Time Temperature mC mC mV mV s °C 0 NaN 414.4 NaN 0 32.77 32.77 415.6 1.2 1 120.459 87.689 416.2 0.6 1 263.462 143.003 411.8 -4.4 2 461.78 198.318 402.8 -9 3 715.412 253.632 399.6 -3.2 4 1024.359 308.947 383 -16.6 4 1388.621 364.262 353.4 -29.6 6 1709.326 320.705 320.2 -33.2 7 1943.5 234.174 274.2 -46 8 2167.237 223.737 238.9 -35.3 9 2377.556 210.319 183.1 -55.8 10 2490.969 113.413 141.4 -41.7 11 2520.893 29.924 140.2 -1.2 12 2565.726 44.833 157.5 17.3 13 2647.831 82.105 150.8 -6.7 14 2722.482 74.651 130.3 -20.5 15 2771.787 49.305 125.8 -4.5 16 2810.657 38.87 130.4 4.6 17 2853.999 43.342 129 -1.4 18

...... … ..… ..… ..…

7029.946 0.107 99 -2.4 894 7029.946 0 102.1 3.1 895 7033.035 3.089 105.2 3.1 896 7040.596 7.561 102.5 -2.7 897 7046.667 6.071 102.8 0.3 898 7048.211 1.544 100.2 -2.6 899 7048.211 0 100 -0.2 900 7048.211 0 100 0 900

Comments - The results of several sample series are plotted in the following diagram:

KF Oven Standard

1.13 1.12 1.11 1.10 1.09 1.08 1.07 1.06 1.05 1.04 1.03 1.02 1.01 1.00 0.99 0.98 Water content / % / content Water 0.97 0.96 0.95 0.94 0.93 0.92 0.91 0.90 0123456789 Sample series / Nr.

- The results show a good agreement with the certified value of the KF standard, i.e. 1.00 ± 0.03 %.

- To improve accuracy, the blank value was also determined as an average of a series of 3 blank vials.

METTLER TOLEDO Page 3 of 4 Titration Application M315-08 Method 001 Title Type Karl Fischer titration 011 Titration stand (Stromboli TTL) Compatible with C30 Type Stromboli TTL ID M315 Titration stand Stromboli TTL 1 Title KF oven standard 1% Oven temperature 180°C (Stromboli) Source for drift Determination Author METTLER TOLEDO Max. Start drift 25 µg/min Date/Time 25.08.2008 15:00:00 Modified at -- 012 Mix time Modified by -- Duration 60 s Protect No SOP None 013 Titration (KFCoul) [2] Sensor 002 Drift determination Type Polarized Wait time [s] 60 Sensor DM143-SC Duration [min] 3 Unit mV Indication Voltametric 003 Sample Ipol 5.0 µA Number of IDs 1 Stir ID 1 Blank 60°C Speed 45 % Entry type Fixed pieces Control Pieces 1 pcs. End point 100 mV Weight per piece 1.0 g Control band 250 mV Correction factor 1.0 Rate Normal Temperature 25.0°C Generator current Automatic Start analysis Manual Termination Entry Arbitrary Type Delay time Delay time 10 s 004 Titration stand (Stromboli TTL) Min. time 900 s Type Stromboli TTL Max. time 900 s Titration stand Stromboli TTL 1 Oven temperature 180°C 014 Calculation R2 Source for drift Determination Result Content blank compensated Max. Start drift 25 µg/min Result unit % Formula R2=(ICEQ[2]/10.712 - 005 Mix time TIME[2]*DRIFT-B[Blank Duration 60 s Stromboli])/(C*m) Constant C= 10000 006 Titration (KFCoul) [1] Decimal places 2 Sensor Result limits No Type Polarized Record statistics Yes Sensor DM143-SC Extra statistical func. No Unit mV Indication Voltametric 015 Record Ipol 5.0 µA Summary Per sample Stir Results Per sample Speed 45 % Raw results Per sample Control Table of measured values No End point 100 mV Sample data No Control band 250 mV Resource data No Rate Normal E-V curve No Generator current Automatic E-t curve No Termination V-t curve No

Type Delay time H2O-t No Delay time 10 s Drift-t No Min. time 900 s H2O-t & Drift-t curves No Max. time 900 s V-t & drift-t No Method No 007 Calculation R1 Series data No Result Stromboli blank value Result unit µg 016 End of sample Formula R1=(ICEQ[1]/10.712 - TIME[1]*DRIFT)/C Note: Constant C= 1 There is no stand-by at the end of the series, i.e. the stand- Decimal places 1 by method function is not defined: Result limits No - The titrator and STROMBOLI are stopped Record statistics Yes - Stromboli stops heating, and the air pump is turned off. Extra statistical func. No - The sample rack turns back to the drift position. - The drift beaker is kept out of the oven. 008 End of sample This is done for security reason. 009 Blank Name Blank Stromboli Value B= Mean[R1] Unit µg Limits No

010 Sample Number of IDs 1 ID 1 KF oven standard 1% Entry type Weight Lower limit 0.08 g Upper limit 0.12 g Density 1.0 g/mL Correction factor 1.0 Temperature 25.0°C Entry Arbitrary

METTLER TOLEDO Page 4 of 4 Titration Application M315-08 METTLER TOLEDO Application M391-08 Determination of the Water Content in Dry Toluene

Determination of the water content of dried toluene by coulometric KF titration.

Sample Toluene, dried with molecular Preparation and Procedures sieve, ~ 2.7 g. 1) Reagent grade toluene is filled into a septum bottle with molecular sieve in it. 2) The toluene is left inside the bottle to dehydrate Compound Water, H2O and stabilize overnight. M = 18.01 g/mol 3) First, the syringe is washed with the sample in order to eliminate residual moisture in the syringe: Chemicals apura® CombiCoulomat Frit - Drawn out approximately 2-3 mL of sample into a (used as both anolyte and 10 mL syringe with needle to rinse it before sample catholyte) analysis. - Discard the sample into disposal container and fill the syringe again with the sample. Titrant N/A - The sample is then injected into the KF titration vessel in aliquots of 2.5-3 mL.

Standard N/A

Indication DM143-SC Remarks Samples with such low water content will have CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 larger srel values, due to this low water content. To → be able to counteract this at least partially, larger (RNH)•SO4CH3 + 2 (RNH)I sample sizes are taken.

Calculation Water content (ppm)

(ICEQ/10.712-TIME*DRIFT) ------C*m

ICEQ: Current consumption at EP TIME: Total time since sample request C: 1 (constant for ppm calc.)

Waste Karl Fischer waste should be disposal disposed of as halogenated organic solvents.

Author, Melanie Nijman (MSG), Version December 2008

METTLER TOLEDO Page 1 of 4 Titration Application M391-08 Instruments - KF Compact Line Coulometers C20D/C30D with generator electrode with diaphragm - XP205 Balance

Accessories - LabX titration pro Software - 10 mL syringe with needle

Results All results Method-ID M391-08 Sample Toluene dry (1/6) R1 (Water content) 2.1 ppm Sample Toluene dry (2/6) R1 (Water content) 1.7 ppm Sample Toluene dry (3/6) R1 (Water content) 2.4 ppm Sample Toluene dry (4/6) R1 (Water content) 1.7 ppm Sample Toluene dry (5/6) R1 (Water content) 2.2 ppm Sample Toluene dry (6/6) R1 (Water content) 2.6 ppm

Statistics Method-ID M391-08 R1 Water content (ppm) Samples n=6 Mean 2.1 ppm s 0.4 srel 17.3 %

Titration curve

METTLER TOLEDO Page 2 of 4 Titration Application M391-08 Table of measured values

Consumption Increment Signal Change Time mC mC mV mV s 0 NaN 152.5 NaN 0 53.708 53.708 141 -11.5 1 111.959 58.251 90.7 -50.3 2 111.959 0 85.5 -5.2 3 111.959 0 96.3 10.8 4 111.959 0 96.9 0.6 5 111.959 0 96.7 -0.2 6 111.959 0 97.4 0.7 7

METTLER TOLEDO Page 3 of 4 Titration Application M391-08 Method

001 Title Type Karl Fischer titration Coul. Compatible with C20 / C30 Method ID M391 Title Toluene dry Author METTLER TOLEDO Date/Time 25.08.2008 15:00:00 Modified on -- Modified by METTLER TOLEDO Protect No SOP None

002 Sample Sample Number of IDs 1 ID 1 Toluene dry Entry type Weight Lower limit [g] 0.0 Upper limit [g] 5.0 Density [g/mL] 1.0 Correction factor 1.0 Temperature [°C] 25.0 Autostart No Entry After addition

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift [µg/min] 25

004 Mix time Duration [s] 15

006 Calculation R1 Result type Predefined Result Content Result unit ppm Formula R1=(ICEQ/10.712- TIME*DRIFT)/(C*m) Constant 1 Decimal places 1 Result limits No Record statistics Yes Extra statistical functions No

008 End of sample Open series Yes

METTLER TOLEDO Page 4 of 4 Titration Application M391-08 METTLER TOLEDO Application M392-08 Determination of the Water Content in Nitrogen Gas

Determination of the water content of nitrogen gas by coulometric KF titration.

Sample Nitrogen gas (N2), 1600 mL Preparation and Procedures 1) Connect the nitrogen gas bottle to the flow meter and through the flow meter with the gas inlet to the titration vessel. Compound Water, H2O 2) Use the flow meter to control the flow of gas. M = 18.01 g/mol With higher water content, the gas flow can be lower and mixing time can be shorter. If the gas contains very little water, the flow of gas should be Chemicals 100 mL larger and the mixing time longer, as the accuracy HYDRANAL® Coulomat-AD will decrease with lower total water content. 3) In general, a gas flow rate between 50 and 200 mL is used. 4) Before a measurement of the nitrogen water Titrant N/A content is performed, the system should first be purged with nitrogen gas and a pre-titration should be performed. 5) Once the pre-titration has finished and the drift is stable, the gas is bubbled through the solution, in Standard N/A this case at 160 mL/min. This is done for 10 minutes, after which the samples size of the gas (here 1600 mL) is input into the titrator. Indication DM143-SC Remarks - A stable and accurate flow meter needs to be CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 used for this application to ensure an accurate → sample volume input into the titrator after the 10 (RNH)•SO4CH3 + 2 (RNH)I minute mixing time.

- In case larger amounts of solvent evaporate, caused by the hot gas flow through the titration Calculation Water content (µg/L) vessel, the solvent needs to be topped up in between measurements. (ICEQ/10.712-TIME*DRIFT) ------C*m

ICEQ: Current consumption at EP TIME: Total time since sample request C: 0.001 (constant for µg/L calc.)

Waste Karl Fischer waste should be disposal disposed of as halogenated organic solvents.

Author, Melanie Nijman (MSG), Version December 2008

METTLER TOLEDO Page 1 of 4 Titration Application M392-08 Instruments - KF Compact Line Coulometers C20X/C30X (without diaphragm) - XP205 Balance

Accessories - LabX Titration pro Software - Flow meter (e.g. from Stromboli drying unit (ME-51108660; whole drying unit)) - Gas inlet and stopper for coulometer (ME-51108669 and ME-51108761)

Results

Results No. Comment / ID Rx Result Unit Name 1/9 NITROGEN R1 = 890.90 µg/L Total water content 2/9 NITROGEN R1 = 841.80 µg/L Total water content 3/9 NITROGEN R1 = 856.00 µg/L Total water content 4/9 NITROGEN R1 = 830.00 µg/L Total water content 5/9 NITROGEN R1 = 848.50 µg/L Total water content 6/9 NITROGEN R1 = 849.50 µg/L Total water content 7/9 NITROGEN R1 = 848.70 µg/L Total water content 8/9 NITROGEN R1 = 846.10 µg/L Total water content 9/9 NITROGEN R1 = 852.10 µg/L Total water content

Statistics Rx Name n Mean value Unit s srel [%] R1 Total water content 9 851.5 µg/L 16.5 1.938

Titration curve

METTLER TOLEDO Page 2 of 4 Titration Application M392-08 Table of measured values

Consumption Increment H2O Online drift Signal Change Time mC mC µg µg/min mV mV s 0.4 NaN 0 0 525.8 NaN 1 32.8 32.4 3.1 13 526.6 0.8 1 120.5 87.7 11.2 94.7 525.6 -1 1 263.5 143 24.6 231.1 526.1 0.5 2 461.8 198.3 43.1 424.3 526.5 0.4 3 715.4 253.6 66.8 648.1 525.5 -1 4 1024.4 309 95.6 868.4 525.9 0.4 5 1388.6 364.2 129.6 1074.6 528 2.1 6 1808.2 419.6 168.8 1288.1 525.9 -2.1 7 2283.1 474.9 213.1 1452.1 526.3 0.4 9 2813.3 530.2 262.6 1421.1 526.7 0.4 10 3398.8 585.5 317.3 1509.1 526.1 -0.6 12 4039.6 640.8 377.1 1643.9 525.1 -1 14 4735.8 696.2 442.1 1720.8 525.8 0.7 16 5464.2 728.4 510.1 1771.9 526.4 0.6 18 6192.6 728.4 578.1 1818.6 525.1 -1.3 21 6921 728.4 646.1 1796.3 524.7 -0.4 23 7649.5 728.5 714.1 1759.4 525.9 1.2 25 8377.9 728.4 782.1 1766.1 524.4 -1.5 27 9106.3 728.4 850.1 1767.9 522.7 -1.7 30 9834.7 728.4 918.1 1812.6 521.2 -1.5 32 10563.1 728.4 986.1 1850.5 519.1 -2.1 34 11291.5 728.4 1054.1 1883 506.7 -12.4 36 12020 728.5 1122.1 1911 493.6 -13.1 39 12748.4 728.4 1190.1 1935.5 471.1 -22.5 41 13476.8 728.4 1258.1 1957.1 426.4 -44.7 43 14205.2 728.4 1326.1 1976.2 297.6 -128.8 45 14427.5 222.3 1346.9 1135.1 190.9 -106.7 46 14543.8 116.3 1357.7 1167.9 134.6 -56.3 47 14563.3 19.5 1359.5 814.9 110 -24.6 48 14572.4 9.1 1360.4 560.6 106 -4 49 14577 4.6 1360.8 406 99.9 -6.1 50 14580 3 1361.1 254.7 107.5 7.6 51 14595 15 1362.5 241.8 101.7 -5.8 52 14595.1 0.1 1362.5 184.2 101.4 -0.3 53 14595.1 0 1362.5 119.5 98.6 -2.8 54 14595.1 0 1362.5 68 100.9 2.3 55 14595.2 0.1 1362.5 53.5 101.9 1 56 14596.7 1.5 1362.6 41.2 95.2 -6.7 57 14596.7 0 1362.6 36.1 103.9 8.7 58 14599.8 3.1 1362.9 39.4 97.5 -6.4 60 14599.8 0 1362.9 19 99.6 2.1 60 14599.8 0 1362.9 12 102.6 3 62 14601.4 1.6 1363.1 5.7 99.6 -3 62 14603 1.6 1363.2 0 99 -0.6 64 14603 0 1363.2 0 99 0 64

Comments Setup for the water content determination gas:

Three-way valve

Flow meter

Gas sample

METTLER TOLEDO Page 3 of 4 Titration Application M392-08 Method

001 Title Type Karl Fischer titration Coul. Compatible with C20/C30 Method ID M392 Title Nitrogen gas Author METTLER TOLEDO Date/Time 25.08.2008 15:00:00 Modified on -- Modified by METTLER TOLEDO Protect No SOP None

002 Sample Sample Number of IDs 1 ID 1 -- Entry type Volume Lower limit [mL] 0.0 Upper limit [mL] 10000.0 Density [g/mL] 1.0 Correction factor 1.0 Temperature [°C] 25.0 Autostart No Entry Arbitrary

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift [µg/min] 25

004 Mix time Note: Duration [s] 600 During a mix time of 600 s the gas is flowing into the titration cell. In this way enough water is accumulated into 005 Titration (KF Coul) [1] the solvent in order to achieve a nice titration curve. Sensor For this reason, the termination parameter is set to relative Type Polarized drift. Name DM143-SC Unit mV Indication Voltametric Ipol [µA] 5.0 Stir Speed [%] 45 Control End point [mV] 100 Control band [mV] 250 Rate Normal Generator current Automatic Termination Type Drift stop relative Drift [µg/min] 3.0 Min. time [s] 0 Max. time [s] 600

006 Calculation R1 Result type Predefined Result Total water Content Result unit µg/L Formula R1=(ICEQ/10.712- TIME*DRIFT)/(C*m) Constant 0.001 Decimal places 1 Result limits No Record statistics Yes Extra statistical functions No

008 End of sample Open series Yes

METTLER TOLEDO Page 4 of 4 Titration Application M392-08 METTLER TOLEDO Application M393-08 KF Coulometric: Determination of the Water Content in Acetone

The water content in acetone (dry) was determined by coulometric Karl Fischer titration.

Sample Acetone (dry), 2 g. Preparation and Procedures Acetone was dried for at least 24 hours over molecular sieve (3 Å) in a septum flask and stored for the analysis as shown below: Compound Water, H2O M = 18.01 g/mol

Chemicals --

Titrant In situ electrochemically Molecular generated iodine, I2 : Acetone sieve - 100 mL HYDRANAL® Coulomat AK; - 5 mL HYDRANAL® Coulomat CG-K

Standard -- In this way, the water content was lowered.

Indication DM143-SC Electrode Remarks - Ketones such as acetone react with methanol CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 forming ketals. In this reaction water is released: →

(RNH)•SO4CH3 + 2 (RNH)I R-CO-R + 2 CH3OH → R-C(OCH3)2-R + H2O

Therefore methanol-free anolyte and catholyte are used. Calculation Water content in ppm: - However, the methanol-free reagents also react (ICEQ/10.712-TIME*DRIFT) slowly with acetone. This leads to an increasing ------(C*m) drift value with increasing number of samples. Thus, it is recommended to replace the anolyte ICEQ: Current at EP after at most 5-6 samples. TIME: Total time since sample request C: 1 (constant for ppm calc.) - With 2 g sample size the drift value increases from 7 μg/min (start) up to about 40 μg/min (5-6 Waste Organic solvents samples). disposal Therefore the max. start drift value has been set to 60 μg/min and the end point to 140 mV with a relative drift stop of 6 μg/min. Author, MSG Anachem Group Version Thomas Hitz, Dec 2008

METTLER TOLEDO Page 1 of 4 Titration Application M393-08 Instruments - KF Compact Coulometer C20D/C30D (D: Generating cell with diaphragm)

Accessories - XS205 Balance - LabX Titration Software

Results All results Method-ID M393 Sample Acetone Dry (1/6) R1 (Content) 20.1 ppm Sample Acetone Dry (2/6) R1 (Content) 26.4 ppm Sample Acetone Dry (3/6) R1 (Content) 21.8 ppm Sample Acetone Dry (4/6) R1 (Content) 24.5 ppm Sample Acetone Dry (5/6) R1 (Content) 25.0 ppm

Statistics Method-ID M393 R1 Content Samples 5 Mean 23.6 s 2.5 srel 10.8%

Titration curve

240

220

200

180

160 Signal / mV Signal

140

120

100 0 100 200 300 400 500 600 700 800 900 Consumption / mV

METTLER TOLEDO Page 2 of 4 Titration Application M393-08 Table of measured values

Consumption Increment H2O Online drift Signal Change Time mC mC μg μg/min mV mV s 0 NaN 0 0 216.8 NaN 0 139.2 139.2 13 386.6 214.6 -2.2 2 242.1 102.9 22.6 781.7 170.4 -44.2 2 276.5 34.4 25.8 578 151.6 -18.8 4 287.1 10.6 26.8 347.6 151.8 0.2 4 305.1 18 28.5 230 154.4 2.6 6 330.5 25.4 30.9 210.4 152.6 -1.8 7 353 22.5 33 208.4 142.4 -10.2 8 357.6 4.6 33.4 106.1 143.8 1.4 9 365.1 7.5 34.1 54.9 146.4 2.6 10 377.2 12.1 35.2 49.9 147.7 1.3 11 392.2 15 36.6 72.9 146.6 -1.1 12 402.7 10.5 37.6 77.4 144.3 -2.3 13 411.8 9.1 38.4 57.7 147.3 3 14 426.8 15 39.8 63.6 143.9 -3.4 15 431.4 4.6 40.3 60.7 144.1 0.2 16 440.4 9 41.1 28.1 146.4 2.3 17 451 10.6 42.1 55.2 148.2 1.8 18 464.5 13.5 43.4 80.1 142.4 -5.8 19 466.1 1.6 43.5 36.4 142.5 0.1 20 472.2 6.1 44.1 0 147 4.5 21

813.8 4.6 76 0 141.8 1.1 67 815.4 1.6 76.1 16.2 144.4 2.6 68 822.9 7.5 76.8 25.4 146.4 2 69 833.5 10.6 77.8 44.6 146 -0.4 70 844 10.5 78.8 63.4 144.9 -1.1 71 848.6 4.6 79.2 55 139.8 -5.1 72

Comments --

METTLER TOLEDO Page 3 of 4 Titration Application M393-08 Method

001 Title Type KF Coulometric Compatible with C20/C30 ID M393 Title Acetone Dry Author METTLER TOLEDO Date/Time 11.11.2008 02:50:12 Modified at -- Modified by -- Protect No SOP None

002 Sample Number of IDs 1 ID 1 Acetone Dry Entry type Weight Lower limit[g] 0.0 Upper limit [g] 5.0 Density [g/mL] 1.0 Correction factor 1.0 Temperature [°C] 25.0 Autostart No Entry After addition

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 60 µg/min

004 Mix time Duration [s] 15

005 Titration (KF Coul) [1] Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol [µA] 5 Stir Speed [%] 45 Control End point [mV] 140.0 Control band [mV] 250.0 Rate Cautious Generator current Automatic Termination Type Drift stop relative Drift 6.0 µg/min Min. time [s] 0 Max. time [s] 3600

006 Calculation R1 Result type Predefined Result Content Result unit ppm Formula R1=(ICEQ/10.712- TIME*DRIFT)/(C*m)

Constant C= 1 Decimal places 1 Result limits No Record statistics Yes Extra statistical func. No

007 Record Summary Per sample Results No Raw results No Table of measured values No Sample data No Resource data No E-V curve No E-t curve No V-t curve No

H2O-t No Drift-t No H2O-t & Drift-t curves No V-t & drift-t No Method No Series data No

008 End of sample Open series Yes

METTLER TOLEDO Page 4 of 4 Titration Application M393-08 METTLER TOLEDO Application M394-08 KF Coulometric: Surface Water in Brown Sugar by External Extraction

The surface water of brown sugar was determined by external extraction with chloroform as a solvent, and subsequent coulometric Karl Fischer titration.

Sample Brown Sugar Preparation and Procedures - To determine the surface water of sugar, the moisture has to be first extracted into an organic solvent which is not dissolving the sample. Compound Water, H2O - In our case, chloroform has been selected. In this M = 18.01 g/mol solvent, brown sugar is (almost) insoluble. - The external extraction is performed into a flask with septum. Chemicals Chloroform (e.g. Fluka – puriss., > 99.8%) External extraction: 22 g sample in 135 g chloroform.

Extraction times: Various extraction times Titrant In situ electrochemically (see Remarks, Comments); generated iodine, I2 : - 100 mL HYDRANAL® Coulomat AD Chloroform aliquot (sample size): 1-4 g.

Standard --

Indication DM143-SC Electrode Remarks - The extracted water content increases with CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 increasing time and reaches equilibrium after → approximately 24 hrs (see graph in “Comments”). (RNH)•SO4CH3 + 2 (RNH)I - To completely extract surface water, an extraction time of 15-30 min is recommended. - A time dependence was noticed: Water content in ppm: Calculation Time/hrs Mean/ppm (n = 6) srel/% R1 = (pw(6)/(pw(6)-C)) * 0.5 56.9 9.5 *(C*msol/mext-B*msol/mext) 1.5 79.2 6.1 C: (ICEQ/10.712-TIME*DRIFT)/m 2.0 103.1 5.4 msol: Solvent weight for titrations of 24.0 286.4 2.5 type KF ext. extraction. mext: Extracted sample quantity for 72.0 300.8 3.1 titrations of type KF ext. extr. B: Blank (water cont. of the solvent) - The blank value of the solvent chloroform was determined prior to the external extraction – Waste Organic solvents Result: 7.3 ppm (n = 4). disposal - For low water content (short extraction time) the generator current (function no. 005: Control) can be set to a fix value of 200 mA to get more data MSG Anachem Group Author, points and avoid overtitration. Version Thomas Hitz, Dec 2008

METTLER TOLEDO Page 1 of 4 Titration Application M394-08 Instruments - KF Compact Coulometer C30X (X: Generating cell without diaphragm)

Accessories - XS205 Balance - LabX Titration Software

Results All results Method-ID M394 Sample Brown sugar (1/6) R1 (Content) 63.9 ppm Sample Brown sugar (2/6) R1 (Content) 55.6 ppm Sample Brown sugar (3/6) R1 (Content) 51.9 ppm Sample Brown sugar (4/6) R1 (Content) 50.7 ppm Sample Brown sugar (5/6) R1 (Content) 60.4 ppm Sample Brown sugar (6/6) R1 (Content) 59.3 ppm

Statistics Method-ID M394 R1 Content Samples 6 Mean 56.9 ppm s 5.1 srel 9.5 %

Titration curve

300

250

200

150 Signal / mV Signal 100

0 0 100 200 300 400 500 600 Consumption / mC

METTLER TOLEDO Page 2 of 4 Titration Application M394-08 Table of measured values

Consumption Increment H2O Online drift Signal Change Time mC mC μg μg/min mV mV s 0.4 NaN 0 0 274.5 NaN 0 228.6 228.2 21.3 317.5 233.5 -41 2 438.9 210.3 41 1063.7 157.1 -76.4 3 480.8 41.9 44.9 1027.3 90.7 -66.4 4 480.8 0 44.9 527.5 85.1 -5.6 5 480.8 0 44.9 283.5 82.1 -3 6 480.8 0 44.9 184.1 84.8 2.7 7 480.8 0 44.9 121.5 85 0.2 8 480.8 0 44.9 81 81.3 -3.7 9 480.8 0 44.9 63.3 86.1 4.8 10 480.8 0 44.9 28.6 85.7 -0.4 11 480.8 0 44.9 13.2 83.1 -2.6 12 480.8 0 44.9 11.2 84.7 1.6 13 480.8 0 44.9 9.6 85.2 0.5 14 480.8 0 44.9 8.3 84.1 -1.1 15 480.8 0 44.9 7.3 87.3 3.2 16 480.8 0 44.9 6.4 85 -2.3 17 480.8 0 44.9 5.7 86.9 1.9 18 480.8 0 44.9 0 85.5 -1.4 19

Comments

350

300

250

200 120 100 80 150 60 40

Water content (ppm) Water content 100 20 0 0123 50

0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Time (h)

- The surface water was determined as a function of the extraction time. It is clearly visible in the graph that the water content is increasing with increasing extraction time. A constant value is reached after approximately 25 h. - The increase can be also explained by partial dissolution of sugar in chloroform. This is supported by the turbid extraction solution.

METTLER TOLEDO Page 3 of 4 Titration Application M394-08 Method 001 Title 007 Record Type KF Coulometric Summary Per sample Compatible with C30 Results No ID M394 Raw results No Title Sugar Table of measured values No Author METTLER TOLEDO Sample data No Date/Time 24.11.2008 12:03:01 Resource data No Modified at -- E-V curve No Modified by -- E-t curve No Protect No V-t curve No SOP None H2O-t No Drift-t No 002 Sample H2O-t & Drift-t curves No Sample V-t & drift-t No Number of IDs 1 Method No ID 1 -- Series data No Entry type Weight Lower limit[g] 0.0 008 End of sample Upper limit [g] 5.0 Open series Yes Density [g/mL] 1.0 Correction factor 1.0 Temperature [°C] 25.0 Autostart Yes Entry After addition

Blank Source for blank Setup Blank Blank KF Unit ppm Entry type Weight Lower limit [g] 0.0 Upper limit [g] 5.0 Temperature [°C] 25.0 Mix time [s] 15.0 Autostart Yes Entry After addition Limits No

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift Online Max. start drift 25 µg/min

004 Mix time Duration [s] 15

005 Titration (KF Coul) [1] Sensor Type Polarized Sensor DM143-SC Unit mV Indication Voltametric Ipol [µA] 5 Stir Speed [%] 45 Control End point [mV] 100.0 Control band [mV] 250.0 Rate Cautious Generator current Automatic Termination Type Drift stop relative Drift 3.0 µg/min Min. time [s] 0 Max. time [s] 3600

006 Calculation R1 Result type Predefined Result External extraction (B in ppm) Result unit ppm Formula R1= pw(6)/(pw(6)-C)* *(C*msol/mext-B*msol/mext) Constant C= (ICEQ/10.712-TIME*DRIFT)/m Decimal places 1 Result limits No Record statistics Yes Extra statistical func. No

METTLER TOLEDO Page 4 of 4 Titration Application M394-08 METTLER TOLEDO Application M395-08 Water Content of PET Granulates (Manual KF Oven)

KF Compact Line: Water content in PET granulates is analyzed with coulometric Karl Fischer titration using gas extraction with the manual oven DO308.

Sample PET granulates Preparation and Procedures (Polyethylene terephthalate) - Fill the titration cell with the anolyte. The nitrogen gas source is connected via the drying unit to the DO308 oven. Compound Water - An insulated transfer tube is used to connect the oven with the titrator. The gas flow rate is adjusted to 120 to 140 mL/min. - The oven temperature is set to 260 °C. It must be HYDRANAL® Coulomat Chemicals heated out during at least one hour before AG-Oven starting measurement.

Carrier gas: dry nitrogen - After starting the method, the reagent gets pretitrated. As long as the status is not “standby” and the drift is higher than 25 µg/min, the Titrant None (coulometric titration) determination cannot be started. - Granulates are weighed in an aluminum insert. This is susequently placed in the glass sample boat. The sample mass is entered at the titrator. - Press “Start sample” at the titrator. During the Standard None 60 s mixing time, move the sample boat into the heated zone. The titration duration is set to a fix time of 10 min. Indication DM143-SC Remarks For different polymers and grain sizes, different CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 temperatures and titration durations are needed: → (RNH)•SO4CH3 + 2 (RNH)I − Polyethylene terephthalate (PET): 260 °C, 10 min − Polyoxymethylene (POM): 180 °C, 15 min Calculation Water content in ppm: − Polyethelene (PE): 200 °C, 20 min (ICEQ/10.712-(TIME*DRIFT) ) ------C*m − Polypropylene (PP): 180 °C, 10 min ICEQ: Current cons. to EP TIME: Total time since sample − Polystyrene (PS): request 160 °C, 15 min C: 1 (constant for ppm calc.) Waste Reagent: KF waste (organic disposal solvents) Sample: domestic waste

Author, Maria-José Schmid (MSG) Version August 2008

METTLER TOLEDO Page 1 of 4 Titration Application M395-08 Instruments - KF Compact Line Coulometer C20X/C30X (without diaphragm) - KF drying Oven DO308 (manual oven) with drying unit (silica gel, molecular sieves) - XP205 Balance

Accessories - USB-P25 Printer - Balance Aluminum inserts for the glass sample boat of DO308

Results All results Method-ID M395 Sample -- (1/6) R1 (Content) 2096.2 ppm Sample -- (2/6) R1 (Content) 1986.1 ppm Sample -- (3/6) R1 (Content) 1936.5 ppm Sample -- (4/6) R1 (Content) 1963.1 ppm Sample -- (5/6) R1 (Content) 1938.5 ppm Sample -- (6/6) R1 (Content) 1938.8 ppm

Statistics Method-ID M395 R1 Content Samples 6 Mean 1976.5 ppm s 61.7 ppm srel 3.124%

Results with POM granulates (oven temperature: 180 °C, fix titration time: 900 s)

R1 Content Samples 6 Mean 3070.1 ppm s 145.9 ppm srel 4.752%

Titration curve

METTLER TOLEDO Page 2 of 4 Titration Application M395-08 Table of measured values

Consumption Increment Signal Change Time mC mC mV mV s 0.000 NaN 485.8 NaN 1 32.770 32.770 487.4 1.6 1 120.459 87.689 484.0 -3.4 1 263.462 143.003 447.7 -36.3 2 461.780 198.318 373.5 -74.2 3 699.514 237.734 324.4 -49.1 4 932.196 232.682 293.7 -30.7 5 1161.896 229.700 242.5 -51.2 6 1381.161 219.265 207.2 -35.3 7 1548.245 167.084 177.4 -29.8 8 1684.021 135.776 199.6 22.2 9 1840.669 156.648 172.2 -27.4 10 1933.210 92.541 155.4 -16.8 11 2028.733 95.523 179.1 23.7 12 2163.018 134.285 156.9 -22.2 13 2255.558 92.540 146.1 -10.8 14 2337.663 82.105 156.6 10.5 15 2428.713 91.050 163.2 6.6 16 2519.763 91.050 142.5 -20.7 17 2613.795 94.032 171.7 29.2 18 2727.208 113.413 165.7 -6.0 19 2807.822 80.614 131.8 -33.9 20 2869.054 61.232 170.3 38.5 21 2994.394 125.340 162.6 -7.7 22 3094.389 99.995 151.6 -11.0 23 … … … … … 24431.941 7.508 98.6 -9.0 595 24433.485 1.544 103.0 4.4 596 24439.556 6.071 102.7 -0.3 597 24441.100 1.544 103.7 1.0 598 24444.189 3.089 101.9 -1.8 599 24453.241 9.052 108.9 7.0 600 24462.240 8.999 101.9 -7.0 600

Comments DO308 KF Oven: Mechanism of the sample transport with magnet

Guide rod with magnet

Slider Glass sample boat

METTLER TOLEDO Page 3 of 4 Titration Application M395-08 Method

001 Title Type Karl Fischer titration Coul. Compatible with C20/C30 ID M395 Title PET granulates (manual oven) Author METTLER TOLEDO Date/Time 25.08.2008 15:00:00 Modified at -- Modified by METTLER TOLEDO Protect No SOP None

002 Sample Number of IDs 1 ID 1 -- Entry type Weight Lower limit 0.7 g Upper limit 5.0 g Density 1.0 g/mL Correction factor 1.0 Temperature 25.0°C Autostart No Entry Arbitrary

003 Titration stand (KF stand) Type KF stand Titration stand KF stand Source for drift online Max. start drift 25 µg/min

004 Mix time Duration 60 s

005 Titration (KF Coul) [1] Sensor Type Polarized Name DM143-SC Unit mV Indication Voltametric Ipol 5.0 µA Stir Speed 45% Control End point 100.0 mV Control band 250.0 mV Rate Normal Generator current Automatic Termination Type Delay time Delay time 10 s Min. time 600 s Max. time 600 s

006 Calculation R1 Result type Predefined Result Content Result unit ppm Formula R1=(ICEQ/10.712-TIME*DRIFT)/ (C*m) Constant C=1 Decimal places 1 Result limits No Record statistics Yes Extra statistical func. No

007 Record Summary Per Sample Results No Raw results No Table of meas. values No Sample data No Resource data No E - C No E – t No C – t No

H2O – t No Drift – t No

H2O – t & Drift – t No C – t & Drift – t No Method No Series data No

008 End of sample Open series Yes

METTLER TOLEDO Page 4 of 4 Titration Application M395-08 METTLER TOLEDO Application M396-08 Automated Coulometric KF Analysis at Different Temperatures: Polystyrene

Polystyrene granules are heated at different temperatures in order to find out the optimum heating temperature for Karl Fischer titration. This is achieved by means of a KF Oven sample Changer.

Sample Polystyrene granules (PS) Preparation and Procedures approx. 2 g - Place a STROMBOLI glass vial onto the balance. - Tare the balance. - Weigh 2 g PS granules into the vial. Compound Polyphenylethene (trivial name: polystyrene, PS), - Close the vial with the aluminum foil, fix the blue rubber and place it on the sample rack. (-CH2-CH(C6H6)-)n - Start the pre-titration, and wait in standby until the drift has decreased to a constant value. Chemicals -- - When the online drift is e.g. in the order of 10 µg/min, confirm the sample determination. Note: Before each sample place an empty vial for the blank determination at the specific temperature. Titrant 100 mL HYDRANAL® Coulomat AG Oven Sample series with different samples: Always place an additional empty glass vial between the last measured sample vial and the subsequent blank vial at higher temperature when running a series with different samples at different Standard -- increasing temperatures. In fact, the subsequent temperature is first achieved in the last sample vial before moving the next one into the oven. This can lead to decomposition, combustion or explosion if Indication DM143-SC Electrode the last sample is thermally not stable. Remarks - Max. 7 PS-samples can be analyzed if for each CH OH + SO + 3 RN + I + H O Chemistry 3 2 2 2 sample a blank vial is used. This means that up → to 7 temperatures can be defined, i.e. one for (RNH)•SO4CH3 + 2 (RNH)I each sample loop. - For PS following temperatures have been selected: 60, 80, 100, 120, 140, 150 and 160°C. Note that for temperatures higher than 160°C PS Calculation Water content in ppm: is decomposing in the presence of air.

(ICEQ/10.712-TIME*DRIFT-B[Blank]) - All sample loops are terminated at fixed time, ------(C*m) e.g. t(min) = t(max). Optimum results are obtained with short mix time (e.g. 60 s), and long ICEQ: Current at EP titration time (termination at fixed time). TIME: Total time since sample request Blank: Blank value - The water contents are plotted as a function of C: 1 (constant) the heating temperature. In this way, the optimum heating temperature for the sample can Waste Organic solvents be selected. disposal - To improve accuracy and reproducibility, the glass vials can be first dried in a drying oven at 130°C over night. The heated vials are cooled to MSG AnaChem, Author, room temperature in a desiccator previous use. Version Cosimo De Caro, Sept 2008

METTLER TOLEDO Page 1 of 5 Titration Application M396-08 Instruments - KF Compact Line C30 Coulometer - STROMBOLI KF Oven Sample Changer - Drying unit with air pump (molecular sieve, silica gel). Note: above 160°C use nitrogen gas. Accessories - XS205 Balance - LabX Titration Software

Results Results No. Comment / ID Start time Rx Result Unit Name

1/1 (1/14) Blank 60°C 14.08.2008 10:45:08 R1 =335.8 µg Blank 60°C 1/1 (2/14) PS 158K 60°C 14.08.2008 10:56:52 R2 =152.9 ppm PS 158K 60°C 1/1 (3/14) Blank 80°C 14.08.2008 11:12:25 R3 =327.3 µg Blank 80°C 1/1 (4/14) PS 158K 80°C 14.08.2008 11:24:02 R4 =214.1 ppm PS 158K 80°C 1/1 (5/14) Blank 100°C 14.08.2008 11:38:29 R5 =369.2 µg Blank 100°C 1/1 (6/14) PS 158K 100°C 14.08.2008 11:50:05 R6 =242.2 ppm PS 158K 100°C 1/1 (7/14) Blank 120°C 14.08.2008 12:05:59 R7 =400.5 µg Blank 120°C 1/1 (8/14) PS 158K 120°C 14.08.2008 12:17:36 R8 =245.5 ppm PS 158K 120°C 1/1 (9/14) Blank 140°C 14.08.2008 12:31:24 R9 =416.6 µg Blank 140°C 1/1 (10/14) PS 158K 140°C 14.08.2008 12:43:01 R10 =292.4 ppm PS 158K 140°C 1/1 (11/14) Blank 150°C 14.08.2008 12:55:48 R11 =434.0 µg Blank 150°C 1/1 (12/14) PS 158K 150°C 14.08.2008 13:07:26 R12 =292.0 ppm PS 158K 150°C 1/1 (13/14) Blank 160°C 14.08.2008 13:20:26 R13 =447.5 µg Blank 160°C 1/1 (14/14) PS 158K 160°C 14.08.2008 13:32:04 R14 =307.3 ppm PS 158K 160°C

Titration curve

METTLER TOLEDO Page 2 of 5 Titration Application M396-08 Table of measured values

Consumpn Increment Signal Change Time Temperature mC mC mV mV s °C 0 NaN 429.9 NaN 0 32.77 32.77 430.4 0.5 1 120.459 87.689 432.6 2.2 2 263.462 143.003 428.9 -3.7 2 461.78 198.318 430.5 1.6 3 715.412 253.632 429.1 -1.4 4 1024.359 308.947 423.6 -5.5 6 1388.621 364.262 420.2 -3.4 7 1808.197 419.576 407.8 -12.4 9 2283.088 474.891 372.4 -35.4 10 2659.108 376.02 287.7 -84.7 12 2887.318 228.21 254.2 -33.5 14 3100.619 213.301 194.3 -59.9 16 3245.34 144.721 180.8 -13.5 17 3345.335 99.995 147 -33.8 18 3385.695 40.36 132.7 -14.3 19 3420.093 34.398 135.2 2.5 20 3466.416 46.323 140.6 5.4 21 3532.122 65.706 137.2 -3.4 22 3591.863 59.741 138.4 1.2 23

...... … ..… ..… ..…

11641.859 1.598 101.3 0.3 593 11644.948 3.089 105.2 3.9 594 11654 9.052 104.7 -0.5 595 11660.071 6.071 101.8 -2.9 596 11660.124 0.053 103.4 1.6 598 11664.704 4.58 103.9 0.5 599 11669.284 4.58 104.8 0.9 600 11673.81 4.526 102 -2.8 600

Comments

Polystyrene

750.0 700.0 650.0 600.0 550.0 500.0 450.0 400.0 350.0 300.0

Water content / ppm content Water 250.0 200.0 150.0 100.0 50.0 0.0 40 60 80 100 120 140 160 180 200 220 Temperature / °C

Termination: fixed time, tmin = tma x= 600 s

METTLER TOLEDO Page 3 of 5 Titration Application M396-08 - The results have been plotted as a function of the heating temperature in the previous diagram. Further, more results from several sample series have been added into it. The diagram shows that

the water content remains approximately the same for a temperature of 160°C (tmin = tmax = 600 s). - At higher temperatures, side reactions occur which lead to a higher water content. These reactions are due to decomposition of polystyrene at high temperatures in the presence of air. Therefore, if the heating temperature of PS samples is increased above 160°C, the use of nitrogen gas is mandatory. - Method M396 consists of 14 sample loops. 7 different temperatures can be tested, since for each temperature one blank vial and one sample vial has to be titrated. In fact, the blank value is dependent on temperature, and therefore it must be determined at each temperature:

Blank value of glass vials for STROMBOLI KF Oven Sample Changer

700.0 600.0 500.0 400.0 g water μ 300.0 200.0

Blank / 100.0 0.0 0 50 100 150 200 250 300 350 Temperature / °C

Termination: fixed time, tmin = tma x= 600 s

- With the KF oven it is recommended to heat the sample using a fixed titration time parameter tmin = tmax, i.e. water is evaporated and transferred in the cell during this time, and immediately titrated until the analysis is stopped at a fixed time tmin = tmax defined in the method. In this way, evaporation of water is avoided which lead to lower water content. - The heating (or extraction) time is also a relevant parameter when working with the KF drying oven. As for temperature, the optimum titration time has to be selected to get accurate and precise results. In this case, the temperature is maintained constant throughout all method, while sample loops have

different fixed titration time, e.g. tmin = tmax = 300 s, 600 s, 900 s, and 1200 s:

15000 341.7 ppm 335.9 ppm

12500

216.7 ppm 10000

7500 Polystyrene 80°C, t=10 min Polystyrene 160°C, t=10 min

Current / mC Current Polystyrene 160°C, t=15 min 5000

2500

0 0 100 200 300 400 500 600 700 800 900 1000 Time / s

- In the above diagram it can be see that approx. the same water content is obtained with two different heating times of 10 and 15 minutes. The water conent doesn’t change with longer times. Thus, to speed up the analysis the shorter time has been selected, i.e. 10 minutes.

METTLER TOLEDO Page 4 of 5 Titration Application M396-08 Method 001 Title 011 Sample Type Stromboli Number of IDs 1 Compatible with C30 ID 1 PS 158K 60°C ID M396 Entry type Weight Title Temperature ramp (Stromboli) Lower limit 0.0 g Author METTLER TOLEDO Upper limit 5.0 g Date/Time 25.08.2008 15:00:00 Density 1.0 g/mL Modified at -- Correction factor 1.0 Modified by ___ Temperature 25.0°C Protect No Entry Arbitrary SOP None 012 Titration stand (Stromboli TTL) 002 Drift determination Type Stromboli TTL Wait time [s] 60 Titration stand Stromboli TTL 1 Duration [min] 3 Oven temperature 60°C Source for drift Determination 003 Sample Max. Start drift 25 µg/min Number of IDs 1 ID 1 Blank 180°C 013 Mix time Entry type Fixed pieces Duration 60 s Pieces 1 pcs. Weight per piece 1.0 g 014 Titration (KFCoul) [2] Correction factor 1.0 Sensor Temperature 25.0°C Type Polarized Start analysis Manual Sensor DM143-SC Entry Arbitrary Unit mV Indication Voltametric 004 Titration stand (Stromboli TTL) Ipol 5.0 µA Type Stromboli TTL Stir Titration stand Stromboli TTL 1 Speed 45 % Oven temperature 60°C Control Source for drift Determination End point 100 mV Max. Start drift 25 µg/min Control band 250 mV Rate Normal 005 Mix time Generator current Automatic Duration 60 s Termination Type Delay time 006 Titration (KFCoul) [1] Delay time 10 s Sensor Min. time 600 s Type Polarized Max. time 600 s Sensor DM143-SC Unit mV 015 Calculation R2 Indication Voltametric Result PS 158K 60°C Ipol 5.0 µA Result unit ppm Stir Formula R2=(ICEQ[2]/10.712 - Speed 45 % TIME[2]*DRIFT-B[Blank Control Stromboli])/(C*m) End point 100 mV Constant C= 1 Control band 250 mV Decimal places 1 Rate Normal Result limits No Generator current Automatic Record statistics No Termination Extra statistical func. No Type Delay time Delay time 10 s 016 Record Min. time 600 s Summary Yes Max. time 600 s Results Yes Raw results No 007 Calculation R1 Resource data No Result Stromboli blank value 60°C Method No Result unit µg Series data No Formula R1=(ICEQ[1]/10.712 - TIME[1]*DRIFT)/C 017 End of sample Constant C= 1 ... Decimal places 1 The first two loops are subsequently repeated 6 times again Result limits No with the same parameters but 6 different temperatures. Record statistics No This would lead to a total number of 94 method functions. Extra statistical func. No To improve visibility in this text, the method structure 008 Record is summarized as it follows: Summary Yes Results No Blank (loop3) + Sample (loop4) 80°C Raw results No Blank (loop5) + Sample (loop6) 100°C Resource data No Blank (loop7) + Sample (loop8) 120°C Method No Blank (loop9) + Sample (loop10) 140°C Series data No Blank (loop11) + Sample (loop12) 150°C Blank (loop13) + Sample (loop14) 160°C 009 End of sample ... 107 End of sample 010 Blank Name Blank Stromboli Note: Value B= Mean[R1] There is no stand-by at the end of the series, i.e. the stand- Unit µg by method function is not defined: Limits No - The titrator and STROMBOLI are stopped - Stromboli stops heating, and the air pump is turned off. - The sample rack turns back to the drift position. - The drift beaker is kept out of the oven.

This is done for security reason.

METTLER TOLEDO Page 5 of 5 Titration Application M396-08 1 Karl Fischer Titration: The Method at a Glance

1.1 Solid samples

1.1.1 Organic Chemicals

Substance Examples Method Substance Examples Method

Acetals R1-CH-(OR2)2 V2, C20 Halogenated R-Cl, R-Br, R-I V2,V4,V5 hydrocarbons C20,C22 Aldehydes Chlorobenzaldehyde, V11 Isocyanates R-NCO V2,V4,V8 nitrobenzaldehyde C0 C20

Mono-/poly- > C12 Stearyl alcohol, V2,V3, Ketones Benzophenone V10, C24 alcohols diphenylmethanol V4 C20

Amines Weakly basic pKa >8: V2 Hydrocarbons > C12 V4,V5 Imidazole, indole, C20 Biphenyl, C20,C22 carbazole anthracene,pyren e, naphthalene

Strongly basic: V16 > C20 V3,V7 aminopyridine C0 Tar, bitumen Reaction with V10 Nitro- di-Nitrobenzene, V2,V3 methanol: C0 compounds nitrotoluene, C20 naphthylamine, nitrochloro- anisidine, toluidine benzene Oxidizable with V0, C0 Peroxides Dialkylperoxide V2,V3,V4 iodine: R-OO-R C20 aminophenol Alkylperoxide R-OOH

Carbamates RO-CONH2 V2, V3, Diacyl peroxide V0, C0 V4

Carboxylic R-CONR´, e.g.: V2, V3, Phenols Low pKa value: V2 acid amides benzamide, V4 phenol, cresol, C20 stearamide C20 salicylic acid

Mono-/poly- Weakly acidic: V2, V3, High pKa value: V15 carboxylic benzoic acid V4 2-chlorophenol, acids C20 o-cresol Strongly acidic: V15 Side reaction: V0, C0 malonic acid, oxalic C20 naphthol, acid aminophenol Oxidizable with V0, C0 Sulfur Aromatic/aliphatic V2,V4,V5 iodine: compounds sulfonic acids C20 ascorbic acid sulfides R-S-R´ disulfides R-SS-R´ thiocyanates R-SCN Ethers Linear and cyclic V2,V4, Mercaptans V0, C0 V5 C20 Esters Carboxylic acid ester V2,V3, R-COOR´ V4 Carbonic anhydride C20 ROCOOR´

METTLER TOLEDO Page 1 of 80 Karl Fischer Titration Applications 1.1.2 Inorganic chemicals

Substance Examples Method Substance Examples Method

Arsenic Na2HAsO3 , V30 Phosphates NaH2PO4 , V2, V31 compounds C30 C20, C32 NaAsO2 Na2HPO4

Bicarbonates KHCO3 , NaHCO3 V34 tert. Phosphate, V15, V31 carbonates C32 e.g.: Na PO C32 Na2CO3 3 4

Boron B2O3 , HBO3 , V31 Silicon Silicon oxide, V2, V32 compounds C32 compounds Silicon dioxide C20, C34 H3BO3 Hydroxides, NaOH, KOH V21, V32 Silanols V10 oxides MgO, CaO C21, R3Si(OH) C34

Halogenides NaCl, KJ, CaCl2 , V2, V32 Sulfates Na2SO4, ZnSO4 V2, V32 C20, C21, C34 MgCl2 C34

Cu(I) salts: CuCl2 V31 Thiosulfates Na2S2O3, NaSO3 V2, V31 C33 C20, C32 Na2S2O5

Nitrates NaNO3 , NH4NO3 V2, V31 Tin(II)salts SnCl2 V31 C20, C32 C32

Nitrites NaNO2 V31 C32

1.1.3 Technical products – organic

Substance Examples Method Substance Examples Method

Agro- Insecticides, V2 Petro- Waxes, paraffins: V12 chemicals fungicides, chemicals shoepolish, C0 herbicides ski wax Dyes Soluble: wool dyes, V2 Tar, bitumen V7 indicator dyes, etc. Insoluble: V2, V8 lubricating grease, V5, V30 pigments, multipurpose dispersion dyes grease Cosmetics Creams, lotion V4, V5 Pharma- Antibiotics, V2 ceuticals disinfectants Lipstick V5 Salves, creams V4, V5 Soaps V2 Suppositories V12 Toothpaste V20 Tablets V2 Adhesives All-purpose glue, V2 Lyophilized V40,C40 glue stick preparations Rubber cement V5 Textile Surfactants powder V2 auxiliaries optical brightener Plastics Polyethylene, PVC, V30, V31 Polymers polypropylene, C31, C32 poly-amides, polystyrene, polyurethane, etc.

METTLER TOLEDO Page 2 of 80 Karl Fischer Titration Applications 1.1.4 Technical products - inorganic

Substance Examples Method Substance Examples Method

Building Mortar, plaster, V32 Minerals Zeolites V32 materials cement (C34) Fertilizers Adherent moisture V6 Detergents Laundry detergent V30, V21 with brightener (hydroxides, perborates) Total water V30 without brightener V2

1.1.5 Technical natural products

Substance Examples Method

Cellulose Paper, wood V30, C30 products Cellulose powder V2 Fibrous Wool, silk V20 materials

1.1.6 Food

Substance Examples Method Substance Examples Method

Protein Cheese, meat V23 Grains and Biscuits, Zwieback, V13, products spread, broth starchy noodles, wheat, (V30) foods rye, corn, potato chips Yogurt, ice cream V2 Potato flakes V22, V23 Dried albumen V13 Dough, noodles, V30, Zwieback (V25) Powdered milk V2, V23 Vegetable Cacao, coffee V22, V23 products beans, tea, instant coffee, tobacco, dried vegetables and fruits Fats Butter, margerine, V4, V5 Hazelnuts,almonds V13 mayonnaise instant coffee Shortening, V5 Sugar and Sugar: total water V13 hardened fat sugar cont., jellied fruits, products gummi bears, candy, caramel, pudding powder, almond paste Spices Pepper, curry, V23 Instant ice tea, V8 mixed spice jam Mustard V24, Honey, V2 (V5) jelly Adherent moisture: V6 Sugar: V6 plain salt, sea salt C23 adherent moisture Total water: V32 Chocolate V5 plain salt, sea salt C34

METTLER TOLEDO Page 3 of 80 Karl Fischer Titration Applications 1.2 Liquid samples

1.2.1 Organic and inorganic chemicals

Substance Examples Method Substance Examples Method

Acetals Acetal, ethylal, V1 Hydrocarbons C1 to C7 V1 methylal C1 hexane, C1 pentane, isobutane, cyclohexene, cyclohexane, benzene, toluene, xylene

Aldehyde Acetaldehyde, V11 saturated and C7 to C14 V4 benzaldehyde C0 unsaturated gasoline, C1 petroleum, dodecane

Alcohols Propanol, hexanol, V1 > C14 V5 benzyl alcohol C1 heavy oil, crude C30 oil, paraffin

Amines Weakly basic: V1 Mineral acids H2SO4, V15 N-methyl aniline C1 C0 HNO3, heterocyclic amines HCl pyridine, chinoline Strongly basic: V16 Nitro- Nitrobenzene, V1 alihatic amines C0 compounds nitrotoluene, C1 N-butylamine, nitrochloro- hexylamine benzene Reaction with V10 Nitriles Acetonitrile V1 methanol: C0 C1 aniline, toluidine, diamine

Easily oxidized V0 Peroxides H2O2 V15 with iodine: C0 hydrazine, hydroxylamine

Carboxylic C1 to C2 V15 Acid amides Dimethyl- V1, C1 acids formic acid, acetic C0 formamide acid

> C2 V1 Sulfur- Sulfides, V1 propionic acid, C1 compounds disulfides C1 acrylic acid sulfonic acids Ethers Dimethylether, V1 Mercaptans V0, C0 dioxane, anisole C1 Esters Methyl benzoate, V1 ethyl acetate C1 Halogenated Methyl chloride, V1 hydrocarbons t-butyl chloride, C1 chlorobenzene, chlorotoluene Ketones Acetone, V10 acetophenone, C10 methylethylketone

METTLER TOLEDO Page 4 of 80 Karl Fischer Titration Applications 1.2.2 Foods and technical products

Substance Examples Method

Aqueous Foods: V1 solutions spiced sauce: soy sauce beverages: beer, wine, liquor Pharmaceuticals: V1 extrakts, tinctures Cosmetics: V1 perfume, shampoo Tensides and V1 detergents Aqueous Milk products: V4 emulsions milk, cream, condensed milk Tensides and V1 detergents Agrochemicals V1 (sprays): herbicides, fungicides Acrylic enamels V29 Emulsions Agrochemicals V4 in solvents (sprays): C1 insecticides Synthetic enamels V7 Vegetable Foods: V4 oils salad oil, sunflower C1 oil Pharmaceuticals V4 and cosmetics: C1 ethereal oils, massage oil Technical Hydraulic oil, brake V5 oils fluid, transformer C1 oil, silicon oil Motor oils V30 C30

For more details on the sample preparation, sample input and titration methods, refer to chapter 11.3.

METTLER TOLEDO Page 5 of 80 Karl Fischer Titration Applications 1.3 Titration Methods

1.3.1 Volumetric methods The methods have been developed using 2-component reagents. These can be also performed using the one-component reagent. In this case methanol is used as a solvent.

Method Procedure V1 Direct titration

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 40 mL KF solvent Stir time: 10 s V2 Direct titration

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 40 mL KF solvent Stir time: 60 - 120 s V3 Direct titration with added 2-propanol

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 40 mL KF solvent / propanol 1:1 Stir time: 60 - 120 s V4 Direct titration with added 1-decanol

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 40 mL KF solvent / 1-decanol 1:1 Stir time: 60 - 120 s V5 Direct titration with added chloroform

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: KF solvent / chloroform 1:1 to 1:2 Stir time: 60 - 120 s V6 Direct titration with added chloroform

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 40 mL KF solvent / chloroform 1:5 Stir time: 60 - 120 s Delay time: 7 seconds. V7 Direct titration with added toluene

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 40 mL KF solvent / toluene 1:1 Stir time: 60 - 120 s V8 Direct titration with added formamide

Titrant: two component reagent 5 mg H2O/mL Solvent: 40 mL KF solvent / formamide 1:1 Stir time: 5 - 10 minutes V10 Direct titration methanol free

Titrant: two comp. reagent 5 mg H2O/mL for ketone und aldehyde Solvent: 40 mL KF solvent for ketone und aldehyde Stir time: 30 – 60 s V11 Direct titration methanol free

Titrant: two comp. reagent 5 mg H2O/mL for ketone und aldehyde Solvent: 40 mL KF solvent for ketone und aldehyde Start immediately (autostart) V12 Direct titration with heat (at 50°C)

Titrant: two component reagent 5 mg H2O/mL Solvent: 40 mL KF solvent Stir time: 5 - 10 minutes

METTLER TOLEDO Page 6 of 80 Karl Fischer Titration Applications V13 Direct titration with added formamide and heat, 50°C

Titrant: two component reagent 5 mg H2O/mL Solvent: 40 mL KF solvent / formamide 1:1 Stir time: 10 - 15 minutes V15 Direct titration with neutralization

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent and 20 mL buffer or 7g imidazole Stir time: 60 - 120 s V16 Direct titration with neutralization

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent and 5 g benzoic acid Stir time: 1-2 minutes V20 External extraction with methanol at room temperature

Titrant: two component reagent 5 mgH2O/mL Solvent: 30 mL KF solvent V21 External extraction with methanol at room temperature in sonicator Remove aliquots of extraction solution over membrane filter with syringe.

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent V22 External extraction with methanol at 50 °C

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent V23 External extraction with formamide

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent V24 External extraction with 1-decanol/formamide 1:1

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent V25 External extraction with 1-decanol/formamide/methanol 8:2:1

Titrant: two component reagent 5 mg H2O/mL Solvent: 30 mL KF solvent V29 External extraction with formamide

Titrant: two component reagent 5 mg H2O/mL Solvent: 40 mL KF solvent / formamide 1:1 V30 With drying oven at 110 - 150°C

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 50 mL KF solvent Stir time: 10 - 20 min V31 With drying oven at 160 - 220°C

Titrant: two component reagent 5 or 2mg H2O/mL Solvent: 50 mL KF solvent Stir time: 10 - 20 min V32 With drying oven at 300°C

Titrant: two component reagent 5 or 2 mg H2O/mL Solvent: 50 mL KF solvent Stir time: 10 - 20 min V40 Special method for lyophilized preparations Dissolve sample in septum flask with 10-20 mL titrated solvent and inject entire solution into titration cell and titrate.

Titrant: two component reagent 2 mg H2O/mL Solvent: 50 mL KF solvent V0 Side reaction: KF titration not possible

METTLER TOLEDO Page 7 of 80 Karl Fischer Titration Applications 1.3.2 Coulometric methods

Method Procedure C1 Direct titration Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent C5 Direct titration with added chloroform Anolyte: 70 mL anode reagent and 30 mL chloroform Catholyte: 5 mL cathode reagent C10 Direct titration methanol free Anolyte: 100 mL anode reagent for ketones Catholyte: 5 mL cathode reagent for ketones C20 External extraction with methanol or external dissolution in methanol Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent C21 External extraction with Methanol in sonicator Remove aliquots of extraction solution over membrane filter with syringe. Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent C22 External extraction with 1-decanol or external dissolution in 1-decanol Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent C23 External extraction with chloroform or external dissolution in chloroform Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent C24 External extraction or external dissolution in chloroform (methanol free) Anolyte: 100 mL anode reagent for ketones Catholyte: 5 mL cathode reagent for ketones C30 With drying oven at 110-130°C Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent Titration time: 10-20 minutes C31 With drying oven at 140-150°C Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent Titration time: 10-20 minutes C32 With drying oven at 160-190°C Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent Titration time: 10-20 minutes C33 With drying oven at 200-250°C Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent Titration time: 10-20 minutes C34 With drying oven at 300°C Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent Titration time: 10-20 minutes C40 Special method for lyophilized preparations Dissolve sample in septum flask with 10-20 mL titrated anolyte and inject entire solution into titration cell and titrate. Anolyte: 100 mL anode reagent Catholyte: 5 mL cathode reagent C0 Coulometric determination not possible

METTLER TOLEDO Page 8 of 80 Karl Fischer Titration Applications 1.4 Sample preparation and input 1.4.1 Solids: Characteristics Property Examples Method Property Examples Method Very hard Minerals, rocks S1 Soft, Butter, margarine S7, S10

e.g.: CaCO3 , SiO2 greasy Hard, Salts, cristalline S1, S2 Soft, Phenols, napthalene, S2, S3 brittle products brittle potato flakes Hard Wheat, noodles, S1, S3 Soft, Tar, bitumen S2, S5 natural pepper, almonds, S10, S11 resinous S10 products coffee, Zwieback Fibrous Dried vegetables and S4, S10 Waxes Paraffins, shoepolish S8 natural fruits, tobacco, tea, S11 products meat Tenacious Jellied fruits, gummi S4, S10 Creams, Salves, creams gel, S9 products bears, paper, wool, silk S11 pastes honey, jelly, ice cream, yogurt Soft Cheese, almond paste, S10, S11 Pulverized Salts, active ingredients S12 products broth soluble Na-tartrate Hard, Hardened fat, chocolate S6, S8 greasy S10

1.4.2 Solids: Sample input Method Sample input Procedure S1 finely ground sample Grind in sealed, cooled analysis mill. with weighing boat S2 finely ground sample Pulverize with mortar. with weighing boat S3 finely ground sample Pulverize with mixer. with weighing boat S4 with spatula Press through small holes e.g. meat grinder.

S5 with spatula or weighing boat Cut in small pieces with scissors or knife.

S6 with spatula or weighing boat Use grate to reduce size.

S7 with spatula Homogenize sample well. After sample removal keep sample Do not use syringe, the moisture sealed. will be pressed out S8 with pre-heated syringe Melt sample. (with needle Ø 0.8 mm ) (Pre-heat syringe with hair dryer or in drying oven.) S9 Fill in backend of syringe Homogenize sample well. After sample removal keep sample sealed. Use syringe with needle Ø 1.5 mm or without needle; for very thick pastes drill hole in bottom of syringe. S10 syringe with needle Ø 0.8 mm In external extraction solvent (methanol, 1-decanol, formamide). Dissolve sample or extract water. Stir, shake or sonicate. At room temp. or up to approx. 60° C. Let solids settle and remove aliquot of remaining solution with syringe. Inject into titration cell. S11 syringe without needle (Fill very Break and disperse with high-speed mixer (up to 24000 cpm) in thick samples in back of syringe) external extraction solvent (methanol, 1-decanol, formamide). S12 weighing boat No further preparation necessary. (Select stir time so that sample dissolves completely.)

METTLER TOLEDO Page 9 of 80 Karl Fischer Titration Applications 1.4.3 Liquids: Characteristics Property Examples Method Liquid Propanol, acetone, gasoline, acetic acid L1 Turbid Silicon oil, crude oil, glycerine L2 Low boiling point 5 - 15°C Acetaldehyde L3

Hygroscopic Methanol, H2SO4, glycerine L4 Low moisture content (< 1000 ppm) Hexane, toluene, petroleum, acetone, salad oil L4 Non-hygroscopic, Perfume, after-shave L1 average moisture content High moisture content (> 50%) Beverages, detergents, milk, aqueous emulsions L5 Inhomogeneous Acrylic enamels, synthetic enamels, cream L6

1.4.4 Liquids: Sample input Method Sample input Procedure L1 Injection with syringe Fluid liquid sample (with needle Ø 0.8 mm) Inject sample into titration cell through septum or needle hole of three-hole adapter with 1 mL or 10 mL syringe. L2 Injection with syringe Turbid liquid sample (without or with needle Ø Inject sample into titration cell through septum or needle hole of 1.5 mm) three-hole adapter with 1 mL or 10 mL syringe. L3 Injection with syringe after Boiling point 5 - 15 °C cooling sample - Cool sample to approx. 0°C in ice bath. (with needle Ø 0.8 mm) - Inject sample into titration cell through septum or needle hole of three-hole adapter with 1 mL or 10 mL syringe. L4 Injection with syringe Hygroscopic sample or low moisture content (< 1000 ppm) - Keep sample in a septum flask. - Inject sample into titration cell through septum or needle hole of three-hole adapter with 10 mL syringe (with needle Ø 0.8 mm). - Rinse syringe with sample 2-3 times (pull up and discard) and condition for 5 minutes. - The pressure loss during sample removal must be compensated with dried air. L5 Injection with syringe Samples with high moisture content (> 50%) - Rinse syringe with sample 2-3 times. - For each titration use 1 mL syringe (with needle Ø 0.8 mm) to remove fresh sample from sample flask and inject this into titration cell through septum or needle hole of three-hole adapter. L6 Injection with syringe after Inhomogeneous emulsions sample homogenization - Homogenize thoroughly by shaking or stirring prior to sample (with needle Ø 0.8 mm) removal. - Remove sample immediately. - Inject sample into titration cell through septum or needle hole of three-hole adapter with 1 mL or 10 mL syringe.

1.4.5 Accessories for sample input Syringe 1 mL ME-71492 10 mL ME-71482 Injection needle 1.2 mm ME-71483 0.8 mm ME-71484

Weighing boat, glass ME-23951 Septum NS24 ME-23950 Visco-Spoon™ ME-51107668 Three-hole adapter ME-23982

METTLER TOLEDO Page 10 of 80 Karl Fischer Titration Applications 2 Food

2.1 Applications: Sugar and Sugar Products Water is released very slowly from undissolved sugar products. Thus, the sample needs to be dissolved completely for the determination of the total water content. This may be achieved by adding formamide to the KF solvents and by heating. Caution: Certain carbohydrates decompose slowly in a process that generates water at temperatures greater than 50°C. Note that the use of more than 50% formamide is not recommended, as the stoichiometry of the Karl Fischer reaction will be altered. On the other hand, the determination adherent moisture of saccharose is generally more interesting than the total water content, as the adherent moisture strongly influences the pourability properties of sugar. In fact, sugar can crystallize to a hard block in storage silos if the surface moisture content is too high. Thus, sugar can not freely flow anymore out of the storage silos. The sample must not be dissolved and the entrapped water should not diffuse out for the determination of the adherent moisture. This may be achieved using a mixture of chloroform/methanol as the solvent, and by rapidly performing a direct titration. As an alternative, an external extraction in e.g. pure chloroform can also be performed. Sugar products containing chocolate require the addition of chloroform to dissolve edible fats. Candy, caramels or toffees, almond paste (marzipan) and gum may necessitate the addition of formamide and/or titration at elevated temperatures. Extensive information on the water determination in these products can be found in: - Zürcher, K. and Hadorn, H., „Störungen bei der Wasserbestimmung nach Karl Fischer“, Mitt. Gebiete Lebensm. Hyg. 72, 177-182 (1981) (in German). - Scholz, E., „Wasserbestimmung in Lebensmitteln“, Deut. Lebensmittel Rundschau 79, 302-306 (1983) (in German). - Schneider, F., Emmerich, A. and Ticmanis., U., „Die Bestimmung der Oberflächenfeuchte von Zucker“, Zucker 28, 349-536 (1975) (in German) - “ICUMSA Methods Book 2007”, International Commission for Uniform Methods of Sugar Analysis (ICUMSA), 2007, see www.icumsa.org . - Kreiser, W.R. and Martin, R.A., “Comparison of Accuracy, Precision and Speed of Three Methods for Determining Moisture in Milk Chocolate”, J. Assoc. Off. Anal. Chem. 60, 303-306 (1977).

2.1.1 Coulometric determinations in sugar products

Sample m n H2O srel Method [g] [ppm] [%]

Sucrose 0.2 6 71.5 4.2 - Sample preparation: External extraction (surface) 10 g in 55 g chloroform, 15 min at 25°C - Sample addition: 3 mL syringe without needle - KF Titration: Cell with diaphragm Stirring time: 0 s, Online drift determination, Autostart, Generation rate: normal, Termination: rel. drift stop Sucrose 0.1 7 533.7 4.7 - Sample preparation: External extraction (total ) 6 g in 60 g formamide, 15 min at 50°C - Sample addition: 1 mL syringe with needle - KF Titration: Cell with diaphragm Stirring time: 0 s, Online drift determination, Autostart, Generation rate: normal, Termination: rel. drift stop

METTLER TOLEDO Page 11 of 80 Karl Fischer Titration Applications Comments - All of the above mentioned applications can also be performed with a generation cell without diaphragm. - Sucrose For the coulometric KF titration, the water content of solids must be determined using external extraction, external dissolution or with the drying oven. Since saccharose is not thermally stable, the drying oven cannot be used. By a short external extraction with chloroform, practically only the surface adherent moisture will be affected. Water diffusion can be neglected. At 50°C saccharose can be dissolved completely in formamide. Using this method, the total water content can be determined.

2.1.2 Volumetric determinations in sugar products

Sample m n H2O srel Method [g] [%] [%]

Cane sugar 4.0 6 0.0503 5.0 - Sample preparation: -- / Direct titration (adherent moisture) - Sample addition: Glass weighing boat - KF Titration: Titrant: Composite 1 Solvent: 10 mL methanol/40 mL chloroform Stirring time: 0 s, Autostart Termination: Delay time 7 s Cane sugar 2.0 6 0.155 5.8 - Sample preparation: -- / Direct titration at 50°C (total ) - Sample addition: Glass weighing boat - KF Titration: Titrant: Titrant 5 (two-component reagent) Solvent: 20 mL KF Solvent/20 mL formamide Stirring time: 900 s Termination: Delay time 15 s French honey 0.1 6 16.2 0.43 - Sample preparation: Homogenize well the sample with a spatula - Sample addition: Fill syringe from behind Syringe without needle - KF Titration: Titrant: Titrant 5 (two-component reagent) Solvent: 30-40 mL KF Solvent Stirring time: 240 s Termination: Delay time 15 s Raspberry jelly 0.03 7 35.4 0.28 - Sample preparation: Homogenize well the sample with a spatula - Sample addition: Fill syringe from behind Syringe without needle - KF Titration: Titrant: Titrant 5 (two-component reagent) Solvent: 30-40 mL KF Solvent Stirring time: 300 s Termination: Delay time 15 s Four-fruit jam 0.07 8 35.5 0.25 - Sample preparation: Homogenize well the sample with a spatula - Sample addition: Fill syringe from behind Syringe without needle - KF Titration: Titrant: Titrant 5 (two-component reagent) Solvent: 30 mL KF Solvent/20 mL formamide Stirring time: 240 s Termination: Delay time 15 s

METTLER TOLEDO Page 12 of 80 Karl Fischer Titration Applications Milk chocolate 0.3 6 1.44 3.4 - Sample preparation: Finely grate sample - Sample addition: Weighing paper - KF Titration: Titrant: Titrant 5 (two-component reagent) Solvent: 20 mL KF Solvent/10 mL chloroform Stirring time: 600 s Termination: Delay time 15 s Milk flavoring 0.05 5 10.1% 2.2 - Sample preparation: Finely grate sample product - Sample addition: Weighing paper Ovomaltine® ® - KF Titration: (Ovaltine ) Titrant: Titrant 5 (two-component reagent) Solvent: 20 mL KF Solvent/20 mL formamide Stirring time: 400 s Termination: Delay time 15 s

Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol as a solvent. - The termination criteria can be also set to relative drift stop. - Cane sugar (adherent moisture): Even with 80% chloroform in the solvent, the partial solubility of the sugar and the diffusion of the water cannot be prevented completely (subsequent consumption). Thus, the titration must be performed rapidly. The delay time is reduced to 7 seconds. - Cane sugar (total water): To ensure that the sugar dissolves completely, the solvent must contain 50% formamide and the titration must be performed at 50°C seconds. Replace the solvent after 3 samples as the dissolving capacity of the solvent will be so low that the 15 minute stir time will not suffice for solubilization of the sugar. Increasing the quantity of formamide in the solvent to enhance the solubility is not recommended as the stoichiometry of the Karl Fischer reaction will be altered. - Honey: Honey dissolves in the solvent, thus formamide need not be added to the solvent. With the indicated solvent amount, 6 determinations have been performed successively. - Raspberry jelly: To ensure that the sugar dissolves completely, the solvent must contain 50% formamide and the titration must be performed at 50°C seconds. Replace the solvent after 3 samples as the dissolving capacity of the solvent will be so low that the 15 minute stir time will not suffice for solubilization of the sugar. Jelly dissolves in the KF Solvent provided that the stir time is sufficient. Using the indicated amount of solvent, 7 determinations have been performed successively. - Four-fruit jam: The jam could be dissolved to completion only by adding formamide and by stirring longer. The time required for one titration was 6-8 minutes. - Milk chocolate: The chocolate could be dissolved only by adding chloroform. The direct titration has a sluggish endpoint and a high subsequent consumption. Titration time: 10-15 minutes. - Milk flavoring product: Ovomaltine® can be dissolved in the presence of formamide. After 3 samples renew the solvent as its dissolving capacity will be exhausted. Direct titration of Ovomaltine shows a sluggish endpoint and a high subsequent consumption. Titration time: 10-15 minutes.

METTLER TOLEDO Page 13 of 80 Karl Fischer Titration Applications 2.2 Applications: Fats, Oils and Dairy Products Water content determinations of fats and oils are unproblematic once the appropriate solvent has been found. Long alkyl chain alcohols -e.g. 1-decanol, 1-octanol- and chloroform have proven to be very useful as auxiliary solvents. The producers of KF reagents have introduced special solvent mixtures which improve the dissolution of fats and oils. For environmental reasons, some of these mixtures are free of halogenated and aromatic solvents. Milk and milk powders can be titrated directly in methanol. Milk powder requires a longer stir time (5-10 minutes), and for high-fat milk powder the addition of chloroform is recommended. Fats (such as butter and margarine) and milk products (i.e., yogurt) generally do not have homogeneous water distributions. To ensure precise measurements, mechanical homogenization of the samples is a must prior to titration. Larger sample sizes also improve the repeatability, however, necessitate an external dissolution so that the amount of water to be titrated can be aliquoted to reasonable sample sizes. When titrating certain vegetable oils, a slightly increased subsequent consumption that may be caused by a side reaction has been observed. Should this be the case, a reduction of the delay time to 5-10 seconds is recommended. The titration of cheese is difficult as the water is distributed unevenly, the sample is barely soluble and it is difficult to extract the water. Scholz recommends titration of the finely grated sample at 50°C in a solvent mixture (methanol/ formamide, see below). External extraction of the sample with propanol in the mixer is another possibility.

Extensive information on the water determination in these products can be found in: - International Standard IDF 23: 1964 (International Dairy Federation, www.fil-idf.org), „Wasserbestimmung in Butteröl mit der Karl Fischer Methode“, Milchwirtschaft 21, 137-139 (1966) (in German).

See also: ISO 5538 / IDF 023:2002, „Determination of Water Content of Butter Oil by Karl Fischer Titration”, - Scholz, E., „Wasserbestimmung in Lebensmitteln“, Deut. Lebensmittel Rundschau 79, 302-306 (1983) (in German). - Zürcher, K. and Hadorn, H., „Wasserbestimmung nach Karl Fischer an verschiedenen Lebensmitteln“, Deut. Lebensmittel Rundschau 77, 343-355 (1981) (in German). - Strange, T.E., “Collaborative Study of Moisture in Cheese by Gas Chromatography and by Karl Fischer Titration”, J. Assoc. Off. Anal. Chem. 53, 865-868 (1970) J. Assoc. Off. Anal. Chem. 55, 507-510 (1972). - Rüegg, M. et al., „Die Bestimmung des Wassergehalts in Milch und Milchprodukten mit der Karl Fischer Methode“, Mitt. Gebiete Lebensm. Hyg. 77, 131-138 (1986) and 139-146 (1986) Mitt. Gebiete Lebensm. Hyg. 77, 446-451 (1986) and 535-543 (1986) Mitt. Gebiete Lebensm. Hyg. 78, 309-316 (1987).

METTLER TOLEDO Page 14 of 80 Karl Fischer Titration Applications 2.2.1 Coulometric determinations in oils, fats and dairy products

Sample m n H2O srel Method [g] [ppm] [%] Olive oil 1.0 5 8.36 0.7 - Sample preparation: -- / Fill syringe from the backside - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm 60 mL Coulomat AG and 40 mL decanol (anolyte) 5 mL Coulomat CG (catholyte) Stirring time: 10 s, Termination: rel. drift stop, Generation rate: normal, Autostart, Online drift determ. Rape oil 1.0 4 424 0.88 - Sample preparation: -- / Fill syringe from the backside - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm 75 mL Coulomat AG and 25 mL chloroform (anolyte) or 60 mL Coulomat AG and 40 mL decanol (anolyte) 5 mL Coulomat CG (catholyte) Stirring time: 10 s, Termination: rel. drift stop, Generation rate: normal, Autostart, Online drift determ.

Comments - All of the above mentioned applications can also be performed with a generation cell without diaphragm. - Edible oil Oils can generally not be drawn into a syringe due their high viscosity. Dispensing with a syringe is still possible: fill the syringe from the backside, or use a thick needle. Special solvent mixtures such as e.g HYDRANAL®Coulomat Oil, or 1-decanol can be used in place of chloroform; the latter, however, better dissolves the oil sample. When using solvents other than chloroform the solvent mixture should be replaced after a few samples. KF coulometry is also well suited for edible oils. In spite of the smaller sample size, the reproducibility is better than that obtained for the volumetric procedure.

2.2.2 Volumetric determinations in oils, fats and dairy products

Sample m n H2O srel Method [g] [%] [%]

Sunflower oil 3.0 7 0.0767 1.2 - Sample preparation: -- / Direct titration - Sample addition: Fill syringe from behind Syringe without needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s Butter 0.2 6 15.3 0.47 - Sample preparation: Homogenize well the sample with a spatula - Sample addition: Spatula - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 25 mL KF Solvent 20 mL 1-decanol Stirring time: 300 s Termination: Delay time 15 s

METTLER TOLEDO Page 15 of 80 Karl Fischer Titration Applications Margarine/Minarine 0.03 6 58.2 0.25 - Sample preparation: Homogenize well the sample with a spatula External extraction: 20 min at 25°C 1.5 g in 50 mL 1-decanol/methanol 4:1, - Sample addition: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30-40 mL KF Solvent Stirring time: 20 s Termination: Delay time 15 s Milk powder 0.05 6 5.09 0.67 - Sample preparation: External extraction 2 g in 50 g formamide, 30 min at 25°C - Sample addition: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30-40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Plain yogurt 0.03 6 86.3 0.22 - Sample preparation: Homogenize 3 min with mixer - Sample addition: Fill syringe from behind Syringe without needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30-40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s Ice cream 0.02 6 66.5 0.28 - Sample preparation: Homogenize well - Sample addition: Fill syringe from behind Syringe without needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30-40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s Casein 0.1 6 10.3 0.46 - Sample preparation: -- - Sample addition: Weighing boat - KF Titration: Titrate at 40-45°C Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent 30 mL formamide Stirring time: 300 s Termination: Delay time 15 s Grated cheese 0.025 6 27.6 0.42 - Sample preparation: Shake well the package External extraction: 2 hours at 25°C 1.5 g in 50 mL decanol/formamide/methanol 8:2:1 - Sample addition: 1.5 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30-40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s

METTLER TOLEDO Page 16 of 80 Karl Fischer Titration Applications - Edible oil Oils can generally not be drawn into a syringe due their high viscosity. Dispensing with a syringe is still possible: fill the syringe from the backside, or use a thick needle. Special solvent mixtures such as e.g APURA® Solvent oils & fats, HYDRANAL® Solvent CM, or 1-decanol can be used in place of chloroform; the latter, however, better dissolves the oil sample. - Butter: Water is distributed very inhomogeneously in butter. For this reason the sample has to be homogenized thoroughly at 25-30°C and is kept in a tightly sealed vessel. Before each determination, the top layer is skimmed off and the sample is removed from the lower product (The water content of the surface butter will decrease slowly, as some of the moisture is released to the ambient). The sample is removed with a spatula and transferred into a titration cell. Do not use syringes or similar devices, as the mechanical shear may drive water drops from the sample. The external extraction with methanol/1-decanol gives better reproducibility (for 6 samples, RSD = 0.29) but requires much greater effort. - Margarine/Minarine: Direct titration in 1-decanol/methanol 1:1 is possible, but the reproducibility is poor. Due to the high water content, using small samples is sensible although even small inhomogeneities will cause great variation in the results. The external extraction method is easily performed and gives reproducible results. - Milk powder: The two-component reagent can be used for a direct titration by adding Triton X-100 (detergent) (RSD = 1.8%). Using the drying oven (20 min. at 120°C) the values obtained are somewhat higher, and the reproducibility RSD is 2.6%. The external extraction works well and gives the most reproducible results. - Plain yogurt: Water is distributed unevenly in yoghurt. Homogenize the sample well with a hand mixer. - Ice cream: Water is distributed unevenly in ice cream. Homogenize the sample thoroughly at room temperature in a closed vessel with a hand mixer. - Casein: This sample can be titrated directly at 40-45°C by adding 50% formamide. After stirring for 5 minutes, the sample is dissolved completely. Replace the solvent mixture after 3 samples, as the dissolving capacity will be exhausted. - Grated cheese: Grated cheese cannot be dissolved in the KF solvents. It is possible to titrate a finely grated sample by adding formamide and titrating at 50°C. Cheese can be dissolved completely in the solvent mixture 1-decanol / formamide / methanol 8:2:1 (1-decanol for dissolution of fats, formamide for proteins, and methanol as the solvating agent). This solvent mixture was used to dissolve the sample. This mixture gives a two-phase solution, and thus it must be stirred vigorously while removing sample aliquots.

METTLER TOLEDO Page 17 of 80 Karl Fischer Titration Applications 2.3 Grains and Starchy Food All of these foodstuffs are insoluble in the usual solvents and release water only very slowly. They are not well suited for the coulometric method since the water content is too high, and the solubility in the KF solvent too poor. Finely milling coarse grains allows for the water to be released more completely. Most of these products may be titrated directly at an elevated temperature (up to 50°C) in the presence of formamide. Also recommended is an external extraction with methanol or formamide (in some cases at higher temperatures). Caution: - Baked goods containing carbohydrates may undergo thermal decomposition at elevated temperatures (greater than 50°C) in a process that generates water. - Using more than 50% formamide is not recommended, as the stoichiometry of the Karl Fischer reaction will be altered. Extensive information on the water determination in these products can be found in: - Zürcher, K. and Hadorn, H., „Wasserbestimmung in Lebensmitteln nach der Methode Karl Fischer“ Teil 1, Deut. Lebensmittel Rundschau 74, 249-259 (1978) Teil 2, Deut. Lebensmittel Rundschau 74, 287-296 (1978) (in German). - Zürcher, K. and Hadorn, H., „Wasserbestimmung nach Karl Fischer an verschiedenen Lebensmitteln“, Deut. Lebensmittel Rundschau 77, 343-355 (1981) (in German). - Hadorn, H., „Ringversuche zur Wasserbestimmung in Lebensmitteln nach Karl Fischer“ Mitt. Gebiete Lebensm. Hyg. 71, 220-235 (1980) (in German). - Scholz, E., „Wasserbestimmung in Lebensmitteln“, Deut. Lebensmittel Rundschau 79, 302-306 (1983) (in German). - ISO 5381:1983 Standard, „Starch hydrolysis products -- Determination of water content -- Modified Karl Fischer method”, see www.iso.org .

2.3.1 Volumetric determinations in grains and starchy products

Sample m n H2O srel Method [g] [%] [%]

All-purpose flour 0.3 5 12.1 0.36 - Sample preparation: -- / Direct titration - Sample addition: Glass weighing boat - KF Titration: Titration at 50°C Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 300 s Termination: Delay time 15 s Dough 0.2 6 20.9 2.8 - Sample preparation: Heat with manual drying oven - Sample addition: --

- KF Titration: 190°C, 200 mL N2/min Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s

METTLER TOLEDO Page 18 of 80 Karl Fischer Titration Applications Rusk / Zwieback 0.5 6 1.89 0.59 - Sample preparation: Heat with manual drying oven - Sample addition: --

- KF Titration: 120°C, 200 mL N2/min Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s Biscuit 0.4 6 5.79 0.87 - Sample preparation: Crush sample with mixer - Sample addition: Glass weighing boat - KF Titration: Titration at 45°C Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 900 s Termination: Delay time 15 s Corn flakes 0.5 6 6.14 0.69 - Sample preparation: Crush sample with mixer Heat with manual drying oven - Sample addition: --

- KF Titration: 140°C, 200 mL N2/min Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s Noodles 0.4 6 10.1 1.3 - Sample preparation: Crush sample with mixer Heat with manual drying oven - Sample addition: --

- KF Titration: 140°C, 200 mL N2/min Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s Instant potato 0.15 6 8.58 0.3 - Sample preparation: Crush sample with mixer flakes External extraction: 2 g in 70 g formamide, 1 hour at 25°C - Sample addition: 5 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 900 s Termination: Delay time 15 s Potato chips 0.3 6 4.83 0.72 - Sample preparation: Crush sample with mixer - Sample addition: Glass weighing boat - KF Titration: Titration at 50°C Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 900 s Termination: Delay time 15 s

METTLER TOLEDO Page 19 of 80 Karl Fischer Titration Applications - All-purpose flour: The direct titration at 50 C using formamide gives reproducible results. A further possibility is an external extraction with methanol at room temperature. - Dough: The water content determination using the manual drying oven at 190°C gave consistently reproducible results. At even higher temperatures (220°C) a slow decomposition of the dough was observed (product turned dark brown to black). The result increased, and the post-consumption was high. - Rusk / Zwieback: Using the manual drying oven at 120°C for the water content determination of the finely crushed sample gave consistently reproducible results. At higher temperatures (140°C) the Zwieback begins to decompose slowly, resulting in a dark brown coloration of the product and in considerably falsified results. After finely crushing the sample, it can also be titrated directly at 50°C in the presence of formamide or as external extraction with 100 mL methanol at 50°C for 60 minutes in the sonicator. - Biscuit: Direct titration of this product in the presence of formamide at 45°C gives reproducible results. External extraction with methanol is a further possibility. As this product contains sugars which decompose at elevated temperatures, the manual drying oven can not be used for this determination. - Corn flakes: The water content determination using the manual drying oven at 140°C gave consistently reproducible results. In spite of the fine milling of the sample, the presence of formamide, elevated temperatures and the longer stir time, water is not released completely from this product. Using a direct titration in methanol / formamide 1:1 at 50°C with a titration time of 1 hour, the water content was found to be only 5.9%. - Noodles: The sample is crushed finely in the mixer. The determination using the manual drying oven at 140°C generates reproducible results. A direct titration at 50°C in the presence of formamide can also be used. - Instant potato flakes: Highly reproducible results can be obtained with an external extraction in formamide for 1 hour at ambient temperature. Do not heat, as this will cause the flakes to swell and the extraction liquid will thicken drastically. A direct titration still takes over 40 minutes even after crushing the sample, adding formamide and raising the temperature. A determination in the manual drying oven with a maximum temperature of 130°C is possible, although it is not recommended since the repeatability was poor: the titration takes approx. 10 minutes, the mean value from 3 samples was 8.2% (srel = 10.6 %). At higher temperatures the product begins to decompose. - Potato chips: The finely ground sample can be titrated directly at 50°C after adding formamide. For the external extraction method, chloroform must be present in the solvent mixture due to the fat content of the chips.

METTLER TOLEDO Page 20 of 80 Karl Fischer Titration Applications 2.4 Applications: Spices Most spices release water very slowly. Therefore, formamide must be added to these products and the temperature elevated when performing a direct titration. External extraction using methanol or formamide has been shown to give good results. Extra care must be taken to grind and homogenize these products thoroughly. Water is enclosed in cellular structures in vegetable based dehydrated foods. The water is released slowly and incompletely. For this reason it is necessary to chop these foods finely and use the external extraction with methanol or formamide at temperatures up to 50°C. Caution: Foodstuffs containing carbohydrates may thermally decompose in a process that generates water at elevated temperatures. Extensive information on the water determination in these products can be found in: - Radar, B.R., „Determination of Moisture in Dried Vegetables“ J. Assoc. Off. Anal. Chem. 50, 701-703 (1967). - Thung, S.B., “Comparative Moisture Determination in Dried Vegetables by Drying after Lyophilisation or the Karl Fischer Method” J. Sci. Food Agric. 15, 236-244 (1964). - Scholz, E., „Wasserbestimmung in Lebensmitteln“, Deut. Lebensmittel Rundschau 79, 302-306 (1983) (in German).

2.4.1 Coulometric determinations in spices

Sample m n H2O srel Method [g] [ppm] [%] Table salt 1.0 5 359.6 4.2 - Sample preparation: Heat with manual drying oven 300°C, 200 mL air/min - Sample addition: -- - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG (catholyte) Stirring time: 60 s, Termination: 600 s, delay time 600 s Generation rate: normal, manual start, Online drift Cinnamon 0.05 7 9.7 % 0.4 - Sample preparation: Heat with STROMBOLI KF Oven powder 180°C, 100 mL air/min (V1.0) - Sample addition: -- - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte, Stirring time: 300 s Termination: Rel. drift 15 µg/min, max. time 500 s Generation rate: normal, manual start, Online drift Garlich 0.05 3 10.1 % 2.2 - Sample preparation: Heat with STROMBOLI KF Oven powder 180°C, 100 mL air/min (V1.0) - Sample addition: -- - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte, Stirring time: 300 s Termination: Rel. drift 15 µg/min, max. time 500 s Generation rate: normal, manual start, Online drift

METTLER TOLEDO Page 21 of 80 Karl Fischer Titration Applications Comments - The applications can also be performed with a generation cell without diaphragm. - The termination parameter can be also set to a fixed titration time (KF drying oven).

2.4.2 Volumetric determinations in spices

Sample m n H2O srel Method [g] [%] [%]

Ground pepper 0.03 6 10.7 0.92 - Sample preparation: External extraction 0.5 g in 55 mL formamide, 1 hour at 25°C - Sample addition: 3 mL aliquot with syringe - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Ground curry 0.07 6 7.59 0.21 - Sample preparation: External extraction 2 g in 50 mL formamide, 1 hour at 25°C - Sample addition: 2 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Sea salt 1.5 7 0.0332 6.7 - Sample preparation: Grind sample in mortar (adherent water) - Sample addition: Weighing boat

- KF Titration: 120°C, 200 mL N2/min Titrant: One-component 2 mg/mL Solvent: 20 mL methanol 30 mL chloroform Stirring time: 250 s Termination: Delay time 15 s Mixed spices 0.2 5 1.83 0.49 - Sample preparation: External extraction 5 g in 45 mL methanol, 45 min at 25°C - Sample addition: 2 mL aliquot with syringe - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Liquid spiced sauce 0.03 6 59.2 0.49 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Liquid soy sauce 0.5 6 63.4 0.41 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s

METTLER TOLEDO Page 22 of 80 Karl Fischer Titration Applications Mustard 0.02 6 72.4 0.59 - Sample preparation: External extraction: 1.2 g in 60 mL mixture 1:1 formamide/1-decanol 1 hour at 25°C - Sample addition: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Ketchup 0.06 6 63.7 0.29 - Sample preparation: Homogenize with mixer - Sample addition: Syringe without needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent Stirring time: 300 s Termination: Delay time 15 s Chives 0.1 7 8.6 4.8 - Sample preparation: External extraction: (dehydrated) 3.2 g in 60 g formamide crush 20 s (manual mixer) extract 30 min at 60°C - Sample addition: 2 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Instant coffee 0.015 7 6.2 0.62 - Sample preparation: External dissolution: 0.5 g in 50 mL formamide Stirring 10 min at 25°C - Sample addition: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s

Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol (or methanol mixtures with auxiliy reagent) as solvent. - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, especially in the case of sluggish end-point. With the oven, you can also set a fixed titrarion time. - Ground pepper and ground curry: These samples release water slowly. They can be titrated directly by adding formamide, increasing the temperature and using a suitable long stirring time. The most reproducible results, however, can be obtained by external extraction with formamide. - Sea salt and table salt (surface moisture): No problems are encountered when these are titrated directly. A coulometric determination can be performed by using the KF drying oven. - Mixed spices: Mixed spices are slightly hygroscopic. When titrating directly, a sluggish endpoint is observed. (Approximate titration time is 20 minutes, water content 3-4 %). The drying oven cannot be used for the determination, as the sample starts to decompose at 130°C. Only the external extraction with methanol gives reproducible results.

METTLER TOLEDO Page 23 of 80 Karl Fischer Titration Applications - Liquid spiced sauce and soy sauce: These samples can be analysed by a straightforward direct titration. Renew the solvent after 3 samples. - Mustard: Homogenize sample well. The sample does not dissolve completely even after the addition of chloroform. Direct titration is thus not an effective method (sluggish endpoint, subsequent consumption). The titration time is approximately 15-20 minutes with srel = 1.1 % for 7 samples. External extraction of the mustard using formamide / 1-decanol 1:1 gives consistently reproducible values. - Ketchup: The thoroughly homogenized sample can be titrated directly without encountering problems. - Dehydrated chives: This product is crushed for 20 seconds in a high-speed hand mixer in formamide prior to being extracted for 30 minutes at 60°C. The thick suspension is filled into the back of a 10 mL plastic syringe. Widening the syringe outlet to 3 mm allows the suspension to be measured out more easily. To prevent moisture uptake from the ambient, the opening is sealed with a stopper after sample input. A further set of 3 measurements resulted in a mean value of 8.7 % (srel = 2.2 %). - Instant coffee: Due to a sluggish endpoint, a direct titration of the coffee is not a suitable method. The finely pulverized sample is completely dissolved in pure formamide after stirring for 5- 10 minutes at 25°C.

METTLER TOLEDO Page 24 of 80 Karl Fischer Titration Applications 2.5 Applications: Various Foodstuffs Water content determinations of vegetable-based products are often difficult, as the water is frequently contained in cellular structures from which it is released slowly and incompletely. The method of choice, therefore, is titration at elevated temperatures up to 50°C, adding formamide to the sample if necessary. External extraction of the samples using methanol at 50-60°C is also recommended. A method that has proven useful in a number of cases is the direct titration in boiling methanol. In any case, thoroughly crushing the samples is a must in all cases. Meat products can be titrated at room temperature in methanol after they have been shredded well. For fatty samples the addition of chloroform is recommended. Shredded meat products can also be dissolved in the solvent mixture 1-decanol / formamide / methanol 8:2:1. This mixture can be used effectively for an external extraction. Note that these products are not well suited for the coulometric method since their water content is generally too high. Extensive information on the water determination in these products can be found in: - Zürcher, K. and Hadorn, H., „Wasserbestimmung nach Karl Fischer an verschiedenen Lebensmitteln“, Deut. Lebensmittel Rundschau 77, 343-355 (1981) (in German). - Scholz, E., „Wasserbestimmung in Lebensmitteln“, Deut. Lebensmittel Rundschau 79, 302-306 (1983) (in German).

2.5.1 Volumetric determinations in various foodstuffs

Sample m n H2O srel Method [g] [%] [%]

Ground 0.4 6 4.84 1.2 - Sample preparation: Crush with mixer hazelnuts - Sample addition: Glass weighing boat - KF Titration: Titration at 50°C Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 900 s Termination: Delay time 15 s Cocoa powder 0.2 6 5.68 0.74 - Sample preparation: External extraction 3 g in 50 mL formamide, 30 min at 25°C - Sample addition: 3 mL aliquot with syringe - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Back tea 0.2 6 7.42 0.26 - Sample preparation: External extraction 6 g in 50 mL formamide, 45 min at 50°C - Sample addition: 2 mL aliquot with syringe - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Ground coffee 0.2 6 2.65 0.28 - Sample preparation: External extraction beans 5 g in 50 mL formamide, 30 min at 50°C

METTLER TOLEDO Page 25 of 80 Karl Fischer Titration Applications - Sample addition: 2 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Liver sausage 0.03 6 61.6 0.29 - Sample preparation: External extraction 3 g in 60 mL mixture 8:2:1 decanol/formamide/methanol 1 hour at 25°C - Sample addition: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s Chicken broth 0.3 5 4.89 0.16 - Sample preparation: External extraction 3 g in 60 mL mixture 8:2:1 decanol/formamide/methanol 1 hour at 25°C - Sample addition: 6 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s Instant pudding 1.5 6 2.24 0.94 - Sample preparation: Shake well the package - Sample addition: Glass weighing boat - KF Titration: Titration at 50°C Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 600 s Termination: Delay time 15 s Chocolate flavor 0.15 5 3.22 1.0 - Sample preparation: -- / Direct titration - Sample addition: Glass weighing boat - KF Titration: Titration at 30-35°C Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 300 s Termination: Delay time 15 s

Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol (or methanol mixtures with auxiliy reagent) as solvent. - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, especially in the case of sluggish end-point. With the oven, you can also set a fixed titrarion time. - Hazelnuts: The results obtained by titrating directly in the presence of formamide with a 15 minute stir time are easily reproducible. After 2 samples the solvent must be replaced as the extraction becomes less efficient (lower values, increased subsequent consumption. - Cocoa powder: In spite of the addition of formamide and the higher temperature, the direct titration takes over 40 minutes. The sample can be dissolved completely at room temperature in pure formamide. - Black tea:

METTLER TOLEDO Page 26 of 80 Karl Fischer Titration Applications External extraction with formamide gives highly reproducible results. A determination using drying oven DO302 should be undertaken at temperatures not to exceed 130°C. More than 15 minutes drying time are required. At higher temperatures, the product decomposes slowly. - Coffee beans: The coffee grounds release water slowly. For this reason the external extraction with methanol at 50-60°C is used, a method which gives reproducible results. Direct titration in boiling methanol may also be used. - Liver sausage: The solvent mixture 1-decanol / formamide / methanol 8:2:1 dissolves the sample to completion (decanol for the fats, formamide for the proteins and methanol as the solvating agent). This solvent mixture is used for to solve the sample. This solution is biphasic and must be stirred vigorously when removing the sample aliquots. The direct titration of this sample in the presence of formamide results in a sluggish endpoint in spite of 10 minutes stir time (incomplete water release). This gives results which are poorly reproducible and water content values which are too low (x = 60.6%, n = 4, srel = 1.5%). - Chicken broth: External extraction using the solvent mixture 1-decanol / formamide / methanol 8:2:1 leads to easily reproducible, reliable results. - Instant pudding: Direct titration at 50°C using the normal KF solvent gives good, reproducible results with a stir time of 10 minutes, even though the sample is not dissolved completely (n = 6, RSD = 0.94%). The total titration time is approximately 12-15 minutes. Upon the addition of formamide, the sample dissolves completely, the water content values obtained, however, are no different. Direct titration of the powder at ambient temperature, on the other hand, clearly results in less reproducible values (n = 6, srel = 6.2%). The titration time is 16-20 minutes. - Chocolate flavor: The sample is completely dissolved after adding formamide and stirring at 30-35°C for 5 minutes. A slow side reaction takes place, as can be seen by the increased drift. (The drift increases by 10-15 µg H2O / min after each sample). To minimize this error, use the smallest sample size possible, renew the solvent after 2 samples, and maintain the temperature at less than 35°C. At higher temperatures, the side reaction increases rapidly.

METTLER TOLEDO Page 27 of 80 Karl Fischer Titration Applications 2.6 Applications: Beverages and Drink Concentrates Fruit juices and alcoholic beverages allow their water contents to be determined readily. Lemonades may also contain carbonates or bicarbonates which react with KF reagents, thereby generating water. This leads to artificially elevated water contents. Determinations using the drying oven may be performed provided that the sample is thermally stable. The samples, however, frequently contain sugars which decompose in reactions that release water at temperatures above about 50°C. Note that these products are not well suited for the coulometric method since their water content is too high. Extensive information on the water determination in these products can be found in: - Zürcher, K. and Hadorn, H., „Wasserbestimmung nach Karl Fischer an verschiedenen Lebensmitteln“, Deut. Lebensmittel Rundschau 77, 343-355 (1981) (in German). - Scholz, E., „Wasserbestimmung in Lebensmitteln“, Deut. Lebensmittel Rundschau 79, 302-306 (1983) (in German).

2.6.1 Volumetric determinations in beverages

Sample m n H2O srel Method [g] [%] [%]

Cognac 0.03 6 60.7 0.19 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Beer 0.02 6 91.1 0.45 - Sample preparation: Sonicate sample (degasing) - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Wine 0.03 7 87.0 0.18 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Egg liquor 0.04 7 54.7 0.29 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Orange juice 0.03 6 61.6 0.29 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s

METTLER TOLEDO Page 28 of 80 Karl Fischer Titration Applications Termination: Delay time 15 s Tomato juice 0.02 6 91.2 0.22 - Sample preparation: Homogenize (e.g. sonicate) - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Black currant 0.05 6 30.8 0.24 - Sample preparation: -- syrup - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Instant ice tea 1.0 5 0.581 2.3 - Sample preparation: -- / Direct titration powder - Sample addition: Glass weighing boat - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 300 s Termination: Delay time 15 s Isotonic drink 0.3 6 1.24 0.49 - Sample preparation: -- / Direct titration powder - Sample addition: Glass weighing boat - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL formamide Stirring time: 300 s Termination: Delay time 15 s

Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol (or methanol mixtures with auxiliy reagent) as solvent. - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, especially in the case of sluggish end-point. - Cognac, egg liquor, orange juice, black currant syrup: These can be titrated directly without encountering any problems. Caution: The dissolving capacity of the solvent will be exhausted after few samples. - Beer: The reproducibility of the direct titration can be improved significantly by degasing the sample in the sonicator for approx. 15 minutes. - Tomato juice: The fibres in the juice may stop the syringe. The sample input will then no longer be representative (filter effect). - Powders for isotonics drinks and instant ice tea: These powders are hygroscopic. The water content cannot be determined with the drying oven as the powders contain sugars and thus are not thermally stable. Adding formamide to the solvent increases the reproducibility of the direct titration (for example: from 2.5% to 0.4% for n = 6 samples).

METTLER TOLEDO Page 29 of 80 Karl Fischer Titration Applications 3 Cosmetics Water or alcohol based cosmetics such as shampoo, liquid soaps, sun lotions, toothpaste, after-shave, perfume or nail polish can be easily titrated using the standard KF solvents. Products containing fats (hand- or body lotion, day-cream, lipstick) and ethereal oils are titrated by adding 1-decanol or chloroform. In some cases, elevated temperatures (50°C) are required.

3.1 Coulometric determinations

Sample m n H2O srel Reagents Method (g) (ppm) (%)

Nail varnish 1.5 7 745.1 0.21 Coulomat AG - Sample addition: remover Coulomat CG syringe with needle (no acetone) - Stirring time: 10 s - Termination: Rel. drift stop - Generation rate: normal - Autostart, online drift Nail varnish 0.5 4 3278.2 0.86 Coulomat AG - Sample addition: remover Coulomat CG syringe with needle, (no acetone) fill syringe from behind - Stirring time: 10 s - Termination: Rel. drift stop - Generation rate: normal - Autostart, online drift Massage oil B 0.3 6 0.1059 0.21 Coulomat AG - Sample addition: % Coulomat CG syringe with needle - Stirring time: 60 s - Termination: Rel. drift stop - Generation rate: normal - Autostart, online drift

Comments - The applications can also be performed with a generation cell without diaphragm. - The termination parameter can be also set to a fixed titration time (KF drying oven). - Nail varnish has a high viscosity and cannot be drawn into a syringe. Dispensing with a syringe is nevertheless possible. The nail varnish must be filled into the syringe from behind. - For nail varnish remover and nail varnish containing acetone, special reagents for ketones and aldehydes must be used.

METTLER TOLEDO Page 30 of 80 Karl Fischer Titration Applications 3.2 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Shampoo 0.03 7 80.3 0.29 - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol (MeOH) Stirring time: 60 s Termination: Delay time 15 s After-shave 0.3 6 32.7 0.39 - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 60 s Termination: Delay time 15 s Day cream 0.015 6 80.5 0.46 - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 20 mL methanol 20 mL chloroform Stirring time: 60 s Termination: Delay time 15 s Sun lotion 0.02 7 71.5 0.54 - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 30 mL methanol 20 mL chloroform Stirring time: 60 s Termination: Delay time 15 s Lipstick 0.1 6 2.13 0.85 - Sample preparation: External dissolution 1.8 g in 50 mL 1-decanol 30 min at 50°C in sonicator - Sample addition: 2 mL with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Massage oil A 2.0 6 0.0461 0.31 - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL 1-decanol Stirring time: 10 s Termination: Delay time 15 s Toothpaste 0.02 6 40.5 0.24 - Sample preparation: External extraction 2 g in 50 mL methanol 30 min at 25°C - Sample addition: 0.5 mL aliquot with syringe - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 60 s Termination: Delay time 15 s

Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol (or methanol mixtures with auxiliy reagent) as solvent.

METTLER TOLEDO Page 31 of 80 Karl Fischer Titration Applications - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop. - Shampoo: Exchange the solvent after 3 samples, as the dissolving capacity will be exhausted. Air bubbles in the syringe can influence the results. - Creams: Renew the solvent after 5 samples, as the dissolving capacity will be exhausted. 1- decanol or other special solvents that dissolve fats and oils can be used instead of chloroform. - Lipstick: Since this product does not dissolve completely even in the presence of chloroform, direct titration is not a suitable method (slow and incomplete water release, long titration times). The sample can be dissolved in 1-decanol at 50°C in the sonicator. - Massage oil A: 1-decanol is used in place of chloroform. The solvent must be replaced after 6 samples as its dissolving capacity will be exhausted. The water content of this sample can also be determined by coulometric KF titration. The reproducibility is clearly better. - Toothpaste: The sample must be homogenized thoroughly. Direct titration is not as well suited for this sample is it does not dissolve completely.

METTLER TOLEDO Page 32 of 80 Karl Fischer Titration Applications 4 Pharmaceuticals 4.1 General remarks In the pharmaceutical industry, moisture contents of active substance, the raw materials required to synthesize these, and the final products themselves are determined. Raw materials: The raw materials used by the pharmaceutical industry are organic and inorganic raw materials as well as solvents. The moisture content determinations of these products are described in the corresponding Chapters of this applications brochure. Active ingredients: The pharmaceutically active ingredients are generally organic or inorganic compounds which are soluble in methanol. These can usually be titrated unproblematically. Aldehydes, ketones and some amides require the use of methanol-free reagents, as their esterification reactions lead to incorrectly high water contents. Lyophilized substances: Injectable substances are sometimes delivered as dried substrates, to prolong the usability of active substance. Here the residual water content is a crucial factor. This water content will be around 100 µg H2O per ampule. Tablets: Tablets may contain substances that will react with KF reagents in side reactions. This is the case for magnesium and aluminum hydroxides, for example. Both compounds are used as antacids. Ointments, oils, suppositories: The solubility of oils and ointments may be improved by adding chloroform or 1-decanol. The United States Pharmacopeia USP XXI recommends a mixture of carbon tetrachloride / chloroform / methanol 2:2:1 for ointments. Suppositories can be dissolved by adding chloroform and elevating the temperature. Extensive information on the water determination in these products can be found in: - Cachet, T. and Hoogmartens, J., „Determination of water in erythromycin by Karl Fischer titration“, J. Pharm. Biomed. Anal. 6 (5), 461-472 (1988) - ISSN 0731-7085. - Müller, M., Deutsche Apotheker Zeitung 127 (41), 2034-2036 (1987) - ISSN 0011-9857 (in German). - Rouf, M.A. and Farrington, K.J., „The use of a simple Karl Fischer apparatus for water determination in lyophilized radiopharmaceutical kits”, Appl. Radiat. Isot. 38 (11), 992-993 (1987) - ISSN 0883-2889. - Lindquist, J., “Determination of water in penicillins using Karl Fischer reagents”, J. Pharm. Biomed. Anal. 2 (1), 37-44 (1984) - ISSN 0731-7085.

METTLER TOLEDO Page 33 of 80 Karl Fischer Titration Applications 4.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Eucalyptus 0.1 6 1267 0.39 - Sample preparation: -- oil - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm 100 mL Coulomat AG, anolyte 5 mL Coulomat CG, catholyte Stirring time: 10 s, Termination: Rel. drift stop Generation rate: Normal - Autostart, Online drift

Sample Batch Bottle H2O Method No. (ppm)

Lyophilized 2B0A 1 771.3 - Sample preparation: plasma 2 699.3 Add 5 mL anolyte into septum flask and dissolve/extract the lyophilized sample. 3 759.2 - Sample addition: Syringe with needle 4 665.7 - KF Titration: Cell with diaphragm Coulomat AG 100 mL Coulomat AG, anolyte RR05 1 720.7 5 mL Coulomat CG, catholyte 2 648.7 Stirring time: 10 s Termination: Rel. drift stop 3 729.3 Generation rate: Cautious 4 619.1 - Manual start, Online drift

Comments - The applications can also be performed with a generation cell without diaphragm. - Eucalyptus oil: Direct titration presents no problems at all. Additional chloroform is not necessary. - Lyophilized plasma: External dissolution of the lyophilized material in methanol is not recommended because of its low water content. In fact, the correction for the blank value of the solvent is too high with respect to the water content of the sample. The following method is used: o Approximately 5 mL anolyte are drawn from the anode compartment of the titration cell using a 10-mL plastic syringe with a long needle and then returned i.e. re-injected into the titration cell. o The syringe is rinsed in this way until it is dry. o Afterward, 5 mL anolyte are drawn into the syringe and injected through a septum stopper into the lyophilized sample vial. o The lyophilized material is suspended (or dissolved) in the liquid by shaking e.g. for about five minutes in an ultrasonic bath. The correct calculations are selected depending if the sample is suspended or dissolved. o The entire suspension is then drawn into the same syringe, injected into the titration cell and the water content determined. o Only dry air may be used for pressure compensation (aeration) when the sample is taken from the sample bottle with septum stopper.

METTLER TOLEDO Page 34 of 80 Karl Fischer Titration Applications 4.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Aspirin 0.2 6 1.53 1.9 - Sample preparation: Crush sample - Sample addition: Weighing boat - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 600 s Termination: Delay time 15 s Antibiotics (dry) 0.2 6 4.56 1.0 - Sample preparation: -- - Sample addition: Weighing boat - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent 2 drops of Triton X100 nonionic surfactant Stirring time: 300 s, high speed Termination: Delay time 15 s Antibiotics 0.03 6 95.8 0.24 - Sample preparation: -- suspension - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Disinfectant 1.0 6 0.0620 1.5 - Sample preparation: -- powder - Sample addition: Weighing boat - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 40 mL KF Solvent Stirring time: 120 s Termination: Delay time 15 s Cough syrup 0.1 6 6.25 0.13 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 10 s Termination: Delay time 15 s Valerian essence 0.1 6 29.3 0.14 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Chamomile 0.03 6 43.2 0.12 - Sample preparation: -- exctract - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s

METTLER TOLEDO Page 35 of 80 Karl Fischer Titration Applications Tonic 0.05 6 57.0 0.26 - Sample preparation: -- - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 5 mg/mL Solvent: 25 mL methanol 15 mL formamide Stirring time: 10 s Termination: Delay time 15 s Gel for sprains 0.7 6 42.3 0.51 - Sample addition: Syringe with needle and infections - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 30 s Termination: Delay time 15 s Ointment 0.03 5 71.2 0.74 - Sample addition: Syringe with needle base - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL Solvent for fats & oils Stirring time: 30 s Termination: Delay time 15 s Ointment 0.03 5 54.7 0.40 - Sample addition: Syringe with needle for burns - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL Solvent for fats & oils Stirring time: 60 s Termination: Delay time 15 s Suppositories 1.0 6 0.0023 13.3 - Sample addition: Input with tweezers - KF Titration: Titration at 50°C Titrant: Two-component 2 mg/mL Solvent: 30 mL KF Solvent 15 mL toluene Stirring time: 60 s Termination: Delay time 15 s

METTLER TOLEDO Page 36 of 80 Karl Fischer Titration Applications - Chamomile extract, cough drops and valerian extract: Titrate directly. - Tonic: When using the normal solvent of the two component reagent as well as methanol, a gel drop forms which immediately deposits itself on the platinum point of the electrode. This results in an over-titration. This effect can be avoided by adding formamide. - Gels, ointments: When using the normal solvent of the two component reagent as well as methanol, a gel drop forms which immediately deposits itself on the platinum point These samples are filled into the back end of a syringe for the sample input. No needle is used due to the sample consistency. The gel dissolves in the solvent of the two component reagent. The complete solvation of fat-based ointments requires the addition of chloroform, 1-decanol or the special solvent for fats and oils. - Suppositories: One complete suppository is used for each titration (The sample size is constant: 1.0 g). In the KF solvent/toluene mixture the suppositories dissolve completely at 50°C. The solvent must be replaced after 2-3 samples, as the samples no longer dissolve completely. This leads to a decrease in reproducibility.

METTLER TOLEDO Page 37 of 80 Karl Fischer Titration Applications 5 Inorganic raw materials 5.1 General remarks Many inorganic salts have low water contents and are thermally stable. They are therefore very suitable for coulometric KF titration using a drying oven Salts: Salts can contain water in different ways: water of crystallization, inclusion water and adherent moisture. In order to quantify water of crystallization and included water, the sample must be dissolved completely. Methanol is usually a suitable solvent. Some substances may require the addition of formamide, or titration at elevated temperatures. For substances that dissolve slowly, finely crushing the sample may be helpful. When determining only the adherent moisture, dissolution of the substance and diffusion of other water into the titration must be prevented. For this reason, a high proportion of chloroform is used in the solvent. Acids and bases: These must be neutralized prior to the titration. Imidazole or pyridine is used for acids, benzoic acid or salicylic acid is added to neutralize bases. Strong acids also tend to undergo esterification, whereby water is formed, for instance, highly concentrated acids such as gaseous HCl or 96% H2SO4. The following inorganic compounds react with Karl Fischer reagents in a side reaction which leads to an incorrect water content: - peroxides, oxides, hydroxides, carbonates, hydrogen carbonates, disulfites, nitrites, sulfites, thiosulfates, hydrazine and derivatives, iron(III) salts, copper(I) salts, tin(II) salts, silanols, arsenites, arsenates, selenites, tellurites and boron compounds. For these compounds a direct KF titration is not suitable. The moisture from thermally stable compounds should be evaporated in a drying oven and the moisture transferred into the titration cell with a dry purge gas. External extraction may also be used; however, the sample must be insoluble in the extraction solvent.

Extensive information on the water determination in these products can be found in: - Bryant, W.M.D. and Mitchell, J., „ Analytical Procedures Employing Karl Fischer Reagents - Determination of Water of Hydration in Salts in Inorganic Oxides and Related Components- “, J. Am. Chem. Soc. 63, 2924-2930 (1941). - ISO 3699-1976 Standard “Anhydrous Hydrogen Fluoride for Industrial Use - Determination of Water Content - Karl Fischer Method”. see www.iso.org . - ISO/DIS 7105 Standard “Liquified Anhydrous Ammonia for Industrial Use - Determination of Water Content - Karl Fischer Method”. see www.iso.org .

METTLER TOLEDO Page 38 of 80 Karl Fischer Titration Applications 5.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Barium 0.03 7 14.77 1.9 - Sample preparation: Heat with STROMBOLI KF Oven chloride 220°C, 80 mL air/min (V1.0) dihydrate - KF Titration: Cell with diaphragm BaCl2•2H2O 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Exp. value: Stirring time: 30 s 14.74 % Termination: Fixed time 600 s, delay time 30 s Generation rate: Normal, manual start, Online drift Cetylpyridi- 0.06 7 4.89 1.6 - Sample preparation: Heat with STROMBOLI KF Oven nium chloride 180°C, 80 mL air/min (V1.0) monohydrate - KF Titration: Cell with diaphragm CPC•H2O 100 mL Coulomat AG Oven, anolyte Exp. value: 5 mL Coulomat CG, catholyte 5.03 % Stirring time: 30 s Termination: Fixed time 420 s, delay time 30 s Generation rate: Normal, manual start, Online drift Magnesium 0.07 5 4.6 2.4 - Sample preparation: Heat with manual KF Oven oxide 250°C, 200 mL air/min MgO - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 400 s, delay time 60 s Generation rate: Normal, manual start, Online drift Sodium 1.0 5 0.0360 4.2 - Sample preparation: Heat with manual KF Oven chloride 300°C, 200 mL air/min NaCl - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 600 s, delay time 60 s Generation rate: Normal, manual start, Online drift Lithium 1.0 5 0.7088 3.4 - Sample preparation: Heat with manual KF Oven chloride 250°C, 200 mL air/min LiCl - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 600 s, delay time 60 s Generation rate: Normal, manual start, Online drift Aluminum 3.0 6 0.0212 3.4 - Sample preparation: Heat with manual KF Oven powder 280°C, 200 mL air/min - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 900 s, delay time 60 s Generation rate: Normal, manual start, Online drift Aluminum 0.45 6 0.603 0.7 - Sample preparation: Heat with STROMBOLI KF Oven oxide 250°C, 80 mL air/min (V1.0)

Al2O3 - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 20 s Termination: Fixed time 360 s, delay time 60 s Generation rate: Normal, manual start, Online drift

METTLER TOLEDO Page 39 of 80 Karl Fischer Titration Applications Comments - The applications can also be performed with a generation cell without diaphragm.. - Barium chloride dihydrate: The recovery of the two molecules of water of crystallization was very good. The theoretical value is 14.737 %. - Cetylpyridinium chloride (CPC) monohydrate: A drying temperature of 180 - 200 °C must be used with STROMBOLI. Above 200 °C, CPC is no longer thermally stable. The theoretical value is 5.028% . - Magnesium oxide: The high water content necessitates the use of a low sample weight. This requires a very homogeneous sample in order to obtain good reproducibility. - Sodium chloride, lithium chloride, aluminum powder, aluminum oxide: The coulometric titration using the drying oven presents no problems at all. 5.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Calcium chloride 0.1 6 26.4 0.46 - Sample addition: Weighing boat dehydrate - KF Titration: CaCl2•Cl2 Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 50 s Exp. value: Termination: Delay time 15 s 24.51 % Phosphoric acid 0.15 6 13.7 0.5 - Sample addition: Syringe with needle

H3PO4 - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent 20 mL buffer solution (for acids) Stirring time: 10 s Termination: Delay time 15 s Sulfuric acid 0.1 6 4.63 1.4 - Sample addition: Syringe with needle approx. 96% - KF Titration: H2SO4 Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent 20 mL buffer solution Stirring time: 10 s Termination: Delay time 15 s Sodium 1.7 6 0.443 3.2 - Sample preparation: Heat with manual KF Oven carbonate 250°C, 200 mL air/min

Na2CO3 - Titrant: One-component 2 mg/mL Solvent 40 mL methanol Stirring time: 700 s Termination: Delay time 15 s Calcium oxide 1.5 5 0.227 10.9 - Sample preparation: Heat with manual KF Oven CaO 250°C, 200 mL air/min - Titrant: One-component 2 mg/mL Solvent 40 mL methanol Stirring time: 600 s Termination: Delay time 15 s Aluminum oxide 1.3 6 0.148 2.2 - Sample input: Weighing boat

Al2O3 - Titrant: Two-component 2 mg/mL Solvent 40 mL KF Solvent Stirring time: 300 s Termination: Delay time 15 s

METTLER TOLEDO Page 40 of 80 Karl Fischer Titration Applications Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol as solvent (or methanol mixtures with auxiliary solvents). - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Calcium chloride: The sample is slightly hygroscopic. All 6 samples could be titrated with the indicated amount of solvent. - Phosphoric acid: This can be titrated directly unproblematically. The sample must be neutralized with a base (such as imidazole), otherwise the endpoint will be sluggish. - Sulfuric acid approx. 96%: The sample is hygroscopic. The syringe must be discarded after 3 samples, as the acid will attack it. Neutralization of the sample with a base (i.e., imidazole) is necessary. - Sodium carbonate: Carbonate reduces iodine to give iodide, which results in an erroneously high water content:

Na2CO3 + I2 + SO2 + CH3OH → 2 NaI + CO2 + CH3HSO4 For this reason the drying oven is used. An external extraction with methanol results in values that are too high, since sodium carbonate is slightly soluble in methanol. - Calcium oxide: The KF solutions are always weakly acidic, thus they may react with oxides in a reaction that produces water:

CaO + 2 HI → CaI2 + H2O This is why use of the drying oven is recommended. An external extraction with methanol results in artificially elevated values, as traces of calcium oxide will dissolve in methanol. - Aluminum oxide: Aluminum oxide is not basic enough to react with the KF solution. Direct titration is possible. The sample is highly hygroscopic, thus contact with the ambient is to be held at a minimum. The sample releases water slowly, therefore it is necessary to stir for 5 minutes.

METTLER TOLEDO Page 41 of 80 Karl Fischer Titration Applications 6 Organic raw materials 6.1 General remarks

Hydrocarbons, halogenated hydrocarbons, alcohols, esters and ether: The water content determination of these substances is unproblematic. Adding propanol or chloroform increases the solubility of long chained compounds. Double bonds rarely cause problems. These compounds generally contain little water, thus coulometry is particularly recommended. Halogenated hydrocarbons may contain active chlorine from the fabrication process. The active chlorine oxidizes iodide to iodine (resulting in erroneously low water content). Phenols: For most phenols the water content determination is unproblematic. In some cases salicylic acid may need to be added as a buffer. Some phenols have been shown to have a high post-consumption (for example, aminophenol). Aldehydes and ketones: These compounds react with methanol to produce acetals and ketals respectively. Water is produced in the reactions. These side reactions may be supressed by using methanol-free titrants and solvents. Special reagents for the water content determination of aldehydes and ketones are available. Bisulfite addition is a further side reaction which aldehydes will undergo. This reaction consumes water and starts as soon as the sample is added to the solvent containing sulfur dioxide. The titration must thus be started immediately to circumvent the bisufite addition. Organic acids: Strongly acidic organic acids should be neutralized to keep the pH value in the proper range for a KF titration. Nitrogen compounds: Strongly basic amines are neutralized with benzoic acid. The solubility of higher amines needs to be improved by adding chloroform. A few amines titrated with methanol as the solvent show an unstable endpoint (such as aniline, toluidine, aminophenol), probably due to a side reaction. This can be eliminated by using a methanol-free solvent. The water content determination of hydroxylamine, hydrazine and hydrazine salts is difficult. Their oxidation by iodine results in erroneously high water contents. Sulfur compounds: The determination of these compounds is unproblematic. Exceptions are mercaptans and thiols. These are oxidized by iodine, resulting in artificially high water contents.

Detailed information on the water determination in these products can be found in: - Scholz, E., „Wasserbestimmung in Carbonsäuren“, Fresenius Z. Anal. Chem. 312, 423-426 (1982), in German. - Scholz, E., „Titration of Aldehydes and Ketones“, Anal. Chem. 57, 2965-2971 (1985). - Scholz, E., „Wasserbestimmung in Phenolen“, Fresenius Z. Anal. Chem. 330 (8), 694-697 (1988), in German.

METTLER TOLEDO Page 42 of 80 Karl Fischer Titration Applications 6.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Napthalene 0.1 6 35.0 10.2 - Sample preparation: External dissolution: 0.4 g dissolved in 25 mL methanol - Sample addition: 5 mL aliquot with syringe - KF Titration: Cell with diaphragm Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. 4-chloro- 0.8 6 30.7 1.7 - Sample addition: Syringe wih needle toluene - KF Titration: Cell with diaphragm dried (1) Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Benzyl alcohol 0.4 6 1273 0.16 - Sample addition: Syringe wih needle dried (1) - KF Titration: Cell with diaphragm Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Phenol 1.0 6 173.7 1.8 - Sample preparation: External dissolution: 17 g dissolved in 60 mL methanol - Sample addition: 1 mL aliquot with syringe - KF Titration: Cell with diaphragm Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Acetophenone 0.3 6 2830 0.46 - Sample addition: Syringe with needle dried (1) - KF Titration: Cell with diaphragm 100 mL Coulomat AK, anolyte 5 mL Coulomat CK, catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Benzaldehyde 0.4 6 242.6 0.9 - Sample addition: Syringe with needle dried (1) - KF Titration: Cell with diaphragm 100 mL Coulomat AK, anolyte 5 mL Coulomat CK, catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Salycilic acid 0.4 6 115.8 2.9 - Sample preparation: External dissolution: 30 g dissolved in 60 mL methanol Titrate immediately - Sample addition: 3 mL aliquot with syringe - KF Titration: Cell with diaphragm Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Methyl- 0.9 6 49.1 2.0 - Sample addition: Syringe with needle benzoate - KF Titration: Cell with diaphragm dried (1) Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Benzamide 0.3 6 117.5 3.6 - Sample preparation: External dissolution: 16 g dissolved in 50 mL methanol - Sample addition: 1 mL aliquot with syringe

METTLER TOLEDO Page 43 of 80 Karl Fischer Titration Applications - KF Titration: Cell with diaphragm Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. n-Butylamine 0.2 6 1.14% 0.8 - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm 100 mL Coulomat AG, anolyte with 20 g benzoic acid 5 mL Coulomat CG, catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Aniline 0.1 1 Not -- - Sample addition: Syringe with needle possible - KF Titration: Cell with diaphragm 100 mL Coulomat AK, anolyte with 20 g benzoic acid 5 mL Coulomat CK, catholyte (1) dried over molecular sieves 3Å

Comments - Naphthalene: Since naphthalene is poorly soluble in methanol, only a small sample amount can be used. The coulometric determination with external dissolution presents no problems at all. - 4-Chlorotoluene, benzyl alcohol, benzoic acid methylester: The coulometric titration presents no problems. - Phenol: Phenol is first dissolved externally in methanol. - Acetophenone: Ketones react with methanol with the formation of water to form a ketal. To prevent this, methanol-free anolyte and catholyte are used. With these reagents the titration of acetophenone can be performed without difficulty and with very good reproducibility. - Benzaldehyde: Aldehydes react with methanol with the formation of water to form an acetal. To prevent this, methanol-free anolyte and catholyte are used. Another side reaction als occurs -the bisulfite addition-, a process where water is consumed. The titration is automatically started using the AUTOSTART parameter immediately after adding the sample. The sample size is entered at the titrator during or after the titration. - Salicylic acid and benzamide: Salicylic acid and benzamide are dissolved in methanol and titrated by external dissolution. The salicylic acid solution must be analyzed immediately because if it is allowed to stand (more than 3 hours) an increasing amount of water is measured (e.g. 250 ppm after 24 hours). This is due to the slow formation of the ester of salicylic acid. - n-Butylamine: n-butylamine is titrated coulometrically after addition of benzoic acid. The values show an increasing tendency within a series, which leads to poorer reproducibility. - Aniline: Aniline does not give a stable end point with methanol as solvent. In spite of the use of methanol- free reagents and neutralization with benzoic acid, a coulometric titration was not possible (no end point was reached).

METTLER TOLEDO Page 44 of 80 Karl Fischer Titration Applications 6.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

4-Chloro- 2.5 8 0.0089 2.3 - Sample addition: Syringe with needle toluene - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 30 s Termination: Delay time 15 s Benzyl alcohol 2.5 7 0.137 0.57 - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 30 s Termination: Delay time 15 s 2-Nitrophenol 2.0 6 0.0549 1.9 - Sample preparation: Crush sample in mortar - Sample addition: Weighing boat - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 30 mL KF Solvent 10 mL buffer solution (acid) Stirring time: 30 s Termination: Delay time 15 s Acetophenone 2.0 8 0.524 0.41 - Sample addition: Syringe with needle - Titrant: One-component K 5 mg/mL Solvent 40 mL solvent for Ketones Stirring time: 30 s Termination: Delay time 15 s Benzaldehyde 1.0 8 0.168 0.32 - Sample addition: Syringe with needle - Titrant: One-component K 5 mg/mL Solvent 40 mL solvent for Ketones Stirring time: 0 s / Start immediately Termination: Delay time 15 s Salicylic acid 1.0 6 0.0140 0.8 - Sample preparation: External dissolution 30 g in 60 mL methanol - Sample addition: 3 mL aliquot with syringe - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 30 s Termination: Delay time 15 s Oxalic acid 0.06 6 27.9 0.43 - Sample addition: Weighing boat - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL buffer solution (acid) Stirring time: 300 s Termination: Delay time 15 s Methyl 3.0 6 0.0083 2.8 - Sample addition: Syringe with needle benzoate - Titrant: One-component 1 mg/mL Solvent 40 mL methanol Stirring time: 30 s Termination: Delay time 15 s Urea 2.0 6 0.123 4.4 - Sample addition: Weighing boat - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL KF Solvent

METTLER TOLEDO Page 45 of 80 Karl Fischer Titration Applications Stirring time: 300 s Termination: Delay time 15 s n-Butylamine 0.5 7 1.12 0.4 - Sample addition: Syringe with needle - Titrant: Two-component 2 mg/mL Solvent 40 mL KF Solvent 4 g benzoic acid Stirring time: 30 s Termination: Delay time 7 s Aniline 2.5 6 0.0819 2.7 - Sample addition: Syringe with needle - Titrant: One-component K 5 mg/mL Solvent 40 mL solvent for Ketones 5 g salycilic acid Stirring time: 30 s Termination: Delay time 15 s K: KF reagents for ketones and aldehydes

Comments - All of the above mentioned applications can also be performed with one-component KF titrant and methanol as solvent (or methanol mixtures with auxiliary solvents). - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - 4-Chlorotoluene, benzyl alcohol: The direct titration is unproblematic. A titrant with a low concentration (2 mg H2O/mL) is used due to the low water content. - 2-Nitrophenol: The sample crushed in the mortar dissolves easily in the solvent. Without addition of a buffer for acids, an elevated post-consumption is observed. - Acetophenone: Ketones react with methanol to give ketals and water. To prevent this, a methanol-free titrant and solvent are used. The solvent should be changed after 4 samples. - Benzaldehyde: Aldehydes react with methanol to form an acetal and water. To prevent this, a methanol-free titrant and solvent should be used. A further side reaction is bisulfite addition, which consumes water. For this reason the titration is started immediately after injecting the sample. The weight is entered on the titrator after the titration. The solvent must be replaced after each sample. When performing several titrations in the same solvent a decrease in the determined water content is observed. - Salicylic acid: Finely pulverized salicylic acid may become electrostatically charged, causing problems during transfer of the sample into the titration cell. The titration cell remains open for different lengths of time. This may lead to considerable scatter in the results when titrating small quantities of water. For this reason, an external dissolution in methanol is produced. This solution must be analyzed immediately, as an increasing proportion of water (i.e., 250 ppm after 24 h) is measurable after the solution has been standing for some time (over 3 hours). The cause is a slow esterification of the acid. - Oxalic acid: The solvent is to be replaced after 2 samples as its buffering capacity will be exhausted. - Methyl benzoate: The direct titration is straightforward. Due to the low water content, a titrant with the concentration of 1 mg H2O/ml was used.

METTLER TOLEDO Page 46 of 80 Karl Fischer Titration Applications - Urea: As urea is slightly soluble in methanol, the solvent must be replaced after each sample. - n-Butylamine: The endpoint is sluggish in spite of the neutralization with benzoic acid. The results are too high and poorly reproducible. Decreasing the switch-off delay to 7 seconds allows reliable and reproducible results to be obtained. - Aniline: No stable endpoint could be achieved using methanol as the solvent. Using a methanol-free solvent eliminates this problem. Neutralization of aniline with salicylic acid is necessary, otherwise the endpoint will be sluggish.

7 Solvents 7.1 General remarks Hydrocarbons, halogenated hydrocarbons, alcohols, esters and ether: The water content determination of these substances is unproblematic. Adding propanol or chloroform increases the solubility of long chained compounds. Double bonds rarely cause problems. These compounds generally contain little water, thus coulometry is particularly recommended. Halogenated hydrocarbons may contain chlorine from the production process. Active chlorine oxidizes iodide to iodine resulting in erroneously low water content. Phenols: For most phenols the water content determination is unproblematic. In some cases salicylic acid may need to be added as a buffer. Some phenols have been shown to have a high post-consumption (for example, aminophenol). Aldehydes and ketones: These compounds react with methanol to produce acetals and ketals, respectively. Water is produced in the reactions. These side reactions may be supressed by using methanol-free titrants and solvents. Special reagents for the water content determination of aldehydes and ketones are available. Bisulfite addition is a further side reaction which aldehydes will undergo. This reaction consumes water and starts as soon as the sample is added to the solvent containing sulfur dioxide. The titration must thus be started immediately to circumvent the bisufite addition. Organic acids: Strongly acidic organic acids should be neutralized to keep the pH value in the suitable range for a KF titration, i.w. pH 5-7. Nitrogen compounds: Nitrogen–containing solvents such as amides (e.g. dimethylformamide), nitrocompounds such as nitrobenzene, and nitriles (acetonitrile) are unproblematic.

Detailed information on the water determination in these products can be found in: - Scholz, E., „Titration of Aldehydes and Ketones“, Anal. Chem. 57, 2965-2971 (1985). - ASTM standard D 1364, 2002 (2007) „Standard Test Method for Water in Volatile Solvents (Karl Fischer Reagent Titration Method)“, see www.astm.org .

METTLER TOLEDO Page 47 of 80 Karl Fischer Titration Applications 7.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Toluene 0.5 5 227.1 1.8 - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm CombiCoulomat Frit Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Toluene I dry(1) 3.0 3 1.42 14.5 - Sample addition: Syringe wih needle - KF Titration: Cell with diaphragm Toluene II dry(1) 3.0 6 4.8 9.8 Coulomat AG 100 mL, anolyte Coulomat CG 5 mL , catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Cautious, Autostart, Online drift n-Hexane I dry(1) 1.0 6 10.5 7.2 n-Hexane II dry(1) 2.0 8 4.6 9.1 Cyclohexane 1.5 4 23.6 7.1 - Sample addition: Syringe wih needle - KF Titration: Cell with diaphragm with e.g. Coulomat AG-K 100 mL, anolyte Coulomat CG-K 5 mL , catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Cyclohexene 0.8 6 78.8 1.2 Isopropyl alcohol 0.2 6 787.6 0.3 Ethylene glycol 6.0 8 588.4 0.4 Ethylene chloride 0.6 6 38.2 3.2 dry(1) Diethylether 0.5 6 40.0 4.0 dry(1) Ethylacetate 0.5 6 34.9 1.2 dry(1) Dimethyl- 0.4 6 347.3 1.0 formamide dry(1) Acetaldehyde 0.5 1 Not -- - Sample preparation: Sample cooled to 0°C possible - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm, with e.g. Coulomat AG-K 100 mL, anolyte Coulomat CG-K 5 mL , catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Acetone 0.7 6 118.0 0.7 - Sample addition: Syringe with needle - KF Titration: Cell with diaphragm, with e.g. Acetone dry(1) 2.0 6 21.3 6.6 Coulomat AG-K 100 mL, anolyte Coulomat CG-K 5 mL , catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Methylisobutyl- 0.4 6 717.0 0.2 ketone (1) dried over molecular sieves 3Å

METTLER TOLEDO Page 48 of 80 Karl Fischer Titration Applications Comments Dried solvents contain only very low concentrations of water, sometimes only traces, so that KF coulometry can be particularly recommended. - Toluene and n-hexane: Toluene and n-hexane sample II were dried for 24 hours over molecular sieves 3Å. n- hexane sample I was dried for 1 hour over molecular sieves 3Å. - Cyclohexene, isopropyl alcohol, ethylene glycol, ethylene chloride, diethylether, ethyl acetate: TThese samples present no problems for coulometric. - Acetone, methyl isobutyl ketone (MIBK): Ketones react with methanol with the formation of water to ketals. To prevent this, methanol-free anolyte and catholyte are used. - Acetaldehyde: The boiling point of acetaldehyde is 15 °C. The sample was cooled to about 0 °C so that it could be added with a syringe. Coulometric determination is not possible because acetaldehyde is oxidized at the anode with the formation of water. - Dimethylformamide: Dimethylformamide can be easily analyzed by KF coulometry. This can also be expected for other acid amides.

7.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Toluene 3.0 6 0.0144 0.37 - Sample addition: Syringe with needle - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 30 s Termination: Delay time 15 s Cyclohexene 2.5 7 0.0080 1.5 - One-component 1 mg/mL

Isoamyl alcohol 2.0 6 0.141 0.25 Ethylene glycol 2.0 6 0.651 0.24 Methylene chloride 3.0 6 0.0115 1.2 - Stir: 60 s Diethylether 1.5 6 0.0553 0.40 Acetonitrile 3.0 7 0.0128 2.3 Acetone 3.0 7 0.0128 0.48 - One-component 5 mg/mL KF reagent K 40 mL KF Solvent K Methylisobutyl-ketone 1.5 6 0.145 0.39 (MIBK) Acetaldehyde 1.3 6 0.0337 4.2 - Cool sample to 0°C and titrate immediately Formaldehyde 0.04 6 52.0 0.58 - Two-component reagent 5 mg/mL 30 mL KF Solvent, 20 mL buffer for acid solution Acetic acid 1.0 6 0.263 0.24 - Two-component reagent 2 mg/mL 30 mL KF Solvent, 20 mL buffer for acid solution Dimethylformamide 1.5 7 0.0867 0.36 (DMF) K: KF reagents for ketones and aldehydes

METTLER TOLEDO Page 49 of 80 Karl Fischer Titration Applications

Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Toluene, isoamyl-alcohol, ethylene glycol, methylene chloride, diethylether: The direct titration is straightforward. Sicen the water content of these compounds is low, a titrant with 2 mg H2O/mL is used. - Cyclohexene: A diluted titrant with a concentration of 1 mg H2O/mL is used to determine the low water content of 80 ppm. To achieve a good reproducibility of 1.5% srel, the following points are important: 1. Condition syringe well with the sample, and 2. Replace the solvent after each 3 samples.

Results using titrant 2 mg H2O/mL: srel = 2.7% , n = 6 samples, sample size 2 g. - Acetone, methylisobutylketone: Ketones react with methanol generating a ketal and water. Methanol-free solvents and titrants must be used to prevent this. - Formaldehyde: Formaldehyde will not react with methanol to form an acetal, thus methanol-containing reagents may be used. The total water content cannot be determined by titrating at room temperature, as a part of the water is bound as paraformaldehyde. Even at 50°C, not all the water will be released (theoretical water content = 55.8%). - Acetaldehyde: Acetaldehyde boils at 15°C. To simplify sample input with a syringe, cool the sample to approx. 0°C. Acetal formation can be detected by the high post-comsumption, even though methanol-free reagents were used. Reducing the swith-off delay to 7 seconds reduces the titration time and increases the reproducibility. Additionally, the sample size was selected to be as small as possible and was kept constant. The solvent was also renewed after each sample. Bisulfite addition occurs as a further side reaction. Water is consumed in this reaction. To circumvent this problem, the titration is started immediately after sample addition. The sample weight is entered on the titrator after the titration. - Acetic acid: Acetic acid has a slight tendency to esterify. Thus it is necessary to add a buffer. - Dimethylformamide: The Karl Fischer titration of DMF is straightforward. This is expected to be true for amides of other carboxylic acids as well. - Acetonitrile: The water content determination of acetonitrile is unproblematic.

METTLER TOLEDO Page 50 of 80 Karl Fischer Titration Applications 8 Petroleum and mineral oil products 8.1 General remarks Oils: Mineral oils can be completely dissolved only in the presence of chloroform. Chloroform may be replaced by 1-decanol or the special solvent for oils and fats when determining light mineral oil products (i.e, benzene, kerosine, diesel oil or heating oil). Crude oils, some of which contain tarry components, are best dissolved in a methanol/chloroform/toluene mixture. Motor oils contain additives that may interfere with the KF titration (such as ketones, keto acids, zinc-dialkyl-dithiophosphates, calcium or magnesium sulfonates). In this case the drying oven can be used. The temperature should be adjusted to 120-140°C, as the additives may decompose. Mineral oils have low moisture contents (frequently less than 100 ppm). For these samples, KF coulometry is particularly suitable. Crude oils contain insoluble impurities that may clog the diaphragm. The drying oven is best used for these samples. Silicone oils: The solubility of silicone oils must be improved by the addition of chloroform or 1-decanol. Fats: Some fats are not rendered soluble even after adding chloroform. In these cases, external extraction in pure chloroform or use of the drying oven (temperature 120-140°C) has proven useful. Paraffins and waxes: Waxes and paraffins may be titrated directly at 50°C in a mixture of methanol/chloroform. The determination using the drying oven is a further possibility. Tar and coals: Tarry products can be dissolved by adding toluene or xylene. The drying oven can also be used for tar and coal.

Detailed information on the water determination in these products can be found in: - ISO 6296 : 2000 „Petroleum products – Determination of water -- Potentiometric Karl Fischer titration method” - ISO 10336 : 1997 “Crude petroleum – Determination of water – Potentiometric Karl Fischer titration method” - ISO 10337 : 1997 “Crude petroleum – Determination of water – Coulometric Karl Fischer titration method” see www.iso.org - ISO 12937 : 2000 “Petroleum products – Determination of water – Coulometric Karl Fischer titration method” see www.iso.org - ASTM Standard D 4377, 1988 „Test Method for Water in Crude Oils (Karl Fischer Titration)“, see www.astm.org .

METTLER TOLEDO Page 51 of 80 Karl Fischer Titration Applications 8.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Gasoline 0.6 13 769.3 0.5 - Sample addition: Syringe with needle super grade - KF Titration: Cell with diaphragm Coulomat A 100 mL, anolyte Coulomat CG 5 mL , catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Kerosene 2.5 7 36.1 1.4 Gasoline 0.5 6 43.4 2.5 dry(1) Brake fluid 0.3 7 1081.2 0.38 Silicone oil 0.5 6 103.5 1.7 Hydraulic oil 0.5 6 579.8 2.0 - Sample addition: Syringe wih needle - KF Titration: Cell with diaphragm with Anolyte: 70 mL Coulomat AG 30 mL Chloroform Catholyte: 5 mL Coulomat CG Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Turbine oil 3.0 3 33.5 1.0 Transformer 10.0 5 30.4 1.9 - Anolyte: 50 mL Coulomat AG oil 50 mL Chloroform Paraffin 3.0 6 31.0 15.9 - Sample preparation: Heat with manual KF Oven granules 150°C, 200 mL air/min - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 900 s, delay time 60 s Generation rate: Normal, manual start, Online drift Crude oil 3.0 5 202.0 6.6 - Sample preparation: Heat with manual KF Oven 130°C, 200 mL air/min - Termination: Fixed time 1200 s Carbon 0.8 5 3582. 1.5 - Sample preparation: Heat with manual KF Oven black 7 200°C, 200 mL air/min - Termination: Fixed time 900 s Motor oil 1.0 5 316.5 4.5 - Sample preparation: Heat with STROMBOLI KF Oven 170°C, 80 mL N2/min - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 700 s, delay time 60 s Generation rate: Normal, manual start, Online drift

(1) dried over molecular sieves 3Å

METTLER TOLEDO Page 52 of 80 Karl Fischer Titration Applications Comments - Gasoline, kerosene, petrol: Since Coulomat A already contains chloroform, these samples can be titrated without the addition of chloroform. After 2-3 samples, the solubility power of the solvent is exhausted and an emulsion is formed. But even under these conditions, coulometric titration yields correct results with good reproducibility. - Turbine and hydraulic oils: With chloroform addition, these oils can be titrated directly as an emulsion. These heavy oils do not emulsify as well. The titration time is longer and the reproducibility is poorer. These oils can also be titrated using the drying oven: the moisture is evaporated at 130°C and the vapor transfered into the titration cell using a dry purge gas as gasoline of petrol. - Transformer oils: This oil can be titrated directly as an emulsion. The very low water content necessitates the use of a large sample weight. - Brake fluid: This sample is soluble in the anolyte. Direct titration is no problem. - Silicone oil: Silicone oil can be titrated directly in Coulomat A as an emulsion without additional chloroform. It emulsifies very well. The titration time is short and the reproducibility good. - Paraffin: Paraffin can only be dissolved in the methanol/chloroform mixture at temperatures of about 50 °C. For instrumental reasons, coulometric determinations at higher temperature are not possible. The drying oven is therefore employed. If air is used as purge gas, the drying temperature must not exceed 180 °C. At 200 °C, oxidation begins after about 10 minutes and water is evolved.The poor reproducibility is mainly due to the inhomogeneity of the sample granules. - Motor oil: Motor oil contains additives such as zinc dialkyldithiophosphate and calcium and magnesium sulfonate to improve its properties. These additives make up a total of 5- 7%. Direct titration with chloroform addition is possible, but the values obtained are too high because side reactions occur with the additives. Water determination using the drying oven at 140 – 180 °C gives reliable values. Nitrogen must be used as purge gas otherwise the additives decompose at these temperatures. This leads to results that are too high and poorly reproducible. - Crude oil: Direct titration with chloroform addition gives wrong results: o the sample is merely emulsified, but the water is not completely released. o the water content is too low (mean value of 6 samples = 128.9 ppm, RSD = 0.7%) o black deposits in the titration cell necessitates cleaning of the cell.

The drying oven is therefore used. The crude oil sample is injected with a syringe through a septum directly into the glass boat of the DO307 drying oven - Carbon black: Coulometric determination using the drying oven is straightforward.

METTLER TOLEDO Page 53 of 80 Karl Fischer Titration Applications 8.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Gasoline unleaded 2.5 6 0.0710 0.26 - Sample addition: Syringe with needle - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 40 mL Solvent for fats/oils Stirring time: 30 s Termination: Delay time 15 s Diesel oil 5.0 6 0.0069 2.1 - Solvent: 20 mL KF solvent, 20 mL 1-decanol Kerosene 4.5 6 0.0077 0.72 - Solvent: 20 mL KF solvent, 20 mL 1-decanol Silicone oil 4.0 9 0.0097 0.56 - Solvent: 20 mL KF solvent, 20 mL 1-decanol Motor oil 0.3 6 0.2261 0.94 - Solvent: 20 mL KF solvent, 30 mL chloroform Stirring time: 60 s Crude oil 2.0 5 0.0206 2.9 - Solvent: 20 mL KF solvent, 30 mL chloroform Stirring time: 60 s Motor oil 2.5 6 0.0721 13.4 - Sample preparation: Heat with manual KF Oven 140°C, 200 mL N2/min - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 600 s Termination: Delay time 15 s Motor oil used 2.5 6 0.0842 9.9 Multipurpose grease 0.2 5 0.338 0.88 - Sample preparation: External dissolution 3 g in 80 g chloroform 20 min. at room temp. - Sample input: 5 mL aliquot with syringe - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 40 mL Solvent for fats/oils Stirring time: 30 s Termination: Delay time 15 s Ski wax 1.5 7 0.0417 1.3 - Sample preparation: Melt sample at 50°C - Sample input: Syringe with needle (warm) - KF Titration: Titrant: One-component 2 mg/mL Solvent: 25 mL methanol 25 mL toluene Stirring time: 10 s Termination: Delay time 15 s Shoe polish 2.0 5 0.0219 4.9

Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Gasoline, diesel oil, kerosene: The solubility was improved by using 1-decanol or the special solvent for oils and fats instead of chloroform. - Motor oil: Motor oil contains additives to improve its usability. These are zinc-dialkyldithiophosphate, calcium sulfonate and magnesium sulfonate. New motor oils contain

METTLER TOLEDO Page 54 of 80 Karl Fischer Titration Applications 5-7% of these additives. Motor oil may be titrated by adding chloroform; however, the resultant values are too high due to side reactions with the additives. Reliable values can be obtained by evaporating the moisture in a drying oven at 140°C and then transfering the vapor into the titration cell with a dry purge gas. At higher temperatures the additives will decompose, which will also generate errors. The sample is injected through a septum directly into the crucible using a syringe. The weight is determined by back-weighing. - Silicone oil: 1-Decanol must be added to the solvent to completely dissolve the sample. After 3 samples the solvent should be replaced, as its dissolving capacity will be exhausted. - Crude oil: The sample can be dissolved completely only by adding chloroform to the solvent. 1- Decanol no longer suffices. Replace solvent after each sample, as its dissolving capacity is quickly exhausted. - Multipurpose grease: This fat is incompletely soluble even after adding chloroform. For this reason an external extraction with chloroform was selected. A determination using the drying oven at 180°C is a further possibility. At higher temperatures, however, the sample will polymerize. - Crude oil: The sample is melted in a sealed flask at approximately 50°C. A pre-warmed syringe is used for sample input. The syringe may be warmed using a hair dryer for instance. This procedure is necessary as the sample will solidify immediately in a cold syringe. The addition of toluene and heating the titration solution to approx. 30°C results in a suspension which can easily be titrated. The solvent must be replaced after 3 samples, otherwise the sample will begin to precipitate. This will clog the electrode. The titration solution was kept at 30-35°C using a double-walled titration beaker maintained in a heated water bath.

METTLER TOLEDO Page 55 of 80 Karl Fischer Titration Applications 9 Plastics, polymers and adhesives 9.1 General remarks Plastics: Water is usually securely entrapped in plastics. It is released very slowly and incompletely by diffusion. Most plastics are not soluble in a KF solvent or solvent mixture. Therefore the moisture is usually evaporated in a drying oven, then the vapor is driven into the titration cell using a dry purge gas. External extraction in methanol, if necessary at 50°C, is also possible. Plastics have low moisture contents (often less than 100 ppm), so that KF coulometry is especially recommendable. Ion exchanger: Ion exchange resins are insoluble in organic solvents. For this reason, the drying oven is used to evaporate the moisture. The vapor is driven into the KF cell using a dry purge gas. Adhesives: The water content of adhesives can be determined directly. In some cases chloroform may be necessary to completely dissolve the sample.

Detailed information on the water determination in these products can be found in: - Muroi, K., “Determination of Water in Plastic Materials by Karl Fischer Method”, Bunseki Kagaku 11, 351 (1962). - Praeger, K. and Dinse, H.D., “Experiences in the Determination of Small Amounts of Water in Polyethylene and Polyamide”, Faserforsch. Textiltechn. 21, 37-38 (1970). - Sharma, H.D. and Subramanian, N., “Determination of Water in Ion-Exchange Resins by Karl Fischer and Drying Methods” Anal. Chem. 41, 2063-2064 (1969). - Sharma, H.D. and Subramanian, N., “Determination of water in ion-exchange resins: anion exchange resins” Anal. Chem. 42, 1278-1290 (1970). - Van Acker, P., de Cote, F. and Hoste, J., “Determination of Water in Strong Base Anion-Exchange Resins by the Karl Fischer Titration”, Anal. Chim. Acta. 73, 198-203 (1974). - ASTM Standard D 6869, 2003 „Standard Test Method for Coulometric and Volumetric Determination of Moisture in Plastics Using the Karl Fischer Reaction (the Reaction of Iodine with Water)“, see www.astm.org . - ISO 15512 : 2008 “Plastics -- Determination of water content” see www.iso.org . - DIN Standard 53715 : 1991-05 “Testing of plastics; determination of water content by titration according to Karl Fischer” see www.din.de .

METTLER TOLEDO Page 56 of 80 Karl Fischer Titration Applications 9.2 Coulometric determination

Sample m n H2O srel Method [g] [ppm] [%] Polyethylene 0.8 3 1858 3.1 - Sample preparation: Heat with STROMBOLI KF Oven granules II 200°C, 80 mL air/min (V1.0) (PE II) - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 1200 s, delay time 60 s Generation rate: Normal, manual start, Online drift Polyethylene 2.0 6 68.7 8.2 - Sample preparation: Heat with manual KF Oven granules I 180°C, 200 mL air/min (PE I) - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 600 s, delay time 60 s Generation rate: Normal, manual start, Online drift Polypropylene 2.0 6 148.3 5.4 granules (PP) Polystyrene 2.0 6 322.2 2.4 - Sample preparation: Heat with manual KF Oven granules 160°C, 200 mL air/min (PS) - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 900 s, delay time 60 s Generation rate: Normal, manual start, Online drift Polypropylene 0.5 3 2743 1.2 - Sample preparation: Heat with STROMBOLI KF Oven film 170°C, 80 mL air/min (V1.0) (PP film) - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 600 s, delay time 60 s Generation rate: Normal, manual start, Online drift PVC film 0.1 6 638.1 2.2 - Sample preparation: Heat with manual KF Oven 150°C, 200 mL air/min - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 30 s Termination: Fixed time 300 s, delay time 60 s Generation rate: Normal, manual start, Online drift

METTLER TOLEDO Page 57 of 80 Karl Fischer Titration Applications Comments - Manual KF oven: When the sample cools, it may stick to the glass boat or vial. When using the manual KF drying oven, place a piece of aluminum foil in the glass boat. This allows you to remove it more easily afterward. - Polyethylene and polypropylene polystyrene granules: If air is used as purge gas, the oven temperature must not exceed 180 °C. At 220 °C, polyethylene and polypropylene begin to oxidize and water is evolved. - Polystyrene granules: If air is used as purge gas, the oven temperature must not exceed 160 °C. In fact, polystyrene starts to decompose at already 170°C, and water is evolved leading to higher results. - PVC and PP film: A 3 cm2 piece is cut from the film so that it easily fits in the glass boat of the manual drying oven. The sample must not touch the sides of the oven otherwise it might stick.

9.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Polyethylene 3.0 6 0.0068 6.9 - Sample preparation: Heat with manual KF Oven granules I 180°C, 200 mL N2/min (PE I) - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 40 mL KF Solvent Stirring time: 600 s Termination: Delay time 15 s Polyamide granules 2.0 6 0.5547 0.67 - Sample preparation: Heat with manual KF Oven 190°C, 200 mL N2/min - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s Epoxy resin 0.05 5 0.258 2.6 - Sample input: Syringe with needle (liquid adhesive) - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 20 mL KF Solvent 20 mL Chloroform Stirring time: 60 s Termination: Delay time 15 s Glue stick 0.05 6 46.3 0.94 - Sample input: Spatula - KF Titration: Titrant: One-component 5 mg/mL Solvent: 20 mL methanol 20 mL Chloroform Stirring time: 100 s Termination: Delay time 15 s All-purpose glue 0.2 7 1.28 2.2 - Sample input: Tube with needle (cyanoacryl glue) - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s

METTLER TOLEDO Page 58 of 80 Karl Fischer Titration Applications Rubber cement - Sample preparation: External dissolution (adhesive) 4 g in 85 g chloroform 10 min. at room temp. - Sample input: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 2 mg/mL Solvent: 10 mL KF Solvent 30 mL chloroform Stirring time: 30 s Termination: Delay time 15 s

Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Polyethylene: The sample melts and, once cooled, sticks to the weighing boat in the manual KF oven. Line the weighing boat with aluminum foil to facilitate sample removal. Due to the low moisture content, the drift must be determined prior to each sample to ensure a good reproducibility. - Polyamide: The use of too high temperatures for polyamides may result in a continual condensation reaction which frees water and artificially elevates the water content. For this reason, the sample was first tested at 120°C for 2 hours; result: 0.5445%. The temperature was then elevated to 190°C and the time drastically reduced; result: 0.5547%. - Epoxy resin (liquid adhesive): Chloroform was added to the solvent to competely dissolve the sample. After 4 samples the solvent should be replaced, as its dissolving capacity will be exhausted. - Glue stick: Chloroform must be added to the solvent to ensure that the sample dissolves completely. All 6 samples can be titrated in the same solvent. - All-purpose glue (cyanoacryl glue): The end of the tube was fitted with an injection needle with a large inner diameter (1.2 mm) for the sample input. - Rubber cement: As the sample is poorly soluble in the solvent/chloroform mixture, it was externally dissolved in chloroform. A solvent/chloroform mixture must be used for titration of the solution. The rubber precipitates in pure solvent and will plate the electrode. The solvent must be replaced and the electrode cleaned (deposit on the electrode) after each titration.

METTLER TOLEDO Page 59 of 80 Karl Fischer Titration Applications 10 Dyes and agrochemicals 10.1 General comments Dyes: A large number of dyes are soluble in methanol; these can be titrated easily. Insoluble dyes, such as pigments, can be titrated directly as a suspension. Using this procedure, only the adherent moisture will be determined. A direct titration of water-based dispersion paints and laquers can also be undertaken. When titrating laquers, care must be taken in the choice of a solvent. Ketones or reactive amines require the use of a methanol-free solvent to prevent interferences caused by side reactions with the KF reagents. Agrochemicals: IAgrochemicals are supplied in solution (with organic solvents), water-based suspensions or as powders. Unpolar solvents are generally used for the solution, thus, the samples can be completely dissolved only by adding chloroform or 1-decanol. The moisture content of these is low so that the coulometric method is feasible. The water-based suspensions are best titrated using a direct volumetric titration. Powders rarely dissolve in most KF solvents. Suspensions of these powders allow the determination of the adherent moisture only.

Detailed information on the water determination in these products can be found in: - ASTM Standard D 4017, 2002 (2008 e1) „Standard Test Method for Water in Paints and Paint Materials by Karl Fischer Method“, see www.astm.org .

10.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Rose fungicide 0.5 6 828.6 0.36 - Sample input: Syringe with needle solution - KF Titration: Cell with diaphragm 80 mL CombiCoulomat frit, anolyte 20 mL 1-decanol 5 mL CombiCoulomat frit, catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det. Comments - Rose fungicide solution: This sample is only soluble with the addition of 1-decanol. If the sample merely emulsifies, lower values are obtained.

10.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Optical brightener 0.3 6 3.87 0.76 - Sample input: Weighing boat - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 300 s Termination: Delay time 15 s

METTLER TOLEDO Page 60 of 80 Karl Fischer Titration Applications Wool dye Orange R 0.25 6 10.99 0.14 - Stirring time: 30 s

Reactive dye scarlett 0.3 6 5.24 0.42 - Stirring time: 60 s F-3G Dye indicator 0.6 6 0.928 1.0 - Stirring time: 120 s methanile yellow Dispersion 0.3 6 6.91 0.22 - Sample input: Weighing boat dye olive - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent 25 mL formamide Stirring time: 300 s Termination: Delay time 15 s Acrylic paint 0.05 6 54.3 0.47 - Sample preparation: External dissolution (water based) 3.6 g in 20 mL formamide 10 min. at room temp. - Sample input: 1 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 25 mL KF Solvent 15 mL formamide Stirring time: 30 s Termination: Delay time 15 s Synthetic enamel 1.5 6 0.0906 3.4 - Sample input: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 20 mL KF Solvent 20 mL toluene Stirring time: 15 s Termination: Delay time 15 s Insecticide 2.0 6 0.0977 0.56 - Sample input: Syringe with needle (biological) - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent 20 mL 1-decanol Stirring time: 15 s Termination: Delay time 15 s Insecticide 2.0 6 0.1461 0.21 - Sample input: Syringe with needle (Diazinon) - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent 10 mL 1-decanol Stirring time: 15 s Termination: Delay time 15 s Rose fungicide 0.03 6 58.3 0.25 - Sample input: Syringe with needle (aqueous solution) - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 30 mL KF Solvent Stirring time: 15 s Termination: Delay time 15 s Herbicide 1.0 6 1.20 0.49 - Sample input: Weighing boat (powder) - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 180 s Termination: Delay time 15 s Insecticide (ant killer) 1.0 6 0.273 2.6

METTLER TOLEDO Page 61 of 80 Karl Fischer Titration Applications Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Optical brightener: This sample is not soluble in KF solvents; it can, however be suspended. A direct titration with a 5 minute stir time in methanol is possible. After 2 samples, the solvent should be replaced and the electrode cleaned (i.e. it forms deposit). - Wool, reactive and indicator dyes: These samples are easily dissolved in methanol. The dissolving capacity of the solvent is limited; replace the solvent after 3-4 samples. - Dispersion dye: This sample is insoluble in KF solvents. It can be titrated directly as a suspension by adding formamide. In pure methanol the water is released slowly and incompletely (result: 6.6% with 30 minutes titration time). - Acrylic paint: A direct titration in a formamide/solvent mixture results in the sample sticking to titration cell and elctrode, and is not a feasible alternative. The sample is dissolved in pure formamide and this solution is added to the formamide/solvent mixture. The resulting emulsion can be titrated straightforwardly (slightly increased post- consumption). Replace the solvent after 2 samples. - Synthetic enamel: The sample is added to a prepared toluene/solvent mixture. The resulting fine emulsion can be titrated unproblematically (slightly increased post-consumption). Replace solvent after 2 titrations. - Insecticide and rose fungicide (solutions): These samples (active ingredient plus solvent) are soluble only in the presence of 1- decanol. If an emulsion is prepared, somewhat lower values will result. Due to the low moisture content, a coulometric determination is a further possibility). - Rose fungicide (aqueous emulsion): The direct titration is unproblematic. - Insecticide and herbicide: These samples are insoluble in KF solvents. A direct titration is possible with a 3 minute stir time. (No elevated post-consumption). After 3 samples have been titrated, the water content values decrease (replace sovent).

METTLER TOLEDO Page 62 of 80 Karl Fischer Titration Applications 11 Detergents and Surfactants 11.1 General remarks For these products, which are solids, liquids or pastes, a direct KF titration can easily be undertaken. Detergents containing hydroxide or perborate result in elevated water content values, as these compounds undergo a side reaction with the KF reagent. 11.2 Coulometric determinations

Sample m n H2O srel Method [g] [ppm] [%] Non-ionic 0.25 6 718.4 0.58 - Sample input: Syringe with needle surfactant: - KF Titration: Cell with diaphragm nonylphenyl- 100 mL CombiCoulomat frit, anolyte ethylene glycol 5 mL CombiCoulomat frit, catholyte Stirring time: 10 s Termination: Relative drift stop Generation rate: Normal, Autostart, Online drift det.

11.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Laundry detergent 0.1 6 13.7 2.3 - Sample preparation: Heat with manual KF Oven 150°C, 200 mL N2/min - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 1200 s Termination: Delay time 15 s Detergent for fine 2.0 6 0.5547 0.67 - Sample input: Weighing boat washings 30-60°C - KF Titration: Titrant: One-component 2 mg/mL Solvent: 40 mL methanol Stirring time: 500 s Termination: Delay time 15 s Liquid detergent 0.03 6 81.7 0.60 - Sample input: Syringe with needle - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL KF Solvent Stirring time: 60 s Termination: Delay time 15 s Liquid cleaner 0.03 5 73.9 0.25 Dish detergent 0.03 6 73.0 0.39 (liquid) Soft soap 0.03 6 76.4 1.3 Fabric softener 0.03 6 84.3 0.37 concentrate Non-ionic surfactant 2.0 6 0.160 0.84 alkylphenylpolyethyle nglycol Impregnation agent 0.03 6 61.4 0.32 - Stirring time: 120 s

METTLER TOLEDO Page 63 of 80 Karl Fischer Titration Applications Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Laundry detergent: This detergent contains perborate, which reacts with the KF reagent in a side reaction, as the whitener. For this reason the moisture is evaporated in the drying oven at 150°C and the vapor is transfered to the titration cell by a dry carrier gas. - Detergents for fine washings: These detergents contain neither hydroxides nor perborates. They can be titrated directly. - Cleaner, dish detergent, soft soap: These samples dissolve in the solvent. The direct titration is straightforward. Caution: Beware of the water dissolving capacity of the KF solvent (rplace the solvent after 3-4 samples). - Fabric softener, impregnation agent: These emulsions can be titrated directly without encountering problems. - Non-ionic surfactant: The direct titration is straightforward. The coulometric determination can be used for samples with a low moisture content.

METTLER TOLEDO Page 64 of 80 Karl Fischer Titration Applications 12 Silk, wool, cellulose, paper and wood 12.1 General remarks Wool and cellulose: Wool and cellulose easily release water. Since these products are insoluble in KF solvents, an external extraction must be performed. Methanol is the usual solvent for the extraction. The drying oven can also be used. Paper: Low moisture content is a criterium of the quality, especially for insulation paper. The KF determination is usually performed by external extraction in methanol or in a methanol/chloroform mixture. If the drying oven is to be used, the temperature should be adjusted to 105-130°C. Wood: The KF moisture determination of wood can be achieved using either an external extraction in methanol, or the drying oven in the 105-130°C temperature range.

Detailed information on the water determination in these products can be found in: - Fujino, H. and Muroi, K., “Determination of Water Content in Electric Insulation Papers by Karl Fischer Method”, Bunseki Kagaku 30, 624 (1981). - ASTM Standard D 1348, 1994 (2008) „Standard Test Method for Moisture in Cellulose“, see www.astm.org .

12.2 Coulometric determinations

Sample m n H2O srel Method [g] [%] [%] Photocopy 0.03 6 4.88 1.0 - Sample preparation: Heat with manual KF Oven white paper 140°C, 200 mL air/min - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 10 s Termination: Fixed time 600 s, delay time 60 s Generation rate: Normal, manual start, online drift Photocopy 0.06 6 4.98 0.51 - Fixed time 900 s, delay time 60 s recycled paper Newspaper 0.04 6 7.04 0.53 - Fixed time 900 s, delay time 60 s Insulating 0.05 6 6.41 0.6 - Fixed time 900 s, delay time 60 s paper

Comments - These applications can also be performed with the STROMBOLI oven sample changer. - Paper: Cut the paper into pieces of approx. 4-8 cm2 and place them in the glass boat of the manual drying oven using tweezers. Store the paper in a closed container under constant conditions. The ambient conditions (humidity) have a strong influence on the water content of the samples.

METTLER TOLEDO Page 65 of 80 Karl Fischer Titration Applications 12.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Wood (pine) 0.5 6 8.28 0.7 - Sample preparation: Heat with manual KF Oven 140°C, 200 mL N2/min - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s Virgin wool 0.08 6 9.84 0.24 - Sample preparation: External extraction 2.1 g in 52 g methanol, 1 h at room temperature - Sample input: 3 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Silk 0.07 6 6.83 0.35 - Sample preparation: External extraction 1.1 g in 42 g methanol, 1 h at room temperature - Sample input: 3 mL aliquot with syringe - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 10 s Termination: Delay time 15 s Cellulose powder 0.2 6 4.23 0.66 - Sample input: Weighing boat - KF Titration: Titrant: Two-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s

Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Wood: Wood releases water slowly. A direct titration is not possible. Thus the moisture is evaporated in the drying oven at 140°C and the vapor is transferred into the KF titration cell using dry nitrogen as the carrier gas. At higher temperatures the wood slowly darkens (decomposes slowly), which leads to an increase in the water content. - Wool and silk: An external extraction is the method of choice for samples of this nature. - Cellulose powder: The finely crushed cellulose powder releases water reluctantly. A direct titration can be performed with a 15 minutes stir time, with the sample suspended in the KF solvent.

METTLER TOLEDO Page 66 of 80 Karl Fischer Titration Applications 13 Building materials 13.1 General remarks Most minerals and building materials such as cement, plaster of paris and lime cannot be titrated directly, as the water is chemically bound. A further hindrance is that oxides or carbonates will undergo a side reaction with the KF reagents. For these reasons, the water is evaporated in the drying oven and the vapor then transfered into the KF cell by a dry purge gas.

Zeolites: Zeolites are silicates with a three-dimensional structure composed of SiO4 and AlO4 molecules with characteristically dimensioned canals and hollow spaces. These occur naturally; most, however, are synthetic. They are used as molecular sieves (for the partitioning of small molecules such as H2O, NH3, H2S, etc.), ion exchangers (for cations), boiling chips, catalysts (for gas-phase reactions) and as detergents. The zeolites tightly bind water in the spaces. Water is released stepwise at successively higher temperatures. The release is complete at 300-350°C.

Detailed information on the water determination in these products can be found in: - Lindner, B. and Rudert, V., “Verbesserte Methode zur Bestimmung von gebundenem Wasser in Mineralien, Gesteinen und anderen Feststoffen”, Fresenius Z. Anal. Chem. 248, 21-24 (1969) (in German). - Farzaneh, A. and Troll, G., „Quantitative Hydroxyl- and H2O-Bestimmungsmethode für Minerale, Gesteine und anderen Feststoffe“, Fresenius Z. Anal. Chem. 287, 43-45 (1977) (in German). - Rechenberg, W., „Bestimmung des Wassergehaltes in Zement“, Zement-Kalk-Gips 29, 512-516 (1976) (in German).

13.2 Coulometric determinations

Sample m n H2O srel Method [g] [%] [%] Cement 0.4 6 0.820 2.2 - Sample preparation: Heat with manual KF Oven 300°C, 200 mL N2/min - KF Titration: Cell with diaphragm 100 mL Coulomat AG Oven, anolyte 5 mL Coulomat CG, catholyte Stirring time: 10 s Termination: Fixed time 1200 s, delay time 60 s Generation rate: Normal, manual start, online drift

METTLER TOLEDO Page 67 of 80 Karl Fischer Titration Applications 13.3 Volumetric determinations

Sample m n H2O srel Method [g] [%] [%]

Plaster of Paris 0.5 6 6.04 0.5 - Sample preparation: Heat with manual KF Oven 300°C, 200 mL N2/min - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 600 s Termination: Delay time 15 s Knifing filler 0.5 5 5.49 1.2 - Sample preparation: Heat with manual KF Oven 300°C, 200 mL N2/min - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 900 s Termination: Delay time 15 s Zeolite S90 0.3 6 12.4 0.43 - Sample preparation: Heat with manual KF Oven ground 300°C, 200 mL N2/min - KF Titration: Titrant: One-component 5 mg/mL Solvent: 40 mL methanol Stirring time: 600 s Termination: Delay time 15 s

Comments - These applications were performed with a fixed delay time to stop the titration (DL35 titrator). The termination parameter can be also set to relative drift stop, if suitable. - Plaster of Paris: Plaster releases the water slowly. A direct titration of the suspension in solvent is still possible though. After a stir time of 25 minutes, nearly the same moisture content (5.8 %) was determined using this method as was obtained with the drying oven (6.0%). - Knifing filler: The filler binds water tightly. The water is not released completely in a direct titration (result of a direct titration with a 75 minute stir time: 2.4 %). Due to the tight binding of the water, the sample must be heated to 300°C for 15 minutes in the drying oven. - Zeolite: Zeolite releases water completely at temperatures greater than 300°C. Drying temperature: 300°C, drying time: 10 minutes. - Cement: Cement contains CaO which reacts with the KF reagent. Therefore the moisture is evaporated in the drying oven at 300°C and the vapor is transfered into the titration cell using a dry purge gas. The coulometric method is used due to the low moisture content.

METTLER TOLEDO Page 68 of 80 Karl Fischer Titration Applications 14 References [1] FISCHER, K. , „Neues Verfahren zur massanalytischen Bestimmung des Wassergehaltes von Flüssigkeiten und festen Körpern“, Angew. Chem. 48, 394 - 396 (1935). [2] WERNIMONT, G.; HOPFKINSON, F.J., „The Dead-Stop End Point – As Applied to the Karl Fischer Method for Determining Moisture“, Ind. Eng. Chem., Anal. Ed. 15 (4), 272 – 274 (1943). [3] See literature references in: EBERIUS, E., „Wasserbestimmung mit Karl-Fischer-Lösung“, 1. Auflage, Verlag Chemie Weinheim (1954) / (in German). [4] MEYER, A.S.; BOYD, C. M., “Determination of Water by Titration with Coulometrically Generated Karl Fischer Reagents”, Anal. Chem. 31 (2), 215 – 219 (1959). [5] BUNSEN, R.W., Liebigs Ann. Chem. 86, 265 (1853). [6] SMITH, D.M.; W. M. BRYANT; J. MITCHELL jr., “Analytical Procedures Employing Karl Fischer Reagent I. Nature of the Reagent”, J. Amer. Chem. Soc. 61, 2407 - 2412 (1939). [7] VERHOFF, J.C.; E. BARENDRECHT, “Mechanism and reaction rate of the Karl Fischer titration reaction. Part I. Potentiometric measurements”, J. Electroanal. Chem. 71, 305-315 (1976). VERHOFF, J.C.; E. BARENDRECHT, “Mechanism and reaction rate of the Karl Fischer titration reaction. Part II. Rotating ring disk electrode meassurement“, J. Electroanal. Chem. 75, 705-717 (1976). VERHOFF, J.C.; E. BARENDRECHT, “Mechanism and reaction rate of the Karl Fischer titration reaction. Part V. Analytical implications”, Anal. Chim. Acta. 94, 395-403 (1977). VERHOFF, J.C.; KOK, W. Th., „Mechanism and reaction rate of the Karl Fischer titration reaction. Part III. Rotating ring disk electrode meassurement - comparison with the aqueous system“, J. Electroanal. Chem. 86, 407-415 (1978). VERHOFF, J.C.; COFINO, W. P., “Mechanism and reaction rate of the Karl Fischer titration reaction. Part IV. First and second order catalytic currents at a rotating disk electrode“, J. Electroanal. Chem. 93, 75-80 (1978). [8] SCHOLZ. E., “Karl-Fischer Reagentien ohne Pyridin“, Fresenius Z. Anal. Chem. 303, 203 - 207 (1980). SCHOLZ. E., „Karl-Fischer Reagentien ohne Pyridin Genauigkeit der Wasserbestimmung“, Fresenius Z. Anal. Chem. 306, 394 - 396 (1981). SCHOLZ. E., „Karl-Fischer Reagentien ohne Pyridin. Einkomponenten Reagentien“, Fresenius Z. Anal. Chem. 309, 30-32 (1981). SCHOLZ. E., „Karl-Fischer Reagentien ohne Pyridin. Neue Eichsubstanzen“, Fresenius Z. Anal. Chem. 309, 123-125 (1981). SCHOLZ. E., „Karl-Fischer Reagentien ohne Pyridin. Zweikomponenten Reagentien mit Imidazol“, Fresenius Z. Anal. Chem. 312, 460-464 (1982). [9] WÜNSCH, G; SEUBERT A., “Stöchiometrie und Kinetik der Karl-Fischer –Reaktion in Methanol als Reaktionsmedium”, Fres. Z. Anal. Chem. 334, 16-21 (1989) (in German). SEUBERT A.; WÜNSCH, G; “Nebenreaktionen in Karl-Fischer-Reagenzien ”, Fres. Z. Anal. Chem. 334, 256-260 (1989) (in German). Both publications from: SEUBERT, A., “Untersuchungen zur Karl-Fischer-Reaktion”, Diploma-Thesis at the University of Hannover / Germany, 1988 (in German). [10] GRÜNKE, S.; WÜNSCH, G., „Kinetics and stoichiometry in the Karl Fischer solution“, Fresenius J. Anal. Chem 368, 139-147 (2000). GRÜNKE, S., “Reaktionsmechanismen in der Karl-Fischer-Lösung“, PhD. Thesis University of Hannover, 1999 (in German).

METTLER TOLEDO Page 69 of 80 Karl Fischer Titration Applications

14.1 Additional literature HYDRANAL®-Manual, „Eugen Scholz Reagents for Karl FischerTitration“, Sigma-Aldrich Laborchemikalien GmbH, D- 30918 Seelze / Germany, 2001. SCHOLZ. E., „Karl Fischer Titration“, Springer Verlag Berlin, 1984. WIELAND, G., „Wasserbestimmung durch Karl-Fischer-Titration - Theorie und Praxis“, GIT Verlag GmbH, Darmstadt / Germany, 1985. SCHÖFFSKI, K., „Untersuchungen einer Karl-Fischer-Reaktion“; Diploma-Thesis at the University of Hannover / Germany, 1989 (German)

14.2 Official Standards The latest (active) versions of the corresponding standards can be found at:

International Standard organization (ISO): www.iso.org European standards (EN): www.cen.eu ASTM standards: www.astm.org German industry standards (DIN): www.din.de

European Pharmacopoeia (Eur. Ph.): www.pheur.org US Food and Drugs Administration (FDA): www.fda.org US Pharmacopoeia (USP): www.usp.org

METTLER TOLEDO Page 70 of 80 Karl Fischer Titration Applications 15 Appendix 15.1 Formula for the external extraction Principle: • The water amount (in grams) after extraction is equal to the water amount before extraction. • The sample P used for the extraction is not dissolved.

Water amount after extraction = Water amount before extraction

WTOT•(mL + (mP•WP) ) = WP•mP + WL•mL where

WP: Water content in the sample P (% or ppm). R This is what we get from the calculation in the KF method.

WTOT : Water content of the supernatant extraction solvent (% or ppm) x

WL : Blank value (water content of the solvent, % or ppm). f1 mL : Amount of solvent (g) after determining the blank value f2 mP : Amount of sample (g) extracted with the solvent f3

Thus:

1) []()32 ⋅+⋅=⋅+⋅ fffRRffx 213

2) 2 3 ⋅⋅+⋅ = ⋅+⋅ fffRRfxfx 213

3) 3 213 ⋅−⋅=⋅−⋅⋅ fxfffRRfx 2

4) ()−⋅⋅ = 2133 ⋅−⋅ fxffffxR 2 1 5) R = ()21 ⋅−⋅⋅ fxff 2 ()3 xf −⋅ )1( 1 1 1 6) R = (⋅−⋅⋅ fffx 212 ) since −= ()3 −⋅ xf )1( 1− xx −1

1 ⎛ f 2 ⋅ ff 21 ⎞ 7) R = ⎜ x −⋅⋅ ⎟ 1− x ⎝ f 3 f 3 ⎠

100 ⎛ f2 ⋅ ff 21 ⎞ For % : R()% = ⎜ x −⋅⋅ ⎟ since 1 = 100%, 100 − x ⎝ f3 f3 ⎠

106 ⎛ f ⋅ ff ⎞ For ppm : ppmR = ⎜ x 2 −⋅⋅ 21 ⎟ since 1 = 106 ppm. ()6 ⎜ ⎟ 10 − x ⎝ f3 f3 ⎠

METTLER TOLEDO Page 71 of 80 Karl Fischer Titration Applications 15.2 Formula for the external dissolution

Principle:

The water amount (in grams) determined by titration is the sum of the water amounts present in the solvent and in the sample. The sample P used for the external dissolution is completely dissolved into the solvent. Thus, the total volume is increased.

Water amount after dissolution = Water amount before before dissolution

WTOT•(mL + mP) = WP•mP + WL•mL where

WP: Water content in the sample P (% or ppm) R This is what we get from the calculation in the KF method.

WTOT : Total water content (sample + solvent), in % or ppm. x

WL : Blank value (water content in% or ppm) of the solvent. f1 mL : Amount of solvent (g) after determining the blank value f2 mP : Amount of sample (g) extracted with the solvent f3

Thus:

1) []32 +⋅=+⋅ ⋅ fffRffx 213

2) 2 3 +⋅=⋅+⋅ ⋅ fffRfxfx 213

3) 3 2 −⋅+⋅=⋅ ⋅ fffxfxfR 213

⎛ + ff 32 ⎞ ⎛ ⋅ ff 21 ⎞ 4) xR ⋅= ⎜ ⎟ − ⎜ ⎟ ⎝ f3 ⎠ ⎝ f3 ⎠

⎛ + ff 32 ⎞ ⎛ ⋅ ff 21 ⎞ For % AND ppm: )(%, xppmR ⋅= ⎜ ⎟ − ⎜ ⎟ ⎝ f3 ⎠ ⎝ f3 ⎠

METTLER TOLEDO Page 72 of 80 Karl Fischer Titration Applications 16 Karl Fischer Reagents 16.1 A choice of manufacturers Here are given the main manufacturers of Karl Fischer reagents. Beside these, there are many small local producers in different regions.

Company Karl Fischer reagents line Link

Sigma-Aldrich HYDRANAL® www.sigmaaldrich.com

VWR/Merck apura® www.vwr.com

EMD Chemicals Aquastar® Karl Fischer www.emdchemicals.com

Mallinckrodt-Baker HYDRA-POINT™ www.mallbaker.com

Fisher Scientific Aqualine™ www.fishersci.com

GFS Chemicals Watermark® www.gfschemicals.com

Mitsubishi Chemical Group AQUAMICRON® www.mcckf.com

Photovolt AQUATEST www.photovolt.com

16.2 Coulometry: A choice of reagents and solvents To determine its total water content, the sample must completely dissolve in the anolyte. If the sample does not completely dissolve, an emulsion is formed. In this case, part of the water content is not measured, i.e. the water content determined is lower than the actual water content of the sample. This means that if an emulsion is formed in the anolyte, the anolyte must be immediately replaced. A number of different anolytes are available to dissolve the various types of samples encountered in practice. Alternatively, more solvent can be added to the anolyte

16.2.1 For samples soluble in methanol or ethanol These include: hydrocarbons (to C10), chlorinated hydrocarbons (to C10), alcohols, ethers, esters, nitrocompounds, acetamide, etc. For cells with diaphragm: - Methanol-based reagents: anolyte: HYDRANAL® Coulomat AG (SIGMA-ALDRICH) apura® combiCoulomat frit (VWR/Merck) catholyte: HYDRANAL® Coulomat CG (SIGMA-ALDRICH) apura® CombiCoulomat frit (VWR/Merck) - Ethanol-based reagents: anolyte: HYDRANAL® Coulomat E (SIGMA-ALDRICH) catholyte: HYDRANAL® Coulomat E (SIGMA-ALDRICH)

METTLER TOLEDO Page 73 of 80 Karl Fischer Titration Applications For cells without diaphragm: - Methanol-based reagents: electrolyte: HYDRANAL® Coulomat AD (SIGMA-ALDRICH) apura® CombiCoulomat fritless (VWR/Merck)

16.2.2 For samples poorly soluble in methanol or ethanol These include: etherial oils, edible oils, ointments, hydrocarbons (C10 to C20), etc. Octanol or hexanol can be added to the anolyte to improve the solubility of these samples. This is limited to a maximum of 30% to prevent the conductivity from falling too low. For cells with diaphragm: - Methanol-based reagents: anolyte: HYDRANAL® Coulomat AG-H (SIGMA-ALDRICH) contains approx. 30% hexanol apura® CombiCoulomat frit (VWR/Merck) + hexanol to max. 30% catholyte: HYDRANAL® Coulomat CG (SIGMA-ALDRICH) apura® CombiCoulomat frit (VWR/Merck) For cells without diaphragm: - Methanol-based reagents: electrolyte: HYDRANAL® Coulomat AD (SIGMA-ALDRICH) + 20% hexanol apura® CombiCoulomat fritless (VWR/Merck) + hexanol to max. 20%

16.2.3 For samples insoluble in methanol or ethanol These include: petroleum oils, transformer oils, silicone oils, hydrocarbons (above C20), etc. Chloroform must be added to the anolyte to ensure the solubility of these samples. This is limited to a maximum of 30% in order to prevent the conductivity from becoming too low. For cells with diaphragm: - Methanol-based reagents: anolyte: HYDRANAL® Coulomat A (SIGMA-ALDRICH) + chloroform to max. 20% HYDRANAL® Coulomat AG (SIGMA-ALDRICH) + chloroform to max. 30% apura® CombiCoulomat frit (VWR/Merck) + chloroform to max. 30% catholyte: HYDRANAL® Coulomat CG (SIGMA-ALDRICH) apura® CcombiCoulomat frit (VWR/Merck) For cells without diaphragm: - Methanol-based reagents: electrolyte: HYDRANAL® Coulomat AD (SIGMA-ALDRICH) + chloroform to max. 30% apura® CombiCoulomat fritless (VWR/Merck) + chloroform to max. 30%

METTLER TOLEDO Page 74 of 80 Karl Fischer Titration Applications 16.2.4 For ketones and aldehydes Ketones and aldehydes react with methanol to form a ketal or an acetal, respectively, with the formation of water.

Acetal formation: CH3COH + 2 CH3OH → CH3CH(OCH3)2 + H2O

Ketal formation: (CH3)2CO + 2 CH3OH → (CH3)2C(OCH3)2 + H2O Special methanol-free reagents must be used for these substances.

For cells with diaphragm: - Methanol-free reagents: anolyte: HYDRANAL® Coulomat AK (SIGMA-ALDRICH) catholyte: HYDRANAL® Coulomat CG-K (SIGMA-ALDRICH) For cells without diaphragm: - Methanol-free reagents: electrolyte: HYDRANAL® Coulomat AK (SIGMA-ALDRICH)

Notes on ketones: - When changing from normal KF anolytes to ketone reagents, the entire titration cell must be thoroughly cleaned because even traces of methanol can cause serious interference. - If you measure ketones regularly, we recommend the use of a second titration cell. - If you have titrated a number of ketone samples, the drift is higher because of the very slow side reaction. After a longer standby period, it is possible that the anolyte is exhausted after a few days even though no further titrations have been performed. - Choose relatively small samples (approx. 1 mL); with reactive ketones such as cyclohexanone use only 0.2 mL to 0.5 mL. The larger the sample, the more important the effect of side reactions becomes, i.e. the drift increases from sample to sample so strongly that the determination of the endpoint becomes difficult. - The special reagent for ketones can also be used for other samples. It is important to note that alcohols that react with ketones should not be titrated in this reagent.

Notes on aldehydes: - Short chain aldehydes (e.g. acetaldehyde) are oxidized at the anode with the formation of water. Coulometric KF titration cannot be used to determine these compounds; the substances can, however, easily be measured with volumetric KF determination. - Aromatic aldehydes (e.g. benzaldehyde) can also be determined by coulometric KF titration. It should be noted that the bisulfide-addition reaction is very strong with aromatic aldehydes. This is why one should wait for the reverse cleavage reaction of the bisulfide compound before aborting the titration (ensure that the drift drops down again to the initial value).

METTLER TOLEDO Page 75 of 80 Karl Fischer Titration Applications 16.2.5 For acids and bases (pH value) A noisy titration or a sluggish end point can indicate a pH shift. In these cases the pH value of the anolyte should be measured. - Measure the pH with a glass electrode adjusted with aqueous buffers. To do this, take an aliquot of anolyte from the titration cell. The measurement should not be performed in the titration cell because the electrode introduces too much water. - Measurement with moistened indicator paper also gives an approximate indication of the pH value of the anolyte. - With Karl Fischer titrations of acidic and basic samples, the pH value of the anolyte must be adjusted to the range 5.5 to . For acidic samples, e.g. acetic acid, formic acid, etc.: - Use the HYDRANAL® buffer (SIGMA-ALDRICH). The use of imidazole leads to a high pH value in the anolyte. - Fill the anode compartment with approx. 80 mL of anolyte and 20 mL HYDRANAL® buffer. For basic samples, e.g. amines: - Basic samples must be neutralized with salicylic acid or benzoic acid. - Fill the anode compartment with approx. 90 mL of anolyte and 5 g salicylic acid or benzoic acid.

16.2.6 Water standards for Karl Fischer coulometry For direct measurements: - VWR/Merck apura® liquid water standard 0.01% - VWR/Merck apura® liquid water standard 0.1% - SIGMA-ALDRICH HYDRANAL® liquid water standard 0.1 mg/g (100 ppm) - SIGMA-ALDRICH HYDRANAL® liquid water standard 1.00 mg/g (1000 ppm)

For measurements with an oven: - VWR/Merck apura® KF oven standard 1% (tungstate) - SIGMA-ALDRICH HYDRANAL® KF oven standard 5.55% (potassium citrate monohydrate)

These standards are not hygroscopic. The water content is declared on a test certificate supplied with every vial.

METTLER TOLEDO Page 76 of 80 Karl Fischer Titration Applications 16.3 Volumetric analysis: titrants and solvents The sample must be completely dissolved in the anolyte to determine the total water content. If it does not completely dissolve, an emulsion is formed. In this case, part of the water content is not measured, i.e. the result determined is lower than the actual water content. In this case, the anolyte must be immediately replaced. A number of different anolytes are available to dissolve the various types of samples encountered in practice. Alternatively, more solvent can be added to the anolyte

16.3.1 For samples soluble in methanol or ethanol These include: hydrocarbons (to C10), chlorinated hydrocarbons (to C10), alcohols, ethers, esters, nitrocompounds, acetamide, etc. One-component reagents: - Methanol-based: Titrant: HYDRANAL® Composite (SIGMA-ALDRICH) apura® CombiTitrant (VWR/Merck) Solvent: Dry methanol Two-component reagents: - Methanol-based: Titrant: HYDRANAL® Titrant (SIGMA-ALDRICH) apura® Titrant (VWR/Merck) Solvent: HYDRANAL® Solvent - Ethanol-based: Titrant: HYDRANAL® Titrant E (SIGMA-ALDRICH) Solvent: HYDRANAL® Solvent E (SIGMA-ALDRICH)

16.3.2 For samples poorly soluble in methanol or ethanol

These include: etherial oils, edible oils, ointments, hydrocarbons (C10 to C20), etc. Octanol or hexanol can be added to the anolyte to improve the solubility of these samples. This is limited to a maximum of 50%. One-component reagents: - Methanol-based: Titrant: HYDRANAL® Composite (SIGMA-ALDRICH) apura® CombiTitrant (VWR/Merck) Solvent: HYDRANAL® LipoSolver CM, MH (SIGMA-ALDRICH) apura® CombiSolvent Fats (VWR/Merck)

Two-component reagents: - Methanol-based: Titrant: HYDRANAL® Titrant (SIGMA-ALDRICH) apura® Titrant (VWR/Merck) Solvent: HYDRANAL® Solvent CM, Oil (SIGMA-ALDRICH) apura® Solvent Oil & Fats (VWR/Merck)

METTLER TOLEDO Page 77 of 80 Karl Fischer Titration Applications 16.3.3 For samples insoluble in methanol or ethanol These include: petroleum oils, transformer oils, silicone oils, hydrocarbons (above C20), etc. Chloroform must be added to the anolyte to ensure the solubility of these samples. This is limited to a maximum of 50% in order to prevent the conductivity from becoming too low. One-component reagents: - Methanol-based: Titrant: HYDRANAL® Composite (SIGMA-ALDRICH) apura® CombiTitrant (VWR/Merck) Solvent: HYDRANAL® Solver (Crude) apura® CombiSolvent Fats (VWR/Merck) Two-component reagents: - Methanol-based: Titrant: HYDRANAL® Titrant (SIGMA-ALDRICH) apura® Titrant (VWR/Merck) Solvent: HYDRANAL® Solvent CM, Oil (SIGMA-ALDRICH) apura® Solvent Oil & Fats (VWR/Merck)

16.3.4 For ketones and aldehydes Ketones and aldehydes react with methanol to form a ketal, or an acetal, with the formation of water.

Acetal formation: CH3COH + 2 CH3OH → CH3CH(OCH3)2 + H2O

Ketal formation: (CH3)2CO + 2 CH3OH → (CH3)2C(OCH3)2 + H2O Special methanol-free reagents must be used for these substances. One-component reagents: - Methanol-free: Titrant: HYDRANAL® Composite 5 K (SIGMA-ALDRICH) apura® CombiTitrant 5 Keto (VWR/Merck) Solvent: HYDRANAL® Medium K (SIGMA-ALDRICH) HYDRANAL® Working Medium K HYDRANAL® KetoSolver apura® CombiSolvent Keto (VWR/Merck) Notes on ketones: - When changing to ketone reagents, the entire titration cell must be thoroughly cleaned because even traces of methanol can cause serious interference. - If you measure ketones regularly, we recommend the use of a second titration cell. - If you have titrated a number of ketone samples, the drift is higher because of the very slow side reaction. - Choose relatively small samples (approx. 1 mL); with reactive ketones such as cyclohexanone use only 0.2 mL to 0.5 mL. The larger the sample, the more important the effect of side reactions becomes, i.e. the drift increases from sample to sample so strongly that the determination of the endpoint becomes difficult.

METTLER TOLEDO Page 78 of 80 Karl Fischer Titration Applications - The special reagent for ketones can also be used for other samples. It is important to note that alcohols that react with ketones should not be titrated in this reagent. Notes on aldehydes: - Short chain aldehydes (e.g. acetaldehyde) are oxidized at the anode with the formation of water. Coulometric KF titration cannot be used to determine these compounds; the substances can, however, easily be measured with volumetric KF determination. - Aromatic aldehydes (e.g. benzaldehyde) can also be determined by coulometric KF titration. It should be noted that the bisulfide-addition reaction is very strong with aromatic aldehydes. This is why one should wait for the reverse cleavage reaction of the bisulfide compound before aborting the titration (ensure that the drift drops down again to the initial value).

16.3.5 For acids and bases (pH value) A noisy titration or a sluggish end point can indicate a pH shift. In these cases the pH value of the anolyte should be measured. - Measure the pH with a glass electrode adjusted with aqueous buffers. To do this, take an aliquot of anolyte from the titration cell. The measurement should not be performed in the titration cell because the electrode introduces too much water. - Measurement with moistened indicator paper also gives an approximate indication of the pH value of the anolyte. - With Karl Fischer titrations of acidic and basic samples, the pH value of the anolyte must be adjusted to the range 5.5 to 7 (see Section 1.2). For acidic samples, e.g. acetic acid, formic acid, etc.: - Use the HYDRANAL® buffer (SIGMA-ALDRICH). The use of imidazole leads to a high pH value in the anolyte. - Fill the anode compartment with approx. 80 mL of anolyte and 20 mL HYDRANAL® buffer. For basic samples, e.g. amines: - Basic samples must be neutralized with salicylic acid or benzoic acid. - Fill the anode compartment with approx. 90 mL of anolyte and 5 g salicylic acid or benzoic acid. 16.4 Water Standards for Karl Fischer volumetric titration For direct measurements: - VWR/Merck apura® liquid water standard 1 % - SIGMA-ALDRICH HYDRANAL®-Water Standard 100 mg/g ( ppm) For measurements with an oven: - VWR/Merck apura® Water standard oven 1% (tungstate) - SIGMA-ALDRICH HYDRANAL®-Water Standard KF-Oven 5.55% (potassium citrate monohydrate).

These standards are not hygroscopic. The water content is declared on a test certificate supplied with every vial.

METTLER TOLEDO Page 79 of 80 Karl Fischer Titration Applications 17 Hazards and waste disposal tips 17.1 One-component reagents Ingredients: Sulfur dioxide, iodine, buffer (imidazole) and solvent (methanol, 2- methoxyethanol or diethyleneglycolmonomethylether). Safety: Flammable to extremely flammable. Irritant when inhaled. Health hazard when inhaled, swallowed or allowed to contact skin. Keep container tightly closed. Keep away from open flames or sparks. Do not let reagent contact skin or eyes. Disposal: Aas organic solvent. 17.2 Two-component reagents: Titrant: Iodine and solvent (methanol, 2-methoxyethanol, xylene or trichloroethaylene). KF solvent: Sulfur dioxide, buffer (imidyzole) and solvent (methanol, 2-methoxyethanol or diethyleneglycolmonomethylether). Safety: Flammable to extremely flammable. Irritant when inhaled. Health hazard when inhaled, swallowed or allowed to contact skin. Keep container tightly closed. Keep away from open flames or sparks. Do not let reagent contact skin or eyes. Disposal: As organic solvent. 17.3 Reagents for coulometry: Reagent: Iodine, sulfur dioxide, buffer (imidazole), and solvent (methanol, chloroform, tetrachloromethane, 2-methoxyethanol). Safety: Highly flammable. Skin irritant, poisonous when inhaled or swallowed. Keep container tightly closed and far removed from open flames or sparks. Do not let reagent contact skin or eyes. Disposal: As organic solvent. 17.4 Safety data for the KF-components and auxiliary solvents:

Ignition MAC value Remarks temperature Sulfur dioxide 200 ppm Iodine 0.1 ppm oxidant Diethyleneglycol- 87°C monomethylether 2-Methoxyethanol 46°C 5 ppm Flammable. Potential teratogenic. Health hazard when inhaled, swallowed or allowed to contact skin. Irritant when inhaled Methanol 11°C 200 ppm Highly flammable. Poisonous when inhaled or swallowed o-xylene 28°C 100 ppm Flammable. Health hazard upon inhalation Chloroform does not burn 200 ppm Health hazard when inhaled. Irreversible damage possible. 1-Decanol 95°C Irritant for skin and eyes Formamide does not burn 20 ppm Irritant for skin and eyes. Potential teratogenic Toluene 6°C 20 ppm Extremely flammable. Health hazard upon inhalation

MAC: Maximum allowable concentration

METTLER TOLEDO Page 80 of 80 Karl Fischer Titration Applications

The methods in this brochure represent selected, possible application examples. These have been tested with all possible care with the analytical instruments mentioned in the brochure. The experiments were conducted and the resulting data evaluated based on our current state of knowledge.

However, the applications do not absolve you from personally testing their suitability for your intended methods, instruments and purposes. As the use and transfer of an application example are beyond our control, we cannot accept responsibility therefore.

When chemicals and solvents are used, the general safety rules and the directions of the manufacturer must be observed.

Application Brochure 38 Selected METTLER TOLEDO Methods for the Compact Karl Fischer Titrators and Titration Excellence Line This brochure contains selected Karl Fischer METTLER TOLEDO methods for the Compact Coulomeric and Volumetric Titrators as well as for the Titration Excellence Line. These applications represent an ideal start for your own method development for the determination of water content in your samples. Thanks to the method concept and the color touchscreen, operation is made easy. A single sample or an entire series of samples can be started with a touch to the ShortCut: One Click™ Water Determination. In addition, these applications contains detailed information on the water content determination in the most different samples such as for instance, plastics, polymers, and pharmaceutical products.

Compact Volumetric and Coulometric Titrators Titration Excellence Line V20/V30 and C20/C30 Karl Fischer dedicated KF analysis and general instruments titration

www.mt.com For more information

Mettler-Toledo AG, Analytical Sonnenbergstrasse 74 CH-8603 Schwerzenbach, Switzerland Tel +41-44 806 77 11 Fax +41-44 806 73 50