The solvent keeps flowing into the titration vessel after the addition, without the add key being pressed.
Set the solvent bottle up in such a way that the fluid level in the bottle is lower than in the titration vessel.

Solvent is added simultaneously during aspriation.
The drying tube in the titration vessel is blocked:

• Increase hole on the drying tube cover to at least 2 mm
• Refill the drying tube, do not pack to tightly, possibly removing some of the molecular sieve.

The pump conveys no fluid.
Press the addition / aspiration key more strongly.

• In most cases the bottle attachments are not screwed tightly enough to the reagent and waste bottles
• Check all connections for leak-tightness
• Check wether the 803 Ti Stand has been set up correctly

It is difficult to make a general statement concerning how often the molecular sieve should be replaced, because this is dependent on the humidity in the laboratory. Our experience has shown us that replacement every 6 weeks is appropriate in the presence of medium humidity. An increase in drift indicates that the leak-tightness of the cell should be inspected and that the molecular sieve should possibly be replaced.

Tip: Write the date on which you last replaced the molecular sieve on the glass of the drying tube; then you will always know when the next replacement is due.

The molecular sieve can be regenerated at 300 degree of celsius in the drying oven. The regeneration time is at least 24 hours. The molecular sieve is then placed in a desiccator for cooling down and afterwards stored in a glass bottle with an airtight seal.

If the drift during pretitration (conditioning) is too high after the addition of fresh working medium, then this is a sign that the titration cell is not protected sufficiently against humidity. Replace the molecular sieve and the septum. Also check to see whether all of the O-rings are in proper condition.

Dosing and Exchange Units require regular checks and must be taken apart from time to time and cleaned if necessary. You will find a detailed description in the corresponding manuals.

Rinse the tubing by means of the Preparation/"Prep" command before the first measurement if the system has not been used over night or for even longer periods. The titrant can change over time in the plastic tubing, which will falsify your results. In addition, the "Prep" procedure removes any air bubbles that may be present in the tubing.

If a titrator with Dosing Unit is not used for several days, then it would be best to empty the Dosing Unit with the Empty command. That way one will avoid the formation of crystals in the Dosing Unit, the tubing and the buret tip (anti-diffusion valve).

If work is being done with an Exchange Unit, then one should immerse the buret tip in methanol in order to avoid crystallization. To accomplish this, one can either fill the storage vessel with methanol or continue to leave the buret tip in the closed cell filled with working medium, where it is immersed in the working fluid.

The Exchange Unit, the tubing and the buret tip should all first be subjected to visual inspection for crystals before the system is returned to operation. If crystals are present, then the tubing should be flushed with methanol or ethanol and then dried afterwards.

If no crystals are visible, then a small amount of reagent should nevertheless be dosed in order to check whether crystals have formed in the flat cock of the Exchange Unit. If reagent escapes from the flat cock, then the Exchange Unit must be taken apart and cleaned (see manual for the Exchange Unit).

Afterwards, flush the Exchange Unit by means of the Preparation/"Prep" command in order to make it ready for operation again.

The balance should have a minimum resolution of 0.1 mg.

Soiled platinum pins can be cleaned with an abrasive agent such as Alox (polishing set 6.2802.000) or toothpaste. Afterwards, they should be flushed with ethanol or methanol. It is often sufficient to wipe off the indicator electrode with a soft paper tissue. This is particularly important when the electrode was immersed previously in an overtitrated solution, because deposits can form there that are not visible but nevertheless lead to overtitrations. If the sample itself adheres to the electrode, clean it off, either with an appropriate solvent or mechanically.

Clean the platinum pins on the indicator electrode with a soft paper tissue and take care to ensure that the pins do not become deformed (pressed together) while doing so; they should always be parallel to one another.

It is difficult to make a general statement. The frequency of titer determination depends on a variety of different factors:

• leak-tightness of the titration system
• selection of the titrant
• humidity in the laboratory
• changes in room temperature
• requirements regarding exactness of the result

In principle, we recommend that titer determination be carried out daily in the presence of high requirements concerning exactness. If no relevant changes are found, then the frequency can be reset to weekly. It should be noted that in general the titer of one-component reagents drops faster than that two-component reagents.

Leading reagent manufacturers offer certified water standards for titer determination. We recommend the use of certified standards; they are simpler to handle, because larger sample volumes can be added here than with titer determination using distilled water.
Standard solutions can be obtained with the water contents 10.0 ±0.1 mg/g and 5 ±0.2 mg/mL, although the latter is not certified.
The possibility also exists of carrying out the titer determination with distilled water or with di-sodium tartrate dihydrate.

See Video 1: Titer determination with standard

The sodium tartrate dihydrate dissolves incompletely when several determinations are carried out in the same working medium because of the limited solvent capacity of methanol. It is recommended that the working medium be replaced at shorter intervals.

 

The di-sodium tartrate dihydrate has probably not dissolved completely. The extraction time should be increased in order to prevent this. In addition, care should be taken to ensure that the sample is not left behind on the side of the vessel or on the electrode. Increasing titer may be traced back to the limited solvent capacity of the methanol. It is recommended that the working medium be replaced at shorter intervals.

See Video 1: Titer determination with di-sodium tartrate dihydrate

Another reason may be that the sample size selected was not optimum.

Recommended substance weight for titer determination

Standard	Sample Size for 5 mL buret
	Titrant 1	Titrant 2	Titrant 5
Standard 10 mg/g	0.05-0.45 g	0.1-0.9 g	0.25-2.25 g
Na Tartrate 156.6 mg/g	–	0.006-0.057 g	0.03-0.27 g
Water 1000 mg/g	–	–	~ 0.02 g
Water	–	–	~ 20 uL

Standard	Sample Size for 10 mL buret
	Titrant 1	Titrant 2	Titrant 5
Standard 10 mg/g	0.1-0.9 g	0.2-1.8 g	0.5-4.5 g
Na Tartrate 156.6 mg/g	0.02-0.057 g	0.013-0.115 g	0.03-0.287 g
Water 1000 mg/g	–	~0.015 g	~ 0.025 g
Water	–	~15 uL	~ 25 uL

Standard	Sample Size for 20 mL buret
	Titrant 1	Titrant 2	Titrant 5
Standard 10 mg/g	0.2-1.8 g	0.4-3.6 g	1-9 g
Na Tartrate 156.6 mg/g	0.02-0.175 g	0.026-0.23 g	0.065-0.58 g
Water 1000 mg/g	–	0.02-0.035 g	0.02-0.09 g
Water	–	15-35 uL	15-90 uL

It is also possible to determine the titer with water. It should be noted in this connection that one must work with very small volumes. The handling takes a bit of getting used to in the beginning and requires a great deal of practice. The detailed procedural method can be seen in the following videos:

Video 1: Titer determination with water (volume addition)
Video 2: Titer determination with water (back-weighing)

If the drift during pretitration (conditioning) is too high after the addition of fresh working medium, then this is a sign that the titration cell is not protected sufficiently against humidity. Replace the molecular sieve and the septum, also check whether all of the O-rings are in proper working condition (damage, cracking).
 

If the drift is too high after the completion of a sample titration, then this is usually because of other reasons:

• The sample has not dissolved completely during titration and continues to release water. An increase of the extraction time is to be recommended or there are sample pretreatment steps to be carried out. Read for more details in the following FAQ: 'What to do to improve the releasing of water?'
• The sample reacts with the KF a side reaction takes place. Depending on the type of the side this can be suppressed not only with special but also by means of appropriate titration methods (KF oven method, titration in the cold or external extraction)
• The sample has shifted the pH value. Read for more details in the following FAQ: 'How can I prevent side reactions or pH value shifts?'

The sample size depends on the water content of the sample. In principle, the sample size should be selected in such a way that the titrant consumption lies between 10% and 90% of the buret volume. This means that if work is being done with a 10 mL Exchange or Dosing Unit, the utilization level of the titrant should be between 1 and 9 mL.

In addition, one should take care to ensure that the sample weight selected is not too low. The weighing error will be too great on the one hand, while on the other the sample will no longer be representative. In cases of high water content, it may be better to use a Dosing or Exchange Unit with a greater buret volume in order to avoid filling the Exchange Unit during the titration.

If the sample does not dissolve completely in the working medium, then the sample size has perhaps been selected too high and exceeds the dissolving capacity of the working medium. This can be remedied through the use of a smaller Dosing/Exchange Unit or of a titrant with a lower water equivalent, e.g. 2 mg/mL or 1 mg/mL. Despite the lower initial weight, the titrant consumption will then achieve 10...90% of the buret volume.

As a general rule: When the titrant consumption is low, one should work with a lower titer (e.g. Titrant 2, Titrant 1), and with a high titer when titrant consumption is a high (Titrant 5).

Recommended sample size

Info: Titer 5 means: 1 mL of the reagent reacts with approximately 5 mg of water, the precise value is established by the titer determination.

In cases of samples with low solubility, one of the following procedures is recommended:

A) Use of a solubility promotor

B) Titration at a higher temperature

C) Mechanical pretreatment

• Homogenizer
• Mortar
• Lab mill (cooled)

These procedures can also be combined with each other. Detailed information as to which procedure is suitable for which sample can be found in the Metrohm KF monograph.

• A side reaction is taking place. Read for more details in the following FAQs: 'How to recognize a side reaction?' 'How to prevent side reactions?'
• The cell is not leak-proof. In this case carry out the following measures: replace molecular sieve, O-rings and septum
• There are deposits on the indicator electrode. Read for more details in the following Installation FAQ: How to clean the indicator electrode?

The following characteristics indicate the presence of a side reaction:

• High drift after the titration has been completed, slow endpoint recognition or no endpoint found at all
• The water contents determined are not independent of the sample weight
• The results are (depending on the type of side reaction) too high or too low
• The water recovery rate found when spiking samples is not within 100 ±3%

See Figure 1:Curve for a titration with side reaction.

The red curve shows the typical curve shape for a a KF titration with side reaction. The curve exhibits a continuous slight rise after the water has been titrated, which means that there is a steady consumption of reagent to be observed as a result of the side reaction, which leads to incorrect results.

See Figure 2: Substances that can cause side reactions.

• Use of so-called K-reagents with samples that contain ketones and aldehydes
• Methanol-free reagents when the sample reacts with the methanol of the reagent
• When the sample causes a pH shift to the acid range, buffering can be accomplished with solvent and imadazole
• When the sample causes a pH shift to the alkali range, adding ca. 5 g salicylic acid or benzoic acid to the working medium helps to adjust the optimum pH range for the KF reaction
• To suppress the side reaction, peroxides can be titrated in conditions of cold, e.g. at -20 degree of celsius to -60 degree of celsius (set Ipol to 5 µA)
• If the sample (e.g. motor oil) contains various substance that react directly with the KF reagent then a KF oven will be of help

You will find more detailed information as to which procedure is suitable for which sample in the Metrohm KF monograph

A) It is usually sufficient to clean the platinum pins of the electrode with a soft paper tissue. One should however take care to ensure while doing so that the platinum pins are not pressed together. They should always be parallel to one another.

B) In addition, the antidiffusion valve on the buret tip should be checked.

See the photo:  good buret tip in comparison with defective buret tips

C) Check if you use the recommended parameters:

D) A further reason for overtitration could be that the initial solution no longer contains sufficient methanol or ethanol. In such a case it is worthwhile to renew the cell solution and/or to check whether or not the proportion of solubilities promoters has not been chosen too high when these are being used. Read for more details in the following FAQ: 'How often must the KF working medium be replaced?'

See Table: Solubility Promoters for volumetric titration

Additional measures:

E) Increasing the stirring rate

F) Check whether the platinum pins of the electrode are immersed completely in the cell solution

G) Check whether the correct reagents were combined: if the titrant of the two-component reagent is used with methanol, the KF reaction cannot take place, because no SO₂ is available for the reaction.

In the case of two-component-reagents, the water capacity is limited by the amount of SO₂ and buffer in the solvent. Decreasing pH values and decreasing SO₂ concentrations result in a drop in titration speed. Reagent manufacturers have published the following guideline: The cell solution should be replaced after approximately 30 mL of titrant have been used in the case of cell solutions with 20...25 mL of solvent.

One-component reagents have no limiting water capacity, because SO₂ and buffer are added with the titrant. Care must nevertheless be taken to ensure that the methanol content does not drop below 25%, because otherwise the endpoint will be shifted.

Info: The water capacity indicates how much water can be determined in a particular titration solution.

In the case of two-component-reagents, the water capacity is limited by the amount of SO₂ and buffer in the solvent. Decreasing pH values and decreasing SO₂ concentrations result in a drop in titration speed. Reagent manufacturers have published the following guideline: In the case of cell solutions with 20...25 mL of solvent, the cell solution should be replaced after approximately 30 mL of titrant has been used.

The following causes can be responsible for poor reproducibility:
Cause	Remedy
Sample amount too small (weighing error)	Increasing the sample amount; in cases of high water content, it may be better to use a Dosing or Exchange Unit with a greater buret volume in order to avoid filling of the Exchange Unit during the titration.
Insufficient use of titrant	Increase sample size and select Exchange Unit or Dosing Unit so that at least 10% of the buret volume is used or use a titrant with a lower titer.
Inhomogeneous sample	Increase sample amount  Homogenize sample prior to the determination
Sample does not dissolve completely	Add smaller amounts of sample or use solubility promoters
Sample remains behind on the wall of the vessel	Slight shaking of the titration vessel
Overtitration	Read for more details in the following FAQ: 'Overtitration took place, i.e. the titrated solution is dark yellow to brown instead of light yellow. How to prevent this?'

Too high results can occur because of the following causes:
Cause	Remedy
Side reaction	Read for more details in the following FAQ: 'How to recognize a side reaction? How to prevent side reactions?'
The sample is hygroscopic	Here we recommend optimization of the sampling and of the sample storage as well as working speedily during the determination
The sample is inhomogenous	Increase sample amount Homogenize sample prior to the determination
The titer has changed since the last determination	Redetermine titer
Overtitration	Read for more details in the following FAQ: 'Overtitration took place, i.e. the titrated solution is dark yellow to brown instead of light yellow. How to prevent this?'
Air bubbles in the tubing	Flush the tubing system with the Preparation/"Prep" command
Non-leakproof tubing system	check for crystals in the tubing system check whether tubing is connected tightly check tubing connections

Too low results can occur because of the following causes:
Cause	Remedy
Side reaction	Read for more details in the following FAQ: 'How to recognize a side reaction? How to prevent side reactions?'
The sample is inhomogenous	Increase sample amount Homogenize sample prior to the determination
The sample dissolves incompletely	Read for more details in the following FAQ: 'What can I do to improve the releasing of water?'
The pretreatment of the sample is not optimal.	Read for more details in the following FAQ: 'What to do to improve the releasing of water?'
Titer is incorrect	Redetermine the titer with certified standards
Titration will stop too soon	Reduce Stop Drift or work with a relative Start Drift
The sample gives off water even during sample preparation	Here we recommend optimization of the sampling and sample storage, as well as rapid working during the determination sequence

Trend within the results
One reason for decreasing water contents in a series consists of the fact that the solvent is exhausted and must be replaced. Decreasing water contents within a series can also occur with moist samples which give off water during sample handling. Here we recommend optimization of the sampling and sample storage, as well as rapid working during the determination sequence.

A further possible reason for apparently decreasing water contents within a series is that the buret has not been prepared. The titrant in the cylinder and in the tubing is exposed to humidity, which can result in increased titrant consumption at the beginning of a series.

Increasing water contents within a series can occur with hygroscopic samples. Here we recommend optimization of the sampling and sample storage, as well as rapid working during the determination sequence.

Tip: you can recognize very quickly whether your sample is either releasing water, is hygroscopic or unproblematic by determining the water content of the sample immediately and then leaving the sample to stand in a beaker in the laboratory air and determining the water content again after 10, 30 and 60 minutes.

* Methanol or Ethanol
** Diethylene glycol monoethyl ether

The following manufacturers offer KF reagents:

Sigma-Aldrich:
www.sigmaaldrich.com

Merck:
www.merck.de

USA: Metrohm USA (Hydranal Reagents)
www.metrohmna.com

Singapore: Metrohm Singapore (Hydranal Reagents)
www.metrohm.com.sg

France: Metrohm France (réactifs Hydranal)
www.metrohm.fr

GLP (Good Laboratory Practice) requires among other things periodic checking of the analytical instruments for their reproducibility and correctness on the basis of Standard Operating Procedures (SOPs). It is recommended that the user validates the Metrohm instruments as a single overall measurement system, i.e. by performing a series of sample determinations with certified standard solutions, and evaluates the results critically with statistical methods.

For more information please contact your Metrohm representative.

It is recommended to repeat annually the testing of the instruments used for determining water content according to the Karl Fischer method. If an instrument is in continuous use, then more frequent testing intervals may be called for, e.g. every 6 months or even every 3 months.

Unscheduled validations are called for when one or more components of the titration system have been replaced.