WHITE PAPER

Moisture in petroleum products according

to ASTM D6304
Knowledge of the water content in lubricating oils, prevent damage to infrastructure and ensure safe
additives, and similar products is important in the operation.
manufacturing, purchase, sale, or transfer of petro-
leum products to help estimate their quality and This White Paper explains the easy determination
performance characteristics. of moisture in petroleum samples by coulometric
Karl Fischer titration according to the three proce-
In lubricating oils, the presence of moisture could dures outlined in ASTM Method D6304. A compar-
lead to premature corrosion and engine wear while ison is given between the procedures to determine
increasing the debris load, resulting in diminished which is most suitable for different sample types.
lubrication and premature plugging of filters. In
addition, moisture can impede the effect of addi-
tives and promote growth of deleterious bacterial
organisms. Knowing about the water content can

1
 MOISTURE DETERMINATION ACCORDING SAMPLING
TO ASTM D6304
Collecting a representative product sample for mois-
ASTM Method D6304 consists of three procedures for ture analysis is essential for accurate and precise water
determining moisture in petroleum samples by coulo- determinations. Sample collection should be
metric Karl Fischer titration. Procedure A describes a performed according to ASTM Practices D4057 and
direct injection where a measured quantity of the D4177, and handled and mixed in accordance with
sample is introduced directly into the coulometric Karl Practice D5854. Care must be taken during sample
Fischer titration cell. This procedure is recommended collection and analysis to avoid moisture contamina-
for samples which are soluble in the titration solvent tion coming from the sample container, the atmo-
but do not interfere with the Karl Fischer reaction. sphere, or transfer equipment.

Procedures B and C can be used to analyze samples Water in hydrocarbon samples can readily separate
that do not readily dissolve in Karl Fischer reagent, into two phases when the solubility limit is exceeded.
including viscous samples or samples with compo- The water solubility depends on the sample matrix,
nents that are expected to interfere with the Karl water concentration, and the temperature. Samples
Fischer reaction. Procedures B and C can additionally must be thoroughly mixed prior to withdrawing an
be used to determine water content in samples that aliquot from the sample container. If phase separation
can also be analyzed by Procedure A. occurs after mixing, then the sample is not suitable
for testing the water content. Samples should be
In both Procedures B and C, the sample is heated in a analyzed for moisture immediately after mixing or
closed system to release moisture, which is transferred homogenizing to minimize any water separation.
to the coulometric Karl Fischer cell by an inert carrier
gas and titrated. The following table (Table 1) lists a Once a sample is taken from the original container,
comparison between the different procedures. the analyst should either use the entire portion of the
test specimen for the analysis, or dispose of the excess.
The portion removed from the sample container
should not be reintroduced back into the original
sample container to avoid sample contamination.

Table 1. Comparison of the different ASTM D6304 procedures.

Factor Procedure A Procedure B Procedure C

Analysis technique Direct injection Oven evaporator Water evaporator
Required sample solubility
Good Independent Independent
in KF reagent
Suitable for samples with
No Yes Yes
matrix interference
Automation possibility No Yes No
Automatic reagent
Yes Yes Only titration cell
exchange
When phase separation
occurs, KF reagents are
Reagent exchange When KF reagents are When KF reagents are
exhausted, or when
frequency exhausted exhausted
sample represents ¼ of
the volume in the cell
All petrochemical All petrochemical
Liquid samples e.g., diesel
Sample type samples with thermal samples with thermal
fuel, aromatics, jet fuel
decomposition > 100 °C decomposition > 90 °C

2
 DIRECT INJECTION A sample for the analysis is then aspirated into the
(PROCEDURE A) same syringe and weighed on the balance. An appro-
priate amount of sample containing at least 50 µg
The direct sample injection into the titration cell water is injected into the titration cell, and the syringe
(Procedure A) is recommended for low viscosity is weighed again to determine the exact sample size.
samples without expected interferences, such as diesel
fuel, aromatics, or jet fuel. An aliquot of known mass
or volume is injected into the conditioned titration
cell of a coulometric Karl Fischer apparatus, where it
is titrated automatically and the results calculated as
mass or volume percent. Method D6304 permits the
use of coulometric generator electrodes with and
without diaphragm (Figure 1).

Before titration of the sample, the titration cell is

conditioned. This ensures that no environmental mois-
ture is present within the titration cell prior to titration.
Conditioning, or titrating the titration cell to dryness,
is carried out while stirring. This conditioning step Figure 1. Coulometric Karl Fischer Generator electrode with
must be performed before each determination. When diaphragm (left) and without diaphragm (right).
the titration system is conditioned, the titration can
begin.
Disposable syringes may be used to introduce sample
To obtain good precision, it is necessary to rinse the into the Karl Fischer cell as long as there is no interac-
syringe with each new sample one or two times prior tion between the syringe material and the sample.
to injecting into the titration cell. For this, a small One disposable syringe should be used per sample.
amount of sample is aspirated into the syringe and
discarded again. This procedure ensures the removal When using direct injection, it is important to inspect
of any water adhering to the syringe. the titration cell to make sure the sample is dissolved
and does not separate into layers. The number of

Figure 2. A coulometric Karl Fischer titration system suitable for ASTM D6304 Procedure A.

3
samples which can be analyzed without phase sepa- EVAPORATION TECHNIQUES
ration is dependent on the volume and type of sample (PROCEDURES B & C)
material injected, the volume of Karl Fischer reagent,
and the sample solubility in the reagent. As a rule of An oven (Procedure B) or water evaporator accessory
thumb, when the volume of sample injected into the (Procedure C) can be used to analyze samples that do
cell exceeds one fourth of the total volume in the not readily dissolve in Karl Fischer reagent, viscous
titration cell, the reagent in the anode compartment samples, and samples with components that are
should be exchanged. expected to interfere with the Karl Fischer reaction.

Karl Fischer reagents may be modified with toluene

or other solvents depending on the sample composi-
tion. Most reagent manufacturers publish acceptable OVEN EVAPORATOR
solvents and the proportion that may be added to the (PROCEDURE B)
anode solution to ensure the release of all moisture
as well as to enhance sample solubility, thus preventing The oven evaporator is an auxiliary device that heats
contamination of the Karl Fischer cell. Often, samples the sample in a heating chamber. The water in the
that may require a co-solvent to improve solubility sample vaporizes and is carried off by a dry non-reac-
also contribute to clogging or coating the diaphragm tive (inert) carrier gas through a heated tube into the
of the generator electrode. These sample types Karl Fischer titration cell. There the vaporized water
are best analyzed using Procedures B and C of is titrated while the sample and contaminants remain
ASTM D6304. in the sample vial.

The Metrohm 874 Oven Sample Processor, 860 KF

Thermoprep, and 885 Compact Oven Sample Changer
are suitable for ASTM D6304 Procedure B. The 874
Oven Sample Processor can perform a temperature
gradient test to determine the optimal temperature

Figure 3. An oven evaporator with a coulometric Karl Fischer titration system suitable for ASTM D6304 Procedure B.

4
Figure 4. Operating principle of the oven evaporation technique.

for removing water without degrading the sample. A water into the titration cell. The temperature of the
detailed description to assist selection of the appro- oven can be varied according to the temperature
priate temperature can be found in the Appendix of stability of the sample. The optimal sample size
test method ASTM D6304. depends on the water content of the sample. The
absolute amount of water transferred to the titration
In Procedure B (Figure 4), a representative portion of vessel is recommended to be in the range of
the sample is placed into a sealed glass vial and heated 300–5000 μg. If the absolute amount of water for a
in an oven to extract any water present into the head- sample cannot be reduced (e.g., smaller sample size),
space of the vial. The vaporized water in the head- also larger amounts of water (> 5000 μg) can be deter-
space is then carried into the conditioned Karl Fischer mined using the oven technique.
titration cell by a dry carrier gas where it is titrated.
For samples which do not readily release their water,
As with Procedure A, conditioning of the titration cell such as crude oil, a co-solvent may be used. Co-sol-
is required to remove any ambient moisture within vents such as toluene enhance water extraction from
the system. Conditioning or titrating the titration cell the sample, resulting in faster analysis time and higher
to dryness is carried out under stirring in the so-called reproducibility. In cases where a co-solvent is used, it
conditioning position. This step must be carried out is necessary to fill the sample vials used for blank deter-
before every determination. The oven is initially mination with the same amount of co-solvent as used
heated up to the defined temperature. After this for the sample determination.
temperature is reached and the titration system is
conditioned, the first measurement can start. The carrier gas transports the released water into the
titration cell. In principle, the gas used is not so
Prior to a titration series, a blank value is determined important. However, due to reactions with oxygen,
using empty sample vials. This is necessary because the thermal stability of organic substances at higher
apart from the water in the sample, the sample vial temperatures is often poor when using dry air as
also contains atmospheric humidity. The blank value carrier gas. Inert gases, e.g. nitrogen, offer more flex-
is considered in the calculation of the water content ibility and reliability. Measurements in liquid samples
(subtracted). If large sample sizes are used, it might (oil, fuel, solvents) should be carried out using gas flow
be necessary to correct the blank value by using a rates of 80 mL/min or higher. Furthermore, it is recom-
relative blank for the sample determination. mended that the inlet needle is immersed in the
sample and the gas is allowed to flow through the
For the sample analysis, the dried carrier gas is passed sample.
through the sample vial and transfers the released

5
 WATER EVAPORATOR PROCEDURE A, B, OR C?
(PROCEDURE C) COMPARISON OF PROS AND CONS

In the water evaporator procedure, an aliquot of All three procedures discussed have advantages and
sample is transferred into a heated chamber containing disadvantages. This section considers the performance
toluene (Figure 5). The water vaporizes together with of the three procedures under the aspects mentioned
toluene in an azeotrope distillation. The azeotrope in Table 1.
vaporizes and is transferred into the Karl Fischer titra-
tion cell by a dry non-reactive carrier gas.
− SAMPLE SOLUBILITY INFLUENCES
CLEANING FREQUENCY
Carrier gas with The solubility of the sample in Karl Fischer reagents
Dry carrier gas
released moisture directly influences the frequency of reagent exchange
and cleaning of the Karl Fischer titration cell. If samples
are not soluble in Karl Fischer reagents, phase sepa-
ration within the cell will occur. If this happens, it is
necessary to replace the Karl Fischer reagent.

Flange
Sample inlet Occasional cleaning of the cell and electrodes may be
Safety plug septum
necessary to remove sample buildup or undesirable
Evaporation coatings. Reagent exchange and cleaning can be
chamber time-consuming due to the time required to fully dry
the cell and remove moisture from newly added Karl
Inlet tube
Fischer reagents. Poorly soluble materials may require
more frequent reagent exchange, leading to increased
Oven unit
chemical usage and costs. Procedure A is therefore
only recommended for samples which have an accept-
able solubility within Karl Fischer reagents, such as
Figure 5. Operating principle of oil evaporator.
aromatics or diesel and jet fuels.

For crude oil samples, heavier distillates, and lubricants

As with Procedures A and B, conditioning of the titra- which have limited solubility in Karl Fischer reagents,
tion cell is required to remove any residual moisture Procedures B and C are more suitable, as they are
within the system. This conditioning step must be independent of sample solubility and cell contamina-
carried out before every determination. Once the oven tion is greatly reduced. Only the evaporated water
is heated up to a defined temperature and the titra- together with the carrier gas is transferred into the
tion system is conditioned, the first measurement can titration cell. Phase separation therefore can no longer
start. occur, and frequency of reagent exchange is greatly
reduced. This results in a higher efficiency and lower
For the sample analysis, the dried carrier gas is passed costs per analysis.
through the distillation chamber and transfers the
azeotrope into the titration cell. The temperature of
the oven is set in dependence of the used solvent. − M AT R I X I N T E R F E R E N C E AND SIDE
REACTIONS
The carrier gas transports the azeotrope into the titra- Additives in lubricants and base oils in the sample or
tion cell. As with Procedure B, the gas used is not so even the sample itself may undergo side-reactions
important. However, inert gases (e.g., nitrogen) offer with the Karl Fischer reagents. For such samples, direct
more flexibility and reliability, as the oxygen content injection into the titration cell according to Procedure
of dry air can negatively influence the thermal stability A may result in an apparent higher water content
of organic substances at higher temperature. caused by side reactions.

In Procedures B and C, the sample does not come into

contact with the Karl Fischer reagent. Therefore, both

6
procedures are unaffected by possible matrix inter- SUMMARY
ferences. Thus, these procedures can provide more
reliable results for samples that cause matrix interfer- The ideal procedure for the moisture determination
ences or side reactions. according ASTM D6304 depends strongly on the
sample. The sample matrix and viscosity are important
aspects when choosing the correct procedure. For
− A U TO M AT I O N F O R E F F I C I E N C Y A N D sample with a low viscosity, such as jet fuel or
LOWER COSTS PER ANALYSIS aromatics, Procedure A is the best option. Using a
Modern laboratory analysts are tasked with performing Metrohm 851 Titrando with the equipment for the
many different tests. If the operator has to spend a lot automatic reagent exchange enables an efficient
of time on a single analysis and is bound to the system, moisture determination.
lab efficiency decreases and impacts operating costs.
Procedure B can be fully automated using the 874 For samples which are not suitable for Procedure A,
Oven Sample Processor. The analyst simply weighs the such as lubricants with additives or base oil, Proce-
samples into the sample vials, seals them tightly, places dures B and C are the solution. In comparison to Proce-
them on the sample rack, and starts the analysis series. dure C, Procedure B has the advantage that no
Procedure B therefore offers simple, walk-away auto- cleaning of the evaporation chamber is required.
mation. Additionally Procedure B can be fully automated,
providing a simple walk-away solution for the mois-
ture determination.
− A U T O M AT E D R E A G E N T E X C H A N G E
DECREASES INSTRUMENT DOWN TIME Using an 874 Oven Sample Processor combined with
When phase separation occurs, the titration cell is an 851 Titrando, it is possible to analyze samples
contaminated, or the Karl Fischer reagent is exhausted, according to Procedure B fully automatically. Addi-
it is necessary to exchange the reagent within the tionally, it is still possible to analyze samples by a direct
titration cell. Opening the Karl Fischer cell for reagent injection according to Procedure A with the 851
exchange introduces moisture into the titration cell, Titrando alone. This allows you to use the system with
which must be titrated before introducing sample. maximal flexibility.
Automated reagent exchange removes the depleted
reagent and replaces it with new reagent without
opening the KF cell. This greatly reduces the amount
of time needed to dry the titration cell before sample
addition, and minimizes user exposure to reagents.

For Procedure C, although it is possible to automati-

cally exchange the reagents in the titration cell, the
evaporation chamber must be emptied manually.

7
 Literature

Metrohm Application Bulletin AB-209 ASTM D6304

Water in insulating oils, hydrocarbons, and their Standard Test Method for Determination of Water in
products Petroleum Products, Lubricating Oils, and Additives
by Coulometric Karl Fischer Titration
Metrohm Application Bulletin AB-280
Automatic Karl Fischer water content determination ASTM E1064
with the 874 Oven Sample Processor Standard Test Method for Water in Organic Liquids
by Coulometric Karl Fischer Titration
Metrohm Application Note AN-K-048
Sample preparation with the oven technique – relative DIN 51777
blank Petroleum products - Determination of water content
using titration according to Karl Fischer
Metrohm Application Note AN-K-070
Water in petroleum products – Fully automated EN IEC 60814
determination according to ASTM D6304 Insulating liquids - Oil-impregnated paper and press-
board - Determination of water by automatic
ASTM D1533 coulometric Karl Fischer titration
Standard Test Method for Water in Insulating Liquids
by Coulometric Karl Fischer Titration EN ISO 12937
Petroleum products — Determination of water —
ASTM D4057 Coulometric Karl Fischer titration method
Standard Practice for Manual Sampling of Petroleum
and Petroleum Products GB/T 11146
Crude petroleum — Determination of water —
ASTM D4177 Coulometric Karl Fischer titration method
Standard Practice for Automatic Sampling of
Petroleum and Petroleum Products ISO 10337
Crude petroleum — Determination of water —
ASTM D4928, API MPMS 10.9 Coulometric Karl Fischer titration method
Standard Test Method for Water in Crude Oils by
Coulometric Karl Fischer Titration

ASTM D5854
Subject to change | Layout by RTS Rieger Team, printed in Switzerland by Metrohm AG, CH-9101 Herisau
WP-061EN – 2021-01 © Metrohm AG. All trademarks are the property of Metrohm and its subsidiaries.

Standard Practice for Mixing and Handling of Liquid

Samples of Petroleum and Petroleum Products

Contact
Lucia Meier
Metrohm International Headquarters; Herisau, Switzerland

Larry Tucker
Metrohm USA; Riverview, FL, United States of America

[email protected]