Viscosity is a measure of a fluid’s resistance to flow. It is one of the most important properties of a fluid and plays a prominent role in the petroleum industry.
The viscosity of crude oil affects our ability to pump it out of the ground; the viscosity and volatility of a fuel affects how easy it is to atomize in the fuel injector; the viscosity of a lubricant affects its ability to protect an engine.
It is an important measurement characteristic in the food, paint, polymer coating and other industries where flow is a critical product or use characteristic.
Small changes in viscosity can have a dramatic impact on the properties of petroleum fluids. The criticality of kinematic viscosity measurement prompted the petroleum industry to develop a precise method, first published in 1937, ASTM D445, Test Method for Kinematic Viscosity of Transparent and Opaque Liquids.
The term “kinematic” simply means that the measurement is made while fluid is flowing under the force of gravity. It is determined by measuring the time it takes for a volume of fluid to flow under gravity through a calibrated glass capillary viscometer.
Although it sounds simple, achieving the high accuracy and precision required by the industry is an extremely formidable task. There are many factors that affect the precision of this test method.
In economic terms, an error of one percent product viscosity that causes a blend adjustment can easily result in increasing product cost by a penny per gallon. For a large lubricant manufacturer, this can amount to $1 million or more in lost revenue per year.
As the saying goes, “The most important factor affecting the value of a home is location, location, location.” In terms of viscosity measurements, “The most important factor affecting the quality of a viscosity measurement is temperature, temperature, temperature.”
Temperature control is the single most important parameter for obtaining accurate and precise kinematic viscosity measurement. This is especially true for petroleum products as their rate of viscosity change per unit temperature is significantly greater than other products.
Thus a slight variation in temperature can have a large effect on the viscosity of a fluid. The bath temperature for the most common measurements, 40°C and 100°C, must be controlled to within +/- 0.02°C. That is an extremely tight window and great care must be taken to achieve this control.
There are a number of factors that impact the temperature:
S Thermometer. A specified thermometer or other temperature-sensing device having the specified accuracy and meeting the test method requirements must be used when measuring viscosity.
Accurately reading a thermometer to 0.01°C can be difficult, so the use of a magnifying lens is recommended. The thermometer must be immersed in the bath at the correct depth. It must be calibrated at least yearly to +/- 0.02°C. In addition, the thermometer ice point should be determined every six months and the correction factor applied.
The procedure for calibrating thermometers is quite involved and is described in ASTM E77, Test Method for the Inspection and Verification of Thermometers. Because temperature control is such an important parameter, the use of two thermometers per bath is recommended.
S Bath temperature uniformity and stability. Bath temperature uniformity and stability are other important parameters. The entire length of the viscometer must be maintained at the appropriate temperature. The type of circulator used, the age of the bath fluid and the bath fluid viscosity impact temperature uniformity.
The circulator and the viscosity of the bath fluid need to be balanced to provide a uniform temperature throughout the bath. The bath fluid needs to be changed well before it begins to discolor, as the discoloration usually indicates the fluid has oxidized and the viscosity probably has increased.
The bath should not be located near a draft, such as in a fume hood, which usually causes excessive temperature gradients in the bath. Temperature stability will be negatively impacted if a second viscometer is added to the test bath during the same time that another adjacent viscometer is being used for a measurement.
S Illumination. There needs to be sufficient illumination of the sample in the tube while in the bath to ensure consistent visual detection of the meniscus crossing the timing lines. Care needs to be taken that this lighting device does not affect the bath temperature control and stability.
These temperature requirements cannot be taken lightly because slight variations in temperature can cause dramatic changes in viscosity for certain types of petroleum products.
Fully annealed, low-expansion borosilicate glass is required for the construction of all viscometers. There are a number of factors to consider when choosing which viscometer size is needed for sample analysis. First, the viscometer must be calibrated to calculate viscosity.
The procedure is described in ASTM D446, Specifications and Operating Instructions for Glass Capillary Kinematic Viscometers. Viscosity standards are used to determine the viscometer constant. Although a single standard is sufficient to obtain a constant, the use of two standards covering the capillary range yields a more robust calibration constant.
Viscosity standards have an expiration date and should not be used beyond that date. A calibrated viscometer can be purchased with constants at appropriate temperatures. Because the viscometer constant varies as a function of temperature for certain types of viscometers, it is important to calculate the viscometer constant for the temperature at which the sample is being analyzed. Viscometer constants should be verified at least yearly.
The size of the viscometer must be chosen such that the flow time is at least 200 seconds for manual determinations; otherwise eye-hand coordination will become a significant factor. The operator waits for the meniscus of the fluid to pass the timing lines and uses a clock or stopwatch to measure the flow time. Flow times greater than 200 seconds are required to eliminate possible operator variation.
It is obvious that a clean viscosity tube must be used for the analysis, but what is actually meant by “clean”? The tube must be free from dust or other particles and the fluid should clearly wet the surface of the glass.
Typically, several rinsings with a sample solvent such as naphtha, followed by a drying solvent such as acetone, and then purging with a dry, dust-free gas such as air or nitrogen, are sufficient. The solvents also must be residue-free on drying. This is of particular concern if commercial- or practical-grade solvents are used.
The residue of previous samples coating the inside of the viscometer can be removed with a chromic acid1 cleaning solution or a nonchromium-containing, strongly oxidizing cleaning solution.2
An alkaline liquid detergent or any cleaning solution with a pH greater than 8.0 should never be used because they will actually change the dimensions of the capillary by dissolving the glass walls, thus changing the calibration constant.
These types of liquids are often found throughout an analytical laboratory and can be used in other applications, but if one is used to clean a tube, the tube will need to be recalibrated. Another way to significantly change the calibration constant of the smallest size viscometers is to soak the tube overnight in boiling water. The tube might be clean, but it needs to be recalibrated before it is used.
The tube must also be dry before using. The use of dry, dust-free air or nitrogen is recommended.
When suspended in the temperature bath, the tube should be in the specified vertical position, free from vibration, and at the specified immersion depth. Follow the manufacturer’s instructions and those in the test method.
The timing device must have an accuracy of within +/- 0.07 percent of the reading and must be capable of taking readings within a discrimination of one part in 2,000 or 0.1 second for a 200-second flow time. Care must be taken when using electrical timing devices because alternating currents may not provide the required accuracy.
Timing devices should be checked against a National Institute of Standards and Technology (NIST) reference or other accepted timing reference on a regular basis. Yearly verification is recommended.
Filter samples should be taken if particles are visible. This is especially important when analyzing used oils. Because the sample should be free of air, it should be allowed to set to disperse the entrained air, if any is present. Just as caustic cleaning fluids will change the calibration constant, the same is true for caustic (high pH) samples.
As the sample pH increases, the change in the constant increases more rapidly. The smaller the capillary, the faster the constant changes. The tube may need recalibration after a caustic sample is analyzed, although very caustic samples are not found often in the petroleum industry.
Samples must be equilibrated at the desired test temperature. Up to 30 minutes may be necessary and possibly longer with some materials.
Two flow time measurements should be obtained for each sample. If the two determinations of kinematic viscosity, calculated from the flow time measurements, agree with the stated determinability limit for the sample type tested, then the average of the two determinations should be reported.
If the two determinations do not agree to within the stated determinability, then the measurements must be repeated after the possible causes have been investigated and corrected. A dirty viscometer, incorrect bath temperature or equilibration time are possible culprits.
Automated viscometers have been used for the determination of kinematic viscosity by D445 in the petroleum industry for more than 30 years. They mimic the physical conditions, operations or processes of the manual apparatus and must meet the precisions specified by the method.
All but one of the requirements discussed above equally apply to automated instruments. The flow-time requirement of greater than 200 seconds was at one time relaxed for automated instruments. The ASTM subcommittee responsible for this test method, D02.07 on Flow Properties, is currently determining the minimum allowable flow time for these automated instruments.
If indeed the flow time requirement is relaxed, then two additional requirements will come into play. First, calculations of the kinematic viscosity from the tube constant and flow time might require an additional term, the kinematic energy correction factor. This factor takes into account the energy that is imparted to speed up the fluid as it passes from the reservoir to the capillary.
This correction is negligible for longer flow times. Second, the timing device will need to maintain the one-part-in-2,000 resolution and detection sensitivity.
A quality control sample should be used routinely to monitor the measurement system and verify that results are within the precision of the test method.
Any sample that is similar in composition to samples typically analyzed should be used. ASTM D6299, Standard Practice for Applying Statistical Quality Assurance Techniques to Evaluate Analytical Measurement System Performance, provides details on a quality control system.
ASTM’s Interlaboratory Crosscheck Programs provide participating laboratories with a statistical quality assurance tool, enabling them to compare their performance in the use of ASTM methods against other laboratories worldwide.
The kinematic viscosity measurement by ASTM D445 is included in most of the programs. More than 700 laboratories participated in the 2003 testing, including more than 45 percent participation by laboratories outside the United States.
ASTM International offers a comprehensive, Web-based training resource providing instruction on ASTM standards and other related technical information. D445 is one of those standards. The course for D445, which is 30 to 45 minutes long, is a step-by-step overview of the test method and can be used to enhance current training procedures.
The Test Method for Kinematic Viscosity of Transparent and Opaque Liquids is one of the most critical tests in the petroleum industry. It has been used since 1937, yet the method is continually being improved. The technique appears to be quite simple, yet many factors can affect the precision of the measurement. Meeting all the requirements for performing the test method will allow laboratories to meet or exceed the stated precision.