Standardization News Search

Magazines & Newsletters / ASTM Standardization News


November/December 2009

Replacing Mercury-in-Glass Thermometers in ASTM Test Methods

Mercury-in-glass thermometerSome Guidelines for a Complex Task

The purpose of ASTM International test methods is to create a “definitive procedure that produces a test result,” according to Form and Style for ASTM Standards. These methods often include special apparatus or techniques, equipment lists and step-by-step instructions on how to use the equipment to achieve the desired test result. For the test method to be of value, the methods must produce a result that is reliable and comparable among users for procurement, laboratory assessment and for evaluation of material (examples include flash point, viscosity, density, etc.). ASTM test methods have precision and bias statements that indicate the expected variability when comparing sets of results between laboratories, buyers and sellers, or other user groups. Temperature is an important variable in many ASTM test methods.

At the time many original manual test methods were being developed, mercury-in-glass thermometers were the gold standard for temperature measurement in industrial testing laboratories. Although platinum resistance thermometers have served as primary temperature standards since 1927, mercury-filled thermometers had compelling advantages for ASTM test methods: they were relatively inexpensive to manufacture, were easy to use and provided sufficiently consistent measurement results to achieve the precision required for the methods. Even today, the mercury-in-glass thermometer is still considered the gold, or referee, standard in some industries.

Of all traditional thermometric liquids, mercury remains the most repeatable and reproducible. New alternative liquids can mimic some of the qualities of mercury but have severe limitations. ASTM standard E2251, Specification for Liquid-in-Glass ASTM Thermometers with Low-Hazard Precision Liquids, has a list of thermometers with alternative liquids that can replace some of the mercury thermometers specified in ASTM standard E1, Specification for ASTM Liquid-in-Glass Thermometers. However, the listed thermometers are limited in scope and use.

Since the start of the ASTM International mercury initiative in 2006, ASTM technical committees have been struggling with the issue of substituting alternative temperature measurement devices for the traditional mercury-in-glass approach without affecting test method results. Among the devices being considered are the ASTM E2251 thermometers and various electronic thermometers with digital readouts.

Most of the mercury-in-glass thermometers found in ASTM E1 were designed for a specific test method or group of methods by Subcommittee E20.05 on Liquid-in-Glass Thermometers and Hydrometers, a part of Committee E20 on Temperature Measurement, in conjunction with the governing ASTM technical committees. The goal of the thermometer designs was often to provide consistent results among the parties. To that end, the designs were often manipulated for optimal repeatability or ease of use within the method, not necessarily for accuracy. Examples of this manipulation include establishing arbitrary emergent stem temperature assignments for partial immersion thermometers (for example, ASTM 5C/5F cloud and pour thermometers), or use of expanded bulb thermometers in tests (thermometers conditioned at their highest temperature before use, for example, ASTM 56C and ASTM 117C calorimetric thermometers).

Still common in many ASTM test methods is the intentional improper use of the mercury-in-glass thermometer. An example of such is in distillation methods. The manual methods call for total immersion thermometers to be used partially immersed. The test methods specifically state not to correct for emergent stem temperature effects, which degrades the absolute accuracy of the thermometer reading. The emergent stem temperature is the ambient temperature surrounding the portion of the liquid column that is not immersed. It is not necessarily the ambient temperature in the room or general area of the test. The magnitude of the error depends on the thermometric liquid used and the temperature difference between the medium being measured and the environment surrounding the stem.

Most electronic thermometers considered as alternatives are minimally or not at all affected by emergent stem temperature. Therefore, in this type of test method, as in many ASTM test methods, the use of an alternative temperature measurement device may provide more accurate temperature measurements but may not reproduce the previously accepted values of the test method. Switching to an electronic alternative would introduce a new bias in the method. In general, because of the unique design manipulations of the ASTM E1 thermometers, results produced by alternative temperature measurement devices in apparatus built for ASTM mercury-in-glass thermometers will not be directly comparable to results obtained using the ASTM E1 thermometer(s) specified in the test method.

immersion depths

The graphic illustrates the difference in the types of immersion depths of partial and total immersion liquid-in-glass thermometers. On the total immersion thermometer, the arrow indicates that the liquid column is at the depth of the medium being measured. The arrow on the partial immersion indicates that it is immersed to a line drawn on the thermometer.

In evaluating an ASTM test method, the technical committee must first determine whether the thermometer specified is designed and manufactured for use at total immersion or a partial immersion depth as provided for in ASTM E1. Then the committee must determine if the specified thermometer is actually used in the test method or test method apparatus at the ASTM E1 specified immersion.

A total immersion thermometer is calibrated during manufacture to give the correct reading when inserted into the medium being measured up to the end of the liquid column. Good practice allows for approximately 10 mm of liquid column to be above the medium level in order to read the thermometer. Changes in the temperature of the medium require an adjustment in the depth of insertion of the total immersion thermometer.

By design, a partial immersion thermometer gives the correct reading when inserted into a medium up to an immersion line or stated immersion depth and when the emergent stem temperature matches the ASTM E1 value. When evaluating the ASTM methods, the technical committee must compare the actual emergent stem temperatures measured around the thermometer at time of use in the test method to those stated in E1. With an electronic thermometer, the minimum immersion length must be less than the depth of immersion of the specified glass thermometer (see E1137/E1137M, Specification for Industrial Platinum Resistance Thermometers, for further details).

After this initial evaluation, the following procedures should be used to complete the committee’s evaluation:

  • For methods specifying ASTM E1 total immersion thermometers used in static temperature measurements (for example, ASTM 62C, 63C, 15C) the method must use the ASTM mercury-in-glass thermometer vertically and at total immersion for the substitution to be simple. In cases where the thermometer is not immersed properly, research will be necessary to determine the temperature difference obtained using the method with the traditional thermometer and with the substitute thermometer. The differences found may need to be incorporated into the revised standard as an adjustment to the readings of the substitute thermometer and/or additional precision and bias statements for the method using the substitute thermometer may need to be developed.
  • For methods specifying ASTM E1 partial immersion thermometers used in static temperature measurements (other than ASTM general use thermometers 1C&F, 2C&F and 3C&F) the partial immersion thermometers in ASTM test methods all have assigned emergent stem temperatures in ASTM E1 that are either assigned for consistency or are measured emergent stem temperatures found over the devices for which they were originally designed. Prior to allowing an alternate thermometer, research will be necessary to determine the difference obtained using the method with the traditional thermometer and the substitute thermometer, including effects on charts, data, and precision and bias statements.
  • For methods specifying ASTM E1 thermometers used in rising temperature measurements (for example, ASTM 9C/9F and ASTM 10C/10F, or Pensky Martens), in addition to the examples above, if the ASTM method utilizes a heating or cooling method, the committee should evaluate the thermal response times of the alternate device(s) as compared to that of the ASTM mercury-in-glass thermometer. In general, electronic devices or other types of liquid-in-glass thermometers will have response curves that are substantially different from that of the specified mercury-in-glass thermometer. Research should be completed comparing data obtained with an alternate device of well-defined geometry and construction and the specified mercury-in-glass thermometers with samples of the same test material. The committee should determine effects on charts, data, and precision and bias statements.

If, after evaluation, the technical committee has confidence that the alternative temperature measuring devices tested may be suitable in lieu of the specified ASTM E1 mercury-in-glass thermometer, a statement to that effect must be balloted as a revision to the test method. If a new precision and bias statement has been developed, a research report will also need to be completed. The technical committee should state the type and required minimum immersion length of alternative measuring device(s) tested (platinum resistance thermometer, thermistor, and thermocouple or ASTM E2251 alternative liquid thermometer). For electronic sensors, sheath length and diameter, along with other relevant information, must be included. For alternative liquid thermometers, contact the mercury initiative chair (see below for contact information) for assistance in the choice and design of the ASTM E2251 thermometer.

ASTM Committee E20 publishes standards relevant to platinum resistance thermometers, thermocouples and non-mercury liquid-in-glass thermometers that may assist committees in properly describing the technical requirements of the alternate devices. The standards include:

  • E1, Specification for ASTM Liquid-in-Glass Thermometers;
  • E608/E608M, Specification for Mineral-Insulated, Metal-Sheathed Base Metal Thermocouples;
  • E1137/E1137M, Specification for Industrial Platinum Resistance Thermometers; and
  • E2251, Specification for Liquid-in-Glass ASTM Thermometers with Low-Hazard Precision Liquids.

For background information on this work, please see “ASTM and the Mercury Initiative,” in the September/October 2008 issue of ASTM Standardization News. If you need assistance with evaluating an ASTM standard, please contact the ASTM E20 mercury task group through Christine DeJong, ASTM International (phone: 610-832-9736), or Deanne Emory, task group chairman (phone: 718-821-7110).

The Committee E20 Mercury Task Group members possess expertise in temperature measurement and various associated measuring instruments. The group assists ASTM committees as they evaluate the possible replacement or removal of mercury or mercury-containing devices from their standards.