Designation: E 1502 – 98 (Reapproved 2003)e1

Standard Guide for

Use of Freezing-Point Cells for Reference Temperatures1
This standard is issued under the fixed designation E 1502; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.

e1 NOTE—Updated caution note in Section 7.2.2 in November 2003.

INTRODUCTION

During freezing, pure material transforms from the liquid state to the solid state at a constant
temperature known as the freezing point. The freezing points of highly purified materials can serve as
reference temperatures, and in fact, the International Temperature Scale of 1990 (ITS-90)2 relies on
the freezing points of some highly purified metals as defining fixed points. Freezing points can be
realized in commercially available systems incorporating freezing-point cells. When the cells are
properly made and used, they establish useful reference temperatures for the calibration of
thermometers and for other industrial and laboratory purposes; with care, the freezing points of highly
purified materials can be realized with an uncertainty of a few millikelvins3 or less.

1. Scope 1.6 This guide does not address all of the details of
1.1 This guide describes the essential features of freezing- thermometer calibration.
point cells and auxiliary apparatus, and the techniques required 1.7 This guide is intended to complement special operating
to realize freezing points in the temperature range from 29 to instructions supplied by manufacturers of freezing-point appa-
1085 °C.3 ratus.
1.2 Detailed design and construction are not addressed in 1.8 The following hazard caveat pertains only to the test
this guide. method portion, Section 7, of this guide. This standard does not
1.3 This guide is intended to describe good practice and purport to address all of the safety concerns, if any, associated
establish uniform procedures for the realization of freezing with its use. It is the responsibility of the user of this standard
points. to establish appropriate safety and health practices and
1.4 This guide emphasizes principles. The emphasis on determine the applicability of regulatory limitations prior to
principles is intended to aid the user in evaluating cells, in use.
improving technique for using cells, and in establishing pro-
2. Referenced Documents
cedures for specific applications.
1.5 For the purposes of this guide, the use of freezing-point 2.1 ASTM Standards: 4
cells for the accurate calibration of thermometers is restricted E 344 Terminology Relating to Thermometry and Hydrom-
to immersion-type thermometers that, when inserted into the etry
reentrant well of the cell, (1) indicate the temperature only of E 644 Test Methods for Testing Industrial Resistance Ther-
the isothermal region of the well, and (2) do not significantly mometers
alter the temperature of the isothermal region of the well by 3. Terminology
heat transfer.
3.1 Definitions:
3.1.1 reference temperature, n—a fixed, reproducible tem-
1
This guide is under the jurisdiction of ASTM Committee E20 on Temperature perature, to which a value is assigned, that can be used for the
Measurement and is the direct responsibility of Subcommittee E20.07 on Funda- calibration of thermometers or other purposes.
mentals in Thermometry.
Current edition approved Nov. 1, 2003. Published November 2003. Originally
approved in 1992. Last previous edition aprpoved in 1998 as E 1502 – 98.
2
Preston-Thomas, H., “The International Temperature Scale of 1990 (ITS-90),”
4
Metrologia, Vol 27, No. 1, 1990, pp. 3–10. For errata see ibid, Vol 27, No. 2, 1990, For referenced ASTM standards, visit the ASTM website, www.astm.org, or
p. 107. contact ASTM Customer Service at [email protected]. For Annual Book of ASTM
3
In this guide temperature intervals are expressed in kelvins (K) and millikelvins Standards volume information, refer to the standard’s Document Summary page on
(mK). Values of temperature are expressed in degrees Celsius (°C), ITS-90. the ASTM website.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.

1
 E 1502 – 98 (2003)e1
3.1.2 Additional terms used in this guide are defined in
Terminology E 344.
3.2 Definitions of Terms Specific to This Standard:
3.2.1 first cryoscopic constant, A, n—a constant of propor-
tionality between the freezing point depression of, and concen-
tration of impurities in, a sample of reference material, given
by the ratio of the molar heat of fusion of the pure material, L,
to the product of the molar gas constant, R, and the square of
the thermodynamic temperature of fusion, T, of the pure
material (freezing point):
L
A5 (1)
RT 2
FIG. 2 Freezing Curve of a Sample of Highly Purified Tin
3.2.2 freeze, n—an experiment or test run conducted with a
freezing-point cell while the reference material in the cell 3.2.8 recalescence, n—the sudden increase in temperature
solidifies. of reference material in the supercooled state upon nucleation
3.2.3 freezing curve, n—the entire time-temperature relation and crystal growth, due to the release of latent heat of fusion of
of the reference material in a freezing-point cell during the reference material.
freezing, including initial cooling, undercool, recalescence, 3.2.9 reference material, n—the material in a freezing-point
freezing plateau, and final cooling to complete solidification. cell that melts and freezes during use, the freezing point of
3.2.3.1 Discussion—Graphic representations of freezing which can establish a reference temperature.
curves are shown in Fig. 1 and Fig. 2. 3.2.10 supercooled state, n—the meta-stable state of refer-
3.2.4 freezing plateau, n—the period during freezing in ence material in which the temperature of the liquid phase is
which the temperature does not change significantly. below the freezing point.
3.2.5 freezing-point cell, n—a device that contains and 3.2.11 undercool, n—the temperature depression below the
protects a sample of reference material in such a manner that freezing point of reference material in the supercooled state.
the freezing point of the material can establish a reference 4. Summary of Guide
temperature.
4.1 A freezing-point cell is used for thermometer calibration
3.2.6 freezing range, n—the range of temperature over
by establishing and sustaining a reference material at the
which most of the reference material in a freezing-point cell
freezing point, to which a value of temperature has been
solidifies.
assigned. The thermometer to be calibrated is inserted into a
3.2.6.1 Discussion—The freezing range is indicated graphi-
reentrant well in the cell; the well itself is surrounded by the
cally in Fig. 1.
freezing reference material.
3.2.7 nucleation, n—the formation of crystal nuclei in liquid
4.2 The cell is heated to melt the reference material. The
in the supercooled state.
temperature of the surrounding environment is then reduced to
about 1 K below the freezing point so that the reference
material cools. Following the undercool, nucleation, and re-
calescence, the well temperature becomes constant during the
freezing plateau. After a time, depending on the rate of heat
loss from the cell, the amount of reference material, and the
purity of the reference material, the temperature starts to
decrease and eventually all of the material becomes solidified.
4.3 Since the temperature in the reentrant well remains
constant during the freezing plateau, one or more test ther-
mometers may be calibrated by inserting them singly into the
well. In some cases the plateau can be sustained for many
hours, and even under routine industrial conditions, the plateau
may be readily sustained long enough to test several thermom-
eters. The duration of the plateau may be lengthened by
A = Stabilized temperature of cell before freezing, typically about 1 K above preheating the test thermometers.
freezing point. 4.4 Measurements are made also during each freeze with a
B = Freezing point of cell.
C = Temperature of cell surroundings during freezing, typically about 1 K dedicated monitoring thermometer. These measurements, to-
below freezing point. gether with other special test measurements, provide qualifi-
D = Maximum undercool. cation test data (see 6.4 and 7.5).
E = Onset of recalescence.
F = Freezing plateau.
G = Total freezing time.
5. Significance and Use
H = Freezing range. 5.1 A pure material has a well-defined freezing behavior,
FIG. 1 Structure of a Typical Freezing Curve and its freezing point, a characteristic of the material, can serve

2
 E 1502 – 98 (2003)e1
as a reproducible reference temperature for the calibration of TABLE 2 Estimated Achievable Uncertainties in Freezing-Point
thermometers. The freezing points of some highly purified CellsA
metals have been designated as defining fixed points on Laboratory
ITS-90. The freezing points of other materials have been Materials Primary, mK Industrial, mK
determined carefully enough that they can serve as secondary GalliumB 0.1 1
reference points (see Table 1 and Table 2). This guide presents Indium 1 10
Tin 1 10
information on the freezing process as it relates to establishing Cadmium 2 10
a reference temperature. Lead 2 10
5.2 Freezing-point cells provide users with a means of Zinc 1 10
Antimony 10 50
realizing freezing points. If the cells are appropriately designed Aluminum 2 20
and constructed, if they contain material of adequate purity, Silver 2 40
and if they are properly used, they can establish reference Gold ... ...
Copper 10 50
temperatures with uncertainties of a few millikelvins or less. A
Values for cells of good design, construction, and material purity used with
This guide describes some of the design and use consider- careful technique. Cells of lesser quality may not approach these values.
ations. B
Realized as melting point.
5.3 Freezing-point cells can be constructed and operated
less stringently than required for millikelvin uncertainty, yet 6.1.2 The deposition of the solid phase from the liquid phase
still provide reliable, durable, easy-to-use fixed points for a requires the presence of liquid in the supercooled state,
variety of industrial calibration and heat treatment purposes. nucleation, and crystal growth. Nucleation may begin sponta-
For example, any freezing-point cell can be operated, often neously in the meta-stable supercooled liquid, or it may be
advantageously, as a melting-point cell. Such use may result in induced artificially. As crystals nucleate and grow, the liberated
reduced accuracy, but under special conditions, the accuracy latent heat of fusion produces recalescence.
may be commensurate with that of freezing points (see 6.2.10). 6.1.3 The undercool of materials may range from as little as
5.4 The test procedure described in this guide produces 0.05 K, for some materials such as zinc, to more than 20 K for
qualification test data as an essential part of the procedure. tin and other materials (see Table 1). The magnitude of the
These data furnish the basis for quality control of the freezing- undercool can depend on the initial temperature, the cooling
point procedure; they provide for evaluation of results, they rate, and the purity of the material.
assure continuing reliability of the method, and they yield 6.1.4 Following recalescence, the temperature remains rela-
insight into the cause of test result discrepancies. The test tively constant for a while during the freezing plateau. The
procedure is applicable to the most demanding uses of temperature associated with the freezing plateau is the tem-
freezing-point cells for precise thermometer calibration; it may perature to which a value is assigned as the freezing point of
not be appropriate or cost-effective for all applications. It is the material.
expected that the user of this guide will adapt the procedure to 6.1.5 As freezing progresses, some trace impurities in the
specific needs. freezing material tend to be swept in front of the advancing
liquid-solid interface and concentrated in the remaining liquid.
6. Principles Since impurities usually depress the freezing point of the
6.1 Freezing Process: reference material, the temperature of the material decreases
6.1.1 Ideally pure material at a given pressure has a unique ever more rapidly until all of the material is solid.
temperature when its solid and liquid phases are in perfect 6.1.6 The effect of small concentrations of impurities may
thermal equilibrium. In contrast, the phase transition of a real be estimated from an approximation rule: the temperature
material from liquid to solid, as heat is released in semi- difference between the start of freezing and midpoint of
equilibrium freezing, exhibits a complex time-temperature freezing (when half the material is solid) equals the tempera-
relation (freezing curve) as shown in Figs. 1 and 2. ture difference between the freezing point of the ideally pure

TABLE 1 Characteristics of Pure Freezing-Point Reference Materials

Typical Pressure Coefficient at Freezing Point First Cryoscopic
Material Freezing Point, ITS-90, °C Constant, K−1
Undercool, K mK/Pa mK/m (of liquid)
GalliumA,B 29.7646 76 − 20 −1.2 0.0073
IndiumA 156.5985 0.1 + 49 + 3.3 0.0021
TinA 231.928 25 + 33 + 2.2 0.0033
Bismuth 271.403 0.19 − 34 − 3.4 ...
Cadmium 321.069 0.05–0.5 + 61 + 4.8 0.0021
Lead 327.462 0.15 + 79 + 8.2 0.0016
ZincA 419.527 0.05–0.1 + 43 + 2.7 0.0018
Antimony 630.630 20 +8 + 0.5 0.0029
AluminumA 660.323 0.4–1.5 + 70 + 1.6 0.0015
SilverA 961.78 1–3 + 60 + 5.4 0.00089
GoldA 1064.18 1–3 + 61 + 10.0 0.00083
CopperA 1084.62 1–2 + 33 + 2.6 0.00086
A
Defining fixed point for ITS-90.
B
Realized as melting point.

3
 E 1502 – 98 (2003)e1
material and the freezing point (at the start of freezing) of the
real reference material (see 8.6.2). The product of this tem-
perature difference and the first cryoscopic constant gives an
estimate of the mole fraction impurity concentration in the
reference material. Conversely, if the impurity concentration is
known, then the temperature difference can be estimated.
6.1.7 The change in temperature during the freezing plateau
due to a change in pressure is generally less than 0.1 µK/Pa
(Table 1). Thus, normal changes in atmospheric pressure have
little effect on the freezing point, but the effect of the pressure
of a head of dense liquid reference material may be significant.
The freezing point is usually taken to be the temperature during
the freezing plateau at a pressure of 101 325 Pa.
6.2 Freezing-Point Cells:
6.2.1 The usual freezing-point apparatus consists of a
freezing-point cell containing the reference material, a means
to melt the reference material and allow the material to freeze
slowly and uniformly, with provision for exposing one or more
test thermometers to the freezing point. A typical cell and
auxiliary furnace are shown in Figs. 3 and 4. Control equip-
ment is not shown.
6.2.2 The freezing-point apparatus must be able to maintain
a freezing plateau of useful duration and must include enough NOTE 1—This example shows an insulated furnace body and two
reference material to establish an isothermal region and depth alternative types of furnace cores. The core on the left is a three-zone
shielded type. The core on the right employs a heat pipe to reduce
of immersion suitable for the intended use. Typically, a mass of
temperature gradients.
reference material of 1 to 1.5 kg (or a sufficient mass of FIG. 4 Example of Freezing-Point Furnace
material to supply 50 to 100 kJ of heat from the latent heat of
fusion) is used. However, carefully designed systems using
half the above mass of some reference materials can produce 6.2.3 The freezing point and its repeatability, as well as the
freezing plateaus longer than 24 h (see 6.2.6, 6.4.3, and 6.5). duration of the freezing plateau for a given rate of heat loss,
depend on the purity of the reference material (6.1.5); material
purity must therefore be adequate for the intended purpose.
Typically, the freezing point of the reference material in a cell
will be within 10 mK of the freezing point of pure material, if
the impurity content of the reference material is of the order of
10 ppm (6.1.6).
6.2.4 The freezing-point cell must be fabricated to prevent
contamination of the reference material during construction
and during prolonged use of the cell. A container (crucible),
made of a material (such as high purity graphite) that is
chemically compatible with the reference material and will not
contaminate it, holds the freezing point reference material. This
container is usually placed inside another vessel, or cell, that
further protects the reference material from contamination and
the container from air. The container and cell must accommo-
date expansion and contraction of the reference material from
ambient to about 10 K above the freezing point.
6.2.5 Cells often have provision for sealing and evacuation
in order to protect the reference materials from contaminants in
the gaseous or vapor phase. For example, oxygen can signifi-
cantly affect the freezing points of some materials by dissolv-
ing in them or by oxidizing them, or both. Some cells have a
close-fitting glass envelope completely surrounding the graph-
ite crucible and well that can be hermetically sealed after the
cell has been purged and filled with an inert gas (usually
argon). The value assigned to the cell freezing point must take
into account the gas pressure inside the cell during freezing.
6.2.6 Under preferred freezing conditions, uniform heat loss
FIG. 3 Examples of Freezing-Point Cells from the container of reference material produces an advancing

4
 E 1502 – 98 (2003)e1
uniform shell of solid on the walls of the container. The freezing point of the cell. The cell must be designed to maintain
liquid-solid interface, thus formed, establishes an isothermal the purity of the reference material with repeated use.
shield around the reentrant well. The freezing-point cell must 6.2.9.2 A major source of error in the use of freezing-point
be designed so that the isothermal region of the well is long cells is the failure of an object under test to attain the reference
enough to accommodate the type of thermometer to be temperature because of unwanted heat flow to or from the
calibrated (see 6.4.3 and 6.5). object. The heat flow depends in part on the characteristics of
6.2.7 For many materials, the duration and repeatability of the object itself. This source of error is minimized by designing
the freezing plateau can be enhanced by inducing freezing, a the cell to (1) provide adequate immersion for the test object in
procedure by which a portion of the liquid metal is rapidly the region of the reference material (see 6.4.3 and 6.5.2), and
solidified by cooling. (2) provide adequate immersion of the cell in the heating
6.2.7.1 For reference materials that exhibit a relatively small device.
undercool (a few kelvins), freezing is induced, after recales- 6.2.10 Users of freezing-point cells interested in using the
cence is observed on a monitoring thermometer, by removing cells to realize melting points should consider 6.2.10.1-
the thermometer and inserting a cool object into the well. The 6.2.10.3. A detailed description of melting-point techniques is
object may be a rod or tube at room temperature, or even the beyond the scope of this guide. For more information, see
cooled monitoring thermometer itself. This procedure, some- Footnote 5.5
times referred to as inside nucleation, results in a thin mantle 6.2.10.1 Plateaus obtained during melting may have practi-
of solid frozen onto the well, forming a liquid-solid interface cal advantages. First, since heat is added to the system during
close to the measuring well. melting, the insertion of a cold test object into the cell tends to
6.2.7.2 For reference materials such as tin and antimony, slow down the phase transition rather than to hasten it. Thus it
which exhibit a deep undercool of many kelvins, it is essential is easier to prolong the melting curve than a freezing curve
that freezing be induced to avoid the unwanted excessive upon multiple insertions. Second, for reference materials such
lowering of the cell heating device temperature. An outside- as tin that exhibit a large undercool, it is necessary to use
nucleated freeze is conveniently induced by removing the cell special techniques in order to initiate freezing in a useful
briefly from the heating device and exposing it to room manner, whereas melting initiation is usually simple.
temperature, or by cooling only the cell while it is in the 6.2.10.2 Impurity segregation upon freezing helps to pro-
heating device with a controlled flow of air or suitable gas. mote reproducibility of the plateau temperature from freeze to
Upon recalescence, observed by a monitoring thermometer in freeze. The melting process does not have this advantage, and
the measuring well, the cell is placed in the heating device, or in fact, the melting curve shape and plateau temperature may
the gas flow is interrupted. depend upon impurity distribution in the solid. Nonetheless,
6.2.8 A value of temperature must be assigned to the melting points may still be useful at reduced accuracy for less
freezing point of a cell; specifically, a value must be assigned demanding applications.
to the reference temperature realized in the isothermal region 6.2.10.3 A freezing-point cell that contains very pure metal
of the well. This value may be assigned by one of two means: (impurity concentration less than 1 part in 107) will produce
melting points that are as reproducible as freezing points and
6.2.8.1 If the purity of the original reference material
that are indistinguishable from them.6 Special techniques are
warrants, if assembly of the cell has maintained the purity, and
required to achieve this as described in Footnote 5.5 For
if subsequent qualification tests so verify, the cell may be
freezing-point cells containing an impurity concentration of
assigned the value of the freezing point of the pure material, as
more than 1 part in 107, the freezing-point method may give
promulgated by appropriate authority (for example, ITS-90). In
more reproducible and accurate values than the melting-point
this case, there is associated with the assigned value an
method, since the melting range is very dependent on the
uncertainty that must be evaluated from knowledge of impurity
method of solidification of the metal prior to the melt.
content of the reference material, augmented by results of
6.3 Auxiliary Apparatus:
qualification tests. See 6.1.6 and 6.4.
6.3.1 Heating devices, such as furnaces (ovens) or baths, are
6.2.8.2 The value of the freezing point may be determined used to heat the freezing-point cells. An important requirement
by measurement with several calibrated thermometers. All of for such devices is temperature uniformity in the region of the
these thermometers must be capable of measurement with cell, so that the reference material will melt and freeze
smaller uncertainty than is required of the freezing-point cell in uniformly. To minimize temperature gradients, furnaces may
its intended application. In this case, the assigned value of be equipped with high-conductivity temperature moderator
temperature and its components of uncertainty are derived blocks or heat pipes, or they may employ multiple zone
from the measurements and from an analysis of errors in the heaters.
complete measurement process.
6.2.9 Important considerations in the design of a freezing- 5
Mangum, B. W., Bloembergen, P., Chattle, M. V., Marcarino, P., and Pokhodun,
point cell include: A. I., Comité Consultatif de Thermométrie, 19th Session, 1996, Document CCT/
6.2.9.1 The use of a reference material of the highest 96–8, entitled “Recommended Techniques for Improved Realization and Intercom-
practicable purity is cost effective and justified. High material parisons of Defining Fixed Points: Report to the CCT by Working Group 1.”
6
Working Group 1 of the Comité Consultatif de Thermométrie (Mangum, B. W.,
purity minimizes variability in the observed freezing point
Bloembergen, P., Chattle, M. V., Fellmuth, B., Marcarino, P., and Pokhodun, A. I.),
caused by variations in operating conditions and procedures, “On the International Temperature Scale of 1990 (ITS-90) Part I: Some Definitions,”
and it reduces the uncertainty in the value to assign to the Metrologia, Vol 34, 1997, pp. 427–429.

5
 E 1502 – 98 (2003)e1
6.3.2 Another important requirement is the ability to control The monitoring thermometer is then removed from the cell
the heating device during melting and slow freezing. Control well, and it is replaced after the last test thermometer has been
may be achieved manually or with automatic controllers that calibrated.
are suitable for the task. In either case, the heating device must 6.4.2.3 If the monitoring thermometer indicates that the
not be operated in a manner that could obscure the normal reference material was initially in the liquid state, that the
freezing plateau, for example, by establishing a period of undercool was not significantly different from previous under-
constant temperature near the freezing point that could be cools, that the first part of the freezing plateau was not
mistaken for the freezing plateau, or by inadvertent remelting significantly different from previous freezing points, and that
after the initiation of freezing. the final observation on the freezing plateau was not signifi-
6.3.3 Auxiliary heating devices are useful for heating ther- cantly different from the initial observation on the plateau, then
mometers to a temperature near the freezing point before they the calibration run is qualified as being valid.
are inserted into the well (see 6.5.4). 6.4.3 Immersion Qualification Test:
6.3.4 A monitoring thermometer is recommended for each 6.4.3.1 The immersion qualification test is performed with
freezing point. The thermometer is used for monitoring and the dedicated monitoring thermometer to determine the uni-
qualification testing at the specific freezing point, and for no form temperature region in the freezing-point cell. The test is
other purpose. The thermometer must be of a quality suitable made when a system is first put into service, and, thereafter,
for the purpose (see 6.4.4); in general, the monitoring ther- when substantial changes are made in the cell heating device
mometer should be more sensitive and stable than the ther- and control system.
mometers to be calibrated in the freezing-point cell. Cells of 6.4.3.2 A freezing plateau is established in the freezing-
the highest quality should be monitored and qualified with point cell, and the temperature profile of the portion of the well
calibrated standard platinum resistance thermometers. surrounded by the reference material is determined with the
6.3.5 A reference temperature such as the ice point or the monitoring thermometer while the plateau is maintained. The
triple point of water may be required for some monitoring uniform temperature region is that region where temperature
thermometers. If the monitoring thermometer is a standard differences are not significant for the intended application.
platinum resistance thermometer, the reference temperature 6.4.3.3 The freezing-point cell is qualified for the calibra-
should be the triple point of water. tion of thermometers that can be accommodated within the
6.4 Qualification Testing: uniform temperature region (see 6.5).
6.4.1 Complete Qualification Test: 6.4.4 Dedicated Monitoring Thermometers:
6.4.1.1 A complete qualification test should be performed 6.4.4.1 A monitoring thermometer suitable for evaluating
each time the equipment is set up, or if the equipment, operator, features of the freezing curve (for example, the undercool, the
or procedure is changed in a significant way, or at any time shape and duration of the freezing plateau, freezing range of a
when an anomalous result is observed during use of the cell. single freeze) must be sensitive enough to show the features
The purpose of this test is to observe whether or not any distinctly, and it must be stable enough to avoid degrading the
changes have occurred in the characteristic features of the observations with thermometer drift. Past performance history
freezing curve that imply a change in the freezing point of the of the thermometer can aid in assessing its suitability.
reference material in the cell. 6.4.4.2 Repeatability of the freezing point from one freeze
6.4.1.2 In a complete qualification test, the entire freezing to the next can be determined with a monitoring thermometer
curve is observed using the monitoring thermometer. Observa- only if it is known that the thermometer does not change
tions are started while the reference material is completely significantly with use. If the monitoring thermometer is a
liquid and continued until all of the material is frozen. precision platinum resistance thermometer, measurements
Observations are made of the magnitude of the undercool, the made at a reference temperature (for example, the triple point
shape and flatness of the freezing plateau, the freezing point, of water or the ice point) before and after the freezing-point
and the range of temperature over which the material freezes. measurements are useful in assessing thermometer stability. If
6.4.1.3 If no significant change from the freezing curve of the monitoring thermometer is a standard platinum resistance
the previous qualification test is observed, the freezing-point thermometer, the assessment should be based on the ratio of the
cell is qualified for use, and the entire system is under thermometer resistance at the freezing point to the resistance at
statistical control. the triple point of water.
6.4.2 Incidental Qualification Test: 6.4.4.3 The thermometer used for determining the tempera-
6.4.2.1 An incidental qualification test is conducted with the ture profile in a freezing-point cell must be sensitive enough for
dedicated monitoring thermometer each time the freezing-point the task, and it must not permit a significant transfer of heat
cell is used for thermometer calibration. The purpose of the test along the length of the well axis. In determining the uniform
is to ensure that the reference material starts in the liquid state, temperature region of the measuring well, the length of the
that all calibration measurements are at the freezing point, and temperature sensitive region of the thermometer must be
that the freezing point has not changed significantly since the accounted for.
previous freeze. 6.4.5 Interpretation of Qualification Test Observations:
6.4.2.2 Observations with the monitoring thermometer are 6.4.5.1 A distinct decrease from previous observations in
started while the reference material is liquid and are continued the magnitude of the maximum undercool may indicate con-
through the undercool to the first part of the freezing plateau. tamination of the reference material. However, recent past

6
 E 1502 – 98 (2003)e1
temperature history of the cell can also influence the maxi- such as resistance thermometers (see Test Methods E 644),
mum. An unusually shallow undercool, or the complete ab- thermocouples, and others.
sence of an undercool, indicates that the reference material was 6.5.2 Thermometers suitable for calibration in a freezing-
probably not completely molten before the freezing cycle was point cell are characterized in 1.5. A thermometer must be long
started. enough to extend fully into the well, and all of the temperature-
6.4.5.2 A distinct increase in the range of temperature over sensing portion of the thermometer must be contained in the
which the entire quantity of reference material freezes probably isothermal region of the well, as determined in 6.4.3. There
indicates that contamination of the material has occurred. It is should be no difference in the indication of a thermometer
useful to verify an increase in freezing range by observing a under test, attributable to unwanted heat transfer by the
corresponding increase in melting range. The amount of thermometer, when its temperature sensing portion is moved
contamination, and the resulting depression of the freezing between extremes in the uniform temperature region of the
point, may be estimated roughly using the method in 6.1.6. well, that is significant in the intended application or use of the
6.4.5.3 A decrease in the duration of the freezing plateau, thermometer.
without a corresponding decrease in the total freezing time, 6.5.3 Heat is transferred between the cell and a thermometer
also indicates that possible contamination has occurred. A in the measuring well mainly by radiation and by conduction
decrease in both plateau duration and total freezing time may through the gas-filled annulus between the well and the
indicate that the reference material is losing heat more rapidly thermometer. Conduction can be enhanced by use of a close-
because of a change in the heating device or its control. fitting metal or graphite bushing in the annulus, but then
6.4.5.4 For the incidental qualification test, two measure- oxygen must be eliminated from the annulus in order to prevent
ments on the freezing plateau are made with the monitoring false readings due to oxidation.
thermometer, one before test thermometer calibration and one 6.5.4 It is usually advantageous to heat thermometers to
after. If the second measurement is significantly lower than the near the freezing point before they are inserted into the
first, this indicates that the plateau duration is not long enough freezing-point cell. This reduces the heat load on the cell; it
for the calibration load. If the second measurement is signifi- helps to prolong the freezing plateau, and it reduces demand on
cantly higher than the first, this indicates that some of the temperature control systems. A thermometer is conveniently
reference material may be remelting, instead of freezing. heated in an auxiliary device held at a temperature above the
6.4.5.5 Failure to observe a uniform temperature region in freezing point. With a little practice, the thermometer can be
the immersion qualification test indicates that the freezing- transferred to the cell without excessive cooling.
point cell does not provide adequate immersion into the 6.5.5 The thermometer temperature must become steady at
freezing reference material for the monitoring thermometer, or the freezing point before the thermometer is calibrated. The
that the heating device is not establishing an adequately temperature is steady when the thermometer indication no
uniform freezing environment for the cell. longer changes significantly with time.
6.4.5.6 If measurements at the freezing point with a stable 7. Test Procedure
monitoring thermometer (see 6.4.4.2) indicate a significant 7.1 Prepare the test equipment.
decrease in the freezing point from one freeze to the next, 7.1.1 Check and adjust all measuring, recording, and con-
contamination of the reference material is the probable cause. trolling equipment for correct operation.
When a freezing-point cell is used at the highest level of 7.1.2 Prepare the monitoring thermometer and make
accuracy, small changes (1 or 2 mK) may be significant, and it reference-temperature measurements. If the monitoring ther-
becomes a difficult matter to determine whether an observed mometer is a standard platinum resistance thermometer, deter-
change should be attributed to the thermometer or the cell, or mine its resistance at the triple point of water (see 6.4.4.2).
both. The recorded trend of complete qualification tests helps 7.1.3 With the freezing-point cell installed, supply power to
to reveal any significant changes in the cell. the heating device and stabilize the temperature several kelvins
6.4.5.7 If repeated measurements at the freezing point with below the freezing point, as indicated by the control system.
the monitoring thermometer indicate no significant change Record control parameters.
from one freeze to the next, then the measurements may be 7.1.4 Establish the temperature of the auxiliary heating
used to derive a value for the precision component of uncer- device about 20 K above the freezing point (see 6.5.4). Record
tainty of the combined thermometer-cell system. The resulting control parameters.
value can be considered an upper bound to the precision 7.1.5 Time each significant event and datum in each proce-
component of the freezing point itself. dure. Record times as real time, or as elapsed time from the
6.4.5.8 If, upon evaluation of all qualification tests, it is time of a reference event.
concluded that a significant change has occurred in the 7.2 Allow the reference material to melt.
freezing-point cell, then the value of temperature assigned to 7.2.1 Insert the monitoring thermometer into the cell well.
the cell or the uncertainty associated with the value, or both, 7.2.2 Adjust the controls to stabilize the heating device at a
must be redetermined. temperature approximately 5 K above the freezing point.
6.5 Thermometer Calibration: Record control parameters. (Warning—Overheating may
6.5.1 The freezing-point cell can be used to realize a damage the cell.)
prolonged and repeatable fixed temperature environment for 7.2.3 Note the indications of the monitoring thermometer at
the calibration of a variety of immersion-type thermometers the onset, during, and at completion of melting.

7
 E 1502 – 98 (2003)e1
7.2.4 Continue to observe the indication of the monitoring indicates a steady temperature, record its indication. If it has
thermometer until all the reference material is molten and the been determined previously that the test thermometer meets the
cell is at the steady temperature of the heating device. Evaluate requirements of 6.5.2, then remove it from the cell. Otherwise,
the setting of the heating device control, based on the indica- raise and hold the test thermometer so that its temperature-
tion of the monitoring thermometer, and note any adjustments sensing region is near the top of, but inside, the uniform
to the control parameters implied by the evaluation. temperature region determined in 7.5.2 (see also 6.4.3). When
7.3 Establish the freezing point. the indication of the test thermometer becomes steady, record
7.3.1 Adjust the controls to stabilize the temperature of the the indication. If the temperature equivalent of the difference
heating device approximately 1 K below the freezing point of between the two indications is not significant, the test ther-
the reference material. Record the control parameters. mometer meets the requirements of 6.5.2.
7.3.2 Observe the indications of the thermometer in the well 7.6.2 Repeat the procedure for the next test thermometer, if
as the temperature decreases into the undercool. If the freeze is any. See 6.5 for details. After calibration of the last test
for a complete qualification test (see 6.4.1), record the indica- thermometer, replace the monitoring thermometer in the cell
tions continuously or at frequent intervals to establish the shape well and proceed as in 7.4 or 7.5.
of the freezing curve. 7.7 Remove the monitoring thermometer from the cell and
7.3.3 If freezing is to be induced by inside nucleation (see make any appropriate low-temperature reference measure-
6.2.7.1), continue to observe or record thermometer indications ments (see 6.4.4.2).
until recalescence is detected. Note and record the maximum 8. Documentation
undercool. Remove the thermometer from the well and insert a
8.1 Purpose and Scope:
room-temperature rod or tube (ceramic or silica glass for
8.1.1 Thorough documentation provides a permanent, com-
temperatures greater than 420 °C) for at least 60 s, then replace
prehensive historical record of the freezing-point cell and its
the rod or tube with the monitoring thermometer.
auxiliary apparatus sufficient to support an estimate of the
7.3.4 If freezing is to be induced by outside nucleation (see
quality of the cell, and an evaluation of the procedure for using
6.2.7.2), remove the freezing-point cell from the heating device
the cell. The documentation system should be designed to meet
when the thermometer in the well indicates that the tempera-
these purposes.
ture is below the freezing point. Keep the thermometer in the
8.1.2 The documentation should include experimental data;
well and continue recording or observing its indications as the
histories of the cell, monitoring thermometer, and auxiliary
cell is held at room temperature. As soon as the thermometer
equipment; and calculations required for evaluating results.
indicates recalescence, replace the cell in its heating device.
8.2 Experimental Data:
Note and record the maximum undercool.
8.2.1 Configuration data should include identification of the
7.4 Observe the indication of the monitoring thermometer freezing-point cell and all other apparatus by unique serial
as its temperature approaches the freezing point. When the number, instrument and control settings, relevant ambient
temperature is steady (see 6.5.5), record the thermometer conditions, narrative description of setup (or departure from
indication, and then proceed to 7.5, 7.6, or 7.7, as appropriate. normal setup), date, and name of operator.
7.5 Qualification Testing: 8.2.2 Measurement data should be recorded in the natural
7.5.1 For the complete qualification test (see 6.4.1), record units (for example, volts, ohms) of the thermometric property
the indication of the monitoring thermometer continuously or whenever possible. The time of each determination should be
at frequent intervals to establish the freezing curve. Continue recorded. Corrections to the data (for example, measuring
recording until all of the reference material is frozen and the instrument calibration corrections) should be shown explicitly.
temperature in the cell approaches the temperature of the 8.2.3 Procedural and incidental data should be recorded as
heating device. Evaluate the setting of the heating device appropriate. These should include the time of all procedural
control, based on the indication of the monitoring thermometer, actions, and the time and a brief description of any observed
and note any adjustments to the control parameters implied by experimental anomalies.
the evaluation. 8.3 Freezing-Point Cell Records:
7.5.2 For the immersion qualification test (see 6.4.3), pro- 8.3.1 Initial Description and Characteristics:
ceed as in 7.5.1 until the monitoring thermometer indicates that 8.3.1.1 Source of the cell,
the freezing plateau has been reached. Raise and hold the 8.3.1.2 Date acquired and placed into service,
monitoring thermometer so that its temperature sensing portion 8.3.1.3 Mass and chemical composition of reference mate-
is near the top of the reentrant well. When the thermometer rial,
indication becomes steady, record the indication. Lower the 8.3.1.4 Critical dimensions, including depth of immersion,
monitoring thermometer a predetermined distance, wait for a and
steady indication, and record the indication as before. Repeat 8.3.1.5 Assigned value of freezing point and associated
this process at five to ten uniformly spaced stations in the uncertainty.
reentrant well until the monitoring thermometer is again fully 8.3.2 History of Cell Use:
immersed. Then continue recording as in 7.5.1. 8.3.2.1 The cumulative time above room temperature,
7.6 Thermometer Calibration: 8.3.2.2 The cumulative time at or near the freezing point,
7.6.1 Remove the monitoring thermometer from the cell and 8.3.2.3 A description of accidents, abuse, and unusual use,
insert a heated test thermometer. When the test thermometer and

8
 E 1502 – 98 (2003)e1
8.3.2.4 A complete description of any cell modification and where the quantity dX/dT is the first derivative of X as a
its purpose. function of T, evaluated at the midpoint of the temperature
8.3.3 History of Cell Performance: increment. The value of dX/dT may be obtained from the
8.3.3.1 The maximum undercool, thermometer calibration or from reference tables.
8.3.3.2 The indication of the dedicated monitoring ther- 8.6.2 Freezing Point Depression:
mometer at the freezing point, 8.6.2.1 If the concentration of impurities in the reference
8.3.3.3 For a calibration test, the difference between the first material is known, and the impurities are known to depress the
and final indications of the monitoring thermometer, and freezing point, the depression of the freezing point, DT, from
that of the ideally pure material may be estimated by the
8.3.3.4 For a complete qualification test, the duration of the
approximation
freezing plateau, the time required for complete freezing, and
the freezing range. C
DT 5 A (3)
8.4 Dedicated Monitoring Thermometer Records:
8.4.1 The thermometer records should include initial de-
scription, characteristics, and history of use comparable to where C is the mole fraction of impurities, and A is the first
those for the freezing-point cell in 8.3.1 and 8.3.2. cryoscopic constant of the reference material (see Table 1).
8.4.2 The records should include the results of all calibra- 8.6.2.2 If the freezing point depression, DT, caused by
tions of the thermometer and all available information dealing impurities is small, it may be estimated from measurements on
with thermometer contributions to uncertainty. the freezing curve using the approximation in Eq 2:
8.4.3 In addition to the data in 8.3.3.2 and 8.3.3.3, thermom- X b 2 Xh
DT 5 dX/dT (4)
eter performance records should include the results of any
measurements at low-temperature reference points. If the
monitoring thermometer is a platinum resistance thermometer, where Xb is the indication of the monitoring thermometer at
performance records should be tabulated in terms of resistance the beginning of the freezing plateau, and Xh is the indication
ratios, for example, R(T)/R(0) or R(T)/R(TP), where R(T) is the when half the reference material is frozen. For a constant
resistance of the thermometer at temperature T, R(0) is the cooling rate, it may be assumed that half the material is frozen
resistance at 0 °C, and R(TP) is the resistance at the triple point when half the total freezing time has elapsed (see Fig. 1).
of water. 8.6.3 Variability:
8.5 Other Auxiliary Equipment: 8.6.3.1 If a set of observations, covering many qualification
8.5.1 The equipment records should include initial descrip- tests, shows no significant change or drift in the indications of
tion, characteristics, and history of use comparable to those for the dedicated monitoring thermometer at the freezing point,
the freezing-point cell in 8.3.1 and 8.3.2. then the variability of the combined thermometer-freezing
8.5.2 The records should include the results of all calibra- point system may be expressed by:
tions of measuring instruments, tabulations of current and past
instrumental corrections, and all available information dealing
with measuring instrument contributions to uncertainty.
S5 Œ (~X̄ 2 Xi! 2
~n 2 1!~dX / dT! 2
(5)

8.5.3 The records should include tabulations of current and where X̄ is the mean of the set of indications, Xi is the i-th
past control settings. indication of the set, n is the number of determinations in the
8.5.4 The records should include a schedule for periodic set, and S is the estimate of the standard deviation of one
maintenance and calibration of equipment. determination in terms of temperature.
8.6 Calculations: 8.6.3.2 If a set of indications, covering many qualification
8.6.1 Temperature and Temperature Difference: tests, of the dedicated monitoring thermometer at the freezing
8.6.1.1 Thermometer indications in terms of the thermomet- point displays a uniform drift with time that the thermometer is
ric property, X, are converted to values of temperature, T, on a at elevated temperature, then a linear regression analysis can
particular temperature scale, through the relationship between yield both a drift rate and standard deviation of the combined
X and T determined when the thermometer is calibrated. The thermometer-freezing point system.
relationship may be expressed by a formula or by a calibration
table, or the conversion may be made automatically by 9. Precision and Bias
recording instrumentation. Note that in the case of a platinum 9.1 A freezing-point cell contributes to the uncertainty of a
resistance thermometer serving as a monitoring thermometer, thermometer calibration because of variability in the cell and
resistance ratio (see 8.4.3) is the usual quantity related to its use, and because of uncertainty in the value assigned to the
temperature. freezing point of the cell.
8.6.1.2 Small temperature differences or increments, DT, 9.2 The precision component due to cell variability and use
may be estimated from small differences or increments in X, must be evaluated from an analysis of measurement variability
DX, by the approximation of the combined system, consisting of the monitoring ther-
DX mometer and the freezing-point cell. Separating thermometer
DT 5 dX/dT (2) variability from cell variability is usually not possible. If the
system is very stable, the evaluation of 8.6.3.1 yields an upper

9
 E 1502 – 98 (2003)e1
bound on the standard deviation of the freezing point of the the value must come from analysis of the measurement process
cell. The best of such systems have standard deviations of less upon which the determined value is based. See 6.2.8.
than 1 mK.
9.3 The uncertainty in the value assigned to the cell freezing 10. Keywords
point contributes a component to thermometer calibration 10.1 calibration; cryoscopic constant; fixed point; freeze;
uncertainty. If the value assigned is that attributed to the freezing curve; freezing plateau; freezing point; freezing-point
freezing point of pure material, then the contribution to cell; freezing range; ITS-90; precision; qualification; recales-
uncertainty caused by impurities in the cell reference material cence; reference material; reference temperature; resistance
may be estimated by the methods of 8.6.2. If a freezing-point thermometer; thermocouple; thermometer; uncertainty;
value is determined by measurement, then the uncertainty in undercool

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or [email protected] (e-mail); or through the ASTM website
(www.astm.org).

10