Polymer Quenchants distortion and quench-cracking problems.

Aqueous
polymer quenchant solutions were developed for
these applications since they typically exhibit quench
severities intermediate between those achievable for
water and conventional oil, as shown by the Grossman
1. Introduction quench severity (H-value) data shown in Table 1.

Aqueous polymer quenching technology has ad-

vanced greatly over more than 40 years of use. How- 2.2 Chemistry of Polymer Quenchants
ever, integration of the fundamental understanding of
this technology into more traditional understanding Aqueous polymer solutions have been used to
of quenchants such as water and oil has been rela- quench metals such as steel and aluminum since the
tively slow. This is evident in proper treatment of 1960s. Some of the more common water-soluble poly-
the use of aqueous polymer solutions as quenchants mers that have been used as quenchants include:
in heat-treating and metallurgy textbooks, standards poly(alkylene glycol) (PAG), more commonly known
development, and industrial process design. The as poly(alkylene oxides) (PAO), poly(vinyl pyrroli-
objective of this article is to summarize the impor- done) (PVP), poly(sodium acrylate) (PSA), poly(ethyl
tant aspects of the chemical and physical pro- oxazoline) (PEOX) and others (Totten 1990). Table 2
perties of polymers that affect process heat transfer provides a selected summary of the various polymers
(quenching), including: polymer structure and quen- that have been reported as quenchants worldwide
chant composition, an overview of the basic surface (Totten et al. 1993a, 1993b).
rewetting process, discussion of the ‘‘problem’’ of Although all of the examples of water-soluble pol-
Newtonian versus non-Newtonian heat transfer, ymers shown in Table 2 may be used as quenchants,
quantification of agitation and the constituent the performance of individual polymers may vary
parameters on heat transfer during quenching, and markedly from others due to the different interfacial
a discussion of the basic surface chemistry of the viscosities as a function of interfacial temperature
quench media itself. and the degree of polymer hydration, polymer film
strength, surface activity, total energy of adsorption,
and factors affecting stability. Therefore, a funda-
mental understanding of the effect of polymer
2. Discussion structure is important in order to understand the
2.1 Alternatives to Water and Oil differences in the performance of a polymer quen-
chant in the heat-treating operation. This article will
One of the most common and traditionally impor- provide an overview of some of the more important
tant quenching and cooling media is water. Water effects relating to polymer structure.
is generally available and nontoxic; however, it is a
relatively severe quench medium as shown in Table 1
(Totten et al. 1993a, 1993b). Many steels undergo 2.3 Polymer Characteristics
cracking, increased distortion, and soft-spotting
when quenched in water. Traditionally, when a less A polymer is a large molecule built up by the rep-
severe quench rate is desired, a quench oil is selected. etition of smaller chemical units called ‘‘monomers’’
Oils, however, possess several substantial disadvan- as illustrated by the PSA example shown in Fig. 1.
tages including: relatively limited variability in Monomers used to synthesize the most commercially
quench rates, fire hazards, smoke emissions, and dis- important quenchant polymers are shown in Fig. 2.
posal problems (Singh and Chakravorty 1987). Many PSA is a ‘‘homopolymer’’ because it is synthesized
applications require intermediate quench severities from only one type of monomer structural unit.
to achieve the desired hardness while minimizing However, a water-soluble quenchant polymer may be
composed of more than one monomer and is then
designated a ‘‘copolymer.’’ An example of a quen-
chant copolymer is PAG, which is synthesized from
Table 1 ethylene oxide and propylene oxide monomers as
Typical Grossman H-values for various common shown in Fig. 2. The simplest structural unit that re-
quench media. peats itself in the long-chain polymeric structure is
designated as a ‘‘repeat unit.’’ The total polymer unit
Quenchant Grossman H-value from beginning to end is called a ‘‘polymer chain.’’
Brine 2.0–5.0 The size of a polymer is defined either by its mole-
Distilled water (room temperature) 0.90–2.0 cular weight (MW) or its degree of polymerization
Aqueous polymer solutions 0.2–1.2 (DP). The MW of a single polymer unit is equal to
Conventional and fast quench oils 0.25–0.80 the number of repeat units (n or n þ m in Fig. 2) times
the MW of the repeat unit. The larger the polymer,

1
Polymer Quenchants

Table 2
Examples of water-soluble polymers used as quenchants.

Polymer Comment Reference

Polyalkylene glycol—PAG Linear with molecular weight (MW) of 7000– Schoch et al. (1984)
(also called polyalkylene 10 000
oxide)
High-MW branched polymers Blackwood and Cheesman (1965)
Glycol blends with PAG Ethylene glycol, propylene glycol, and Totten (1993a, 1993b, 1993c)
hexamethylene glycol blends with PAG
Polyacrylamide Terekhova et al. (1984)
Zakamaldin et al. (1983)
Polyethylene oxide MW ranges 500 000–5 000 000 Bedarev et al. (1978)
Chase et al. (1962)
Considerably lower MW than those in Hibi and Shori (1985)
(Bedarev et al. 1978, Chase et al. 1962)
Cellulosic derivatives Quenching performance compared with water Wyszkowski and Sobol (1986)
and oil
Carboxymethyl cellulose (CMC), methyl Gordon (1956)
cellulose (MC), hydroxyethyl cellulose Trusculescu et al. (1986)
(HEC), etc. Ezhor et al. (1988)

Vinyl copolymers Polyvinyl alcohol, polyvinyl acetate Cornell (1952)

Polyethyloxazoline MW ranges from 50 000 to 500 000 Warchol (1985)
Polyvinyl pyrrolidone MW ranges from 5000 to 400 000 Meszaros (1975)
Polysodium acrylate Viscosities of 20% solutions may vary from Kopietz and Munjat (1978)
5000 to 100 000 cps at 25 1C (77 1F)
Polyisobutylenemaleic acid Anionic copolymer of isobutylene and maleic Nakamura and Junkatsu (1987),
acid (MW approx. 60 000) Murakami and Shori (1985)

Polyurethane Prepared by reacting a low-MW PAG with a Knopf (1984)

diisocyanate (e.g., isophorone diisocyanate)

Catalyst
CH2 CH + CH2 CH (CH2CHCH2CH)
C O C O C O C O
ONa ONa ONa ONa
Sodium acrylate
monomer
CH1 CH
(n − 3)CH2 CH
C O
C O ONa
ONa
(CH2CH)n (CH2CHCH2CHCH2CH)
C O C O C O C O
ONa ONa ONa ONa
Poly(sodium acrylate) polymer
Figure 1
Synthesis of PSA from sodium acrylate monomer.

the greater the MW. The total MW of the polymer increases as the DP increases. Polymers that are most
chain divided by the MW of a repeat unit is the DP, typically used as quenchants have DPs varying from
which is also equal to the average number of repeat approximately 100 to more than 10 000. The MW of
units per chain. Therefore, the size of a polymer polymers in quenching applications is of particular

2
 Polymer Quenchants

Poly(sodium acrylate) (PSA)

mCH2 CH —(CH2CH)n —

C O C O

ONa ONa
Sodium acrylate PSA
Figure 3
Poly(vinyl pyrrolidone) (PVP)
Schematic illustration of polymer configuration:
(a) linear and (b) branched.
nCH2 CH [CH2CH]n

N N
O O
iron (Fe þ 3), copper (Cu þ 2), aluminum (Al þ 3) which
PVP
may be present due to corrosion to form water-in-
Vinyl pyrrolidone
soluble, ionically cross-linked network structures that
Poly(ethyl oxazoline) (PEOX) take the form of intractable slimes or gels (coacerv-
ates) or precipitates (Goddard et al. 1982, Totten et al.
N 1986). Thus, it is essential that ionic polymer be used
[NCH2CH2]n with distilled or deionized water with effective cor-
nC2H5C
O rosion inhibition systems.
C O It is also noteworthy that the unfortunate use of
C 2 H5
the term ‘‘glycol quenchants’’ is often misapplied to
Ethyl oxazoline PEOX
PAG quenchants. Most importantly, PAG polymers
Poly(alkylene glycol) (PAG) are NOT ‘‘glycols,’’ in the usual sense which most
O O typically indicate at least two hydroxyl groups/mol-
ecules such as ethylene glycol or propylene glycol.
nCH2 — CH2 + mCH2 — CH [(CH2CH2O)n(CH2CHO)m]—
Furthermore, glycols possess low MWs (usually
CH3 CH3 o100) and PAG polymers are of much higher MW
Ethylene Propylene PAG (typically in the range of 10 000–25 000). This is a
oxide oxide problem since there have been occasions where heat-
Figure 2 treaters have attempted to use glycol solutions, such
Examples of the most commonly encountered as antifreeze or coolants, instead of the proper use of
water-soluble polymers currently being used as aqueous PAG polymer solutions. Generally, glycols
quenchants. are not effective quenchants and furthermore, owing
to their relative volatility in quenching operations,
their use is not safe!

importance since the viscosity of a polymer and its

aqueous solution increase exponentially with increas- 2.4 Polymer Water Solubility
ing MW. Cooling rates decrease and interfacial film
strength increase with increasing MW. Polymers are dissolved in water by hydrogen-bonding
In addition to the polymer size, polymer con- interactions of the repeat units with water (homo-
figuration is also important. Polymers used for geneous solution). Typically, hydrogen-bonding
quenchant formulations are either linear or branched interactions can be broken by increasing the solu-
as illustrated in Fig. 3. Although both types of tion temperature. At a critical point, designated as
polymers can be used, the actual properties (such the ‘‘cloud point,’’ the degree of hydrogen bonding
as viscosity–MW, viscosity–temperature, stability, is insufficient to facilitate a true solution and the
and others) vary with polymer configuration. polymer chain or partially hydrated polymer
One illustration of the effect of polymer struc- chain coils upon itself, resulting in phase separation
ture is the problem of the use of ionic polymers as (a heterogeneous solution). It should be noted that
quenchants. For example, a number of well-known all polymers separate from aqueous solution if the
polymer quenchants are anionic in character. How- phase temperature is sufficient. In the laboratory,
ever, it is well known that anionic molecules (inclu- this is sometimes accomplished with polymers that
ding polyanionic polymers such as PSA and exhibit a cloud point 4100 1C by heating the aque-
carboxymethyl cellulose (CMC)) react with di- and ous polymer solution in a sealed tube to the cloud
trivalent metal ions (such as calcium (Ca þ 2), mag- point.
nesium (Mg þ 2), manganese (Mn þ 2), and iron Figure 4 illustrates the thermal separation process
(Fe þ 3)) that are typically present in hard water and for a common aqueous PAG polymer solution. The

3
Polymer Quenchants

water content of the separated phase depends on the

separation temperature and the ratio of ethyleneoxy
units (n) and propyleneoxy units (m) in the copolymer
structure. Typically, the cloud point decreases as the
number of propyleneoxy units increase and is ap-
proximately equal in 1C to the wt.% of ethyleneoxy
units in the PAG polymer (assuming no strong
additive effects are present). Most PAG polymer
quenchants contain 65–85% by weight ethylene
oxide. Quenching properties are very sensitive to
the specific ethylene oxide content, degree of random
versus block structure of the monomer constituents,
polymer MW, and molecular configuration (linear or
branched).
The thermal separation properties of a PAG quen-
chant can be used in some cases, such as small tanks,
to remove ionic contaminants such as hard metal ions
or heat-treating salts from the quenchant solution
(Jarvis et al. 1989).

2.5 Polymer Quenchant Formulations

Typically, the heat-treater never uses the water-sol- Figure 4
uble polymers described above directly as quenchants PAG quenchants typically exhibit cloud points
because of the extreme difficulty in handling them in o100 1C. The degree of hydration of the separated
this form in the heat-treating shop. For example, polymer solution decreases with increasing separation
polymers in their anhydrous form are either extre- temperature. This process can be used in some cases to
mely viscous liquids with a consistency similar to remove ionic contaminants from a quenchant bath.
molasses or they are solids, which may have an
appearance of hard rock candy or a powder/dust.
Although these are ultimately water-soluble, achiev-
ing clear homogeneous solutions is often not a in Fig. 5(b) (Bozhko et al. 1983). Although these data
routine task since effective solubilization procedures are illustrative, it would also be informative to sim-
may require a specific order of addition and heating ilarly compare data under isoviscous conditions
sequence or even the use of a cosolvent such as (the same viscosity).
isopropanol. Therefore, it is not generally practical
for the heat-treater to attempt to use these polymers
in their anhydrous form. 2.6 Metal Quenching Mechanisms
This problem is addressed by the use of aqueous
concentrates. Polymer quenchants, as purchased One of the objectives of the quenching process is to
and used by the heat-treater, are typically aqueous mediate heat transfer from the hot metal to the cooler
concentrates containing polymer, water, corrosion quenchant to control the formation of the desired
inhibitors, antifoams and bactericide/fungicide com- microstructure and related as-quenched properties.
positions, and other additives as needed for the The performance of an aqueous polymer quenchant,
specific polymer being used. This concentrate is then that is, its ability to mediate heat transfer and facil-
further diluted with water, either distilled or deion- itate uniform quenching, is also dependent on the
ized, to the desired concentration. Thus, what is organic and physical–chemical properties of the
actually being used is not a ‘‘polymer’’ but an ‘‘aque- quenching medium. These include the effect of pol-
ous polymer concentrate (solution).’’ ymer composition on interfacial film-forming (and
Bozhko et al. studied the effect of MW of a pol- breaking) properties, viscosity, and interfacial wet-
ymer designated as ‘‘PPS’’ (the actual chemical name ting performance.
of the polymer was not reported) on diluted solution In Fig. 6, the cooling process at the hot metal in-
viscosity and cooling curves produced by 0.3% solu- terface is compared for water (Fig. 6(a)), an aqueous
tion of this polymer. Clearly, increasing the polymer PAG polymer quenchant (Fig. 6(b)), and oil (Fig.
MW produces increasing solution viscosity as shown 6(c)). It should be noted that both the water and
in Fig. 5(a) and cooling rates produced by a fixed oil quenching media, although exhibiting very differ-
concentration, in this case 0.3%, produce decreasing ent quench severities, exhibit similar quenching
cooling rates with increasing polymer MW as shown mechanisms with a transition of film boiling (FB) to

4
 Polymer Quenchants

28 900
Molecular weights (Da)
1
24
(1) 1 428 600 800
(2) 789 400
(3) 535 700
(4) 298 800
700
20
2 (5) 158 500

Temperature (°C)
600
Viscosity 16
500
12 400
3
8 300
4 200
4 1
5 100
4 3 2
0 0.2 0.4 0.6 0.8 1.0 0 10 20 30 40 50 60
(a) Concentration (%) (b) Time (s)
Figure 5
(a) Dependence of aqueous solution viscosity of a water-soluble polymer designated as ‘‘PPS’’ as a function of the
following MWs(Da): (1) 789 400; (2) 428 600; (3) 535 700; (4) 298 800; and (5) 158 500. (b) Illustration of cooling curves
obtained with 0.3% solutions of each of these polymers (Totten et al. 1997a, 1997b: reproduced by permission of
ASM International, 1997).

nucleate boiling (NB) and then to convective MW of the polymer used to formulate the quenchant
(CONV) cooling occurring over the cooling surface (Totten 1990).
simultaneously. This is called ‘‘non-Newtonian’’ In addition to the three well-known cooling
cooling and is particularly significant because of the processes discussed here, the presence of shock-film
difference in heat transfer rates between the different boiling that typically occurs within 0.1 s after the hot
cooling mechanisms that are typically approximately: metal is first immersed into the quenchant is also im-
aFB (100–250 W m
2 K
1), aNB (10–20 kW m
2 K
1), portant. Shock-film boiling is followed by the FB,
and aCONV (700 W m
2 K
1) [x]. The time-dependent NB, and CONV cooling processes listed above. How-
differences in the surface-cooling mechanisms (wet- ever, experimental studies have shown that the
ting phases) and the enormous differences in their cooling potential of a quenchant is defined by the
values of aFB, aNB, and aCONV will affect the related first critical heat flux density (transition of shock-film
time-dependent temperature distribution within the boiling—the point where bubbles are first formed
metal and will significantly impact the magnitude and and depart from the hot metal surface—to full-film
uniformity of the resultant residual stresses and po- boiling), and second critical heat flux density
tential for distortion of the metal being quenched. (minimum heat flux at the transition from full-film
The mechanistic behavior of a polymer quenchant boiling to nucleate boiling), which are designated as
is fundamentally different, as shown in Fig. 6(b). qcr1 and qcr2, respectively. It is difficult to measure
When hot steel is quenched into an aqueous polymer qcr1 experimentally. However, Kobasko et al. and
solution, the metal is surrounded by a polymer film, others have shown that the ratio qcr2/qcr1 ¼ a con-
or membrane, which forms by polymer dehydration stant (0.204–0.207) (Kobasko et al. 1997). Using this
at the relatively high temperatures encountered at the principle and cooling curve analysis with a 50 mm
hot metal interface producing an essentially Newton- probe, Liscic determined the values of qcr1 and qcr2
ian cooling process, without the enormous variability for water, conventional unadditized oil, and a
in heat transfer rates due to the different and simul- 25% aqueous PAG quenchant solution as shown in
taneous cooling processes described above for water Table 3 (Liscic 2003). It is noteworthy that under
or oil quenchants. (These temperatures are similar to these conditions, the full-film boiling process is
those encountered for superheated steam.) Lainer significantly more stable than that exhibited by a
and Tensi (1996) have determined the vapor-film conventional unadditized quench oil.
thickness for a PAG quenchant to be 0.35–0.44 mm Water is often replaced by a dilute solution (5–8%)
during quenching. This means that the cooling proc- of an aqueous polymer quenchant to prevent crack-
ess for a polymer quenchant is fundamentally more ing with immersion quenching of a crack-sensitive
uniform than typically observed for water and oil. carbon steel or for induction hardening applications.
In this situation, heat transfer is controlled by the Cracking is prevented by enhancing the uniformity of
viscosity, film thickness, and film strength of the surface cooling by utilizing a Newtonian cooling
polymer membrane. The nature of the polymer film process. Figure 7 illustrates that a dilute solution of
that is formed will vary with the composition and the aqueous polymer quenchant provided cooling

5
Polymer Quenchants

produced essentially uniform surface cooling and

thus much lower thermal gradients than obtained
with water. Note: temperature measurement taken at
the center of instrumented probes provide relatively
poor information about the actual quenching process
and is therefore often insufficient to adequately
characterize the overall hardening process.

2.7 Summary of Quenching Variables

The impact of bath temperature, agitation, and tem-
perature on the interfacial viscosity and film strength
of the polymer film is reflected in the Grossman
quench severity (H-) factor. For example, the inter-
facial viscosity of the polymer film would be expected
to decrease with increasing temperature, thus increas-
ing quench severity. Conversely, quench severity
would be expected to decrease with increasing poly-
mer MW which will increase the interfacial viscosity,
thus inhibiting rupture of the film that is formed.
Increasing polymer MW and variation of polymer
structure may also lead to an increase in film strength
and a substantial decrease in the ability of a polymer
film to rupture, or rupture uniformly. In fact, some
polymers are not readily removed from the workpiece
except by procedures such as sand-blasting.
Increasing agitation rates would be expected to
both decrease the interfacial film stability and sweep
away the hotter quenchant solution faster, thus in-
creasing quench severity. Agitation rates are often
considered as mass flow; however, other equal factors
must also be considered. For example, Hilder showed
that the turbulence of the flow at the metal interface
must also be considered and that quench severity in-
creased with turbulence (Hilder 1988). Similarly, the
direction of flow at the metal interface is vitally im-
portant. All of these concepts affect the ‘‘quality of
flow’’ and it is important, not only for aqueous pol-
ymer quenchants but for all fluids (gas and liquid).
Many quench-related problems are directly attribut-
able to flow nonuniformity around the workpiece
during quenching. It is therefore vitally important to
optimize the uniformity of the quenching process if
cracking and distortion control problems are to be
Figure 6 minimized.
Illustration of the process of transition between full-film Figure 8 provides a comparison between the
boiling, nucleate boiling, and convective cooling of Grossman quench severity values as a function of
cylindrical bar probes (25 mm dia.  100 mm) at varying concentration, bath temperature, and agita-
850 1C quenched into: (a) water at 30 1C at 0.3 ms
1; tion (mass flow) rate for a PAG and a PVP polymer
(b) an unagitated 10% aqueous PAG quenchant quenchant (Totten et al. 1993a, 1993b). This figure
solution at 25 1C, and (c) unagitated oil at 601. clearly shows that although quench severity decreases
with increasing bath temperature and concentration
and decreasing agitation rate for both types of
properties very similar to water when measured at the polymer quenchants, the actual response of each pol-
center of a 25 mm dia  100 mm CrNi–steel probe ymer to these variables is very different. These dif-
(Tensi et al. 1994). However, the surface-cooling ferences are due to differences in interfacial cooling
properties illustrate that water produces substantial properties due to the differing physical and chemical
thermal gradients whereas the polymer quenchant properties of the two quenchants. However, it is also

6
 Polymer Quenchants

Table 3
Comparison of qcr2/qcr1 ratio for water, oil, and an aqueous polymer solution.

1st critical heat flux density 2nd critical heat flux density Time at transition
Quenchant qcr1 (MW m
2) qcr2 (MW m
2) for qcr2 (s)
Unagitated tap water at 20 1C 5.0 1.0 0.4
Unagitated unadditized 2.5 1.5 0.25
conventional quench oil at
20 1C
25% aqueous PAG quenchant 1.5 0.3 30
solution at 40 1C and
0.8 ms
1

the cooling metal surface. Variables affecting this

3 mediation process include: interfacial film viscosity
2 (including viscosity–temperature behavior), film
Temperature (°C)

750 C 1 thickness, agitation, and additives. Of these, perhaps

1
2 the most important is viscosity (Luty 1991). In gen-
500 eral, cooling rates decrease with increasing interfacial
Water 3 film viscosity (Totten et al. 1993c). Since maintaining
Polymer polymer viscosity during use is critically important,
250
three important processes that lead to polymer deg-
1_3
radation and viscosity loss during use will be
0 discussed: polymer drag-out, mechanodegradation,
0 10 20 30 40 0 5 10 15 20
(a) Time (s) (b) Time (s) thermal/oxidative degradation, and bacterial degra-
dation (Totten et al. 1997a).
Figure 7 Polymer chemistry may exhibit significant impact
Comparison of the cooling performance of water (slow on performance by the amount of polymer drag-out
wetting process) and a dilute aqueous PAG quenchant on the part surface upon removal from the quench
solution (sudden wetting process) using a preheated tank. As the polymer is progressively removed, solu-
CrNi–steel probe at 850 1C with thermocouples placed tion concentration of the polymer in the quenchant
at (a) the center and (b) surface of the probe. will decrease, leading to a corresponding increase in
quench severity. The relative susceptibility of a
polymer to drag-out is dependent on various factors
including polymer structure, molecular weight, and
important to note that conditions exist for both solution viscosity (higher solution viscosities produce
quenchants where quench severities varying from a higher drag-out rates). Some illustrative drag-out
slow oil (o0.2) to greater than water (41.0) can be properties reported earlier by Hilder for different
obtained. polymers are provided in Fig. 9 (Hilder 1988). One
The influence of the physical properties of the problem with this method of illustration of the data
quenchant and the material being quenched on the is that relative drag-out rates should be shown to
wetting process, and thus on the heat transfer coef- account for different polymer compositions in the
ficients aFB, aNB, and aCONV, and therefore on the quenchant solutions. Thus, the same absolute drag-
quenching process, is very strong (Tensi et al. 1994). out rates on the steel surface may correspond to
These can be summarized by wetting kinematics vastly different relative depletion rates from solution.
which include the variables: starting time of the wet- Therefore, changes in polymer quenchant concentra-
ting process (ts), time at the end of the surface wetting tion should be reported as a relative figure such as %
process (tf), and the difference (tf
ts) which is the loss.
wetting time tw. The influence of these parameters on One polymer degradation mechanism that is often
the wetting of the quenching process is summarized in ignored when considering the potential utility of a
Table 4. polymer for use as a quenchant is ‘‘mechano-
degradation,’’ which occurs during agitation of the
polymer and is dependent on both the polymer struc-
3. Quenchant Stability ture and molecular weight (Zarkhin et al. 1989). The
relative stabilities of three polymers using a standard
Heat transfer rates throughout the quenching process ASTM shear stability test are illustrated in Table 5
are mediated by interfacial films that are formed at (Liscic et al. 1992). Before identifying a polymer to be

7
Polymer Quenchants

140 0.2 140 0.6

0.4

Bath temperature (°F)

Bath temperature (°F)

0.6 0.8
130 0.4 130
0.8
0.6 1.0
Circulation
Circulation
0.8
120 rates 120 1.0
0.6
50
50 ft/min
ft/min 1.0
0.8
1.0 75 ft/min 1.2
110 0.8 110
1.0 100 ft/min
100 ft/min 1.21.2
1.2
100 100
15 20 25 15 20 25
(a) Quenchant concentration (%) (b) Quenchant concentration (%)

Figure 8
Comparison of the sensitivity of quench severity to polymer type and process parameters: (a) a PAG aqueous
quenchant and (b) a PVP aqueous quenchant.

Table 4 of this discussion.) There are various methods of

Effect of fluid and metal property variation on quench monitoring a polymer for oxidative degradation. One
severity. method is gel permeation chromatography (which
quantifies changes in polymer distribution). Another
Effect on property method is solution viscosity. Of course, other meth-
variation ods may be used as well. Figure 10 illustrates rather
(m ¼ increasing, severe degradation that had occurred with a PAG
k ¼ decreasing) polymer quenchant after many years of use, with
however only minor effect on quenching performance
tS, tf DtW a in this particular instance. Unfortunately, effective
antioxidants for aqueous polymer quenchants, simi-
Fluid property
lar to those used for oils, have not been identified.
Type of quenchant mk mk mk Because of potential changes in concentration due
Addition of additives mk mk mk to drag-out or polymer degradation due to thermal/
Increasing agitation (v) k k m oxidative degradation, or both, it is necessary
Increasing bath temperature (Tb) m m k to monitor polymer quenchant baths during use by
Increasing concentration m m k following well-established procedures such as those
Metal property described in ASTM D 6666-01.
Increasing thermal diffusivity (a) m m m Quenchant polymer stability can be assessed by
Increasing cross-section size m m k repeatedly quenching in a given volume of quenchant
Increasing surface roughness k k m solution and monitoring change in solution viscosity
Increasing surface oxidation k k k due to degradation normalized to the starting solu-
tion concentration (to correct for polymer drag-out
Parameters: tS, time when wetting starts; tf, time when wetting is fin-
ished; Dtw, time interval of wetting (s); a, heat transfer coefficient.
effects). The data shown in Fig. 11 were obtained in
this manner for experimental quenchant solutions
based on three PAG polymers with the same chemical
composition but with different MWs of: M, 2 M, and
used as a quenchant it is necessary to assess its sta- 5 M (Totten et al. 1997a). The results show that deg-
bility to mechanodegradation. The Waring blender radation rates increase substantially with increasing
test shown is one test that may be used. Another is to polymer MW. These results are important because
simply recirculate the diluted quenchant concentrate typically slower, more ‘‘oil-like’’ polymer quenchants
in a closed-loop pumping system at constant temper- are formulated with polymers of increasing MWs
ature and monitor solution viscosity change with (see Fig. 5). However, the data in Fig. 11 show that
time. increasing the polymer MW will produce a corre-
Polymer quenchants, like all organic materials, are sponding decrease in the lifetime of the quenchant
susceptible to thermal and oxidative degradation. (It solution during use.
is often not possible to differentiate which process is Polymer quenchants may also degrade by anaero-
occurring since air is almost always present. There- bic biodegradation processes (Blackwood et al. 1985,
fore, they will be considered together for the purposes Kramer 1992). Such degradation processes are often

8
 Polymer Quenchants

600 10 mm 600

500 500

400 Static
Drag-out (mg)
400
Specimen

Drag-out (mg)
300 Agitated
0.5 m s−1 300

200
200
100
100
0 5 15 25
5 15 25 5 15 25
PAG Polyacrylate PVP
0
0 2 4 6 8 10 12 14 16 18 20 22
(a) Concentration (vol.%) (b) Kinematic viscosity at 40°C (mm2 s−1)

Figure 9
(a) Hilder’s drag-out results for various polymer quenchants, concentrations, and agitation conditions; (b) Effect of
polymer solution viscosity on drag-out under unagitated solution conditions.

Table 5
Effect of polymer structure on stability to 800
mechanodegradation (ASTM D3519 Waring Blender
test). GPC trace or fresh
Temperature (°C)

600 and used PAG quenchant

Absorbance

Fresh
Viscosity Used
(mm2 s
1 at 50 1C) 400

Total shear time (min)

Polymer 1 2 3 Change 200 Retention time
Used
Polyacrylamide Fresh
20% hydrolyzed 6.3 6.2 6.0
0.3 0
0 20 40 60 80 100
50% hydrolyzed 6.3 5.9 5.6
0.7 Time (s)
Polyalkylene glycol 6.0 6.0 6.1 0.1
Figure 10
Illustration of the effect of severe oxidation on cooling
curve behavior.
caused by contamination by fluids such as coolants
containing bacterial and/or fungal contaminants 4. Conclusion
known as ‘‘bugs.’’ The potential occurrence of bio-
logical degradation is minimized by daily agitation of One of the most important functions of a quenchant
the tank for at least 20–30 min. If bacterial contam- is mediation of heat transfer and facilitation of uni-
ination occurs, it may be treated with the addition of form heat transfer throughout the quenching process.
a bactericide or fungicide. Appropriate recommen- The information provided here, although certainly
dations can be obtained from the quenchant supplier. not comprehensive, clearly shows that various aspects
However, it is important to note that under abso- of the organic and physical–chemical properties of a
lutely no conditions should an active chlorine bacte- quenchant may exhibit dramatic effects on both cool-
ricide (also known as ‘‘swimming pool chemicals’’) be ing rates and potentially on quench uniformity.
used with PAG polymers since spontaneous poly- Therefore, substantially greater concern for the im-
mer degradation will occur, thus damaging the pact of these properties on the quench process must
quenchant. be considered. Additionally, the variation of the

9
Polymer Quenchants

Ezhor N M, Semakin S V, Svetkichnyi V V, Mazaer A M,

30 Soboler Yu V 1988 Ebergomaschinostroenie 1, 35–8
Bath composition
Goddard E D, Matteson G H, Totten G E 1982 Counterion
effects in a cationic monolayer. J. Coll. Interface Sci. 85,
Polymer MW = 5M (15−24%) 19–27
Gordon M 1956 U.S. Patent 2,770,564
(% of change by adjusted viscosity)

Hibi T, Shori N 1985 25 (1), 8–13

Relative quenchant degradation

Hilder N A 1988. The behavior of polymer quenchants. PhD

Thesis, University of Birmingham, Aston
20 Jarvis L M, Blackwood R R, Totten G E 1989 Thermal sep-
aration of polymer quenchants for more efficient heat treat-
ments. Ind. Heat. November, 23–4
Polymer MW = 2M (20%) Knopf R J 1984 U.S. Patent 4,481,367
Kobasko N I, Moskalenko A A, Totten G E, Webster G M
1997 Experimental determination of the first and second
critical heat flux densities and quench process characteriza-
10 tion. J Mat. Eng. Perf. 6 (1), 93–101
Kopietz K-H, Munjat F S 1978 U.S. Patent 4,087,290
Kramer J F 1992 The use of glutaraldehyde for microbiological
Polymer MW = M (15−20%) control in polymer quenchants. In: Totten G E (ed.) Conf.
Proceed. Quenching and Distortion Control. ASM Interna-
tional, Materials OH, pp. 141–5
Lainer K, Tensi H M 1996 Determination of vapor film thick-
ness during immersion cooling in aqueous polymer solutions.
0 1 2 3 Totten G E, Funatani K, Howes M A H, Sjostrom S (eds.)
Repeat quench time (d) Conf. Proceed. 2nd International Conference on Quenching
and Control of Distortion. ASM International, Materials OH,
Figure 11 pp. 75–82
Relative degradation of three experimental PAG Liscic B 2003 Critical heat flux density, quenching intensity and
polymer quenchants with the same ethyleno oxide heat extraction dynamics during quenching in vaporizable
content (75% by weight) but with different MWs liquids. In: Proc. of 22nd Heat Treating Society Conference
designated as: M, 2 M, and 5 M. and the 2nd Surface Engineering Congress, September 15–17,
2003. Indianapolis, IN, ASM International, Materials Park,
OH, pp. 161–8
Liscic B, Tensi H M, Luty W (eds.) 1992 Theory and Technology
chemistry of the quenching process during use must
of Quenching. Springer-Verlag
also be monitored if optimal quality control is to be Luty W 1991 Types of cooling media and their properties.
achieved. In: Liscic B, Tensi H M, Luty W (eds.) Theory and Technol-
ogy of Quenching. Springer-Verlag, Berlin, Germany,
pp. 248–340
Meszaros A G 1975 U.S. Patent 3,902,929
Bibliography Murakami Y, Shori 1985 Netsu 25 (1), 14–5
Nakamura E, Junkatsu 1987 32 (8), 570–4
ASTM D 6666-01 Standard Guide for Evaluation of Aqueous Schoch E, Reyrol F, Beck G, Moreaux F 1984 Brit. Patent
Polymer Quenchants. ASTM International, 100 Barr Harbor 2,132,638A
Drive, West Conshohocken, PA Singh R K, Chakravorty 1987 Tool Alloy Steels 21 (5), 145–7
ASTM D 6482-99 Standard Method for Cooling Curve Analysis Tensi H M, Stich A, Totten G E 1995 Fundamentals of
of Aqueous Polymer Quenchants by Cooling Curve Analysis quenching. Metal Heat Treating Mar./Apr., 20–8
with Agitation (Tensi Method). ASTM International, 100 Terekhova N F, Tanicheva O N, Tiunuva N M, Lobanova N M
Barr Harbor Drive, West Conshohocken, PA 1984 Soviet Patent 1,250,586
ASTM D 6549-01 Standard Method for Cooling Curve Analysis Totten G E 1990 Polymer quenchants: the basics. Adv. Mat. &
of Aqueous Polymer Quenchants by Cooling Curve Analysis Proc. 137 (3), 51–3
with Agitation (Drayton Unit). ASTM International, 100 Barr Totten G E, Bates C E, Clinton N A 1993a Chapter 5—polymer
Harbor Drive, West Conshohocken, PA quenchants. In: Handbook of Quenchants and Quenching
Bedarev A S, Ye G, Il’Yushko G, Beloborodov G I, Ko- Technology. ASM International, Materials Park, OH, pp.
nyukhov G P, Al’tergot E V 1978 Soviet Patent 600,190 161–90
Blackwood R R, Cheesman W D 1965 U.S. Patent 3,220,893 Totten G E, Bates C E, Clinton N A 1993b Chapter 9—quench
Blackwood R R, Lorbach M M, Mueller E R 1985 Is something system design. In: Handbook of Quenchants and Quenching
bugging your polymer quenchant. Heat Treat July, 40–2 Technology. ASM International, Materials Park, OH, pp.
Bozhko G T, Bannykh O A, Tropkina M N, Manikhin P I, 339–411
Popov A V 1983 Effect of aqueous polymer solution viscosity Totten G E, Narumi T, Jarvis L M 1993c Quenchant chemistry
on cooling capacity. Mat. Sci. and Heat Treat. 25 (11–12), holds the key to improved shop operations. Heat. Treat.
810–3 December, 16–9
Chase F L, Ewing D, Ewing C W 1962 U.S. Patent 3,022,205 Totten G E, Goddard E D, Matteson G H, Wanchisen M L
Cornell E R 1952 U.S. Patent 2,600,290 1986 Counterion effects on the aqueous solution

10
 Polymer Quenchants

viscosity of cationic surfactants. J. Am. Oil Chem. Soc. 63, Totten G E, Webster G M, Kang S H, Han S W 1997b Direct
1586–9 forge quench quenching with poly(alkylene glycol) polymer
Totten G E, Webster G M 1996 Quenching fundamentals: quenchants. In: Asfahani R (ed.) Accelerated Cooling/Direct
maintaining polymer quenchants. Adv. Mat. & Proc. June, Quenching of Steels. ASM International, Materials Park,
64AA–64DD OH, 15–18 Sept., pp. 207–16
Totten G E, Webster G M, Blackwood R R, Jarvis L M, Trusculescu M, Trusculescu D, Serban D, Raduta A 1986
Narumi T 1995 Designing chute quench for conti- Cercet. Metal. 27, 223–8
nuous furnace heat treating effectively. Ind. Heat. Nov., Warchol J F 1985 U.S. Patent 4,528,044
49–52. Wyszkowski J, Sobol St. 1986 Proc. 5th Int. Congr. Heat Treat.
Totten G E, Webster G M, Jarvis L M, Kang S H, Mater. Budapest, 3, 1836–43
Han S W 1997a Thermal/oxidative stability and polymer Zakamaldin A A, Kochurkina Yu I, Tikhonov G I 1983 Met.
drag-out behavior of polymer quenchants. In: Milam D, Sci. Heat Treat. 255 (1–2), 13–8
Poteet D, Pfaffmann G, Albert W, Muhlbauer A, Rudnev V Zarkhin L S, Sheberstov S V, Panfolovich N V, Manevich L I
(eds.) 17th Heat Treating Society Conference Proceedings 1989 Russ Chem. Rev. 58, 381
Including the 1st International Induction Heat Treating
Symposium. ASM International, Materials Park, OH,
pp. 443–8 G. E. Totten and L. C. F. Canale

Copyright r 2005 Elsevier Ltd.

All rights reserved. No part of this publication may be reproduced, stored in any retrieval system or transmitted
in any form or by any means: electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or
otherwise, without permission in writing from the publishers.
Encyclopedia of Materials: Science and Technology
ISBN: 0-08-043152-6
pp. 1–11

11