THERMAL STABILITY OF BENTONITES IN

FOUNDRY MOLDING SAND*

hy
GEORGE J. VINGAS and ARTHUR H . ZRIMSEKJ
Magnet Cove Barium Corporation, Arlington Heights, Illinois

ABSTRACT
T H E EFFECTS of heat on the behavior of sodium and calcium bentonite in bonding
foundry sands were studied.
Tests such as differential thermal analysis, viscosity, and melting or vitrification
temperatures do not indicate the suitability of a bentonite for foundry use.
The deterioration temperatures of sodium and calcium bentonite are 1180° and 600°F
respectively.
Reactivities of bentonites with cereal as affected by temperature and mulling were also
studied. Flowability of sands, defined as mold densiflcation in this paper, is better with
heated calcium than sodium bentonite.
Heating bentonites alone does not simulate the conditions encountered in heating
bentonite-sand mixtures in a mold.

INTRODUCTION
SYNTHETIC molding sands, used by the foundry industry to contain and shape
liquid metal, are essentially composed of a pure quartz sand, bentonite, and
an additive, such as pulverized coal or cereal, and water.
The molding sand is mixed in a sand muUer and is used to prepare a mold.
Molten metal is poured into the cavity. After the metal solidifies, the resulting
casting is removed from the mold for final cleaning.
Most iron foundries re-use the sand indefinitely, but steel foundries favor
the use of facing sand, prepared from all new materials, for the surface against
the casting, with the bulk of the mold made with the re-used sand.
The foundry literature (Zrimsek and Vingas, 1960, 1961a, b and c; 1962;
Vingas and Zrimsek, 1961; Heine et al., 1959; Ojala, 1962) contains consider-
able information on the properties of bentonite-bonded new sand mixtures.
Few data have been presented on the rate and nature of changes produced
by re-use.
The high temperature of the metal during casting heat-treats the molding
* The writers thank Magnet Barium Corporation for permission to conduct these
experiments and publish the results.
t P.O. Box 355, Arlington Heights, Illinois.
367
368 THIRTEENTH NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

sand to temperatures ranging from room to pouring temperature of the metal.

At some temperature, the bonding agents in the sands become inactive, and
are replenished by new materials. Knowledge of the effects of temperature on
foundry molding materials is, therefore, important, to determine the amount
of new materials to be added, and to know the rate of accumulation of in-
active materials within the sand system as it recirculates. Such materials can
promote many defects in the final casting (Ojala, 1962).
The experiments reported were concerned mainly with the behavior of
bentonites heat-treated at various temperatures. The studies of heat transfer
through molding sand conducted by Paschkis, 1945; Pellini, 1951; Ruddle,
1957; Marek, 1963, will be useful in applying these findings. Previously
published data (Zrimsek andVingas, 1960,1961a,bandc.VingasandZrimsek
1961, 1962; Heine, et al., 1959; Ojala, 1962) on new sand mixtures provide
an understanding of the design of this experiment. The effects of heat-treat-
ment systems that include combinations of variables encountered in regular
foundry practice were investigated.
Some specifications for bentonites for foundry use call for tests for vis-
cosity, liquid limit, gel strength, water repellency, fusion or vitrification
temperature, differential thermal analysis, or ion exchange ability, which may
relate to the mineralogy of the bentonites, but do not predict directly their
suitability for making a satisfactory mold and a satisfactory casting.
Clays in general and bentonites in particular are added to foundry molding
sands primarily to bind the aggregate together when making a mold. The
mold must maintain its shape while the metal is poured and until the casting
solidifies. The adhesive and cohesive forces of the sand-bentonite-water
mixtures are evaluated by the resistance of the mixture to compression or
shear in the green (undried) or dry state. Vingas and Zrimsek (1963) showed
that the mechanical properties alone cannot predict the performance of
adequately bonded aggregate for the production of defect-free castings.
The effects of heat-treated bentonites, when mixed with sand, on the
mechanical and physical properties and on bonding ability and viscosity have
been systematically evaluated. The temperature at which bentonites cease
to act as bonds was determined. The limitations of the mechanical and physi-
cal properties to predict and evaluate the quality of re-used sands was
shown. Foundry molding sands are complex aggregates, and heat treating
the bentonites separately does not simulate the heat treatment induced by
the casting process or their reactivity to additives such as cereal.
The tests used are described in the Foundry Sand Handbook, 7th edition,
1963.

EXPERIMENTAL
Two bentonites were investigated: a sodium (western) and a calcium
(southern) bentonite, mined and processed in Greybull, Wyoming and
Kosciusko, Mississippi respectively. The bentonite was dried to 7 per cent
 THERMAL STABILITY OF BENTONITES IN FOUNDRY MOLDING SAND 369

water content and ground so that 90 per cent was finer than a no. 200 sieve.
The behavior of the two bentonites after heat treatment was evaluated,
with the water content of the sand mixture, the mulUng response and ram-
ming as variables. Tests were performed to evaluate the effects of cereal
flour on heat-treated western bentonite and the effects of heat-inactivated
western bentonite on new western bentonite.

Heat Treatment of Bentonites

The bentonites were heat-treated in a 10 x 9 x 13-in. electrically heated
furnace. After the temperature was stabilized at the desired temperature, the
bentonite contained in a 5|-in. i.d. by 10-in. high clay-graphite crucible was
placed in the furnace.
A chromel-alumel thermocouple connected to a potentiometer was im-
mersed in the bentonite sample. The sample was left in the furnace for at
least 3 hr after the immersed thermocouple reached the stabilized temperature
of the furnace.
Each of the bentonites was evaluated as received and after heat treatment
to the various temperatures shown in the illustrations. The evaluation was
performed on sand mixtures prepared by two methods of mixing, an efficient
method and another that involved only stirring the aggregate.
The sand was a 51.3 AFS FN (Foundry Sand Handbook, 1963) Wisconsin
pure silica sand, the sieve analysis of which is shown in Table 1.

T A B L E 1.—SCREEN ANALYSIS OF B A S E
SAND U S E D

U.S. standard sieve no. Retained %

20 0.0
30 2.6
40 19.3
50 30.1
70 24.4
100 15.1
140 5.6
200 2.4
Pan 0.6

Mulling
In the first method of mulling, 1250-g sand and 100-g bentonite to be tested
were placed in a 12-in. Cincinnati muUer and mulled dry for 15 to 20 sec.
Water was added, and the mulling continued for 6 min more.
For the second method, 1250-g sand was placed in the muUer, the wheels of
which were removed, water was added, stirred with the plows only until
distributed (30-35 sec) and the 100 g of bentonite was then added and stirring
continued for an additional four minutes.
370 THIRTEENTH NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

Each sand-bentonite-temperature-mulling combination was tested at five

levels of water content, ranging from 2-5 per cent for the sand mixtures
mulled with the wheels, and 3-6 per cent for those stirred.

Tests
The sand mixtures were discharged into polyethylene bags and tested for
moisture content, green compression strength, green shear strength, dry
compression strength, and dry shear strength at 3 rams. The density of the
sand mixture was also checked at 1, 3, 5, 7 and 10 rams (Zrimsek and Vingas,
1961b). The sand mixtures were also riddled through a J-in. screen into a
250-cc volume and the density was calculated and reported as riddled density
in lb. per ft*.
Cereal Effects
The western bentonite, heat-treated to 1120°F., continued to act as a bond
in sand. To study the reactivity of this heat-treated bentonite to cereal and
compare it to the reactivity already established (Vingas and Zrimsek, 1963)
in new sand mixtures, some limited data were collected.
A mixture of 1250-g sand, 62.5-g (4.75 per cent) heat-treated and as-
received bentonite, and 12.5-g (1 per cent) cereal were mulled (with muUer
wheels on) as described above, at five moisture levels. The mechanical
properties of these systems were used for comparison and determination of
the effects of heat treatment on the reactivity of western bentonites to
cereal.

Effects of Heat-Inactivated Bentonite

The western bentonite heated to 1180°F or above ceased to react as a bond
when mixed with sand. Five per cent of the heat-inactivated bentonite at
1700°F was mixed with sand and new western bentonite (7.45 per cent),
mulled as described earlier (with muller wheels on) at five water levels, and
the physical properties checked. Subsequently, mulled sand containing 5 per
cent new western bentonite and water was heated at 1700°F for three hours.
This sand was mulled (with wheels on) with 7.45 per cent new bentonite at
five water levels and the physical properties were checked. The physical
properties of the two systems containing the heat-inactivated bentonite were
compared to the system containing only new western bentonite (at the same
level, 7.45 per cent).

Viscosity Tests
Sodium (western) bentonite when mixed with water exhibits viscosity,
and it is common for some foundries to evaluate this property. The sodium
bentonite as received and all heat-treated samples were also checked for
viscosity. This was done by mixing 24-g (6 per cent) bentonite with 376 cm*
distilled water in a Hamilton Beach model 3D mixer for 6 min. The suspension
 THERMAL STABILITY OF BENTONITES IN FOUNDRY MOLDING SAND 371

was then checked for viscosity with a Fann V.G. Viscosimeter at 600 rpm, and
calculated in centipoises.

RESULTS
Because of the volujne of data collected, all results are presented in graphic
form. Not all the data on cereal are presented since doing so would be repe-
titive.

DISCUSSION
We studied the deterioration of bentonites and the behavior of molding
sands containing heat-treated and deteriorated bentonites mixed with new,
unused materials.
Several investigations (Hofmann, 1958; Grim and Cuthbert, 1945) have
been conducted to establish the deterioration temperature of various ben-
tonites. In Europe, where sodium bentonites are scarce, the investigators
usually attempt to alter the calcium bentonite to sodium by treatment with
sodium salts. In most investigations, the major variables encountered in
regular foundry practice, such as water content, mulling response and
ramming energy, are maintained constant. As shown previously (Zrimsek and
Vingas, 1%0, 1961a, b, Vingas and Zrimsek, 1961, 1962; Heine, et al., 1959)
molding-sand data, unless collected on a system basis, are almost impossible
to analyze accurately.
This study shows the differences between sodium and calcium bentonites
when heated and mixed thoroughly with sand and water.
Figs. 1 and 2 show the data on sands bonded with the western and southern
bentonites before and after heat treatment, with the efficient mulling method
employed.
The curves obtained with the 7.45 per cent bentonite-sand-water mixture
before heat treatment are typical curves obtained with foundry molding sands
(Zrimsek and Vingas, 1960, 1961a, 1961b, Vingas and Zrimsek, 1961, 1962;
Heine, et ah, 1959).
Figs. 1 and 2 show that, as the heat-treating temperatures increase, the
peak green compression strength (Figures 1(a) and 2(a)), green shear strength
(Figures 1(b) and 2(b)), and the minimum density point (Figures 1(f) and 2-f))
shift to higher water levels.
When peak green compression strength (Fig. 3) is plotted versus tempera-
ture, the strength of the western bentonite bonded sand drops beyond
1150°F and that of the southern, beyond 600°F. The drop in strength beyond
these temperatures is drastic for the western and more gradual for the
southern.
Similar indications are reached if the other strength values are plotted as a
function of temperature and at other relatively equal water levels.
Figs. 4 and 5 show representative properties obtained at relatively low
 372 T H I R T E E N T H NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

^.2' /.'/'
S 22 -£. 200_ />'/"
-<% k 180 /•'•
•= 20

;\l\
- 18 S" 16(- / ''''
CO i 7
W

- 14t_ / '•'*

16 / '"'
Sl4 •a 6 12(1
B
5 - \ **^v\ .2 lot
S 12 £
K 4 - ^0^*\X •S. 80
/ 'i' /
1' //i / /
5 8 s 1 ^ ^ • v ^ ^ ^ . - ^ - *
E 60 ,//
" 6_ s 2 ' /-^-ti « 40 _ ^-^^
s 1 __ ST 20
o
/ "
*• 2 1,'... 1 - 1
0 0 1 - - i " ' 1
\ i \ I i i 5
3 4
% WAT E R % WAT E R i ?• W A T E R

(a) (b)

R.T. 10 rams,/'
230°F
510
870
1120
1210
c 70
B 1290
i
5?
60 1\\\
i
a
50 \^\

1
S
40
\v\
\\.\
\LX
i 40
30
?0 ,'>' :-'-'
c'
3e s 10 ' •
1 1 f VJ
0
3 4 S i 3 4 S 3 4 5
="
%.W A T E R * WAT E R ^0 W A T E R

(d) (e) (f)

F I G . 1. Water content effects on green compression strengths (a), green shear strength (b),
dry compression strength (c), riddled density (d), dry shear strength (e), and rammed
density (f) of 7.45% western (sodium) bentonite bonded sands; western bentonites
checked a t room temperature and a t the heat-treated temperature indicated; all proper-
ties shown a t 3 rams except rammed density, which was also checked at 1 and 10 rams.
Sands mulled for 6 min after water was added with wheels of muUer on.
 THERMAL STABILITY OF B E N T O N I T E S IN F O U N D R Y MOLDING SAND 373

3 4 5
% W Jl T E R

(a) (b) (c)

- lOram /
1 X. ;•-'
1 /,• ,V

-—

J; 3 ram /
- -1.^
_-v
-^y
\ j . ~/
-\
-
1 '~'-"r' 1
3 4 5
•?. W A T E R

(f)

FIG. 2. Water content effects on green compression strength (a), green sliear strength (b),
dry compression strength (c), riddled density (d), dry shear strength (e), and rammed
density (f) of 7.45% southern (calcium) bentonite bonded sands; southern bentonites
checked a t room temperature and a t the heat-treated temperature indicated; all
properties shown at 3 rams except rammed density, which was also checked at 1 and
10 rams. Sands mulled for 6 min after water was added with wheels of muUer on.
374 THIRTEENTH NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

200 400 600 Tm WOO

Temperature— "F
F I G . 3. Peak green compression strength as a function of the temperature
a t which the bentonites were heat-treated; mulled and stirred sands as
indicated.

R.T.
870 F
1120

CO

G3

''• 1 i i i
% WA T ER
(a)
FIG. 4. Water content effects on green compression strength (a), and riddled density (b),
of 7.45% western (sodium) bentonite bonded sands.
 THERMAL STABILITY OF BENTONITES IN FOUNDRY MOLDING SAND 375

mulling efficiency. The peaks exhibited by the green strengths, compression

and shear, are reduced and the influence of water content is diminished. The
resulting curved maxima occur at substantially higher water levels. At this
low mulling efficiency level, if any of the green strengths are plotted at the
same relative wetness versus temperature, the apparent deterioration tem-
perature of southern begins at a lower temperature than when the sand mix-
ture was thoroughly mtdled. The ability of the bentonite after heat treatment
to rehydrate, plasticize, and act as a bond is readily measured by the mulling
response.
R.T.
230°F 80 - \
510 e \
s
738 . \
70
^ *.
^
CO
60 \ \ '.
a>

1 \V"'.
10 /
-X"^\.
/'"T-'-^'-^-^'-i^ ^
50
' \\"•••-.
\ \
8 / -' 40
«« 6 / /
Z^ ~" ,<--
3.
2 •''** ^ , , ,
I I
\ 4 5
Vo W A T E R %W A T E R
{b)
F I G . 5. Water content effects on green compression strength (a), and riddled density (b),
of 7.45% southern (calcium) bentonite bonded sands.

The green strengths of the stirred western bentonite systems, although

substantially reduced, still break at the same temperature, 1150°F, as did
those efficiently mulled.
Density minima are also moved to higher water levels, and the replacement
of the sharp minima by curves occurs at lower temperatures than with the
high efficiency mulling.
The data on mulling indicate that temperature influences the rehydration
of southern considerably more than western bentonite, as measured by
25
376 THIRTEENTH NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

mulling response. The same data, however, show that with the low mulling
efficiency new as-received western bentonite responds to mulling more readily
than southern. This is contrary to what was previously reported (Vingas and
Zrimsek, 1961) when the low mulling efficiency was designed differently. In
that work, the low muUing efficiency was simulated by placing the sand and
bentonite in an 18-in. muUer and mixing for 30 sec. Water was added, and the
mulling was continued for 2 min more.
The direct relationship between 1- and 10-ram density differential (Zrimsek
and Vingas, 1961b, 1963; Vingas and Zrimsek, 1962) and the percentage of
bentonite present in a sand mixture, when checked at any water content
beyond the minimum density point, is substantiated by this experiment.

BD I B P MOO
Temperature — °F
F I G . 6. Density differential as a function of the temperature at which the
bentonites were heat-treated; mulled and stirred sands as indicated.

Fig. 6 shows the density differential plotted against temperature, and it

substantiates the conclusions reached when strengths were considered in
judging the deterioration temperature. The differential also shows that
bentonite content increased with heat treating temperature because both the
free and combined water in the bentonite were removed by heating, a variable
not considered in the design of this experiment. The differential also shows
the response to mulling, substantiating what was shown by the strength.

R I D D L E D SAND D E N S I T Y
It was shown previously (Zrimsek and Vingas, 1964), when molding sands
were investigated for their ability to reproduce a particular casting dimension
from that of the pattern, that the riddled density of the sand is the most signifi-
cant property. In addition, they showed that the main disadvantage of
bentonites in molding sands is the very narrow range of water after which a
major change in riddled density occurs. Modern high-production molding
machines do not consider this variable in their design. Riddled density is here
 THERMAL STABILITY OF BENTONITES IN FOUNDRY MOLDING SAND 377

considered synonymous to fiowability, a term used by the foundry industry

to describe response of mold densification.
Figs. 1(d), 2(d), 4(b), and 5(b) show the effects of water content on the
riddled density of sands. They show that the density drops drastically over
a very narrow water range as water is increased. As the heat-treat temperature
increases, the drop becomes more gradual, particularly with southern ben-
tonite.
TABLE 2 . — W A T E R R A N G E R E Q U I R E D TO R E D U C E R I D D L E D D E N S I T Y FROM 65 LB/FT^ TO
40 LB/FT3

Water %
Bentonite Temp. °F From To Water range

Western
Mulled Room 2.45 2.90 0.50
510 2.55 3.05 0.50
870 2 50 3.15 0.60
1120 2.60 3.65 1.05
Southern
Mulled Room 1.35 3.20 0.85
510 2.55 3.65 1.10
738 3.00 4.40 1.40
Western
Stirred Room 2.70 3.60 0.90
870 3.15 4.15 1.00
1120 3.70 5.60 1.90
Southern
Stirred Room 3.20 4.20 1.00
230 3.40 4.90 1.50
510 3.60 Never reached
738 4.90 Never reached

Table 2 shows the range of water required to reduce the riddled density
from 65 lb/ft* to 40 lb/ft*. At all conditions, the southern bentonite shows a
larger water range than the western. This substantiates the statement made
by foundrymen that southern bentonites are more flowable than western.
Heat treatment of the southern bentonite not only reduced the sensitivity
of moisture to riddled density, but simultaneously increased the density
achieved at minimum. For example, the minimum riddled density of the
738°F heat-treated, southern bentonite-bonded sand (Figure 5(b)) reached is
63 lb per ft^ versus 33 lb per ft^ for the new, as-received bentonite.
It should be remembered that many additives put in molding sands affect
the fiowability of bentonites. For example, Vingas and Zrimsek (1963)
showed that cereal, when mixed with western bentonite, eliminates the narrow
water range at which these sands are workable.
Effects of Cereal
The western bentonite heated to 1120°F was used to prepare sand mixtures
378 THIRTEENTH NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

with cereal flour in order to investigate the reactivity of cereal to the heat-
treated western bentonite.
Fig. 7 shows the influence of 1 per cent of cereal on green compression
strength for the 4.75 per cent bentonite system. Green strengths at low water
content were reduced, and the comprehensive tests showed that the minimum
density and comparable dry compression and dry shear strengths were
realized at about 0.5 per cent higher water contents.
The systems at lower heat-treating temperatures showed no major effects
on the reaction of cereal to the bentonites, and they are not included in this
report. The shifts in water requirement closely followed the systems con-
taining no cereal.

10
8
6 R.T.
CO
—II70°F
? 1 1 1
2 3
4 5
%W A T ER
F I G . 7. Water content effects on green compression strength of 4.75% west-
ern (sodium) bentonite and 1 % cereal flour bonded sands.

Effects of Heat-Inactivated Bentonite

As described in the experimental section, the western bentonite heat-
treated to 1700°F, thus completely heat-inactivated, was used to prepare
several mixtures. Five per cent of this heat-inactivated bentonite and 7.45
per cent new bentonite were mulled (wheels on) at five water levels. These
mixtures were used for comparison with sand mixtures in which western
bentonite was inactivated by heating a mixture of 5 per cent western ben-
tonite and 95 per cent sand to 1700°F. The resulting sand, containing 5 per
cent heat-inactivated bentonite and 7.45 per cent new bentonite, was also
mulled (wheels on) at five water levels.
The mixture with bentonite heat-inactivated on the sand grains and subse-
quently bonded with new bentonite is not very different from the new sand
mixed with only new bentonite. On the other hand, the sand mixed with the
separately heat-inactivated bentonite increased the water requirement for
peak green compression strength and minimum density by 0.7 per cent. This
indicates that heat treating the bentonites separately and then adding them
to sands does not exactly simulate what happens in practice (Zrimsek and
Vingas, 1963).
Molding sands are complex aggregates. To be properly and comprehensibly
investigated, each of the components must be varied independently and in
 THERMAL STABILITY OF BENTONITES IN FOUNDRY MOLDING SAND 379

conjunction with each other. Furthermore, heating the bentonite separately

from the sand is not the perfect method for investigating re-used sands. The
data presented should, therefore, be used with discretion.
It should be stated, however, that 1180°F and 600°F, which were estab-
lished as the points at which western and southern bentonite respectively
deteriorate as bonds, are below the vitrification temperature that has been
reported (Grim and Cuthbert, 1945) to be the deterioration point.
Viscosity
Table 3 gives the viscosity in centipoises of the heat-treated western
bentonite. It shows that viscosity quickly drops to 4.0 at 810°F from 20.0
at room temperature, although the strengths exhibited in the sand mixtures
are unaffected. Judging utility of the bentonite as a bond by viscosity is,
therefore, impossible.

T A B L E 3.—^FANN VISCOSITY IN CENTIPOISES OF H E A T T R E A T E D

B E N T O N I T E AT VARIOUS TEMPERATURES (6% SOLIDS)

Heat-treat temperature, ° F Viscosity (centipoises)

Room temp. 20
510 17.5
870 4.0
1120 1.5

SUMMARY
The difference between sodium (western) and calcium (southern) bentonites
and their reaction to heat have been shown. The evaluation made of their
behavior in bonding foundry molding sand was established by the physical
and mechanical properties they impart. Although both physical and mechani-
cal properties could help to establish the mold-making ability of the aggregate,
they cannot be used to predict the probability of making a defect-free casting.
The end product of a foundry is a casting, not a mold. Discretion is needed,
therefore, when applying the principles developed. It should not be taken for
granted that an adequately bonded sand that will produce a good mold will
necessarily produce a good casting.
Other tests developed to establish the mineralogy of bentonites, such as
differential thermal analysis, viscosity, and melting or vitrification tempera-
ture, cannot be used by the foundry industry to check the suitability of
bentonites, even from the bonding standpoint.
The conclusions reached are as follows:
1. The deterioration temperature of western and southern bentonite is
1180°F and 600°F respectively.
2. The drop in strength beyond the deterioration temperature is gradual
for southern and abrupt for western.
380 THIRTEENTH NATIONAL CONFERENCE ON CLAYS AND CLAY MINERALS

3. The mulling response of heat-treated southern bentonite is substantially

poorer than western and, at low mulling, the apparent deterioration tem-
perature of southern is reduced.
4. Heating bentonites separately from the sands does not simulate the
heating condition encountered in a mold.
5. The reaction of cereal to western bentonite remains unaffected at
temperatures below 900°F. At 1120°F, however, there are indications that the
cereal-bentonite combination starts to lose its ability to impart dry strength.
6. The riddled density of the sand can be used as a measure of flowability
of sands and ease of mold densification.
7. Southern bentonite used as a bond in re-used sand, and without additives,
yields more flowable sands than western-bonded ones.

REFERENCES
BISHOP, H . F . , and PELLINI, W . S . (1951) Solidification of steel from and and chill
walls, U.S. Naval Res. Lab. NRL Rep. 3785.
Foundry Sand Handbook, 7tli ed. (1963) American Foundrymen's Society, Des Plaines,
Illinois.
GRIM, R . E . , and CUTHEKET, F . L . (1945) The bonding action of clays. Part I, Clays in
green molding sands, Illinois State Geol. Survey Rep. Invest. 102.
H E I N E , R . W . , PARKER, J. J., K I N G , E . H . , SCHUMACHER, J. S., and BOSWORTH, T . J.
(1959) Sand movement and compaction in green sand molding, Trans. Am. Foundry-
men's Soc. 67, 47.
HoFMANN, F . (1958) Investigations on the effect of heat on the bonding properties of
various bentonites, Trans. Am. Foundrymen's Soc. 66, 305.
MAREK, C. T . (1963) Transformation zones in green sand, Mod. Castings 4 3 , 185.
OjALA, T. (1962) The equilibrium of foundry system sands. Part I, Cast Iron Melts and
Methods, 5, 2.
PASCHKIS, V. (1945) Studies on solidification of castings, Trans. Am. Foundrymen's
Soc. 53, 90.
RUDDLE, R . W . (1957) The Solidification of Castings, Institute of Metals, London.
ViNGAS, G. J., and ZRIMSEK, A. H . (1961) Systematic approach to sand design and
control, Progress Report 3, The Mulling Effect, Trans. Am. Foundrymen's Soc.
69, 101.
ViNGAS, G. J., and ZRIMSEK, A. H. (1962) Systematic approach to sand design and
control. Progress Report 6, Trans. Am. Foundrymen's Soc. 70, 321.
ViNGAS, G. J., and ZRIMSEK, A. H. (1963) Steel foundry green facing sands. Mod.
Castings 4 3 , 50.
ZRIMSEK, A. H., and VINGAS, G . J. (1960) Systematic approach to sand design and
control. Trans. Am. Foundrymen's Soc. 68, 449.
ZRIMSEK, A. H., and VINGAS, G . J. (1961a) Systematic approach to sand design and
control. Progress Report 2, Clay bonded sand dry properties. Trans. Am. Foundry-
men's Soc. 69, 33.
ZRIMSEK, A. H., and VINGAS, G . J. (1961b) Systematic approach to sand design and
control, Progress Report 4, Wood flour: Trans. Am. Foundrymen's Soc. 69, 141.
ZRIMSEK, A. H., and VINGAS, G . J. (1961c) Systematic approach to sand design and
control. Progress Report 5, Interpretation of data. Trans. Am. Foundrymen's Soc.
69, 229.
ZRIMSEK, A. H., and VINGAS, G . J. (1963) Mechanics of rework sand, Mod. Castings 4 3 , 36.
ZRIMSEK, A. H., and VINGAS, G . J. (1964) Factors controlling casting dimension, Trans.
Am. Foundrymen's Soc. 72, To be published.