.

CUTTING FLUIDS
By Eugene C. Bingham

ABSTRACT
Cutting fluids are used both to cool and lubricate. When lubrication is the more
important, it is generally recognized that fatty oils are superior to mineral oils, al-

though the reason has never been clearly explained. The evidence appears to be
that the value of fatty oils is due to their residual valence or acidity which causes their
adhesion to metal to be greater than is the case with mineral oils. If this should prove
to be the case, it may yet be possible to synthesize an oil which has all of the virtues
of lard oil without its obvious defects.

CONTENTS
Page
Part i. Theory of cutting fluids 35
I. General 35
II . Conclusion 54
Part 2. Tests 55
I. Determination of specific heat, density, surface tension, viscosity,
and lubricating value of typical oils 55
Part 3 . Practice 64
I. Purposes 64
II. Substances used 65
1. Oils 65
2 Air 69
3 Aqueous solutions 69
4. Emulsions 69
III. Operations and machines 71
IV. Materials Z 72
V. Formulas 73
VI. Application of the fluid 74
VII. Kind of cuts 75
VIII. Hygienic aspects 75

PART L THEORY OF CUTTING FLUIDS

I. GENERAL
It is a curious fact that so many of the very common opera-
tions of our ordinary life are without adequate explanation. A
case in point is the use of soap as a cleansing agent. It used to
be said that the hydrolysis of the soap produced an alkali which
saponified the grease, but this explanation is not in accord with
the known fact that soaps act very well as detergents even when
35
 36 Technologic Papers of the Bureau of Standards
[Voi. l6

the grease one not capable of being saponified, for example,

is

a paraffin oil; to state that the soap emulsifies

the grease is
strictly in accord with the facts, but it in no
wise explains the
—
phenomenon it merely describes it.
So again, the use of lard oil or other liquid in various cutting
operations in the machine shop is universally recognized as bene-
but there is general disagreement as to the cause of the
ficial;

improvement which is brought about.

(a) Cooung. —In
various operations, such as sawing, milling,
planing, turning, boring, reaming, tapping, and parting off, the tool
becomes more or less heated in use, and to prevent the overheat-
ing of the tool a stream of liquid over the work is naturally bene-
ficial; stated by many that the main purpose of a cutting
* it is

a refrigerant or "coolant." Following up this explana-

fluid is as
tion we should expect that water with its very high specific heat
and latent heat of vaporization would be the best cutting fluid.
Water actually used for this purpose, but only to a limited
is

extent. For
drilling ship plates and boiler plates, water is said
to give just as good results as cutting solutions or soluble oils.
It is used for rough turning and for grinding. But it is generally
objected to because pure water rusts the machines, and the work
as well, if the latter is of iron or steel. Naturally this difficulty
has nothing to do with the theory which we are discussing and
it can be overcome by adding to the water substances like soda
or soap which tend to prevent the formation of rust. Such
solutions are used satisfactorily in drilling and grinding; they are
also used on brass, copper, and aluminum, although a cutting
fluid is generally regarded as unnecessary for these metals. The
fact gradually emerges that with difficult work, water and simple
aqueous solutions do not give a smooth finish and they slow
down production because the tools become quickly dulled.
(b) Lubrication. —
It has been observed that the greatest wear of
a tool in cutting operations is often not at the point of the tool,
but on its face at some distance back from the point. This is a
striking and very significant fact. It is illustrated by a single
figure (Fig. i) taken from Taylor's Art of Cutting Metals. This
is proof of the very great friction between the chip and the face

of the tool at some distance back from the point. The chip
'

often adheres to the tool so firmly that a bead of metal appears '
'

'

1
P. W. Taylor, Art of Cutting Metals, p. 15. The effect of cooling the tool through pouring a heavy
stream of water onit permits an increase of 40 per cent in the cutting speed.
Technologic Papers of the Bureau of Standards, Vol. 16

Fig. i.
—Showing tool with "Cutting edge still

good but deep groove worn or guttered in the lip

surface by the pressure of the chip,' 'from Taylor's

Art of cutting metals, Fig. ije. In other cases,
small particles of the chips scraped from the
forging are pressed or if elded into a compact
pile on the lip surface of the tool
Bingham] Cutting Fluids 37

to be "welded" to the face of the tool so that it is necessary to

remove it forcibly.
It has therefore been suggested that the cuttingfluid acts prima-
rily as a lubricant. Since the lubricant reduces the friction, the
heat which otherwise would have been developed is avoided.
If the lubricant also removes heat by conduction, so much the
better. According to this theory water is unsatisfactory because
its viscosity is so low that it has very little lubricating value, and
"cutting fluids" and "cutting lubricants" become synonymous
terms.
This theory has the obvious merit of explaining why no lubri-
cant is necessary in cutting certain brittle materials such as cast
iron. Cast iron has very great compressive strength, but since
it is also very brittle a chip does not form; hence lubrication is
unnecessary. 2 It also explains why it is impossible to get a
smooth surface machining certain metals without oils, even
in
when the speed and depth of cut are reduced as far as practicable,
whereas a perfect surface is obtained as soon as an oil is employed.
Under such conditions it appears probable that the cooling action
of the lubricant is unimportant, and thus in the extreme case it
becomes a question of lubrication pure and simple.
But lubrication is another of those common operations for
which there is no adequate theory up to the present time. Thus
the theory gives little to guide one in the selection of an appro-
priate cutting oil. In taking a heavy cut at high speed, the
theory would seem to call for an oil of high viscosity at room
temperature, so that at the elevated temperature the oil would
still have sufficient viscosity to carry the necessarily heavy load.

If the cooling effect of the oil is taken into consideration, one

must also take into account the specific heat of the oil. It is a
fact, however, that these two properties are practically never
given any serious attention in the selection of a cutting oil.
Whether these properties should be so generally neglected remains
to be discovered.
There are properties of oils which have no relation to the
theory of cutting fluids, but which nevertheless have importance
in practice, (i) Cutting fluids should not separate out solids

at ordinary working temperatures; (2) their components should

not become immiscible with each other in use; (3) they should
be stable and neither oxidize in the air, give rise to gummy de-
2 Graphite, present in cast iron, can of course act as a lubricant.
38 Technologic Papers of the Bureau of Standards [Vol. 16

posits, nor carbonize as they pass over the work; (4) they should
not be liable to spontaneous combustion as, for example, when
exposed on cotton waste (combustible fluids should not be easily
volatile because this adds seriously to the fire hazard) (5) they ;

should not give off unpleasant odors when heated, even after con-
tinued use; (6) they should not injure the skin of the workmen
either directly or by means of disease germs; (7) they should not
corrode the work or the machines; (8) they should not injure the
machine even if the cutting oil becomes mixed with the lubricating
oil of the machine itself; (9) for some operations the cutting oils

should be fairly transparent; (10) they must be readily obtainable.

Were these the only things to be considered, it would be easy
to provide a cutting oil which would meet all of the require-
ments. One would select a petroleum oil of low cold-test and
fairly high viscosity, having regard for the temperature of the
oil in use. Most machine shops would effect a great apparent
saving were they to use a straight mineral oil in place of the oils
now employed for many classes of work. Before, however, we
conclude that the practice of modern machine shops is unduly
bound by tradition, it is well to inquire into the theory of lubrica-
tion, and particularly into the theory of cutting lubricants, to see
whether there are any other possible factors which can affect
the lubricating value of an oil.

(c) Adhesion. — It is well-nigh universal experience that pure

mineral oils, regardless of their viscosity, have a very limited
usefulness as cutting lubricants. Lard oil, on the other hand, is

the cutting oil par excellence. Nevertheless sperm oil is practically

the equivalent of lard oil, and this is the more remarkable because
sperm oil is much more fluid than lard oil; it does not resemble
lard oil in its chemical composition, since it is not an oil at all in
the strictly chemical sense but a liquid wax. According to general
testimony, olive oil and neatsfoot oil are nearly equal to lard oil;

and aside from their tendency to oxidize in the air, rapeseed,

cottonseed, corn, peanut, and linseed oils would serve very well.
Aside from their objectional odor fish oils would be useful. There
is therefore a fairly sharp differentiation between the so-called

fixed or fatty oils and the mineral oils.

This experience of the machine shop demonstrates that viscosity
is not the only measure of the " lubricating value" of an oil. But
this conclusion is directly contradictory to the results of Tower's
investigations on very well lubricated bearings, which showed
Bingham] Cutting Fluids 39

is dependent only upon the viscosity of the given

that the friction
oil. However, Tower's conclusions concerning ideal lubrication
may be inapplicable to cutting oils, since these oils are used under
conditions which are nearly the worst conceivable for effective
lubrication.
The conclusion that viscosity is not the only measure of lubri-
cating value is so highly important, not only in choosing a suitable
cutting oil but in problems of lubrication in general, that it should
not rest on hearsay evidence, no matter how widespread. We
must inquire: What is the evidence upon which the machinist
judges the fixed oils to be superior to others ? Can this difference
be measured quantitatively? Is there any similar preference for
fixed oils in general lubricating practice or elsewhere ? Is there
any known property of fixed oils which can conceivably give them
the supposed advantage ?
In answer to the first of the above queries, it is very easy to
demonstrate qualitative differences between the mineral and
the fixed oils in a variety of ways. One finds differences in the
smoothness of the finish of the work, in the speed of production,
in the power consumption, in the life of the tool, and after a little
experience one can judge the merits of the oils by the appearance
of the shavings.
A mechanic employed at this Bureau noted that in
skilled
cutting off a rod of mild steel i }4 inches in diameter once, it was
necessary to sharpen the tool 10 times when mineral oil was used;
whereas with lard oil he could cut off the rod 8 times without
resharpening. Moreover the cut surfaces of the latter were very
smooth while the former were unspeakably rough.
This conclusion was confirmed by the author, by cutting off
a from a bar of mild steel i^i inches in diameter,
series of disks
using an Armstrong cutting-off tool on a direct-drive lathe, giving
constant speed and feed. A mineral oil was used whose viscosity
at room temperature was identical with that of a given sample of
lard oil. The oil was dropped into the groove just ahead of the
tool by means of a pipette in such a way as to obtain a fairly
uniform flow. After some preliminary experiments which proved
the striking superiority of lard oil, the author cut off 20 disks
with lard oil in the effort to find out how many disks could be
obtained with a single sharpening. This attempt was abandoned
after examination of the edge of the tool with a compound micro-
scope showed that the dulling was still not perceptible. The
 —

40 Technologic Papers of the Bureau of Standards [Vol. 16

surface of even the last disk was very smooth, as is shown in

Fig. 2, a. After the tool was sharpened to remove any dulling
that might have escaped detection and given as nearly as possible
the same set as before, another disk shown in Fig. 2, b was cut off.
All of the conditions were the same except that mineral oil was
employed. The tool was badly dulled in cutting the first disk
and broke in two on the fourth disk. The surface was very
rough, as can be seen readily in the photograph. The chips were
thrown out with violence and there was evidence of heating and
heavy consumption of power. As to the nature of the individ-
ual chips, with lard oil the metal cut much like butter, the under
side of the chip being perfectly
smooth and the upper side finely
corrugated. With mineral oil the
corrugations were so deep that
they extended completely through
at the middle of the chip. This
can easily be seen in the figure.
To grasp the problems con-
nected with this form of lubrica-
tion, it becomes necessary to con-
sider in detailhow the shaving is
produced. Let a in Fig. 3 repre-
sent the body of metal from which
the layer b is being removed by
Fig. 3. Diagram to illustrate the relation
the tool c. The pressure of the
of chip (d) to tool (c)
tool at the point d wedges off the
material and causes a fracture to extend into the work ahead
of the tool at e. The point of the tool cleans up the rough
surface,but the greater part of the pressure is exerted on the face
of the tool at some distance back from the edge where, as has
already been pointed out, the greatest wear occurs. The material
in the region / is subjected to severe compression, and when the
pressure reaches a certain critical value there takes place mo-
mentary plastic flow. The direction of flow fg is at an angle of
approximately 45 ° with the pressure, which is taken to be normal
to the surface ad.
As a result of the plastic flow the chip becomes about twice its
normal thickness. 3 Since, however, the plastic flow can not begin
3 The
amount of thickening presumably depends on several factors, notably upon the friction of the
chip on the tool as well as on the softness of the metal, cf. Taylor, Art of Cutting Metals, p. 119.
 —

Technologic Papers of the Bureau of Standards, Vol. 16

a

Fig. 2. Showing the difference in lubrication between lard oil (a) and, mineral oil
(b) in parting off mild Heel rod

The "bead" of metal which gouged into the surface of (b) may be actually seen at the middle of
(e), where it formed on the tool. The chips with lard oil (c) are less deeply serrated than when using
mineral oil (d)
48 Technologic Papers of the Bureau of Standards [Vol. 16

surface of the glass was scratched perceptibly in the process."

In other words there was seizure. Extending the work of Miss
Pockels and Tord Rayleigh, I,angmuir has devised a "talc test"
by means of which the contamination of a surface may be deter-
mined, and he has studied the sliding friction with various kinds
of contamination. Using a glass slider on a glass surface it was
found that the latter could be tilted usually to an angle of 70
from the horizontal, often to 75 ° and in some cases 90 or even 92 °,
before the slider would begin to move.
The " sliding angle" of glass on glass was found to be 6o°;
that of glass on mica and of platinum on mica was io°. Similarly
the sliding angle of platinum on platinum was 35 °, whereas for
glass on platinum it was 30 In each case the sliding angle was
.

lower when the materials were different. Of course, the results

are too few to admit of generalization. In every case the sliding
angle was greatly lowered by adding a monomolecular layer of
oleic acid.
In lubrication it is of paramount importance to keep the surfaces
from coming into contact, for wherever the solid surfaces touch
adhesion will occur. The problem of lubrication is theoretically
one of great simplicity. The two metals are prevented from seizure
by means of a third substance which may be a solid such as
graphite or talc, a liquid such as lard oil, or even a gas such as air.
If the third substance is a solid, it must be soft so that it will readily
undergo plastic flow. If it is a liquid, it must adhere strongly and
it must not be too fluid, lest it squeeze out.

The cutting of metals offers the most severe conditions for lu-
brication. The tool is continually taking off fresh metal, often
at a high rate of speed and at high temperature. The pressure
of the chip upon the tool sometimes excessive. All of these
is

causes tend to make adhesion between the chip and the tool very
great and the need for a lubricant to prevent seizure may be im-
perative. That lubricant is best which has the greatest adhesion
for the metal, since it is most strongly drawn into the space between
the tool and the metal being removed.
Caution needs to be taken not to draw too broad inferences from
the above statement. Thus in cutting brittle substances, such as
glass or very hard steel, a shaving is not formed and lubrication is
unimportant. In fact, seizure is apparently desired in order to
get the tool to "bite." What is needed therefore is something
which will clean the surface and cool the tool. Turpentine, either
 —

Technologic Papers of the

Fig. 4. Shoiving a piece of cast

iron whose surface was smeared
with lard oil (a), castor oil (b),
oleic acid (c), and turpentine

Where the oil was the work appears

lighter colored, owing presumably to
smoother In the case of turpen-
finish.
tine there was no observable effect
 1

Bingham] Cutting Fluids 5

can not claim a direct proportionality between vapor pressure and

adhesion. The fixed oils are characterized by a very small vapor
pressure as compared with even the heavy mineral oils. Thus
lard oil has a probable boiling point of 580 C, while a mineral
engine oil was found to have a boiling point around 330 C. There
must be a lowering of the vapor pressure of the absorbed film
which is a function of the adhesion.
5. In grinding a given pigment in oil it has been noticed that
the amounts of different oils required to give a paste of a given
consistency vary very widely, and in a manner which can not be
accounted for by the viscosity. It is quite possible that the
effects are due to the different adhesion of the pigment for the
various oils. It is suggested therefore that a given powder be
ground in a similar volume percentage of oil and that the yield
value 21 be measured.
6. In addition to the above there are practical methods. It
appears that the so-called friction testing machine can sometimes
be used to measure adhesion. It would appear probable that with
a given bearing and two oils of the same viscosity, the oil having
the lower adhesion would squeeze out first. It has often been
stated that glycerol and molasses are absolutely devoid of lubri-
cating value, although they have the requisite viscosity. If this
is the case, then it ought to be readily possible to make a set of

similar bearings and, having "run them in" on a friction-testing

machine, to test different lubricants for the maximum load which
each would carry before seizure took place. The author has
devoted a large amount of labor in following up this lead. The
only important result of his study is to prove that, with well-
fitting bearings bathed in lubricant, glycerol, sugar solution, and
gelatin solution will undoubtedly function as lubricants. We
continued to dilute these materials with water until we obtained
a viscosity of only 45 seconds, Saybolt, without obtaining seizure
at a load of 800 lbs. /in.
2
The machine used was of the " Cornell"
.

type, with a journal $ 3 A

inches in diameter and 3X inches long.
The bearing block of lead-calcium alloy was 2 inches wide, thus
providing a projected area of test bearing of about 7 square
inches.
7. The flotation methods seem to offer at least a qualitative
measure of adhesion.
8. It is possible to get comparative results from experiments in

cutting a given metal under constant conditions of speed, feed,

»Cf. B. S. Bull., 13, p. 309; 1916.

52 Technologic Papers of the Bureau of Standards [Vol. 16

rate of application of cutting liquid, etc. Power consumption,

torque, finish on the work, and speed of production, and dulling
of the tool all offer possibilities of measurement. Taylor 22 has
worked out a " standard cutting speed" which should be given
careful consideration by any one proposing to measure adhesion
by this method. He used the cutting speed in feet per minute
which would cause the tool to be completely ruined in 20 minutes
under standard conditions.
9. A number of variations of the above methods will readily
suggest themselves. It has been shown that when solutions of
various electrolytes are filtered through clay, a portion of the elec-
trolyte is So, if a metal powder were shaken up in a
adsorbed.
solution of some solvent like ether and then centrifuged, it
oil in
might be found that the amount of oil retained was a function of
the adhesion between the metal and the oil. The author has shown
elsewhere 23 that if a mixture of two liquids is forced through long,
fine, capillary tubes or pores, such as are found in a column of
clay, the rate of fiow will be affected by the relative adhesion of
the two liquids for the solid material. There will, therefore, -result
a partial separation of the components of the mixture. Such a
separation has been demonstrated by Gilpin and his coworkers in
connection with petroleum. This method might be applied to
lubrication by forcing the two oils to be compared through pow-
dered metal until an equilibrium is established, when the adhering
oil could be extracted and analyzed.

10. In connection with the measurement of adhesion we may

refer to the change of surface energy when 1 cm of interface is
2

formed between two liquids, as measured by Harkins. He has

obtained the following values, which are of great interest: 24
Ergs
Water vs. water 145- 6
Ricinoleic acid vs. water 94.. 9
Caprylic acid water
vs. 93. 7
Octyl alcohol vs. water 91.8
Oleic acid vs. water 89.o
Castor oil vs. water 87.7
Ethyl ether vs. water 79. 2

Benzene vs. water 66. 2

Paraffin oil vs. water 47- 8

Hexane vs. water 41-2

The adhesion of liquids to water seems remote from their adhe-

sion to steel and other metals. But according to the views of

22 Art of Cutting Metals, p. 41.

*3 B. S. Bull., 13. p. 324; 1916.
M J. Amer. Chern. Soc, 39, pp. 354, 541; 1917. See also Southcombe, Engineering, 109, p. 184; 1920.
 :

Bingham] Cutting Fluids 53

Langmuir and also of Harkins, adhesion is dependent upon the

presence of active groups in the molecules, the hydroxyl group,
carboxyl group, double bond, etc. When a layer of oil only i
molecule deep is placed on water or a solid surface there is a dis-
tinct change of properties, as shown by Miss Pockels and Lord
Rayleigh. Results privately communicated by Dr. Langmuir
show that a layer of paraffin oil can be readily washed off from a
surface of glass, mica, calcite, platinum, sphalerite, galena, pyrites,
or magnetite by simply passing the contaminated surfaces through
a gentle stream of running water. But in every case it was found
to be impossible to remove the greasiness due to oleic acid by
repeatedly passing through a stream of water. He says:
We may assume that the attractive force between hydrocarbon molecules and the
solid surface is greaterthan that between hydrocarbon molecules, but active groups
like those contained in water or oleic acid are attracted to the solid surface very much
more than are hydrocarbon molecules. The paraffin oil thus readily wets the solid,
if brought into contact with it, but the hydrocarbon molecules are readily displaced

from a surface layer when water or oleic acid is present.

This is exactly in accord with our own ideas arrived at from a

different point of view.
ii. Langmuir has noted that fairly constant results were
obtained in the use of the slider referred to above by starting it
in motion down the inclined plane of glass and noting the angle
at which motion just stops. For example, if oleic acid or other
oily substance is put on and then thoroughly wiped off, the sliding
angle is reduced from 50 down to 6 to io°. Paraffin oil lowers the
sliding angle nearly as effectively as oleic acid, presumably owing
to the fact that the active group of oleic acid is in contact with
the glass, so that the free ends of the molecules are the same in
both cases. Whatever the explanation, this is evidently not a
possible method for the measurement of adhesion.
Langmuir has also measured the contact angle of a drop of
water with various surfaces contaminated with oleic acid.
Thus he gets
Degrees
Mica 24
Quartz 31
Glass 42
Platinum 72
Calcite 82
Sphalerite 92
Galena 106

These results are specific, but it is too soon to say exactly what
is their significance.
62649°—22 4
54 Technologic Papers of the Bureau of Standards [Vol. 16

II. CONCLUSION
Cutting oils are used partly on account of their cooling value,
partly on account of their value as lubricants. But the value
of an oil as a cutting lubricantnot measured by any known
is

test. Apparently a high adhesion of the oil for the metal is

necessary. A number of methods are suggested for the measure-
ment of this quantity, but its maximum development seems to
be dependent upon residual chemical activity, such as is present
in oleic acid. According to this view the prevalent practice of
excluding as far as possible free oleic acid from a cutting oil is
shortsighted. We can even go further and predict that for
many, if not all purposes, it may yet be possible to synthesize an
oil which has all of the virtues of lard oil without its obvious de-

fects. Thus sulphur in the free condition or as sulphide possesses

a great amount of residual affinity. Many oils when heated with
sulphur dissolve it to a greater or less extent. Pine oil, which is
largely used in flotation, has residual affinity which causes it to
absorb sulphur readily. The product we have found to possess
great virtue as a cutting oil either by itself or diluted with large
amounts of mineral oil. For it is to be noticed that,
our argu- if

ment is correct, small quantities of a substance of high adhesive

qualities will entirely alter the properties of an inferior oil. Others
have noted the fact that the properties of an oil are greatly
altered by small amounts of impurities. Miss Pockels showed
that the behavior of purified oil is quite different from that of
the common oil. Richter states that the tendency of oil to
spread on water is due only to the free oleic acid contained in
it. Ivord Rayleigh showed that when sufficiently purified "the
drop remains upon the water as a lens, and flattens itself out, if
at all, only very slowly."
To test this further the author made a series of experiments.
In correspondence and conference with a large number of prac-
tical users of cutting oils in quantity, it was found that some
users were convinced that regardless of cost, or convenience,
certain oils sold under trade names were indeed superior to even
the best lard oil. Some of the best oils according to general
consensus of opinion contain sulphur in large quantity. On
analysis one showed 5.5 per cent sulphur, and not only blackened
copper but evolved large quantities of hydrogen sulphide when
boiled. Apparently it contained a considerable amount of pine
Bingham] Cutting Fluids 55

oil. There was an odor of phenol which was probably used in

very small quantity merely for the sake of its odor. Even
without phenol it would probably be germicidal, and testimony
is that in working with such an oil the men have been unusually

free of the customary infections. With a knowledge of the proper

composition of cutting oils, users should be able to write specifica-
tions more intelligently, thereby securing better oils, and with
the possibility of supplying their own needs they may be able to
secure better terms.

PART 2. TESTS
I. DETERMINATION OF SPECIFIC HEAT, DENSITY, SUR-
FACE TENSION, VISCOSITY, AND LUBRICATING VALUE
OF TYPICAL OILS
(By A. W. C. Menzies, W. G. Kleinspehn, G. Q. Lewis, and Eugene C. Bingham)

In the study of cutting oils in particular and the question of

lubrication in general, it seems necessary to determine the differ-

ent properties of fixed and mineral oils which may possibly pro-
duce the effects upon their value as lubricants, which we have
been discussing. A. W. C. Menzies has determined the specific
heat, surface tensions, and densities of certain oils. He used
cottonseed oil, castor oil, and a commercial brand of naphthene-
base oil "U." The samples were furnished by this Bureau.
In the case of the first two, the oils themselves were used as
calorimeter fluid, and the rise in temperature was observed when
a perforated brass cylinder of known heat capacity and at known
temperature was dropped into the calorimeter. In the case of
castor oil, the viscosity was so great that the above method was
not satisfactory. The oil was therefore placed in a 25 cc cylin-
drical glass bulb furnished with a calibrated capillary so as to
form a very sensitive thermometer, whose readings were com-
pared with those of the calorimetric thermometers. The castor-
oil thermometer, taken from a chamber at o° C was immersed

in a water calorimeter and readings made at simultaneous tem-

peratures of calorimeter and castor-oil thermometer. Due cor-
rection was made for the heat capacity of the glass of the bulb.
The calorimetric thermometers were supplied with Reichsanstalt
certificates of corrections rounded to fiftieths.
56 Technologic Papers of the Bureau of Standards [Vol. 16

TABLE 1.— Specific Heats of Oils

Temperature
Kind of oil
rise in de- Specific
grees centi- heat
grade

(1)20.6-28.6 0. 4537
(2)21.3-29.3 .4502

Average specific heat at 25.0 ° C .452

«*U"oil (1)19.1-27.9 .4141

(2)21.4-29.9 .4122

Average specific heat at 24.6 ° C .413

Castor oil (1) 0.0-20.5 .487

(2) .0-20.5 .478

.483

Each of the specimens of oil was previously freed from water

and dissolved gases by heating at 180 or higher in a vacuum.
The densities were determined in a quartz glass dilatometer.
Temperatures were measured with a set of Allihn type thermom-
constant zero points, furnished with Reich-
eters, of satisfactorily
sanstalt certificates to o.i° up to 200 and to 0.2 above 200 .

The threads were completely immersed. Surface tensions were

measured in a glass capillary tube of radius 0.3254 mm, and the
data assume a zero angle of contact with the glass. The lower
meniscus had a diameter of over 34.2 mm. Due regard was given
25
to the remarks of Richards in reference to the measurement of
surface tension. Violently stirred baths of a lard substitute and
of fused nitrates were used to maintain uniform temperatures.

TABLE 2. —Density and Surface Tension of the Furnished Specimen of Cotton-

seed Oil

Surface Surface
Temperature in degrees Density Temperature in degrees Density
tension tension
centigrade g/cm 3 centigrade g/cm 3
dynes /cm 2 dynes/cm2

.20. 0.920 32.7 130 0.847 24.4

30. .913 31.0 140 .840 23.7

40. .907 31.1 150 .834 23.1

50. .900 30.2 150 .827 22.5

60. .893 29.4 170 .821 21.9

70. 28.6 130 .814 21.3

89. 27.9 190 .808 20.7

90. .873 27.1 290 .801 20.1

100 .867 26.4 210 .795 19.5

110 .860 25.7 220 .788 18.9

120 .854 25.0 230 .782 18.4

Note.— The oil was freed from dissolved gases and water by heating at 200 C in a vacuum.

« J. Am. Ch. Soc., 37, p. 1674; 1915-

Bingham) Cutting Fluids 57
TABLE 3. —Density and Surface Tension of the Furnished Specimen of Naphthene-
Base Oil, "U"

Surface Surface
Temperature in degrees Density Temperature in degrees Density
tension tension
centigrade g/cm 3
centigrade g/cm 3
dynes/cm 2 dynes/cm 2

. . .. 0.942 32.5 140 0.849 21.1

10 .940 31.5 150 .842 20.5
20 .929 30.6 160 .835 19.8

30 .922 29.6 170 .829 19.2

40 .915 28.7 180 .823 18.6
50 .908 27.8 190 .817 18.0
60 .901 27.0 200 .811 17.4
70 .895 26.2 210 .805 16.8
80 .888 25.4 220 .799
90 .881 24.6 230 .792

100 .874 23.9 240 .786

no .868 23.2 250 .780
120 .861 22.5 260 .774
130 .855 21.8

Note. — The oil was freed from dissolved gases and water by heating at i8o° C in a vacuum.

TABLE 4. —Density and Surface Tension of the Furnished Specimen of Castor Oil

Surface Surface
Temperature in degrees Density
tension
Temperature in degrees Density
tension
centigrade g/cm 3
centigrade g/cm 3
dynes/cm 2 dynes/cm 2

20 0.957 34.0 130 0.879 24.9

30 .950 33.0 140 .872 24.2
40 .943 32.1 150 .865 23.5
50 .936 31.2 160 .858 22.9
60 .929 30.3 170 .851 22.3
70 .922 29.5 180 .844 21.7
80 .915 28.7 190 .837 21.1
90 .908 27.9 200 .829 20.5
100 .901 27.1 210 .822 19.9
no .894 26.3 220 .815 19.3
120 .886 25.6 230 .808 18.8

Note. —The oil was freed from dissolved gases and water by heating at 200 C in a vacuum.

The surface tension curves for these three oils are nearly
identical.
W. G. Kleinspehn has measured the fluidities of cottonseed oil,

linseed and other oils. The method is that already described

oil,

by Bingham. 26 The measurements were in duplicate. The data

are given in Table 5.

« B. S. Bull., 14, p. 59; 1917.

 — 6
3

58 Technologic Papers of the Bureau of Standards [Vol. 16

TABLE 5.—Fluidities of Several Oils

COTTONSEED OIL

Temperature in Limb Time in Specific

degrees centi- of vis-
seconds volume
Density Po P Viscosity Fluidity
grade cometer

cm 3
/g g/cm a g/cm 2 g/cm 2
0.7 L 807.0 1.073 0.8070 289. 77 289. 77 184.29 0. 5425
R 807.8 289. 59 289.59 184.36
10 L 482.2 1.080 .9260 289. 73 289. 72 110.11 .9079
R 482.7 289. 57 289. 56 110. 16

20 L 615.9 1.088 .9190 139. 39 139.39 67.66 1.478

R 616.4 139. 26 139. 39 67.65
30 L 316.1 1.096 .9123 178. 07 178.05 44.36 2.255
R 316.3 177.86 177.84 44.34
40 L 352.0 1.104 .9055 110.69 110.68 30.70 3.257
R 352.5 110. 55 110.54 30.71
50 L 386.6 1.113 .8988 72.82 72.81 22.18 4.508
R 388.0 72.60 72.59 22.20
60 L 596.2 1.121 .8924 35.74 35.74 16.79 5.957
R 597.0 35.68 35.68 16.79
70 L 469.4 1.129 .8860 35.10 35.09 12.98 7.709
R 464.0 35.45 35.44 12.96
80 L 363.8 1.123 .8904 30.83 30.80 8.84 11.32
R 366.3 30.68 30.62 8.85
90 L 300.9 1.132 .8835 30.71 30.67 7.28 13.70
R 303.0 30.62 30.60 7.32
100 L 253.2 1.141 .8765 30.82 30.78 6.15 16.26
R 255.0 30.71 30.67 6.17

U" OIL

10.. L 6054.4 1.069 0.9353 411.70 411.30 1,964.50 0.051

R 6067. 411.30 410.25 1,965.70
20. L 2086. 1.077 .9286 410.56 409.02 675.15 .148
R 2086.8 410. 26 409.85 674.75
30. L 856.4 1.084 .9225 409.85 409.43 276.62 .362
R 856.2 409. 73 409.34 276.48
40. L 404.0 1.091 .9163 409.77 409. 36 130.47 .749
R 403.9 409. 65 409.25 130.40
50. L 348.6 1.099 .9102 251.68 251.41 69.14 1.446
R 348.8 251.60 251.33 69.16
60. L 332.8 1.106 .9040 154. 14 153.95 40.42 2.474
R 333.0 154.04 153. 86 40.42
70. L 442.2 1.114 .8978 72.85 72.78 25.40 3.934
R 442.5 72.79 72.71 25.38
80. L 446.6 1.122 .8913 48.13 48.08 16.94 5.901
R 448.4 47.99 47.95 16.95
90. L 487.6 1.130 .8848 31.03 30.99 11.92 8.384
R 490.7 30.86 30.82 11.93
100 L 360.4 1.139 .8783 31.08 31.05 8.83 11.297
R 365.9 30.79 30.74 8.87

Note .
Po is the true average pressure producing the flow and of this pressure, the part
, P is used up
solely in overcoming the viscous resistance. Cf B. S. Bull., 14, p. 70; 1917.
.
Bingham) Cutting Fluids 59

TABLE 5.—Fluidities of Several Oils—Continued

LARD OIL
[Stearin crystallizes out at low temperatures. The effect of heating the oil is shown by the marked change
in fluidity after cooling to 50 and 20° C]

Temperature in Limb Time in Specific

degrees centi- of vis- Density Viscosity Fluidity
seconds volume
grade cometer

cm3/g g/cm3 g/c m2 g /cm2

20.. 391.5 1.096 0.9126 266.99 266. 98 82.38 1.214

391.6 266.89 266.88 82.37

30.. 405.0 1.104 .9058 165. 89 165.88 52.96 1.890

405.0 165 76 165. 75 52.91

40.. 509.3 8933 89.00 88.99 35.72 2.780

510.1 88.89 88.88 35.73
50.. 457.7 1.120 .8926 70.13 70.12 25.29 3.955
458.4 69.99 69.98 25.28
60.. 495.6 1.129 47.79 47.78 18.66 5.354
497.6 47.67 47.66 18.69
70.. 377.8 1.137 ,8794 47.78 47.77 14.22 7.028
379.0 47.66 47.65 14.23
80.. 426.0 1.146 .8726 33.24 33.23 11.16 8.953
428.6 33.12 33.11 11.18
90.. 409.0 1.155 .8657 27.79 27.78 8.98 11.17
410.7 27.66 27.65 8.95
100. 335.0 1.165 .8586 27.79 27.78 7.33 13.62
337.2 27.67 27.66 7.35
50.. 506.2 64.99 64.98 25.92 3.855
508.0 64.86 64.85 25.96
20.. 520.7 210.40 210.39 86.34 1.157
521.7 210.31 210.30 86.46

Bach
of the fixed oils here tested has a considerably higher
heat than does the single mineral oil, which may probably
specific
be taken as typical. Even at that the highest specific heat is a
little less than one-half that of water.
The fluidities of the two fixed oils are generally greater (Fig. 5)
than the fluidities of the mineral oil at the same temperature.
The characteristic feature, however,
is that the increase of the

fluiditywith the temperature is so much greater with the mineral

oil. The contrast between the two classes of oils is brought out
much more clearly when we plot the fluidities against the specific
volumes, Fig. 6. The viscosities in centipoises are plotted in Fig.
7. Only a small portion of the curve for the mineral oil can be
shown on the plot.
Although the specific heat and the slower change of fluidity with
the temperature are both in favor of the fixed oils, we have tried
to show that the adhesion is probably the main reason for the
superiority of the fixed oils as lubricants.
G. Q. Lewis of the U. S. Navy Yard at Washington has made an
elaborate study of the cutting properties of fixed oils, compounded
 —

6o Technologic Papers of the Bureau of Standards \voi. &

oils, and trade preparations. He employed for the tests the actual
machines and materials which are to be used in practice. He has
found several trade preparations to be superior to even lard oil
in boring chrome nickel steel propeller shafts with an elastic limit
of ioo ooo lbs. /in.
2
He has secured an increase of production of
.

50 to 100 per cent over the earlier practice, where a mixture of

10 per cent rapeseed oil and 90 per cent paraffin oil was used. A
saving of 13 per cent in the cost of the oil was also obtained. One

40* J0"° 60* 70 s 80° 90° 100

c

TEMPERATURE
Fig. 5. The mineral oils increase in fluidity more rapidly with the temperature than
do the fixed oils. A single asphalt-base oil "U" is here compared with lard, linseed,

and cottonseed oils

particular mixture increased production on a turret lathe 100 per

cent;on small brass work it greatly increased the life of the tool
and the cutting speeds on a small lathe cutting gimble centers for
;

torpedo gyroscopes from "Ketos" tool steel it doubled production

and increased the life of the tool about 300 per cent. In an auto-
matic screw machine, cutting alloy steel screws for securing air
flask heads, a 1 to 6 mixture, costing 35.6 cents per gallon, gave
results slightly better than those obtained with lard oil at $1.52
 —

Bingham] Cutting Fluids 61

per gallon. When both oils were used under the same conditions,
pieces cut from the same bar were smoother and therefore more
desirable when the new preparation was employed. In the tool
shop a i to 6 mixture at 35.6 cents per gallon, and later a 1 to 10
mixture at 29.2 cents per gallon, replaced sperm oil at $1.79 per
gallon.
The numerous other tests are also given in the report.
results of
We merely add Fig. 8, which is an example of a graphical com-

SPEC/FJC VOLUME
Fig. 6. The contrast between the mineral and fixed oils is most clearly brought out
when the fluidity in absolute units is plotted against the specific volume in cubic centi-
meters per gram

parison of different oils so diluted with paraffin oil that each of the
mixtures would cost 35 cents per gallon. The machine was run
by its own motor which was connected with an ammeter and volt-
meter. The machine was run at a rate of 120 r. p. m. with a feed
cut of 8 inches per hour on chrome nickel steel. The watts and the
temperature of the were measured and were plotted against the
oil

elapsed time. With

oil or pure mineral oil the tool would
lard
break down at once. With two of the mixtures tested, Nos. 13
 . —

62 Technologic Papers of the Bureau of Standards \voi. 16

and 1 8, the tool worked steadily without any balk whatever.

With the other mixtures the machine balked and had to be started
over again, as can be seen by the sharp maxima appearing in the
curves. Mixture 13 is the one which had been found in the earlier
tests to give such superior results. The temperature curves do

1 1 1

1
\
\ \
150 \ \
\ \
\
\ 1

\
V
\ \ \
*L \ \
\ \
\
£100 \ \
v VA
1

>-
H V *«
V*. ~V*-
O ^ V >

CO \ V

>* \ \

\N N
s
^
CP
o°c. io° ZQ C 30° 40° 50° 60' 70° 80< 90 c
100°
TEMPERATURE
Fig. 7. It is more usual to plot vicosity against temperature, but due to the hyperbolic
character of the curves, the figure would need to be more than tenfold its present height
to show the complete curve for the mineral oil

not seem to show any relation to the value of the oils, so they are
not reproduced.
The author desired to test this oil on other operations. For
this purpose a number of samples of oil were made up. A few
of these were:
1 The preparation which had given best results at Navy Yard.
2. I^ard oil.

3. A mixture of 90 per cent of paraffin oil and 10 per cent of

oleic acid.
The oleic acid had been heated with sulphur.
 —

Bingham] Cutting Fluids 63

4. Pine oil, sulphur, and turpentine.

These samples were placed in plain bottles with numbers only,
sufficient dye and pyridine being added to the lard oil to disguise
it. They were then given to a skilled mechanician to be used in
threading wrought iron. The resulting threads were examined
with a microscope by several independent observers. There was

/ooo

/o 20 30 40 SO CO TO do
£LAPS£D T/M£ //V M/A/UT£S
Fig. 8. Navy Yard tests showing fluctuations in power consumption when using various
oils for lubrication.

general agreement that the lard oil gave the best results and the
new preparation, No. 1, the worst results. The results fell in the
order, Nos. 2, 4, 3, 1. They were very decisive for this particular
operation. Pure mineral oil was, of course, inferior to all of these.
It appears likely that lard oil is better than anything which has
been devised for certain operations. It may be, however, that for
64 Technologic Papers of the Bureau of Standards [Vol. 16

certain other operations, when a good lubricant is necessary, a

substitute may
do as well as or better than lard oil, and at a frac-
tion of the cost. The prediction seems to be amply justified that
oils with high residual affinity, such as those containing oleic acid,

or pine oil with sulphur, will be found best. But it is noted that
the trade preparation No. 1 also contained several per cent of
sulphur.
So many persons have contributed to this bulletin in a variety
of different ways that it is pleasant to have this opportunity to
express thanks to all. It is hoped that bringing together these
results of experience maynot only be of practical benefit but that
it may lead to a systematic investigation of the subject which
shall be of far-reaching value.

PART 3.—PRACTICE
I. PURPOSES
The purpose of a cutting fluid is (a) to cool the work; (b) to
lubricate, thereby increasing the speed of production; (c) to- lessen
the wear of the tool; (d) to diminish the energy consumption;
(e) to insure a good finish; and (/) accurate dimensions; (g) to
wash away the chips; and (h) occasionally to prevent the forma-
tion of metallic dust. Materials are often added to cutting fluids
for the sole purpose of preventing the rusting or corrosion that
would otherwise occur, but it is hardly correct to say that cutting

fluids are used to prevent rusting, since that result, though im-
portant, is incidental. In the same way it is incorrect to say that
cutting fluids are used to prevent disease merely because many
of them contain germicidal substances.
There are two ways in which the use of a cutting fluid assists
in getting accurate measurements: (1) The work may be unduly
heated and thereby expanded unless a good cutting oil is employed,
and (2) the " finish" obtained by such an oil is often essential.
Cooling is said to increase the cutting speed on wrought iron
and steel from 30 to 40 per cent and on cast iron from 16 to 20
per cent. It is to be noted that the most effective cooling is
obtained by directing the cooling fluid on the shaving at the point
where it is being formed, because thereby, through cooling, the
plastic flow of the material is lessened as well as the tendency of
the shaving to seize on the tool. At the same time the fluid can
be drawn in readily between the shaving and the tool, as the
pressure fluctuates, and act as a lubricant.
Bingkam) Cutting Fluids 65

Washing away the chipsis quite important in such operations

as deep example, in the manufacture of rifle barrels.

drilling; for
Securing a strong, continuous shaving is an advantage in such
work and it is necessary to have a lubricant with sufficient ad-
hesion. The stronger the force of the stream of cutting liquid
the better it will carry away the chips, as well as cool and lubri-
cate the parts. It sometimes happens that metallic particles are
not carried away by the emulsions and they cause the dulling
of the tool. This condition is avoided by using a stronger emul-
sion.
If is tough, a lubricant is necessary, but if brittle,
the material
it may be dispensed with. With cast iron, for example, no
lubricant is necessary, but water in the form of an alkaline solu-
tion or emulsion may be used to cool the work, wash away the
chips, and prevent the dust from getting into the air. The
lubricant used in a given case will naturally depend upon the
finish desired and the tolerances allowed. For roughing opera-
tions on a given material a cheap lubricant may suffice, whereas
for fine operations on the same material, such as in the manu-
facture of micrometer screws, the best grade of lubricant is de-
manded regardless of price. There is the further fact that with
some machines such as turret lathes and automatics, the cutting
fluid and the oil used to lubricate the machine are liable to inter-
mingle. In such cases an emulsion which would be serviceable
as a cutting fluid would be worthless as a lubricant in the machine.
It is perhaps needless to add that a cutting liquid must be used
which will also serve to lubricate the machine.

II. SUBSTANCES USED

The substances used as cutting fluidsmay be classified under
the heads (a) oils, (b) air, (c) aqueous solutions, and (d) emulsions.

1. OILS

We may subdivide the oils used into fixed oils, mineral oils,

and compounded oils.

(a) Fixed Oils. —
The fixed oils may again be subdivided into
animal oils, fish oils, and vegetable oils.
—
Animal Oils. Under animal oils we have the several varieties
of lard oil, No. 2, No. 1, extra No. 1, and winter strained prime,
tallow oil, neat's-foot oil, sperm oil, wool fat, horse oil, and whale
66 Technologic Papers of the Bureau of Standards [Vol. 16

oil. Lard oil is the cutting oil par excellence. It is used on the
most difficult work such as cutting delicate threads, tapping and
cutting gears, deep hole drilling in steel and forming work such
as the making and balls on the turret lathe. It is not
of handles
suitable for cutting very hard steel and glass, or where the heat
production is excessive. Although lard oil does not vaporize,
it carbonizes and gives rise to offensive odors. Small amounts
of free acid present in the oil are probably beneficial and they
do not cause corrosion of brass parts. There should, however,
be not more than 15 per cent of free acid, as an excess of acidity
will cause gumming and the formation of verdigris. In the great
majority of cases substitutes for lard oil can be used to advantage.
The very high cost of the fixed oils as a class makes it desirable
that the purchaser make suitable tests to determine that the
material is both unadulterated and sanitary. Horse oil, in
particular, seems to have been severely criticized from a sanitary
point of view, but this is perhaps unwarranted.
Not only should the free acid be not too high, but the stearin
should be kept as low as possible. During the winter months
the lubricant stands in relatively cold shops overnight or over
week ends, and if any material extent,
stearin separates out to
the oil is practically inoperative and means must be provided
for heating it considerably above ordinary room temperature
until the stearin goes back into solution.
Sperm oil is practically interchangeable with lard oil as a
cutting oil, in spite of the fact that it does not resemble it in chem-
ical composition and that it has a much higher fluidity.

—
Fish Oils. Several users have experimented with fish oils.
The report is that the results were satisfactory but the odor was
so objectionable that their use was abandoned. Fish oils, how-
ever, can be deodorized, and it is understood that these deodor-
ized oils are already available. Whale oil has been abandoned
on account of its odor, but otherwise it would be satisfactory.
Vegetable Oils. —
Under vegetable oils, olive oil, rapeseed oil,
mustard-seed oil, castor oil, peanut oil, soya-bean oil, cocoanut
oil, cottonseed oil, corn oil, and linseed oil have been more or less

extensively used. Under the heading of vegetable oils we may

also include creosote oil, turpentine oil, turpentine, pine oil, and
rosin oil, although they are distillates and therefore are not
"fixed oils."
Bingham) Cutting Fluids 67

From tlie iodine numbers of these oils it can be seen that the
tendency to oxidize and, therefore, to gum is very great in linseed
oil, and it is moderate in some of the other oils such as cotton

seed. The tendency of the oil to give trouble by gumming, and

the fire hazard as well, can be reduced by diluting it with mineral
oil.

In continuous use it is obvious that all cutting oils will become

contaminated. They should be sterilized frequently by heating,
all solid particles should be given an opportunity to settle out,

and after a certain time the whole system should be cleaned out
and the oil purified or replaced.
The objection to the fixed oil on account of expense does not
apply to the distilled oils of vegetable origin. The tendency of
pine oil to produce a foam is objectionable, but when this oil is
compounded with sulphur and a mineral oil this tendency is
reduced. These oils appear not to be very widely used except
in emulsions, where rosin oil and soluble creosote are apt to cause
foaming. They come to the top, adhere to the chips, and fre-
quently choke up the system.
Cottonseed oil is much used in the South, but many have re-
jected it, owing to its gumming. It is suggested that there are
many operations, such as threading and tapping by hand, where
cottonseed oil could be used either alone or in mixtures. But it
will not do in place of lard oil for very fine work. Corn oil has
been suggested for use on account of its low cold test.
The use of white lead and linseed oil in tapping and threading
by hand seems to be very common. A number of other solids
are used to a greater or less extent. Graphite, sulphur, and
ground mica are examples.
In drawing fine lead wires cold it is creditably reported that
palm oil, which contains free palmitic acid, works perfectly,
whereas vaseline of almost identical consistency is no good. If
the metal is drawn warm enough, beef tallow, which contains
stearic acid, works well, but paraffin is not good. This is evidence
of the effect of adhesion.
Turpentine is very useful for certain classes of work, as in cut-
ting aluminum, very hard steel, and glass. But the fire hazard is
great with its use. It is often compounded with lard oil. By
itself it is too fluid to act as a lubricant. It is recommended
where lard oil causes a "glaze."
 :

68 Technologic Papers of the Bureau of Standards [Vol. 16

(6) Mineral Oils. —The mineral oils are naturally cheaper

than the fixed oils. In fluidity they range from kerosene, which
is used in cutting brass and aluminum, to heavy engine oils.
They prevent rusting and are very stable, so that corrosion,
carbonization, and gumming are small factors. They are not
cheap enough to compete with emulsions on the one hand; and
on the other hand, because they do not have sufficient adhesion
to those steels which are difficult to work with emulsions, their
use is not very satisfactory. On aluminum, kerosene is prefer-
able from every point of view to a fixed oil. On brass, a paraffin
oil of 25 to 28 Baume mixed with 5 per cent of 300 fire-test
kerosene is described as ideal. On these metals many use no
cutting fluid whatever. On account of the fire hazard emulsions
might better be used.
id) Compounded Oils. —Compounded oils are the natural
result of the desire to secure the adhesion of the fixed oils com-
bined with the stability and cheapness of the mineral oil. Thus
compounded oils are being supplied under a great variety of
trade names, many of which suggest the presence of lard oil.
Some of the compounded oils actually do contain considerable
amounts of lard oil, while others contain none at all. A typical
formula may be given as follows
Per cent
Lard oil 25
Corn oil 15
Natural West Virginia petroleum 60

Many users state that fixed oil must be present to the extent of
from 10 to 50 per cent, and that thereby the valuable advantages
of lard oil are gained in full measure. On the other hand, for
some classes of work there is abundant testimony that even 50
per cent of lard not sufficient. One large user testifies that
oil is
10 per cent of rapeseed oil added to a paraffin oil caused a positive
loss in cutting value.
It should be observed that " cutting compounds," which are
very satisfactory to many large users, apparently contain no fixed
oil whatever. Thus there is possible a great saving in fats by a
knowledge of the best usage in regard to cutting oils. It appears
probable that the necessary adhesion can be gained by thoroughly
incorporating sulphur in oils, such as red oil (oleic acid) and pine
oil, which themselves have considerable adhesion. It will not
do, however, to have the sulphur merely in coarse suspension, as
it will settle out. It should be chemically united.
 :

Bingham] Cutting Fluids 69

2. AIR

Air can not be considered as a lubricant for the purposes which

we are herediscussing, and it has very little cooling effect, How-
ever, with a brittle metal like cast iron it is used successfully to
take away the chips, and thus it prevents the loading and choking
of the tools. The flying of the chips makes guards necessary, and
the dust is also an inconvenience.

3. AQUEOUS SOLUTIONS
Water alone has been found satisfactory in some cases, as in the
and boiler plates. Water does not generally
drilling of ship plates
give a smooth finish, nevertheless it is sometimes used to give a
bright, smooth surface to steel, which is called a " water-cut"
surface. It may also be used on brass and aluminum, but water
is in disfavor, owing to the fact, noted by every machinist, that

it rusts the machines and the work as well if this is either iron

or steel.
To prevent rusting, an alkali may be added to water. Thus for
some operations, such as grinding and drilling, a i to 2 per cent
solution of sodium carbonate has been found satisfactory. Caustic
soda, borax, sodium silicate, and sodium resinate are all used.
A soluble soap is added with the same end in view. Thus a
solution of 50 pounds of sodium carbonate and 25 pounds of soft
soap in 200 gallons of water is said to have given satisfactory
results in boring deep holes, as in gun tubes. Soap, moreover,
has the advantage that if oil is added it will readily form an
emulsion which will possess the cooling properties of water and
will lubricate betterthan water, either alone or containing soda
or soap. Soap solutions have the disadvantage that as the water
evaporates they become sticky and may clog the feed pipes, etc.

4. EMULSIONS
There are three types of material used for making emulsions
1. One class of "soluble oil" is made from mineral oil com-

pounded with a neutralized sulphonated oil. The fixed oils when

treated with sulphuric acid unite with the acid to form so-called
sulphonic acids, which after neutralization with alkali are readily
soluble in water. Rosin oil, though a less valuable product, is
sometimes used. The mixture with mineral oil is, of course, not
soluble in the strict sense, but it does emulsify in the heated
solution.
 : 5 :

70 Technologic Papers of the Bureau of Standards [Vol. 16

2. The second type of soluble oil is made from a mineral oil

compounded with an alcoholic solution of soap. Potassium or

"soft" soap is preferred. Both of these are marketed as clear
oils. In use the oil is stirred into water, usually i part of oil to
4 parts of water.
3. The third variant marketed as paste made of a thick soap
is

solution and mineral oil. These are known as cutting pastes or

compounds.
Of the three forms the last is the least desirable, and the second
is perhaps the most so, as it may be made without expensive

apparatus and requires much less technical skill to fabricate. The

chances of its being made properly are therefore greater. One
very large user has developed the following formula, which is
reported to give very satisfactory results and may be taken as
typical
Gallons Per cent
Water 2. OO 6.0
Alcohol I.50 4-5
Caustic solution (30 B) •75 2.
Kerosene 6.00 17-5
Oleic acid 3.00 9.0
Machine oil 2
21.00 61.

Total 34-25 100. o

This amount of soluble oil is then diluted to make 860 gallons

for average work. Another typical formula is made up as follows
Per cent
Caustic soda , o. 65
Alcohol and water 5.8
Resin acid 1. 64

Fatty acids 11.76

Mineral oil (22. B) 80. 15

Such an oil is diluted with water according to the kind of metal

to be machined or the character of the operation. In a dilution of
one to three parts of water, it is said to do the same work under
any and all conditions that is required of pure lard oil. It should
be observed that neither of these formulas contains either a fixed
oil or a disinfectant. Soluble oils will take care of the most diffi-
cult as well as the lightest work at a far less cost for the same treat-
ment than any other cutting lubricant. They are particularly
recommended where the heating is excessive. The emulsion must
be exceedingly fine grained, so that it will penetrate readily and
not tend to settle out. Emulsions give trouble on automatics or
machines with turret heads because the emulsion gets under the
 1

Bingham) Cutting Fluids 7

turret and displaces the lubricant. In this case the turret will
stick and the machine has to be taken apart and cleaned. If the
operator will frequently squirt lubricating oil through the turret,

tend to obviate the difficulty.

it will

In filtering through a column of chips, as in a deep bore hole, it

has been repeatedly noticed that the character of the emulsion
becomes changed. Presumably the oil is held back and the lubri-
cating property of the emulsion is lost. The very virtue of an
emulsion, that its composition can be readily changed to suit con-
ditions, is thus a disadvantage, since it is also continually
changing
during use. Thus, the system must be watched and occasionally
it should be thoroughly cleaned out and a new emulsion put in.

This should be done anyway, from sanitary reasons. The oil can
be recovered by the use of a rotary separator.
Some waters are sufficiently hard or salty in character to render
an emulsion unstable. Strong acids are, of course, destructive to
emulsions in even exceedingly small amounts. If the water of the
locality is suspected of causing instability of the emulsion, rain
water may be tried, and if this removes the difficulty, it may be
found possible to soften the water. In making an emulsion it is
well to follow the exact procedure which has been found to be suc-
cessful. Too much mineral oil in the emulsion is a cause of insta-
bility. Ammonia is often present in soluble oils but it is not to be
recommended, since it will tend to make an unstable emulsion and
to promote corrosion.
Gum arabic and starch are sometimes added to cutting oils,
presumably in order to increase the stability of the emulsion.
Phenol and cresol are frequently used because they give the mix-
ture a distinctive odor and render it less habitable by disease-
producing bacteria. The use of phenol, even in the form of its
salts, is objected to, and it would appear that recently developed
disinfectants are far less likely to be injurious to the workmen and
are at the same time better germicides.

III. OPERATIONS AND MACHINES

The character of the operations performed, the shape and adjust-
ment of the tool, the speed, feed, and depth of cut have more to
do with the choice of a cutting than the character of the steel.
fluid
management to predict beforehand
It is therefore feasible for the
the most suitable lubricant and save time and materials. For the
 —

72 Technologic. Papers of the Bureau of Standards \voi.x6

great majority of operations aqueous solutions and emulsions

made up of suitable strengths over quite a wide range should un-
doubtedly be used.
For drilling, an emulsion (say, i 20) is most commonly used,
:

but for light, slow work a compounded oil, or even lard oil, is
resorted to. For rough drilling, water containing a little soda,
soap, or borax is employed. Much soda will give rise to foaming.
For reaming, an emulsion (1:3) is generally used, although the
work is often done dry. For a high polish and for extremely
heavy work, compounded oil or lard oil is resorted to.
For milling, emulsions (1 20) are almost universally used, with
:

alkaline solutions, compounded oils, and lard oil as variants.

Planing is often done dry. Sometimes a lubricant is applied with

a brush. In many cases, however, an alkaline solution or a dilute
emulsion (1 120) is used, especially if a high polish is desired.
In tapping and threading, emulsions are not so often successful
as in the above cases, but they are used nevertheless in a large
minority of instances. Lard oil and compounded oils are resorted
to generally, with or without the addition of white lead to prevent
the binding of the taps and dies and to secure a smooth, steam-
tight finish.
Turning practice in different shops and for different operations
shows the widest variance. Some workmen turn dry, others with
alkaline solutions, emulsions, compounded oils, corn oil and
—
where a high finish is desired lard oil and turpentine.
On automatic screw-cutting machines and machines with a
turret head it is best to use a compounded oil with as small a
proportion of fixed oil as will serve the desired ends.

IV. MATERIALS
The nature of the material should be taken into account in
determining the most suitable oil. There is a consensus of opin-
ion that soft steel and wrought iron are difficult metals on which
to get a good surface without the use of lard or sperm oils. Some
of this difficulty can be overcome by the shape of the tool, one
with more rake being required for the soft metals. A change of
speed or feed also helps at times. Without suggesting a cause,
we may be permitted to merely report the fact that speeding up
the operation has often decreased the number of failures from
"bugging " and also the excessive wear on the tools. In threading
wrought iron in dies it is almost impossible to get a good thread
 :

Bingham) Cutting Fluids 73

without the use of lard oil or sperm oil. In drilling deep holes in
tough steels at high speed of production, as in making rifle barrels,
lard oil or its equivalent is positively necessary. Of course, it
is possible that pine oil united with sulphur and compounded with

a suitable mineral oil will in time replace lard oil, which is needed
for other things.
Lard oil or some
with good adhesion will be used where a
oil

long chip is obtained, and it will not be needed where the chips
are short, which indicates a brittle material. In the latter case,
cooling is the more important function of the cutting oil.
With a very hard, brittle steel, turpentine is called for, to cause
the tool to "bite," since lard oil would merely produce a glaze.
Forming tools sometimes have a tendency to "hog" the metal,
which is an indication that a better lubricant is required. It
must be remembered that a very thin film of good oil is often
sufficient to give the smooth surface that is desired, whereas it
can not be obtained by the use of a flood of water or other sub-
stitute.
V. FORMULAS
It is advisable for the engineer of tests in each large organization
to give careful study to the subject of cutting oils rather than
accept too readily the customary substitutes supplied by the trade.
Of the several hundred materials on the market some are good
while others are positively harmful. Small concerns buy up
odd where they can be obtained cheaply, but they are
lots of oil
not always able to keep their products uniform. The freight
rates to and from their plants have to be added to their costs.
It is, after all, not difficult to manufacture a cutting oil, but it is
difficult to analyze a compounded oil so as to be able to state with
certainty whether it is constant in composition and whether the
components are healthful and pure. Moreover, the user can, by
studying his own needs, change his formulas until he finds that
his needs are being met in the most economical manner.
The number of formulas required by a large works need not be
very great. The practice of a single concern, taken as typical,
may be given in detail
i. Pure Oils. —
(a) Pure lard oil, off prime, free fatty acids, determined as oleic acid
not over 5 per cent. Restricted to cutting nickel steel and other steel of high cutting
hardness. Its delivery is controlled under rather rigid restrictions.
(b) Pure mineral spindle oil, viscosity 100 seconds, Sayboit Universal at 40 C,
open cup, 185 C. Used in many classes of work, both in automatic
flash point in
machines and on turret lathes and on other machines where a light-bodied oil is
74 Technologic Papers of the Bureau of Standards [v i.i6

adapted. It is used in turning, shaping, and cutting both steel and copper and in all
operations on brads in automatic machines.
2. Compound Oils. —
(a) Equal parts of lard oil and mineral spindle oil. Used
generally on automatic machines and on other lathes in heavier classes of work for
processes which include tapping and threading. It is also used on machines which
are changed frequently from operations for which the pure mineral spindle oil would
be used to operations involving tapping and threading for which the lard-oil mixture
is required.
(6) Equal parts of mineral spindle oil and heavy mineral oil, viscosity of the latter
730 seconds, Saybolt Universal at 40 C, flash point, 205-2 io° C. This mixture is
substituted for the spindle-lard-oil mixture wherever practical experience shows that
it is economically possible.

3. Water Compounds. — (a) Solution of water and soda. Used in turning of large
shaft forgings.
(&) Solution of soap and water. Used on simple milling and drilling operations
where a water lubricant of light body is satisfactory.
(c) Emulsion of water, soap base, soda, and lard oil. Used where a water lubricant
of good body is required. It is generally in use on boring mills, Gisholt lathes, cold
cutting-off saws, milling machines, etc., and in the cutting of metals in general.
(d) Emulsion of water, soap base, soda, lard oil, and mineral spindle oil. Used
solely for the grinding and finishing of small steel shafts.

The above table gives a clear idea of the number of cutting

fluidswhich are deemed necessary to meet all the needs of a large
plant, but obviously the amounts of the different fluids which
are required will be very different. It is interesting to note the
experience of a large manufacturer on this point. He finds that
a pure paraffin oil (25 ° Be) can be used for 98 per cent of the
work requiring a cutting oil. For the large portion of the remain-
der he uses 40 per cent of off-prime lard oil or sperm oil com-
pounded with 60 per cent of paraffin oil (25 ° Be) For only a very .

small portion of the work he uses off-prime lard oil or sperm oil.

VI. APPLICATION OF THE FLUID

Taylor 27 seems to have been the first to recognize the very
great gain in cutting speed which can be made by having the
cutting fluid applied in a heavy stream with as much force as
practicable directly upon the chip. From what has already been
said it is obvious why this should be so. It has been noticed
that many of the machines work better in winter than in summer,
which can be explained by the more effective action of the cooling
fluids. With the better lubricants, such as lard oil, it is probable
that no corresponding gain would be obtained by the use of a
heavy stream even were practicable. if it

The use of white lead combined with the oil used in threading
and tapping is readily justifiable. By itself the oil would quickly
27 Art of Cutting Metals, p. n.
Bingham) Cutting Fluids 75

run off, but by adding a finely divided solid the fluidity of the
mixture is lowered and it "stays put." If white lead were
objected to on account of the danger of poisoning, it could prob-
ably be replaced by zinc oxide or defiocculated graphite. Un-
fortunately, workmen find graphite objectionable on account of
the color.
VII. KIND OF CUTS
Where the heating is necessarily very great, an emulsion is
always to be preferred. If for some reason a better lubricant seems
required, one would naturally suppose that the lighter the cut the
less viscous the oil need be, and, conversely, with a deep cut one
might expect to have to use an oil with high viscosity. This,
however, is not the case, for if the viscosity of the oil is high it
will be slow in reaching the surfaces to be lubricated and the
cooling effect of the oil will also be reduced. Hence, an oil of
low viscosity is preferred for all high-speed work. To emphasize
the importance of cooling, it may be said that in milling the teeth
of steel gears, the thinnest soap solution that can be used with-
out rusting the machine will do better work than pure lard oil.
In using pure lard oil for this purpose, the friction of the tool is
so great as to burn the oil and fill the room with smoke. The
life of the cutters is no longer with pure lard oil than when a water

compound is used.

VIII. HYGIENIC ASPECTS

The used oil contains sharp metallic particles which are suffi-

ciently small to enter the hair follicles of the skin. not washed
If
out, they may block up the openings to the sebaceous glands,
produce irritation and abrasions, and therefore a favorable environ-
ment for pus-forming bacteria. The oil itself should not produce
any irritation to the skin, but it is suited to catch germs which are
floating about and retain them. The clothing has this same
tendency where it becomes moistened with the cutting fluid.
The main precautions to be taken are (1) the removal of the fine
particles of metal from the used oil, particularly those which are
less than 0.04 mm
in diameter; (2) cleanliness on the part of the
workmen, with abundant facilities therefor, also preventing the
cutting fluid from coming into contact with the clothes and person
more than is necessary; and (3) the avoidance of the use of strong
acids in sulphonated oils or from the pickling liquor and too great
76 Technologic Papers of the Bureau of Standards \Voi.i6

an excess of fatty acids or of alkali, or the use of irritating anti-

septics, etc., in the composition of the oils.

The metallic particles can best be removed by settling in a large

tank. This is easier to accomplish if the oil is not too viscous
and therefore if the oil is hot. It may be cleared by filtration or
by means of a centrifugal separator.
It is an excellent practice to have available in the wash room
a soap emulsion of a light petroleum oil to rub on the hands and
arms before using the soap and brush. This makes the greasy
layer more fluid and thereby facilitates its removal by means of
soap and water.
The United States Public Health Service is making a study of the
hygienic aspect of cutting fluids, hence a more detailed discussion,
of the subject would be out of place here.

Washington, April 20, 1921.