EEO 2 UwE:m--EwE:uon_ ?.3022;

:QExm.

 

 

 

 

 

 

o

~\'.·;~·:· · · o·
t:~~~.,:

.

.. •'l"~'
,.';.-,

'0' :

Protected Document--Subject to Protective Order

JNJAZ55_000001033

 I
I
I
I
I
I
I
I
I
I

PROGRESS REP OR T

on

STUDIES OF THE PHYSICAL PROPER TIES
OF TALC, THEIR MEASUREMENT,
AND COMPARISON
to
JOHNSON AND JOHNSON

October 15, 1957

I

I
I
I
I
I
I
I
I
I
I

.l)

Protected Document--Subject to Protective Order

by

W. L. Smith

BATTELLE MEMORIAL INSTITUTE
505 King Avenue
Columbus 1, Ohio

JNJAZ55_000001034

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
'I
I
I
I
I
II

K-3262-2
cc:

OK 1 d by O. F. Tangel and A. C. Richardson before typing.

-(?) F. TaI1f5f1 (3) MA' C. Richarts1

.N(ll~tt@n

emoria

R. D. Macdonald

'

W. L.,Smith (-3)

nstitute "

/

\
K

50S

1

N

A

G

V

E

N

u E

COLUMB

U 5

OHIO

I,

October 25, 1957

Dr. W. H, Lycan
Director of Re,search
Johnson and Johnson
New Brunswick, New Jersey
De ar Dr. Lyc an:
This letter transmits six copies of our report "Studies of the Physical Properties
of Talc, Their Measurement, and Comparison".
At the pre sent stage of this inve stigation it can be seen that the lubricity of the
Italian talc is closely related to its purity,crystalline habit, and particle-size distribution and is expressed in bulk density, surface area, porosity, and average diameter
measurements. The acceptable Italian talc was found to fall within a small range of
physical measurements., Lubricity was found to be controlled by the shape of the relatively small content of comparatively larger particle s in the otherwise finer mixture.
It appears feasible that the slip of the Italian talc may be improved by the removal
of the coarser mineral contaminants.

Your comments on the findin'gs o'f this investigation will be appreciated.
Very truly yours,

W. L. Smith
Principal Geologist
Minerals Beneficiation Division'

WLS:rr
Enc. (6)

R

E

S

Protected Document--Subject to Protective Order

E

A

R

c

H

F

o

R

N

o

u

s

T

R

y
JNJAZ55_000001035

 I
I
I
I
I
I
I
I
I
I
I
I

TABLE OF CONTENTS

SUMMARY.
INTRODUCTION
2

DISCUSSION OF LUBRICITY

.

,THE ROLE OF MINERALOGICAL PURITY IN LUBRICITY.

:.

3

THE ROLE OF THE CRYSTALLOGRAPHIC HABIT OF TALC IN LUBRICITY

3

THE MEASUREMENT OF LUBRICITY

4

4

Discussion.
The LUb~icity Board
The Lubricity of Talc Sample s

4

7
9

THE RELATIONSHIP OF LUBRICITY TO PARTICLE-SIZE DISTRIBUTION

9

Discussion.

Particle-Size Distribution in Italian Talc.
Correlation of Lubricity With Particle-Size Distribution Data.

9
·11

MOISTURE CONTENT •

15

MEASUREMENT AND CORRELATION OF OTHER PHYSICAL PROPERTIES RELATED TO LUBRICITY

15

Surface Area Determinations by Nitrogen Adsorption
Average Diameter of Partic~e8 .

"

15
15

Specific Surface Calculated FraIn Average Diameter.

17

Porosity
Bulk Density

18
20

CONCLUSIONS

21

BIBLIOGRAPHY

23

APPENDIX A
DESCRIPTION OF LUBRICITY BOARD AND TECHNIQUE OF OPERATION

A-I

APPENDIX B

I
I
I
I
I
I
I
I
I

PROCEDURE FOR PARTICLE-SIZE ANALYSIS

B-1

LIST OF FIGURES

Figure

1.

The Lubricity Board, Showing Descent of Steel Puck

Figure

2.

Underside of Lubricity Board, Showing Microswitchea and Electric Tim.er
Connected tu Lock-in Relay

Figure B-1.

•

Standard Flowsheet for Sizing of Talc Samples

6

B-2

LIST OF TABLES

Table 1.

Typical Data From Lubricity-Board Measurement of Italian Talc Sample (Cranford, 12/22/56)

Table 2.

Lubricity_Board Measurements

~nd

Per Cent Contamination o£ Talc Sa1Tlpled at Cranford,

New Jersey Showing Relation of Lubricity to Purity of Sample

BATTELLE

Protected Document--Subject to Protective Order

7

8

MEMO R I A L

JNJAZ55_000001036

 LIST OF TABLES
(Continued)
Page
Table

3.

Lubricity-Board Data on Flotation
of Contamination on Lubricity.

Produ~ts 'of

Italian Talc, Showing Deleterious Effect
'

8

Table

4.

Previously Reported Particle-Size Distribution Data on Italian Talc

10

Table

5,

Particle-Size Distribution of Three Samples of Italian Talc "

10

Table

6.

Particle-Size Distribution of Talc Sampled 'at Cranford Plant

l<!

Table

7.

Deviation in Particle-Size Distribution in Weight Per Cent of Talc Sampled at the Cranford Plant

13

Table

8.

Relationship of Lubricity to Particle Size .

13

Table

9.

The Lubricity of Mixtures of Coarse and Fine Sizes of Talc, Demonstrating the Control
to be in the Coar,se Fractions.

14

Lubricity Measurements of Italian Talc Samples From Which Specific Particle-Size'
Fractions Have Been Removed

14

Table 10.

Table 11. Surface Area Measurements of Italian Talc, Compared With L':'bricity-Board Measurements
Table 1 <!.

Comparison ,of Theoretical Average-Diameter Measurements and Specific Surface to
Lubricity-Board Measurements of Cranford Samples .

15

16

Table 13. Relationship of Lubricity-Board Measurements to Theoretical Average Diameters on Sized
Fractions of Italian Talc
'

17

Table 14.

18

Correlation of Specific Surface and Lubricity-Board Measurements of Sized Fr,actions of Italian Talc.

Table 15. Porosity, Bulk Density, and Lubricity of Cranford Samples

19

Table 16.

Correlation of Porositr and Lubricity-Board Measurements of Sized Fractions of Italian Talc

ZO

Table 17.

The Relationship of Bulk Density to Lubricity-Board Measurements of Sized Fractions of Italian Talc

zo

Table 18.

Summary of Physical Properties of Sized Fractions of Italian'Talc

zz

~A

Protected Document--Subject to Protective Order

T TEL L E

M E M ,0 R I A L

NSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I'
I
I
I
I
I
I
I
I
I
JNJAZ55_000001037

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
,I

STUDIES OF THE PHYSICAL PROPERTIES OF TALC,
THEIR MEASUREMENT, AND COMP ARISON
by
W. L. Smith

SUMMARY

In order to improve the physical properties of talc it is necessary to be able to
measure the differences in talc and to establish a basis for the determination of improvement. To study the lubricous property of talc, an experimental lubricity measuring device was built, ,and the behavior of different talc samples was compared with
their other physical properties. The comparative physical measurements were made
, upon sized fractions and whole sample s of Italian talc with conventional laboratory
devices. It was found that the ,acceptable Italian talc fell within a small range of the
physical measurements and that the samples with the more desirable slip have the
greater surface area, the smaller average particle diameter, the' greater ratio of· voids
to total volume, and the lesser bulk density. Lubricity was found to be controlled by
the shape of the relatively small content of comparatively larger particle's in an otherwise finer mixture. At the pre sent stage of the inve stigation, the improvement of the
slip of the Italian talc appears feasible by the removal of the coarser mineral
contaminants.

INTRODUC TION

The talc currently used by Johnson and Johnson, obtained from Pinerolo, Italy, is
believed to be a blend of five or more different grades of ore which gives a high quality,
,lubricous powder. In a proposal for research on the improvement of the properties of
talc, dated June 4, 1956, it was proposed to study the basic properties of the acceptable
Italian talc and to determine if and how the quality might be improved. The development
of objective te sts which might serve in the evaluation of talc was recommended, with
the initial workto be done upon the product now used by Johnson and Johnson.
, In order to determine improvement in the quality of talc, however, it is necessary
to measure the apparently small differences between acceptable talc and talc of lower
quality. Previously, measurements have been made by subjective methods only, which
were thought to be of insufficient precision to measure sUlaH difference s. The development and correlation of physical measurements has been undertaken to perUlit the
measurement of improvement. The measurements of the physical properties ofacceptable talc, and their range, have been made upon a series of one kilogram samples of
grade "EGT Extra 00000'· taken at weekly intervals from the conveyor at the Cranford,
New Jersey, plant just before the talc enters the ribbon blendors. An additional large
sample of talc was obtained for tests requiring larger volumes of material.
As a test for the comparative lubricity of talc samples, a lubricity board was constructed. This device, though not providing absolute values, gives reproducible
BAT TEL L E

Protected Document--Subject to Protective Order

M EM 0

RIAL

INSTITUTE

JNJAZ55_000001038

 2
relative figure s, with which the other physical propertie s, which can'be more easily
measured, may he correlated. This study" when coupled with proposed work on abrasiveness and other properties, should indicate the course to follow in beneficiation, the
primary work on which Battelle reported in a letter dated June 12, 1957. The flotation
work to date has been a laboratory expedient of producing talc of superior quality for
physical te sts.

/ ' DISCUSSION OF LUBRICITY

This report deals with lubricity and other physical properties including particlesize distribution and surface area,' which are pertinent to lubricity. The desirable
quality in talc, however, is orily partly a matter of lubricity. That is, talc with the
de sired "feel" (as sensed subjectively) is not dete·rmined either by very high or very
low lubricity (diminution of friction) but by a balance of physical properties which
produce s a particular sensory effect. This quality is referred to in this report as slip.
The primary determining factor is the platinessof fine grained particles sliding over
one another under slight pre ssure -not being lubricous in the sense of bearings which
cut down friction by transmitting the moving forces to the rolling of an intermediate
body, producing point friction; not being lubricous in the sense of'a viscous fluid which
buffers contact; but being lubricous in the sense of a series of leaves which impart the
relative movement of two bodies along several planes parallel to the contact, producing
the sensation of softness of surface contact.
The nature of the sliding of the platelets is a matter of kinetic s, the change s in
types of motion produced by applied forces. A certain intensity of force i.s required to
maintain sliding between any two surfaces, and this varies with the nature of the surfaces. When the applied force is distributed alo~g numerous planes rather than between
two surfaces, the resistance to relative motion between the two bodies is distributed
among a series of translation movements rather than in a rotational movement or ~n the
overcoming of inertia in one plane. Inasmuch as the coefficient of sliding friction is
apparently much less between talc platelets than between talc and flesh, the total friction
resulting from the sum of translation movements is necessarily less than that of flesh in
contact with fIe sh, and thus the lubricous property is sensed.' The ,force producing the
relative motion of two bodies is applied to'.. . the several planes of free moving talc platelets, which orient their greatest surfaces normal to the force applied; the contact of
this serie s of parallel talc plane s with fIe sh produce s the silky or smooth sensation
desirable in high quality talcmTI. powder.
Grit (granular and acicular particle s), where pre sent, introduce s point fr'iction as
in bearings, or plowing and thus is the primary objectionable' contaminant in talcum
powder.
'
The lubricou,sness or slip of talcum powder is determined by its mineralogical
purity, the crystallographic habit of the talc ,the size distribution of the powder, its
moisture content, and, the nature of the ,contaminants. Most of these factors must be
determined petrographically, often on separated fractions of the powder. Other
physical properties, such as surface area, average diameter, porosity, and bulk
density, may be measured mechanically. Such physical measurements have been made,
and the data have been correlated to determine which propertie s are significant in
ground talc which has the desired slip. Measurements have been made to establish the
BAT T E L.L. E

Protected Document--Subject to Protective Order

M E'MO R I A'·L.

INSTITUTE

I
I
I
I
I
I
I
I
I
I

'I
I
I
I
,I
I
I
I
I
I
I
I

JNJAZ55_000001039

 I

'I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

r,

3
optimum limits of many of the propertie s relevant to lubricity. 'Data relevant to color,
reflectance, moisture content, alkalinity, and abrasiveness will appear in a subsequent
report.

THE ROLE OF MINERALOGICAL PURITY IN LUBRICITY

The Italian talc contains from about 97 to more than 99 per cent pure mineral talc.
The predominant contaminant is carbonate, which is present in all size fractions, being
slightly more abundant in the fines. Among other contaminants, pre sent in trace
amounts, are amphiboles, rutile, zircon, apatite, and titanite. The Italian talc is
essentially free of opaques. The contaminants are generally prismatic or angular
particles which act as grit and introduce point friction. A few such equidimensional or
acicular particles present in an otherwise platy talcum, particularly if the grit is present in the coarser sizes, may be easily noticed subjectively. They diminish the
lubricous feeling by the introduction of plowing, bearing-like particles, and the disruption of the lamellar movement of the talc particle s. Inasmuch as the contamtnants
generally have diameters ·considerably greater than the thickness of talc platelets, their
removal wQuld improve the slip of any platy talc. in which they occur in an e.ffective
amount. The small percentage of contamination in the Italian talc is an effective
amount, as demonstrated by lubricity te sts Oil a sample upgraded by froth flotation.
Further discussion of the nature of the impurities of talc was reported in earlier
Battelle reports to Johnson and Johnson dated May 11, 195s(l)*, February 29, 1956(2),
May 28, 1957(3), and July 25, 1957(4).

v/THE ROLE OF THE CRYSTALLOGRAPHIC HABIT OF TALC IN LUBRICITY

. Platy talc is the most desirable for the purposes of the Sponsor. Whereas
acicular and granular talc particles plow or roll, producing point friction, platy particle s slide over one anothe r producing the soft lubricous sensation de sirable in talcum
powder. The Italian talc averages about 10 per cent fibrous or acicular p'articles and
about 90 per cent platelets. The amount of granular talc particle s is negligible.
Fibrous and granular particles of talc, though physically softer than the foreign
mineral contaminants of talc ore, are none the less undesirable - if to a lesser degree.
Such particles are most undesirable when present in the larger grain sizes where these
crystals or aggregates may act as bearings or irritants.
Whereas differ.ent minerals may be separated from one another by physical
proce s se s, such as the removal of carbonate s from talc by flotation, more difficulty is
involved in separating particles of monomineralic, impalpable powder on the basis of
their crystallographic habit, except where the crystal types concentrate in specific size
fractions. Until beneficiation procedure s for concentrating talC with the de sired
crystallographic habit are developed, talc which has the crystallographic habit preferred must be obtained by selective mining.
A detailed discussionof the various crystallographic habits of talc appears in a
Battelle report to Johnson and Johnson dated February 29, 1956(2}.
• References are given on page 23.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001040

 4

THE MEASUREMEN T OF LUBRICITY

Discussion

Although the desired property of talc, the slip or optimum balance of sizedistribution and friction is only partly a matter of lubricity, it is correlative within certain
limits of lubricity. A standard method of objectively measuring the lubricousness of
talc has not been devised previously. The lubricity has been determined comparatively
by feeling a pinch of powder between the fingers. People experienced in so testing talc
subjectively are able to distinguish line difference s in quality. Since the de sir able and
undesirable qualities of talc are a subjective matter" it is likely that subjective testing
is preferable. However, since the subjective tests are a matter of sensation, involving
human reaction to several physical properties, such tests are of little help in devising
ITlethods of iITlproving the physical properties of talc or of measuring small differences
in particular properties. Because of this it was necessary to build a device to objectively test and measure lubricity, apart from the other properties which contribute to
the de sirability of talc.
It is not to be inferred that an objective test can replace the subjective test or that
pleasantness of sensation is mechanically measurable; however, the physical properties
which contribute to the unctuousness of talc can be. measured and their optimum limits
can be determined. Thus the means of improvement of tales can be visualized. The
figures obtained on the lubricity-board experiments are compared with the more easily
made measurements of other physical properties in order to determine if there is a
correlation and to establish the desirable liITlits of particular propertie s in acceptable
Italian talc.

The Lubricity Board

In order·to obtain quantitative measurements of talc samples, against which the
ITleasurements of other physical properties could be compared, a simple ITlachine was
constructed with a minimum of interacting physical ffictor~. This device consists of a
wooden plane inclined at 25 degrees, which is covered with talc (Figure 1). The
1ubrictty is determined by measuring the time it takes a ZZ6-gram steel puck to slide
over two microswitche s spaced 5 feet apart (Figure 2). The microswitche s actuate an
electric timer.
The lubricity board was designed as a preliIninary device in making lubricity
experiments; however, a routine method of measurement has been established and the
device has demonstrated a reproducibility with an accuracy of more or less 1 per cent.
Although more precise machines might be built, the lubricity board has proven to be an
adequate ITleans of measuring the comparative lubricity of talc saITlples and to be
adequately precise for the comparison of data from other physical measurements. The
measurements made on the lubricity board are' pre sented in terms of . xxx second, the
figures representing the average of fifty readings. A typical setof figures is shown in
Table 1. A description of the lubricity board and. the technique of its operation comprise s Appendix A.

BAT TEL L

Protected Document--Subject to Protective Order

E

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
f;l

JNJAZ55_000001041

 5

N40207
FIGURE 1.

THE LUBRICITY BOARD, SHOWING DESCENT OF STEEL PUCK

BAT TEL L

Protected Document--Subject to Protective Order

E

MEMORIAL

INSTITUTE

JNJAZ55_000001042

 6

N40208
FIGURE 2.

UNDERSIDE OF LUBRICITY BOARD, SHOWING MICROSWITCHES AND
ELECTRIC TIMER CONNECTED TO LOCK-IN RELAY
BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001043

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
11

7
TABLE 1.

TYPICAL DATA FROM LUBRICITY-BOARD MEASUREMENT
OF ITALIAN TALC SAMPLE (CRANFORD, 12/22/56)

Series 1

Series 2

0.96(a)
0.98
0.95
0.95
0.97
0.97
0.96
0.95
0.97
0.94

0.97
0.96
0.96
0.95
0.96
0.95
0.96
0.97
0.96
0.97

Measurement in Seconds
Series 3
Series 4
0.98
0.97
0.97
0.97
0.96
0.96
0.97
0.98
0.97
0.97
Average 0.965 second

0.97
0.96
0.96
0.98
0.98
0.97
0.96
0.98
0.97
0.99

Serie s 5
0.97
0.97
0.96
0.95
0.96
0.94
0 .. 97
0.98
0.95
0.97

. (a) Lubricity board newly covered for each series of ten slides.

Contrary to preconceived ideas about the behavior of solid lubricants, the puck
was found to slide faster on poorer grades of talc than on cleaner samples with better
slip. The controlling factor is the presence of contaminants or equidimensional particles which act as bearings while purer talc, within the limits of the particular size
distribution, presents a surface composed of flat platelets, producing more friction and
slowing the de scent of the puck.
It is not suggested at this time that Johnson and Johnson conduct similar experi-

ments on a lubricity board. Until the lubricity studie s are completed, it is intended
that the lubricity board serve only as a basis for comparison with other measurements
by which the physical properties which control lubricity can be evaluated.

The Lubricity of Talc Sample s

Lubricity-board measurements were ma.de on 15 sample s of talc obtained from the
Cranford plant of Johnson and Johnson r collected at regular intervals from August 10 to
December 22, 1956. Table 2 lists the figures obtained for each sample. The readings
ranged from 0.936 second for the sample which permitted the fastest descent of the
puck to 1.083 seconds for that sample which most slowed the de scent.
To test the hypothesis that the faster descents were due to bearing-like contaminants, the lubricity figures were compared to those of the percentage of contamination. Table 2 lists the amount of contamination as compared to the lubricity of the
sample s . The contamination figures repre sent microscopically identifiable particles
and do not include the impalpable fines. Although correlation was not perfect, the
relationship of contamination to lubricity is clear in the extreme instance s. The talc
containing the greater amount of contaminants permitted the faster de scents on the
lubricity board. The slight difference s in contamination were not discernible
subjectively.

BAT T E L L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001044

 I
8
TABLE 2.

LUBRICITY-BOARD MEASUREMENTS AND PER CENT
CONTAMINATION OF TALC SAMPLED AT CRANFORD,
NEW JERSEY, SHOWING RELATION OF LUBRICITY TO
PURITY OF SAMPLE
Contamination(a) ,
per cent

Date Sampled

Lubricity-Board Measurements,
seconds

<1

9-6-56
11-6-56
9-12-56
9-19-56
10-18-56
8-10-56
9-27-56
8-28-56
10-29-56
10-4-56
8-20-56
10-12-56
12-22-56
11-30-56
11-15-56

1.083 (slowest)
1.053
1. 030
1. 028
1. 025
1. 021
1. 017
1. 007
1. 006
0.982
0.971
0.968
0.965
0.952
0.936 (fastest)

1

<1
<1
1
1

1
2

1-2
1-2
2
2-3
2
2-3
2

(a) Determined petrographically.

Inasmuch as the contamination figures were small, were close together, and
could be prejudiced, the contaminants were removed from a sample of talc by froth
flotation and the products were tested on the lubricity board. The test results, which
are also noticeable subjectively, are given in Table 3.
TABLE 3.

LUBRICITY-BOARD DATA ON FLOTATION PRODUCTS
OF-ITALIAN TALC, SHOWING DELETERIOUS EFFECT
OF CONTAMINATION ON LUBRICITY

Product

Lubricity-Board Measurement,
seconds

Starting sample
Float product(a)
Nonfloat product(b)

0.990
1. 046 (superior)
0.873 (inferior)

(a) Essentially pure talc, representing 90 per cent of starting sample.
(b) 85 per cent talc, 15 per cent contaminants, representing 10 per cent of starting sample.

The essentially pure talc product produced a slower de scent of the steel puck than
did the unseparated sample, and the flotation tailings permitted a descent considerably
faster than did the unseparated sample. It may be concluded, on the basis of several
experiments on the lubricity board, that the purer talc with the better slip require s a
longer time for the puck to slide, while the samples more contaminated with granular
"bearings " permit faster descents. Although the details of practical beneficiation of
this talc by froth flotation have yet to be worked out, the amenability of the talc to
BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
r~

JNJAZ55_000001045

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

II

9
flotation and the obvious improvement in the purity and slip of the product indicate that
beneficiation is a feasible consideration for the improvement of the Italian talc.

c..

THE RELATIONSHIP OF LUBRICITY TO
PARTICLE-SIZE DISTRIBUTION

Discussion

The desirable slip in talcum powder does not depend alone on the physical lubricity of the mineral but on numerous interdependent physical properties, one of the more
important being particle-size distribution. The relative amount of grains in different
size fractions, the extreme sizes, and the crystalline habit of grains of different sizes
playa major role in lubricity. Comparatively larger grains in an otherwise fine
powder may roll like bearings, plow, or act as barriers to the free movement of
smaller platelets. Too large an amount of very fine grains may behave as "flour"
de spite their particle shape and may disrupt or may clog the movements of larger
platelets.
Size distribution is reflected in bulk density, porosity, surface area, and average
dianleter nleasurenlents. The samples of Italian talc were found to have physical
properties which fall within a small range of many of these nleasurenlents and which
can be related to lubricity.in some instances. Too nlany variable properties exist in
talc to assess specific requirements for many physical measurenlents; however, in
testing for acceptable tales, those ores with properties similar to the Italian talc
should also be expected to have measurements within or close to the range of those
obtained on the Italian talc. The measurements should be a useful guide in the blending
of tales and for the rejection of inferior grades.
Battelle wishes to emphasize that immediate c'onclusions should not be drawn
fronl the following physical measurements alone, inasmuch as they represent but half
of the story. A forthconling report which will deal further with lubricity and other
physical properties such as whiteness, abrasiveness, and nloisture content, will expand
the list of the properties required for an acceptable talc. Although nlany tales may be
rejected because they fail to meet certain physical requirements pre sented here,
acceptability involve s additional factors.

Particle-Size Distribution in Italian Talc

In a previous report' to Johnson and· Johnson( 2), Battelle reported a dry screen
analysis and a particle-size distribution of Italian talc based on both dry screening and
sedimentation in water. These findings are repeated in Table 4. Further, extensive
particle-size distribution studie s were made showing that three replicate analyses of a
large saITlple of Italian talc were closely reproducible and in close agreement with the
earlier analyses, although obtained from a separate sample (Table 5). Appendix B
contains a description of the procedure for particle-size analysis~

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001046

 10

TABLE 4.

PREVIOUSLY REPORTED PARTICLE-SIZE
DISTRIBUTION DATA(2) ON ITALIAN TALC

a.

Dry Screen Analysis

Tyler Mesh Size

Weight Per Cent

t150
-150t200
-200+270
-270t325
-325

O. 1
1.8
2.1
13.5
82.5
100.0

b.

Approximate Particle-Size Distribution Based on Dry
Screening and Sedimentation in Water
Approximate
Weight
Per Cent

Size
-100t200 Mesh
-200t325 Mesh
-325 Mesh t 15 Microns
-15 Microns t 10 Microns
-10 Microns

TABLE 5.

2

16
62
9
11
100

PARTICLE-SIZE DISTRIBUTION OF THREE SAMPLES
OF ITALIAN TALC

Size
t200 mesh(a)
-200 mesh t 325 :mesh
-325 mesh t 400 :mesh
(38 :microns)
-38 :microns t 30 microns
-30 :microns t 15 microns
-15 :microns

Average Weight
Per Cent

Per Cent
Deviation From Mean

1
10
7

0.12
0.97
1.3

57
12
13
100

5.18

0.8
3.1

(a) Tyler.

8

A T TEL L

Protected Document--Subject to Protective Order

E

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
IJ

JNJAZ55_000001047

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

II

11
To check the variation in particle-size distribution of the talc samples from
Cranford, a series of measurements by dry screening ,and sedimentation were made on
12 sample s. Table 6 lists the weight per cent of each size fraction. One of the
samples, Cranford 10/4/56, might be eliminated statistically from the sample population; however, the variation is real and its data are included in the weight per cent
deviation from the mean (Table 7). The effect of the variant sample shows clearly in
the measurements of average particle size, specific surface, bulk density, and porosity
(Tables 12, 15)., The cause of the variation is not obvious petrographically, the
sample being similar to the others except in size distribution. This variant sample
demonstrates that the Pinerolo product is not uniform. Also, since the larger coarse
fraction does not cause the expected effect on the lubricity measurement as do variations within the normal size distribution population, it shows that the matter of lubricity
is more complex than is indicated by variations within a small range in particle-size
distribution.
The Cranford samples have minor variations in particle-size distribution. In all
of the samples, however, the -400 mesh (38 micron) + 30 micron fraction constitutes
about one-half of the sample, with minor amounts in the smaller and larger fractions.
This distribution should be kept in mind should platy tales of other distributions be considered. Inasmuch as the size distribution may be as much a matter of the grinding
and blending of ore s as of the physical nature of the are, fabrication of acceptable
talcum powder from lower grade ore might be accomplished by a proper blend of sized
fractions of platy talc.

Correlation of Lubricity With Particle-Size Distribution Data

In order to correlate the particle-size distribution and lubricity data, a large
sample of Italian talc was sized and the size fractions were te sted on the lubricity
board. Experiments clearly showed that the coarser fractions permitted a faster
de scent of the puck while the finer fractions produced a slower de scent. This is apparently due to the more equidimensional bearing-like particles in the coarser fractions. Table 8 demonstrate s the relationship of particle size to lubricity, showing the
larger, more desirable, lubricity figures for the fines, the lower for the coarser,
gritty fraction.
Inasmuch as the lubricity of talc involve s the physical propertie s of material of
various grain sizes, lubricity measurements were made on various proportional mixtures of specific particle-size fractions and on powders from which specific particlesize fr.actions were removed. In order to determine if the over-all lubricity was controlled by the coarse or by the fine sizes, a series of lubricity measurements was made
on proportional mixtures of the fine (-400 mesh) and coarse (+250 mesh) sizes.
Table 9 clearly shows that the control is in the relative amount of coarser to finer
grains. That is, a small amount of coarse particles added to an otherwise fine powder
has a pronounced adverse effect on lubricity, whereas a similai- percentage addition of
fine particles to an otherwise coarse grained powder has comparatively little effect on
lubricity. It may be concluded from this study that the removal of the coarser particles, which tend to be more equidimensional, will improve the slip. On the other hand,
the addition of fines to gritty or granular powders makes comparatively little
improvement.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001048

 "'0

o
It
n

It

C-

O
n

o

l:

3

III

::!.

in
l:

.E

PARTICLE-SIZE DISTRIBUTION OF TALC SAMPLED AT CRANFORD PLANT

TABLE 6.

III

!:lei
"'0

o

It
!:l-

<'
o

ID

~

>

a.

-30 Microns
+15 Microns

-15 Microns

65.72

10.76

10.78

7.75

56.66

16.49

13.46

8.30

7.85

60.76

10.23

11. 94

0.84

5.54

7.65

55.79

16.42

13.88 -,'

10-18-56

0.96

7.86

7.27

58.07

13.32

12.52

9-27-56

0.59

4.48

6.77

56.52

15.93

15.71

8-28-56

0.50

4.34

6.89

56.39

16. 18

15.71

10-4-56(a)

1. 22

6.38

9.36

40.02

31. 74

11.28

8-20-56

0.65

6.08

7.98

57.53

16.64

11. 12

12-22-56

0.88

3.42

12.30

52.27

20.01

11. 12

11-30-56

0.72

6.38

10.33

57.86

11. 21

13.50

11-15-56

0.88

6.93

9.91

48.72

22.08

11. 48_

+200 Mesh

9-6-56

0.47

5.44

6.83

11-6-56

0.51

5. 13

9-12-56

0.92

9-19-56

Date Collected

III

~

Weight Per Cent of Size Fractions
-400 Mesh
-:325 Mesh
+400 Mesh
+30 Microns

-200 Mesh
+325 Mesh

~

11'1

r
r
11'1

~

11'1
~

0
lJ

-

>
r

-

z

CIl
-f

.....
tv

-f

c
-f

11'1

(a) See comment under "Particle-Size Distribution in Italian Talc".

c...
Z
c...

~

I~
o
o
o

~~

<D

-

-

-

-

-

-

-

-

- -

-

-

-

-

-

-

-

-

-

-

-

I

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
11

13

TABLE 7.

DEVIATION IN PAR TICLE-SIZE DISTRIBUTION IN WEIGHT PER
CENT OF TALC SAMPLED AT THE CRANFORD PLANT

Size

Deviation, weight per cent
(Excluding Sample 10/4/56)

Deviation, weight per cent
(Including Sample 10/4/56)

0.26
2.44
1. 78

0.39
2.44
1. 78

8.50
5.42
3.72

12.85
10.75
3.72

+200 mesh(a)
-200 me sh + 325 me sh
-325 me sh + 400 me sh
(38 microns)
-38 microns + 30 microns
-30 microns + 15 microns
-15 microns
(a) Tyler.

TABLE 8.

RELATIONSHIP OF LUBRICITY TO PARTICLE SIZE

TyIe r Me sh Size

Lubricity-Board Measurement,
seconds

Unseparated

0.990

+200
-200+250
-250+270
-270+325
-325+400
-400

0.889
0.951
0.980
1. 030
1. 043
1.099

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001050

 I
14

TABLE 9.

THE LUBRICITY OF MIXTURES OF COARSE AND FINE SIZES
OF TALC, DEMONSTRATING THE CONTROL TO BE IN THE
COARSE FRAC TIONS

Per Cent Coarse
(+250 Mesh)

Lubricity-Board
Measurement,
seconds

a

100

0.951

10

90

0.951

25

75

0.960

50

50

0.970

75

25

0.986

90

10

1. 038

Per Cent Fine s
(-400 Mesh)

Difference in
Lubricity

.000
.009
.010
.016
.052
.061
100

a

1.099

Further measurements of the effect of particle-size distribution on lubricity were
made by te sting whole powder from which different size fractions had been removed.
The measurements, Table la, demonstrate that removal of the fines decreases the
quality of the powder, whereas removal of the coarse fractions improves it. It would
have to be determined by further tests of beneficiation products whether it is most
advisable to remove entire size fractions or merely the small percentage of coarse
contaminants.
TABLE 10. LUBRICITY MEASUREMENTS OF ITAliAN TALC SAMPLES FROM WHICH SPECIFIC PARTICLE"SIZE
FRACTIONS HAVE BEEN REMOVED
X Represents Fractions Removed From Whole Powder
U Represents Fractions Tested

Tyler
Mesh Size

Lubricity -Board
Measurement ot<a)
Size Fractions

Test
1

Test
2

Whole
Powder

Test
3

Test
4

+200
-200+250
"250+270
-270+325
"325+400
-400

0.889
0.951
0.980
1. 030
1. 043
1. 099

U
U
U
X
X
X

U
U
U
U
U
X

U
U
U
U
U
U

X

X

X
U
U
U
U

X
X

U
U
U

0.945

0.963

1. 038

1. 068

97.

82.

2.

3.

Lubricity-Board Measurement
Approximate Weight Per Cent of
Fractions Removed

0.990
Increase in slip
O.

(a) Repeated from Table 8 for comparative purposes.

BAT T E L L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

~.

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
(~

JNJAZ55_000001051

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

15
MOISTURE CONTENT

Of considerable importance to the lubricity of talc is its moisture content. This
topic is more thoroughly treated in a forthcoming report. It is important, however, to
note here that an increase in moisture content slows the descent of the puck on the
lubricity board and falsely indicates superior lubricity. All of the Italian talc was
found to contain a moisture content in the hundredths of one per cent. Analyse s of
various .size fractions show that the moisture content is higher in the fine sizes,
possibly due to adsorption on the greater surface area.

MEASUREMENT AND CORRELATION OF OTHER
PHYSICAL PROPERTIES RELATED TO LUBRICITY

Surface Area Determinations by Nitrogen Adsorption

The relationship of lubricitY,to particle-size distribution has been shown to be a
matter of friction and surface area, the finer platelets having the greater surface area
per unit of weight.
Four sample s of Italian talc from Cranford were measured for their surface area
using the Brunauer, Emmett, Teller technique of nitrogen adsorption at liquid nitrogen
temperature s (Table 11).
TABLE 11.

SURFACE AREA MEASUREMENTS OF ITALIAN TALC, COMPARED WITH LUBRICITY-BOARD MEASUREMENTS

Cranford
Sample Date

Lubricity-Board
Me asuretnent,
seconds

BET Surface Area
Measuretnent,
m 2 /g

9-6-56
9-12-56
9-19-56
10-4-56

1.083
1. 030
1. 028
0.982

3.18
2.93
2.26

3.57

The te sts show a relationship between surface area and lubricity, the sample s
with the greater surface area also having the larger lubricity measurements. The
values are believed to be accurate to within 5 per cent.

Average Diameter of Particles

An easily operated instrument for rapid particle-size determinations is the Fisher
Subsieve Sizer. The instrument measure s average particle size by determining the
resistance to the flow of air by a weighed sample of powder under standard packing conditions. On the basis of the principle that a fluid meets less resistance to flow while
8

A T TEL L E'

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001052

 I
I

16

TABLE 12.

COMPARISON OF THEORETICAL AVERAGE-DIAMETER
MEASUREMENTS AND SPECIFIC SURFACE TO LUBRICITYBOARD MEASUREMENTS OF CRANFORD SAMPLES
Lubricity-Board
Measurement,
seconds

Theoretical Average(b)
Particle Diameter

9-6-56

1.083

2.60

8392

11-6-56

1. 053

2.65

8233

9-12-56

1. 030

2.60

8392

9-19-56

1.028

2.45

8905

10-18-56

1. .025

2.60

8392

8-10-56

1. 021

2.80

7792

9-27-56

1. 017

2.80

7792

8-28-56

1.007

2.75

7934

10-29-56

1. 006

2.80

7792

10-4-56(a)

0.982

3.30

6612

8-20-56

0.971

2.90

7524

10-12-56

0.968

2.90

7524

12-22-56

0.965

3. 10

7038

11-30-56

0.952

3.30

6612

11-15-56

0.936

3.20

6818

Cranford
Collection Date

Specific Surface,
crn 2 /g

(a) See comment under "Particle Size Distribution in Italian Talc".
(b) Determined on Fisher Subsieve Sizer.

BAT T E L L

Protected Document--Subject to Protective Order

E

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000001053

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

17
penetrating a bed of coarse particles than while penetrating a bed of fine particles, a
figure is derived which disregards the shape of the individual grains, porosity, size
distribution and other variables. Inasmuch as the average diameter figure represents
a theoretical sphere, the data are of relative rather than actual value. Average diameter measurements made on the Fisher Sub sieve Sizer are shown in corn.parison with
lubricity rn.easurements (Table 12), dern.onstrating the correlation of small average
diameters to talc with the larger lubricity rn.easurement and larger average diarn.eters
to talc with the lower, Ie s s de sirable, lubricity rn.easurements. A clear-cut correlation of the theoretical average-partic1e-diarn.eter rn.easurern.ents with the lubricityboard measurements is shown in Table 13.
TABLE 13.

RELATIONSHIP OF LUBRICITY-BOARD MEASUREMENTS
TO THEORETICAL AVERAGE DIAMETERS ON SIZED
FRACTIONS OF ITALIAN TALC

Tyler Me sh Size

Lubricity-Board
Measurements,
seconds

Theoretical Average
Particle Diarn.eter,
rn.icrons(a)

0.990
0.889
0.951
0.980
1. 030
1. 043

2.60
7.40
3.60
2.50
2.35
2.25
2.10

Unseparated
+200
-200+250
-250+270
-270+325
-325+400
-400

1.099

(a) Determined on Fisher Subsieve Sizer.

Specific Surface Calculated From Average Diameter

The average particle diarn.eter as deterrn.ined on the Fisher Subsieve Sizer rn.ay,
by use of a sirn.ple equation, be expressed in terms of specific surface in square
centimeters per grarn. of dry powder. This is a sirn.pler, less expensive method than
nitrogen adsorption. ' Specific surfaces, as calculated frorn. the average-particlediameter rn.easurern.ents, are pre sented in Table 12 in corn.parison with lubricity. The
calculated specific-surface figures, because of their derivation frorn. average-particlediameter rn.easurern.ents, are inversely correlative with particle size. The sarn.ples
with the greater specific surfaces are those which irn.pede the slide of the puck on the
lubricity board, and which have better slip, while the samples with the srn.aller specific
surfaces, those containing the larger particles, are the sarn.ples perrn.itting faster
de scents of the puck on the lubricity board.

*

As in the case of the average particle-diarn.eter rn.easurement, the specificsurface calculations repre sent theoretical sphere s which, since the powder is corn.posed
of platelets, are of relative rather than exact value. Whereas the surface area as determined by gas adsorption is relatively exact, the value of the surface -area figure s
derived frorn. the theoretical average-particle-diameter rn.easurern.ents is purely
corn.parative .
•Specific surface (cm2/g)

=

.
6 x 10 4 ..
average diameter (p.) x specIfiC graVity of talc

BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001054

 I
18

I

The relationship of specific surface to the lubricity of sized fractions of Italian
talc is presented in Table 14, which shows that the fractions with the better lubricity
also have the greater surface areas.
TABLE 14.

CORRELATION OF SPECIFIC SURFACE AND
LUBRICITY-BOARD MEASUREMENTS OF
SIZED FRACTIONS OF ITALIAN TALC

Tyler Mesh Size
Unseparated
+200
-200+250
-250+270
-270+325
-325+400

-400

Lubricity-Board
Measurements,
seconds
0.990
0.889
0.951
0.980
1.030
1. 043
1.099

Specific Surface,
cm 2 /g
8392
2948
6061
8727
9284
9697
10390

Porosity

A measurement of porosity, independent~£ the other measurements, may be
made on the Fisher Subsieve Sizer. The porosity figure represents the ratio of voids
to the total volume of the packed sample, in a range of 0.40 to 0.80. Inasmuch as part
of the test involves a manual operation, the results are subject to a human error. The
porosity figures, however, are reproducible through the second decimal place. The
range in porosity, 0.448 to 0.490, determined on the Cranford samples, is a relatively
small range and should, by its close limits alone, be of assistance in evaluating Italian
talc (Table 15). The more porous powders, those with the greatest amount of asymmetrical, platy grains, are also those with the larger lubricity-board measurements,
hence the better slip. Correspondingly, the samples with the lower porosity, those
containing the greater amount of equidimensional grains, are the powders which have
the lower lubricity-board measurements.
Table 16 shows the porosity of sized fractions of Italian talc as compared with its
lubricity-board measurements. The porosity is clearly shown to be less in the coarser
fractions with the poorer slip and greater in the finer fractions.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000001055

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

19

TABLE 15.

POROSITY, BULK DENSITY, AND LUBRICITY
OF CRANFORD SAMPLES

Porosity
Ratio

Bulk Density,
Ib/cu ft

Lubricity-Board
MeasureIIlents,
seconds

9-6-56

0.490

22.942

1.083

11-6-56

0.480

22.958

1.053

9-12-56

0.475

23.259

1. 030

9-19-56

0.456

23.437

1. 028

10-18-56

0.452

22.860

1. 025

8-10-56

0.460

23.528

1.021

9-27-56

0.464

23.096

1. 017

8-28-56

0.470

22.642

1.007

10-29-56

0.461

22.491

'1.006

10,:",4-56(a)

0.475

24.061

0.982

8-20-56

0.460

23.756

0.971

10-12-56

0.455

23.429

0.968

12-22-56

0.452

22.616

0.965

11-30-56

0.448

22.725

0.952

11-15-56

0.450

22.583

0.936

Cranford
Collection Date

(a) See comment under "Particle-Size Distribution in Italian Talc".

BAT TEL L E

Protected Document··Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001056

 20
TABLE 16.

CORRELATION OF POROSITY AND LUBRICITYBOARD MEASUREMENTS OF SIZED FRAC TIONS
OF ITALIAN TALC

Tyler Me sh Size

Lubricity-Board
Measurements,
seconds

Porosity Ratio

Unseparated
+200
-200+250
-250+270
-270+325
-325+400
-400

0.990
0.889
0.951
0.980
L 030
1.043
1.099

0.448
0.401
0.426
0.439
0.446
0.442
0.455

Bulk Density

The. bulk density of ground talc may be measured on a Scott Volumeter. The
Cranford talc samples were found to have bulk densities ranging between 22 and 25
pounds per cubic foot. Table 15 lists the bulk densities of the Cranford talc samples.
The bulk-density measurement is not precise enough to accurately compare small
differences but is valuable in establishing a range of acceptability. When sized fractions are tested, the bulk density is seen to have inverse relationships with porosity
and specific surface and to have a direct relationship with average particle size. Thus,
as shown in Table 17, bulk density is inver sely correlative with lubricity as a function
of particle size. The coarser fractions, with poorer slip have the higher bulk density;
the finer fractions having the lower bulk density. The relationship of bulk density,
moisture content, and lubricity is pre sented in a forthcoming report.
TABLE 17.

THE RELATIONSHIP OF BULK DENSITY TO
LUBRICITY-BOARD MEASUREMENTS OF
SIZED FRACTIONS OF ITALIAN TALC

Tyler Me sh Size

Lubrici ty-Board
Measurements,
seconds

Unseparated
+200
-200+250
-250+270
-270+325
-325+400
-400

0.990
0.889
0.951
0.980
L 030
1.043
1.099

8 A T TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

Bulk Density,
lb/cu ft
23.030
34.261
26.645
20.721
19.335
19.139
16.894

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000001057

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

21
CONCLUSIONS

Because this study represents but part of the picture of evaluating acceptable talc
by means of its physical properties, it is not possible to state final conclusions without
qualifications. Several relationships between physical properties, however, have been
established for acceptable Italian talc and the range of their variations have been
measured. A forthcoming report including studies of other physical properties will
add to the picture and will indicate the course to follow for beneficiation of the Italian
talc in order to improve its physical properties.
The physical properties of the Italian talc samples have been measured and the
following range s in value s were obtained:
(1)

Contamination: from less than 1 to more than 3 per cent.

(2)

Crystallographic habit:
10 per cent fibrous.

(3)

Lubricity-board measurement: from 0.936 to 1. 083 seconds.

(4)

The ratio of voids to total volume: from 0.45 to 0.49.

(5)

Theoretical average particle diameter:

(6) Bulk density:

more or less constantly 90 per cent platy,

from 2.45 to 3.30 microns.

from 22.49 to 24.06 lbs/cu. ft.

(7)

Specific surface: from 6612 to 8905 cm 2 /g.

(8)

Moisture content:

hundredths of one per cent.

(9) Particle-size distribution:
+200 mesh
-200+325 me sh
-325+400 mesh (3~ microns)
-38 microns + 30 microns
-30 microns + 15 microns
-15 microns

0.47 to 1. 2Z per cent
3.42to8.30
6.77 to 10.33
40.02 to 65.72
10.23 to 22.08
10.78 to 18.23.

Physical measurements on sized fractions of Italian talc showed the coarser
particle fractions to have lower, less desirable, measurements on the lubricity board
and the finer fractions to have the larger, more de sirable, measurements. The sized
fractions with the preferable lubricity were found to have the higher porosity and
specific surface and the smaller particle size and bulk density (Table 18).
Lubricity-board studies on Italian talc fabricated to particular size distributions
show that the lubricity is controlled by the relatively small amount of comparatively
larger grains in an otherwise finer mixture. Lubricity-board studies also show that the
lubricity of the Italian talc may be improved by the removal of the coar,ser size fractions. This is not a simple matter, however, as it involves the variation in size of the
abrasive particle s.
.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001058

 "'0

0

It
n
It

C-

O
0
n
l:

3

III

::!.

in
l:

.E
III

!:lei
"'0

0

It
n
~

III

0

a.
~

[Jl

>

TABLE 18.

-l

SUMMARY OF PHYSICAL PROPERTIES OF SIZED FRACTIONS OF ITALIAN TALC

-l

m
Tyler Mesh Size

Lubricity-Board
Measurements,
seconds

Average
Diameter,
microns

Unseparated

0.990

2.60

+200

0.889

II

-200+250

>

r
r

m
~

m
~

0

Specific
Surface,
cm 2 /g

Porosity
Ratio

Bulk
Density,
Ih/cu ft

8392

0.448

23.030

7.40

2948

0.401

34. 261

0.951

3.60

-6061

0.426

26.645

-250+270

0.980

2.50

8727

0.439

20.721

-270+325

1.030

2.35

9284

0.446

19.335

-325+400

1.043

2.25

9697

0.442

19.139

-400

1.099

2.10

10390

0.455

16.894

N
N

-

r

Z
l/l
-l

-

-l

C
-l

fTI

c...
Z
c...

~

I~
o
o
o

~-

'"
<D

- - - - - - - - - - - - - - - - - - - --

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

L~

23 and 24
Measurements on flotation products show 'that the removal of the small per cent
of contaminants improves the lubricity of the talc .
The physical propertie s of the Italian talc can be improved, with the least possible
loss of sample, by removing the mineral contaminants from either the coarse fractions
or from the sample as a whole. This appears to be one clear cut course to follow in
improving the Italian talc.

(The original notes on the laboratory work described in this report are in Battelle
Laboratory Record Books No; 12667, pages 1 through 71, and No. 13034, pages 1
through 77. The work was done in the period from November 7, 1956, to September 30,
1957.)

BIBLIOGRAPHY

(1) Sclar, G. B., et al., "A Review and Appraisal of the Literature on Talc Depo sits

of the United States", Battelle Report to Johnson and Johnson (May 11, 1955).
( 2)

Sclar, G. B., eta!., "An Investigation of Selected Talc Deposits of the United
State s", Battelle Report to Johnson and Johnson (February 29, 1956).

( 3)

Smith, W. L., and Snider, R. H., "Investigation of the Salgada and Gasa Nova Talc
Deposits of Brazil", Battelle Report to Johnson and Johnson (May 28, 1957).

( 4)

Smith, W. L., Letter Report to Johnson and Johnson on the Talc Deposits of
Madoc, Canada (July 25, 1957).

WLS:rr

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INS T IT UTE

JNJAZ55_000001060

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

~

APPENDIX A

DESCRIPTION OF LUBRICITY BOARD
AND TECHNIQUE OF OPERATION

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001061

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1

A-I and A-2

APPENDIX A

DESCRIPTION OF LUBRICITY BOARD
AND TECHNIQUE OF OPERATION

The experimental device described as the lubricity board in this report consists
of a wooden plane inclined at 25 degrees, which is lightly, but completely, covered with
talc. The lubricity is determined by measuring the time it takes a steel puck to slide
over two microswitches which actuate an electric timer.
The inclined plane is made of 6-ply birch plywood, 3/4 inch thick, 6 inches wide,
and 6 feet long. The even-grained wood was sanded smooth to prevent the grain from
influencing the descent of the puck. Twenty-five degrees was selected as the inclination
from horizontal after much experimentation which showed it to be the minimum angle at
which a sustained slide could be made on all of the Italian talc samples.
The microswitches are located 6 inches from each end of the slide, in the middle
of the board, making the measured path a length of 5 feet. The microswitches are connected to a double-pole, lIS-volt, Struthers-Berm lock-in relay which actuates a
Standard Electric Time Company electric timer. The steel puck weighs 226 grams, is
3/4 inch by 2-1/2 inches, has rounded edges, and presents a circular sliding surface
of 1-3/4 inches diameter. Such are obtainable from amusement equipment distributors
as a piece used in the game of American Shuffleboard. One flat surface of the puck was
ground smooth and polished for the lubricity experiments.
The talc is applied to the lubricity board from a 9 -ounce Johnsons I Baby Powder
can until a thin even layer is present over the measured path. The puck is manually
released from a dead start from the top of the slide, 6 inches above the first microswitch. For purposes of eliminating errors of freak descents, lubricity is measured
as the average of 50 runs. The board -is newly covered with talc after each 10 runs,
although no difference in lubricity measurements could be accounted for between those
early and late in a series. The puck was washed in warm water and thoroughly dried
between runs.

8 A T TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001062

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

APPENDIX B

PROCEDURE FOR PARTICLE-SIZE ANALYSIS

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001063

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

B-1

APPENDIX B

PROCEDURE FOR PARTICLE-SIZE ANALYSIS

In a previous report to Johnson and Johnson from Battelle(2) a procedure for
sizing Italian talc was outlined. For purposes of the present investigation, the following procedure was developed by D. A. Jacobs of the Battelle staff.
One hundred grams of talc is wet screened at 325 mesh with water. The +325
mesh product is dried and dry screened on a Ro-Tap for 30 minutes producing +200,
200 x 325, and -325 mesh fractions. The -325 mesh fraction of the dry screening is
combined with the -325 mesh product of the wet screening. A suspension of -325 mesh
material is allowed to settle through a la-centimeter column in a 4-liter beaker to
which sodium silicate has been added in the amount of 1 pound per ton, and agitated for
a period of 10 minutes. The sodium. silicate is added to the first 3D-minute settling of
each sample only. At the end of the first 30-minute cycle, the supernatant column of
liquid is siphoned off. This liquid contains the -15 micron fraction. Four cycles are
required to remove the -15 micron fraction entirely. Another series of 4 cycles with
settling times of 5 minutes produces the 15 x 3D-micron fraction, which is siphoned off,
plus sands of 30 microns x 325 mesh. The sands are dried and dry screened at 400
mesh on a Ro-Tap for 30 minutes, producing 325 x lOa mesh and 400 mesh x 30 micron.
fractions. The sizing flowsheet is presented as Figure B-1.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000001064

 I
I

B-2

1oo- grl,a1c s"",ple
Wet screen at 325 mesh with water

t------t------.
t

+325 mesh

-325 mesh

T
Dr

screen (Ro-Tap

+200 ITlesh

I

nu
3:_m_i__ _t_e_s)

200 x 325 ITlesh

,

-325mesh

Add 1 pound of sodium silicate per

'on and agUate for 10 minnr'
Sedimentation in water, 4 cycle s,
cohuun depth 10 centiITleters

,

t

s1'"e for 30 minotes,

4

CYclr

Supernatant,
-15 microns

Sands

+

I

Settle for 5 ITlinutes,

4

cycles

superna,L"
15 x 30 microns

Sands,
30 microns x 325 mesh

~

Dry

+

I:-::~o:s:o

Dry screen at 400 mesh

miour

.

400 me sh x 30 microns
(38 microns)

FIGURE B-l.

STANDARD FLOWSHEET FOR SIZING OF TALC SAMPLES

BATTELLE

Protected Document--Subject to Protective Order

I
I
I

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000001065

  

 

 

 

Plaintiff? 
Exhibit
1 14

*5 

JN 

 

 





 

?vv? 

 

 

 

 

 

 

 

 

 

 

Protected Document--Subject to Protective Order 

I
I
I
I
I
I
I
I
I
I

PROGRESS REPORT

on

FURTHER STUDIES ON THE MEASUREMENT
AND CORRELATION OF THE PHYSICAL
PROPERTIES OF TALC

to
JOHNSON AND JOHNSON

~C~
May 9, 19~8·

I

I
I
I
I
I
I
I
I
I
I
I

by

w.

L. Smith

BATTELLE MEMORIAL INSTITUTE
505 King Avenue
Columbus 1, Ohio

Battelle is not engaged in research [or advertising, sales promotio~, or publicity
purposes, and this report may not be reproduced in full or in part [or such purposes.

Protected Document--Subject to Protective Order

JNJAZ55_000000907

 I
I
I
I
I
I
I
I
I
I
I
I
I

I
I
I
I
I
I
I
I
I

K-3262-2 OK'd before typing by R. D. Macdonald for O. F. Tangel and A. C. Richardson,
cc: W. L,.- Smith (3)
R. D. Macdonald
Files
~. Tan~91 (3)
~
c. Richardson

rJA·ttefle
5

0

5

~

I

N

M

t·
emOridJl in~tt\en1e

G

A

V

E

1\1

U

E

COLUt4BUS

I,

0

H l

0

July 18, 1958

Dr. W. H. Lycan
Director of Research
Johnson and Johnson
New Brunswick, New Jersey
Dear Dr. Lycan:
This letter transmits six copies of our report "Further Studies on the Measurement and Correlation of the Physical Properties of Talc ll •
This report, plus that of October 25, 1957, demonstrates that lubricity may be
improved and abrasiveness lessened by the removal of the mineral contaminants from
talc. A minimum number of physical-property measurements are recommended as
important in the comparison of high-grade tales and in the evaluation of improvement
of physical properties through beneficiation.
Because of urgency on other phases of Battelle's investigation of talc, further
studies on the physical properties are' postponed.
We would be pleased to have your comments on our findings.
Very truly yours,

Wm. L. Smith
Principal Geologist
Minerals Beneficiation Division
WLS/djo
Ene. (6)

DEDICATED

Protected Document--Subject to Protective Order

TO

THE

ADVANCEMENT

OF

SCIENCE

JNJAZ55_000000908

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

TABLE OF CONTENTS

ABSTRACT.

1

INTRODUCTION AND SUMMARY.

1

DISCUSSION OF ABRASIVENESS.

3

THE MEASUREMENT OF ABRASIVENESS.

5
5

Discussion
The Abrasion Machine
The Abrasiveness of Talc Samples

6
12

THE RELATIONSHIP OF ABRASIVENESS TO PARTICLE-SIZE DISTRIBUTION
AND CONT AMINA TION.

14

Discussion
Correlation of Abrasivene ss With Particle-Size Distribution and
Contamination.

14

MEASUREMENT AND CORRELATION OF OTHER PHYSICAL PROPERTIES

18

14

18
20
21
22

Moisture Content
Absorptive Power •
Alkalinity
Acid Solubility •

23

REFLECTANCE AND WHITENESS

23
27

Discussion
The Measurement of Reflectance and Whiteness

THE DUST COMPONENT

28

APPRAISAL OF PHYSICAL-PROPERTY MEASUREMENTS IN THE
EVALUATION OF ORES AND BENEFICIATION PRODUCTS

30

FUTURE WORK

33

REFERENCES .

33
APPENDIX A

DESCRIPTION OF ABRASION MACHINE AND TECHNIQUE OF OPERATION.

A-I

APPENDIX B
DETERMINATION OF EQUIVALENT DOLOMITE CONTENT IN ITALIAN TALC
BY VOLUMETRIC ANALYSIS.
BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

B-1

INSTITUTE

JNJAZ55_000000909

 ~
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

FURTHER STUDIES ON THE MEASUREMENT
AND CORRELATION OF THE PHYSICAL
PROPERTIES OF TALC
by
W. L. Smith

ABSTRACT
To establish the interrelationships of the physical properties of talc and
to be able to visualize the means of their improvement, it has been necessary to
devise means of measuring small differences in properties. To determine the
nature and effects of grit, a chemical analysis for small concentrations of carbonate
minerals and a machine for measuring the relative abrasiveness of talc samples
were contrived, and the measurements were compared with those of other physical
properties. The various other measurements were made on standard laboratory
instruments, using both sized fractions and whole powder. Physical-property
measurements of the talc demonstrated that the samples which produced the least
abrasion were those with the greater platy talc component and those with the least
amount of contaminants. It is concluded that the improvement of slip and the
lessening of abrasiveness may be accomplished by the removal of the mineral
contaminants, but not by the removal of size fractions.
Preliminary work on color and reflectance properties is presented, and
demonstrates a relationship to particle size and, hence, secondarily to other
physical properties.
The report includes an appraisal of the various physical-property measurements employed in the evaluation of improvement of talcs. A minimum ,nu~ber
of measurements are recommended as important in the consideration of benefIciation for improvement and for comparison of natural high-grade ores.

INTRODUCTION AND SUMMARY

This report is a continuation of the studies of the physical properties of talc,
their measurement, and comparison(l )*, previously reported to Johnson and Johnson.
The first Progress Report dealt with petrography, lubricity, and such physical measurements as average diameter, bulk density, porosity, and surface area. It was concluded from the previous study that the acceptable Italian talc fell within a small range
of physical measurements and that the samples with the more desirable slip have the
greater surface area, the smaller average particle diameter, the greater ratio of voids
to total volume, and the lesser bulk density. Lubricity was found to be controlled by
the shape of the relatively small content of larger particles in an otherwise finer mixture. Removal of the coarser contaminants, or preferably of all of the contaminants,
was concluded to be a means of improving the slip of the talc.

'. References appear a r end of report.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

NSTITUTE

JNJAZ55_000000910

 2

The conclusions of the previous progress report have been further substantiated
by the following studies; however, since lubricity is but one of many of the properties
of talc, the previous work represented but part of the picture. Whereas the previous
report dealt primarily with the physical measurement of areas, diameters) weight, and
directly related characteristics, this report deals with refl"ectance, color, moisture
content, abrasiveness, alkalinity, and acid solubility - properties related to lubricity
only through their common correlation through surface area, size, and component
contamination.
As in the previous report, both particle size and shape and the amount and nature
of the contaminants are investigated to determine the contributin& factors to physical
measurement variations. It is recommended as a result of these studies that the
following measurements are sufficient to determine satisfactory talc within the range
in composition of Italian talc. These are the determination of the mineralogy and
particle size distribution, volumetric analysis of the carbonate component, and the
measurements of the bulk density, moisture content, reflectance, and whiteness. This
should also serve as a basis for the determination of acceptability in other tales and
beneficiated products, taking into consideration the differences in size distribution,
crystallographic habit, and mineral contamination.
Because size distribution, as it is reported, is dependent upon the analytical
procedure, the measurement of physical properties is made only on closely sized .
fractions in the coarser size ranges. Division into size fractions of the finer size portion of a powder which is composed of platelets is inaccurate by standard laboratory
procedures and requires detailed petrographic examination of the products. A method
proposed by R. W. Schatz(2) which reports size distribution on the basis of theoretical
spheres rather than on the basis of actual petrographic measurement serves as an excellent comparative measurement for powders with similar mineralogical and crystallographic composition. Of necessity, these figures show little resemblance to the
measurements of the greater dimensions of the talc platelets. Consistent with the
previous progress report, the particle-size-distribution data here presented are based
on screened size fractions and the sizes given are those measured on a petrographic
microscope.
As in the study of lubricity, in order to determine the improvement of specific
physical properties, objective tests had to be devised to measure the small differences
between acceptable talc and talc of lower quality. Until now, acid solubility was
determined gravimetrically on the various talc samples. The small differences in per
cent composition and per cent incidence of the carbonate component of Italian talc, and
,its close relationship to both abrasiveness and lubricity, required the development of
an analysis for equivalent dolomite in low concentrations (Appendix B).
Ab~asiveness has previously been measured subjectively, similar to lubricity.
However, because subjective measurements are not correlative, and because small,
often significant differences cannot be so measured, an abrasion machine was built.
This device, does not give an absolute value to abrasion, however, it provides reproducible figures which are of relative value and which are correlative with measurements
of other physical properties.

Except where otherwise noted, the measurements presented in this report ,were
made on the same samp,~es of "EGT Extra 00000" talc, obtained from the Cranford,
New Jersey, plant, which were used in the' work reported in the previous Progress
Report on physical properties.
BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

11
I
'I
I
I
I,

I

-I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000911

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

3

DISCUSSION OF ABRASIVENESS

A highly undesirable property of a talcum powder is abrasiveness or grittiness_
Grit is undesirable in that it may scratch or otherwise irritate the skin, and even very
small amounts of grit may quickly be noticed subjectively.
Grit consists of that portion of ground talc which is angular, or oversize, particularly in thickness. Grit includes both oversize and nonplaty talc particles as well
as mineral contaminants. It occurs as aggregates of talc and contaminants, as acicular
and fibrous particles of talc and amphibole, as shards and granules of amphibole or
carbonate, and as prismatic grains of titanite, rutile, zircon, apatite, and other
accessory minerals.
Where the grit is other than oversize talc, it has hardness and angularity sufficient to scratch. Talc which is oversize in its greater dimensions is rare in the.
samples studied. It is the product of incomplete grinding and may easily be removed
on a ISO-mesh screen. Talc which is oversize in thickness is of the nonplaty variety,
the result of the incomplete alteration of pre-existing minerals or the formation of
pseudomorphs after more equidimensional species. Such particles serve less as
abrasives than as deterrents to proper slip. The 8 to 10 per cent of nonplaty talc in the
Italian material is presumed to be derived from tremolite or enstatite. This mechanism
is discussed in the reports on the Brazilian(3) and Canadian(4) talc deposits.
Whereas friction, as expressed in the sense of the translation movements of talc
platelets over one another, produces the desirable property of slip, such friction is not
a disruption of the free lamellar movement of the component particles of the powder nor
a disruption of the free movement of the surfaces in contact •. When, however, a lubricant fails to mask irregularities in the contacting surfaces or introduces asperities
of its own, then point friction or plowing is initiated. Point friction and plowing are the
sources of irritation or grittiness. Grit permits wear between the contacting surfaces
by abrasion, either in the plowing or scratching mechanism of oversize and angular
particles, or by the dis ruption of lamellar movement of the platelets which leaves areas
unlubricated or introduces a damming-up and rolling of particles.
Although lubricity and abrasiveness may seem to be relative, or the presence of
one may seem to preclude the presence of the other, no direct correlation should be
expected between the two properties inasmuch as both are the functions of several
variable factors. A decrease in grit, however, is certain to improve the lubricity of
whole powders where particle size is not a controlling factor.
Idealized talc particles are rounded platelets which may be thought of as es sentially two dimensional, the thickness being about 1/8 to 1/15 of the greater dimension,
depending upon the crystalline nature of the mineral and the degree of subdivision
attained. In the better grades of talc the greater dimensions of a platelet are nearly
equal.
The Italian No. I talc contains from less than I per cent to about 3 per cent of
contaminants. The contamination is natural and consists mostly of carbonate with
minor amphibole and rare accessory minerals. The carbonate component has been
identified petrographically as primarily dolomite (CaO· MgO· 2C0 ) plus a minor amount
2
of probable m.agnesite (MgO' CO 2 ), No calcite (CaO· CO 2 ) was identified. The amphibole
component has been established to be the variety tremolite (2CaO- 5MgO' 8SiOZ' HZO).
BAT T E L L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000912

 4
Table 1 based on Table 2 of the previous Progress Repord l ) lists the incidence of contaminants in the Cranford samples. Table 2 shows the distribution of the major contaminants in the different size fractions of Italian talc.
TABLE 1.

PREVIOUSLY REPORTED{l) PER CENT CONTAMINATION IN
TALC SAMPLED AT CRANFORD, NEW JERSEY

Incidence of Contaminants (a),
per cent

.<1

Date Sampled
9-6-56, 9-12-56, 9-19-56
8-10-56, 9-27-56, 10-18-56, 11-6-56
10-4-56, 10-29-56
8-20-56, 8-28-56, 11-15-56, 12-22-56
10-12-56, 11-30-56

1
1-2
2
2-3

(a) Determined petrographically.

TABLE 2.

THE DISTRIBUTION OF THE MINERAL CONTAMINANTS IN
THE DIFFERENT PARTICLE-SIZE FRACTIONS OF
ITALIAN TALC
Incidence of Contaminants{a), per cent
Dolomite
Tremolite

Tyler Mesh
Size

Total

Unseparated

:2

<2

+200
-200+250
-250+270
-270+ 325
-325+400
-400

<1
1
1-2
2
2
>2

< 1
1
1-2
2

Trace

2

>2

O-trace
O-trace
Trace
Trace
Trace 1
<1

(n) Incidence determined petrographica lly.

Grit is present in all size fractions, being somewhat more abundant in the fines.
In the coarser fractions the mineral contaminants and the talc particles which are oversize in thickness are most readily sensed subjectively. The presence of grit in the fines
is largely masked to the senses by the presence of larger platelets. In this regard no
solution to, abrasiveness lies in the removal of entire coarse size fractions, inasmuch
as the grit in the then remaining coarser fractions would be as readily noticeable subjectively and more abundant percentagewise. To remove the abrasive particles, it is
necessary to remove both nonplaty talc and the mineral contaminants from the whole
powder by such beneficiation methods as flotation(5) and classification by cycloning(6),
the initial studies on which have been reported or are in preparation (Table 3).

B A T TEL L E:

Protected Document--Subject to Protective Order

'M E M OR , A L

~

INS T I. T UTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000913

 I
I
I
I
I
I
I
I
I
I
I
I
I
I

I
I
I
I
I
I
I
I

5

TABLE 3.

Talc
Sample

THE EFFECT ON LUBRICITY OF THE, REMOVAL OF MINERAL
CONTAMINANTS AND NONPLATY TALC FROM ITALIAN
TALC SAMP LES
Incidence of Indicated Particle Type(a), per cent
Platy
Nonplaty
Talc
Talc
Dolomite
Tremolite

Italian No. 1
Feed
Float

88-90
95

Italian No. 2
Feed
Float

90
98

8-9

<2

4

Trace

Lubricity-Board
Measurement, sec

<1

0.990

Trace

1. 046

5

3

2

0.926

<1

Trace

Trace

1. 051

(a) Mlneralogicallnc1dence determined petrographically.

It is important to emphasize the difference between the incidence or frequency of
contaminants and their per cent of total composition. The per cent incidence is determined petrographically by grain count. It is a two-dimensional measurement approximating area and does not consider the thickness of the particles observed. The incidence of a min~:I:al or crystal type is of primary importance inasmuch as a powder
consists of a mixture of discrete grains, each with its particular size, shape, and other
physical properties.' In eonsidering the behavior of a powder as a lubricant, we are
dealing with the ;mechanical interactions of individual grains in lamellar movement and
thus are concerned with the frequency of types of grains, not with their per cent of total
composition. That is, for example, in considering lubricity or abrasiveness we must
deal with the incidence of individual particles of dolomite rather than with the total
volume oxt weight per cent of the sample which is dolomite, except when dealing with
closely sized samples. Conversely, when considering acid solubility, moisture content,
and the analysis and evaluation of beneficiation products, we, of neces sity, deal with
total components; not the incidence of particles. While small differences in per cent
incidence of contaminants in a powder may influence the physical properties of mechanical movement, in no case described here is the per cent incidence different from the
weight per cent or the chemically analyzed component by more than 1 per cent of the
whole sample.

THE MEASUREMENT OF ABRASIVENESS

Discussion

A standard method of measuring the abrasiveness of high-quality talc has not been
devised previously. Abrasiveness or grit has 'been measured subjectively by testing
samples between the fingers or teeth. As in the case of lubricity, the final analysis of
acceptability in regard to abrasiveness is subjective: consumer reaction. Objective
tests are not designed to replace the subje~tive tests; however, to be able to determi~e
BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000914

 6
improvement in beneficiation procedures and to determine the correlative relationships
of the physical properties of talc, it is necessary to be able to measure small differences in the phy~ical properties and to be able to compare them to other quantitative
'measurements. Knowiedge of these interrelationships s~rv:es as the basis for interpretation of improvement in quality and thus serves to make it po'ssible to visualize
methods of beneficiation.
Since the 'subjective tests are of little help in measuring small differences in one
of the many physical properties encountered, and since such tests have no basis for
correlation, a machine was built to measure objectively, or test for, abrasiveness,
apart from other physical properties.

The Abrasion Machine

Because it was necessary to measure small differences 'in the abrasiveness of
talc, a machine was built to test the wear effect of small concentrations of grit on
standard material. The machine was built of a 1/20-hp l725-rpm electric motor'
mounted vertically and fitted with a 5-inch lap covered by a Buehler Microcloth held in
place by a rubber belt. The lap portion of the machine is set into a steel bowl and
covered with a plastic lid. Mounted on a ringstand over the lap a SOO-ml open separatory funnel with stopcock is connected by a rubber tube with an adjustable pinch clamp
to a feed spout. The separatory funnel contains the sample of talc to be tested in a
slurry of 3 grams of talc to 350 ml of water. The feed spout and a cylindrical pellet
holder are mounted in a removable crossbar over the lap. Accessibility to these parts
is afforded through a hole in the plastic cover. $tandard 1 /2-inch-diameter pellets are
held in the sample holder by a 16. I-gram weight to prevent their skipping or floating on
the lap. A 1000-ml beaker mounted under a drain in the steel bowl catches the tested
,slurry. Figure 1 shows the over -all apparatus. Figure Z shows the detail of the feed
and abrasion mechanism. A detailed description of the abrasion machine and the
technique of its operation are found in Appendix A.
In order to measure the abrasiveness of the talc in the slurry a test had to be
designed where the object abraded wOlild have a great enough loss to be measured
physically. Since the abrasiveness to be measured was that of a powder containing
generally from only 1 to 3 per cent of abrasive gangue particles, the material to be
abraded had to have a hardness greater than that of the talc, less than that of the grit,
and also had to be coherent and homogeneous. After testing a large number of materials
it was decided to perform the bulk of the tests on pellets made of minus 400-mesh
Italian talc pressed under'50, OOO-psi pressure. The pellets average 5.20 grams and
have dimensions of 1/2 by 7/10 inch. The pellets have a hardness greater than that of
the raw talc and less than that of the contaminants (Table 4). Carbonate pellets were
made to test specifically for the rarer, harder components, in a similar manner, but
using alcohol instead of water in the slurry.

,

BAT TEL L E

Protected Document--Subject to Protective Order

~

,E M 0

!

R, I AL

tNSTITUTE

I
I
I
I

I'
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000915

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

7 and 8

N45893
FIG URE 1.

THE ABRASION MACHINE SHOWING RESERVOIR CONTAINING
SAMPLE IN SLURRY TO BE TESTED FOR ITS ABRASIVENESS

BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000916

 a_
It
n
It

c-

o
o

- - - - - ------- - - - - - - - - -

n

l:

3

III
:::l

~l:

.E
III

!:lei
"'0

~

n

~

tD

o

»

III

a.
~

...

~i!;:

ITl

r
r

.

ITl

__-II-Ii

(.~.~

,=,,~,.,,=,"~ ~f~J~_,t=·""",:
~~~',-','<~~"<1
. . . . ~~v--.' .

~

ITl
~

o

~.

1

:u

','

»
r

,

~ ~::;/~ ~ I

~ J~

n
'"

~

If)
)

n
,

[

-

I

o

Z

.
.
III

C
-4

ITl

I

----~

FIGURE 2.
c...
Z
c...

I:~
o
o
o
o
o

~

.....

N45892
DETAIL VIEW OF ABRASION-MACHINE LAP SHOWING FEED SPOUT, CYLINDER IN WHICH PELLETS
ARE HELD ON THE LAP, AND STANDARD PELLETS OF PRESSED TALC

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

11
TABLE 4.

RELA TIVE HARDNESS OF THE TEST PELLETS AND THE
GRIT PRESENT IN ITALIAN TALC

Mineral

Moh Hardness

Talc

I

Pressed-talc test pellet

± 2 (scratches talc)

Magnesite
Dolomite

3. 5 (scratches talc pellet)
4
>4 (scratches dolomite)

Pre s sed-carbonate te st pellet
Apatite
Titanite
Tremolite
Rutile
Zircon

5 (scratches carbonate pellet)
5
6
6

7. 5

The abrasion-machine operation is timed electrically. The pellets are measured on a micrometer caliper before the test and afterward, after drying. The abrasion measurement is reported in decimal fractions of an inch per second. Although
there are limitations to the use of a micrometer, the samples which were compared
demonstrated differences in measurement large enough to be significant. Measurements based on weight were found to be entirely unsatisfactory inasmuch as some weight
loss was due to spalling and abrasion of the pellet by the walls of the sample tube on
portions other than that exposed to the lap and abrasive. This indicated losses which
were no indication of the degree of loss due to action on the tested surface alone.
The abrasion machine subjects the standard pellet to abrasion by the sample of
talc being studied, at a high rate of speed. It has been calculated that the pellet
receives wear equivalent to being rubbed over more than 1800 ft/min of surface of the
talc being tested. As expected, the abrasion machine demonstrates that the slurry
samples with the greater incidence of mineral contaminants produce the greater amount
of abrasion on the pres sed pellets. It is also shown that those samples with primarily
platy habit are less abrasive than those containing effective amounts of nonplaty talc.
More precise abrasion machines could be built; however, the device used is satisfactory for the purpose of comparing samples within the. range of those tested and is an
adequate means of obtaining comparable measurements of the effect of grit. Typical
figures obtained by the abrasion experiments are shown in Table 5.

BAT TEL L

Protected Document--Subject to Protective Order

E

MEMORIAL

INSTITUTE

JNJAZ55_000000918

 12
TABLE 5.

TYPICAL FIGURES OBTAINED BY ABRASI()N TESTS ON
FOUR SAMPLES OF ITALIAN TALC

Test

Incidence of
ContalUinants(a),
per cent

1
2
3
4

2
2
2
2

Pellet
MeasurelUents, in.
Before
After
O. 6453
O. 6435
O. 6184
O. 6442

Diffe renee in
Measurements,
in.

Time,
sec

Abrasion(b),
10- 3 in./sec

0.0887
O. 0952
O. 0810
0.0743

38
41
36
32

2. 33
2. 32
2.25
2. 32

o. 5566
0.5483
0.5374
0.5699

(a) Determined petrographically.
(b) Significant figure: 2.3 x 10- 3 in. /sec.

The Abrasiveness of Talc SalUples

Fifteen salUples of talc collected at the Cranford plant of Johnson and Johnson,
the same samples used in the previously reported lubricity experiments(l), were tested
on the abrasion lUachine. The results of the lUeasurelUents are shown in Table 6. The
lUeasurements show a range of from 1.62 to 2.69 ,x 10- 3 in. I sec wear on the standard
pellets. These figures are generally correlative with the incidence of contalUinants, as
deterlUined petrographically, as reported in the Progress Report dealing with lubricity(l). As it will be shown further in the report, this relationship only holds in whole
unseparated powder where particle size is not a fundamental controlling factor. The
correlation of lubricity-board'lUeasurelUents with contalUination reported in Table 2 of
the previous Progress Report(l) shows a silUilar general relationship between contalUination and lubricity. Where the lUbricity experilUents concluded that the samples containing the greater alUount of contaminants demonstrated the poorer lubricity, the'
abrasion-lUachine experiments show that the, samples with the greater contamination
produce the greater alUount of abrasion. Thus, when dealing with whole, UIlscreened
powders, the relUoval of grit should also serve to ilUprove lubricity.
Although the relUoval of grit ilUproves the lubricity of whole powders, the relationship of lubricity to abrasiveness cannot be considered to be lUathelUatically inverse.
The properties which control abrasiveness are the size and shape of the contaminants
and their incidence in the coarser fractions; whereas, the properties which cont'rol
lubricity are the over-all size distribution, those previously described properties
directly related to surface area, plus the incidence of the coarser components. When
lUeasuring screened fractions of powders both abrasiveness ,and lubricity are influenced
by the specific particle size, and abrasiveness will be directly related'to the grit COlUponent, whereas in whole powders the finer abrasive particles will be in part lUasked by
the coarser platelets.
To test further the effect of contaminants upon abrasiveness, the contaminants
were removed from a sample of talc by froth flotation and the products were tested on
the abrasion machine. The same samples had previously been tested for IUbricity(l).
The test results, which are noticeable subjectively, are reported 'in Table 7.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000919

 I
I
I
I
I
I
I
I
I
I
I

13
TABLE 6.

Date Sampled
9-12-S6
8 -1 0-S6
9-19-S6
9 .. 6-56
10-18-56
9-27 -56
10-4-S6

RELATION OF PURITY OF SAMPLE TO ABRASIVENESS
AND LUBRICITY IN WHOLE POWDER

Abrasiveness,
10- 3 in. {sec

Incidence of
Contaminants (a),
per cent

Lubricity-Board
Measurement,
sec

1, 62

<1

1, 030

1. 70

1

1. 021

1. 84
1. 87

<1
<1

1. 028
1. 083

1.88

1
1

2.32
2.32

2
1-2
2-3

1, 025
1, 017
0.982
0.971
1,'007
1. 053
0.936
1. 006
0.952

2.59
2.69

2-3
2

0.968
O. 965

1, 90
1, 90

8-20-56

1-2
2

1. 91
1. 97

8-28-S6
11-6-S6

2
1

2. 15

II-IS-56
10-29-56
11-30-56
10-12-s6
12-22-56

2.30

(a) Previously reported( 1). determined petrographically.

I

-I
J
I
I
I
I
I
I
I
I

TABLE 7.

ABRASION AND LUBRICITY MEASUREMENTS ON FLOTATION
PRODUC TS OF ITALlAN NO. 1 TALC
Abrasiveness,
10- 3 in. {sec

Product

Lubricity-Board
Measurement, sec

Starting sample

2. 14

O. 990

Float producd a )

1. 50 (superior)(C)

1.046 (superior)

Nonfloat product{b)

3.03 (inferior)

0.873 (inferior)

(a) Essentially pure talc. representing 90 per cent of starting sample.
(b) 85 per cent talc. 15 per cent contaminants, representing 10 per cent of starting sample.
(c) Less abrasive float products have been made from Italian No.2 talc.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000920

 14
The float products clearly demonstrate superiority over the starting sample in
regard to both abrasiveness' and lubricity. The deleterious effect of contaminants is
shown by the inferior measurements derived from testing the reject product of the
flotation process.

THE RELATIONSHIP OF ABRASIVENESS TO PARTICLE-SIZE
DISTRIB UTION AND CONTAMINATION

Discussion

The abrasivenes s of the talc studied is determined by its component grit .. When
dealing with the mechanics of a powder in lamellar motion, we are dealing with the interrelationship of individual particles, and thus are concerned with the per cent
incidence and particular sizes and shapes of individual~. The distribution of the contaminants and nonplaty talc in the different size fractions is thus a primary consideration. The coarse contaminants are those which scratch and are quickly noticed subjectively. The finer contaminants clog the lamellar movement of the talc platelets and
initiate rolling of the powder, introducing aggregate asperities.
In the previous reports to Johnson and Johnson(l, 2,7) the problem of particlesize distribution has been thoroughly discussed. It does not seem requisite here but to
re-emphasize the importance of establishing the particle-size distribution of a powder
when studying its physical properties or the means of their improvement. Since the
size of the abrasive particles, as well as their incidental abundance, contributes to
abrasiveness, it was necessary to determine the size distribution of the contaminants.

Correlation C?f Abrasiveness With Particle-Size
Distribution and Contamination

Portions of the same samples which were used to test for lubricity and other
physical properties were used in the following experiments on abrasiveness. Size
fractions were made of Italian talc and were measured for their abrasiveness on the
abrasion machine. Results of the experiments show that the finer particle-size fractionsare more abrasive than the coarse. This is in agreement with the incidence of
grit determined petrographically. Chemical·analyses for equivalent dolomite on these
samples are also in agreement. These analyses appear in the section of this report
dealing with acid solubility.
Table 8 shows the results of abrasion tests of size fractions on standard talc
pellets. When comparing size' fractions a parallel relationship exists between lubri'city
and abrasiveness, as a function of particle size. Thus, were only the fines, the most
lubricous fraction, used for baby powder, this fraction would also be the most abrasive.
In order to retain the more lubricous particles yet remove the more abrasive, it is
necessary to remove the contaminants only. Both lubricity and grittiness cannot be
improved by the removal of particle ~s ize fractions.

B A. T TEL L .E

Protected Document--Subject to Protective Order

M.E M 0

~d

I A ,L

INSTITUTE

I
I
I
I
I'
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000921

 I
I
I
I
I
I
I
I
I
I
I
I
I
I

I
I
I
I
I
I
I
I

15
TABLE 8.

RELATIONSHIP OF ABRASIVENESS AND LUBRICITY TO PARTICLE SIZE
AND CONTAMINATION IN SIZE FRACTIONS OF ITALIAN NO.1 TALC

Lubricity-Board
Measurement, sec

Incidence of Contaminants(a),
per cent
Total
Dolomite
Tremolite

Tyler Mesh Size

Ab r a s i ve ne s s,
10- 3 in. I sec

Unseparated

2. 14

0.990

±2

<2

Trace

+200
-200+250
-250+270
-270+325
-325+400
-400

1. 30

0.889
0.951
0.980
1.030
1. 043
1. 099

<1
1
1-2
2
Z
>Z

<1
1
1-2
2
Z
>2

O-trace
O-trace
Trace
Trace
Trace-l
< 1

1. 59
1.72
2. 00
Z.33
Z.48

(a) Determined petrographically.

In order to show which effects are primary, and which are secondarily related
because sized materials are analyzed, tests were made on the plus ZOO-mesh fraction
of the talc as received from Italy, on the minus 400-mesh fraction, and on the plus
200-mesh material after it was crushed to pass a 400-mesh screen. The natural plus
200-mesh material had but a trace of contaminants compared with the more than 2 per
cent present in the natural minus 400-mesh fraction. Abrasiveness and lubricity tests
(Table 9) show clearly that the abrasiveness is controlled primarily by the contamination and only secondarily by the size fraction analyzed. It also shows that the lubricity
is controlled primarily by the particle-size fraction tested, and only secondarily by the
contamination present in the specific size range. This experiment serves as the basis
of interpretation for relating the physical properties of sized material, and also establishes the controlling factors behind lUbricity and abrasiveness in whole powders.

TABLE 9.

COMPARISON OF LUBRICITY AND ABRASIVENESS TO GRAIN
SIZE AND CONT AMINA TION

Abrasivene ss,
10- 3 in. I sec

Sample
Natural +ZOO-mesh fraction
Natural -400-Ine sh fraction
+200-mesh fraction ground
to -400 mesh

Lubricity-Board
Measurement, sec

1. 30
2.48
1. 12 (b)

0.889
1. 099
1. 10S(C)

Incidence of
ContaIninants(a),
per cent
Trace
>2
Trace

(a) Determined petrographically.
(b) The abrasiveness of this sample is only slightly less than that produced by the same sample prior to grinding, much less than
the natural minus 400-mesh fraction which contains more grit.
(c) The lubricity of this sample is greatly improved by regrinding, but it is essentially like the natural minus 400-mesh sample
despite the difference in the grit present,

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000922

 "'0

0

It
n
It

C-

O
0
n
l:

3

III

::!.

in
l:

.E
III

!:lei
"'0

0

It
!:l-

TABLE 10.

<'

III

0

0~

ABRASIVENESS AND LUBRICITY MEASUREMENTS OF IT ALIA..~ TALC SAMPLES FROM WHICH
SPECIFIC PARTICLE-SIZE FRACTIONS HAVE BEEN REMOVED

m
X Represents Fractions Removed From Whole Powder
U Represents Fractions Tested

>
~
~

IT!

r
r

Tyler Mesh Size

Incidence of
Contaminants (a),
per cent

+200
-200+250
-250+270
-270+325
-325+400
-400

<1
1
1-2
2
2
>2

IT!

3:
IT!

3:
0

:u

-)r

Z

III

Lubricity -Board(b)
Measurement of
Size Fractions., sec
0.889
0.951
0.980
1.030
1. 043·
1. 099

Abrasiveness,
10-3 in . /sec

1. 30
1. 59

1.72
2.00
2.33
2.48'

Test

Test 2

Whole Powder

Test 3

Test 4

'. U

U
U

U
U
U
U
U
U

X
X
U

X
X
X

U
U

U

U
U
U

0.990

1.038

1.068

U
U
X
X
X

Lubricity-Board
Measurement, sec

0.945

Abrasiveness, 10- 3
in. / sec

1. 88

u·
U
U
X

0.963

--

- - .-IncreaSe in slip and abrasiveness

1. 88

....0"-

2.14

2.27

2.34

0.0

2.0

3.0

~

-

~

.C

Approximate Weight
Per Cent of
Fractions Removed

97.0

82.0

~

IT!

(a) Detennined petrographically;
(b) Based on Table 10 of previous report(l).

c...
Z
c...

~

I~
o
o
o

~-

IV

'"

---------------------

 I
I
I

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

17
In the previous study of lubricity(l), measurements were made on powders from
which different size fractions had been removed, and on mixtures of specific size
fractions. The experiment demonstrated that the over -all lubricity was influenced
primarily by the coarser particles. Similar experiments on abrasiveness have been
made and show that it is not possible to increase the lubricity by removing total size
fractions without increasing the abrasiveness (Table 10).
Because talc contains tremolite and rare accessory minerals as well as carbonate
as abrasive components, a test was devised to measure the abrasiveness of the harder
contaminants. Carbonate pellets were made by fusing a three-to-one mixture of sodium
carbonate and sodium borate into a melt. The fusion product was crushed and pressed
into pellets under 15, 000 psi. The resulting pellets were harder than the talc or the
carbonate contaminants but softer than the tremolite and accessory minerals. The pellets were slowly soluble in water; therefore, alcohol was used as the fluid in the test
slurries. The test results are not necessarily correlative with measurements made on
talc pellets, but demonstrate the distribution of the harder contaminants (Table 11).
All other abrasive data contained in this report have been determined on pressed talc
pellets.

TABLE 11.

Tyler Mesh
Size

THE DISTRIBUTION OF THE CONTAMINANTS HARDER THAN
CARBONATE IN ITALIAN TALC AS SHOWN BY ABRASION
TESTS MADE ON CARBONA TE PELLETS

Abrasiveness
(on Talc Pellets),
10- 3 in. I sec

Abrasiveness
(on Carbonate Pellets),
10- 3 in. I sec

Incidence of Contaminants(a),
pe r cent
Total Dolomite
Tremolite

Unseparated

2. 14

0.9

±2

<2

Trace

+200
-200+250
-250+270
-270+325
-325+400
-400

1. 30
1. 59

O. 5
O. 7

<1

1.72
2.00
2.33
2.48

0.6

<1
1
1-2

O-trace
O-trace
Trace
Trace
Trace-l
<1

0.6
0.8
0.9

1
1-2
2
2

>2

2
2

>2

(a) Determined petrographically.

To demonstrate the effect of the rarer contaminants on the abrasion of pressed
talc pellets, a series of sized fractions of Italian talc were leached free of the carbonate
components and measured on pressed talc pellets. Table 12 shows the degree of abrasiveness produced by the leached samples.

8

A T TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000924

 18
TABLE 12.

ABRASION TESTS USING LEACHED AND UNLEACHED SIZED
FRACTIONS, DEMONSTRATING THE EFFECT OF DOLOMITE
ON ABRASIVENESS
, Abrasiveness, 10- 3 in. / sec
Unleached Powder
Leached Powder

Tyler Mesh Size
U~se:parated

+200
. -200+250
~250+270

-270+325
-325+400
-400

2. 14

1. 50

1. 30
1. 59
1. 72
2.00
2. 33
2.48

1. 60
1.72
1. 71
1. 90
1. 75

1. 34

An additional test, not relative to the beneficiation of Italian talc, but designed as
an experiment to serve as a basis for study of lower grade talc, was made on a series
of mixtures of Italian talc and minus 400-mesh calcium carbonate (Table 13). Although
the data are not to be considered correlative withthose of Italian talc, they demonstrate
clearly the increasing effect of contamination on abrasion.

TABLE 13.

Per Cent
Italian Ta1c(a)

ABRASIVENESS OF MIXTURES OF ITALIAN TALC
AND CALCIUM CARBONATE

Per Cent CaO· C02
(-400 Mesh)

o

100
90
50
10

2. 14
2.89
5.09
8. 17
13. 65

10
50
90
100

o

Abrasiveness,
10- 3 in. /sec

Diffe renee in
Abrasivenes s,
10- 3 in. /sec
0.75
2.20
3.08
5.48

(a) Contains about 2 per cent native carbonate.

MEASUREMENT AND CORRELATION OF OTHER PHYSICAL PROPER TIES

Moisture Content

The previous Progress Report(l) introduced the problem of moisture content in
talc, sugge sting that the fine-grain- size fractions should adsorb more moisture on its
greater surface area per unit of weight. Table 14 shows the moisture content of size
fractions of Italian talc, demonstrating an increase in moisture content with decreasing
particle size. Because of the relationship of particle size to other physical properties,
BAT TEL L E

Protected Document--Subject to Protective Order

M E M O.R I A L

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000925

 I
I
I
I
I
I
I
I
I
I

19
the moisture content of the s~zed fractions was found to be apparently correlative with
a number of other measurements, indicating coincidental relationships which would not
hold true in unsized sample s.

TABLE 14.

RELATIONSHIP OF MOISTURE
CONTENT TO PARTICLE SIZE
IN IT ALlAN TALC

Tyler Mesh Size

Per Cent Moisture{a)

Unseparated

0.05

+200
-200+250
-250+270
-270+325
-325+400
-400

0.01
0.03
0.03
0.04
0.05
0.06

(a) Moisture content determined by method outlined in Johnson and lohnson's
Raw Materials Specifications sheet.

I
I
I
I

The lubricity of talc is related to its moisture content insofar as the moisture content of the finer fractions is higher. In lower grade talc the moisture content was found
to be much higher. Six domestic tales, fabricated to a particle- size distribution similar
to that of the Italian talc, showed from 0.08 to O. 20 per cent moisture on analysis. The
higher moisture content of some inferior tales requires that moisture be determined on
sample s prior to te sting for lubricity. Moisture tends to make talc pasty, producing a
false indication of superior lubricity on the lubricity board. To demonstrate that the
0.05 per cent moisture content of Italian talc did not affect the lubricity, tests were run
on talc from which the moisture had been driven off. A nonreproducible difference of
only 0.006 second was recorded, and is not considered to be a significant figure.

I
I
I
I
I
I
I
I

To further test the correlation of moisture content and particle size, the talc samples collected at Cranford were analyzed and the data compared with the percentage of
fines in the sample. Table 15 compares the moisture content and the percentage of the
powder finer than 400 mesh, as compiled from Table 6 of the previous Progress
Report(l). The samples with the greater component of fines were found to contain the
greater moisture content. No absolute interpretation should be given this relationship.
The figure s are all very close and their similarity is of greater importance than the
correlation. However, there is a theoretical basis for the variance, and the data are
presented for whatever they may be worth in the light of future studies. The correlation
seems more than coincidental. Moisture content shows no other correlation in whole
powders.
Inasmuch as the lubricity-board studies(l) showed that the lubricity variations
depended upon the coarser fractions, the small difference in the fine component as re1ated to moisture content should have no expression in lubricity.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000926

 20

TABLE 15.

THE RELATIONSHIP OF MOISTURE CONTENT TO PER CENT OF MINUS
400-MESH PARTICLES IN WHOLE SAMPLES OF TALC COLLECTED AT
CRANFORD

Per Cent Fines{a)
(-400 Mesh, Tyler)

Per Cent Moisture

Date
Sampled

88.28
88. 16
87.26
86.61
86.09
85.29
83.91
83.40
83.04
82.93
82.57
82. 28

0.06
0.07
0.06
0.06
0.05
0'; 05
0.05
0.05
0.05
0.03
0.03
0.04

8-28-56
9-27-56
9 -6 -56
11-6-56
9-19-56
8-20-56
10-18-56
12-22-56
10-4-56
9-12-56
11-30-56
11-15-56

(a) Repeated from previous report(l).

Johnson and Johnson's Raw Materials Specifications sheet states O. 15 per cent
moisture content to be the tolerable upper limit. The Crarlford samples, as well as the
size fractions, contain considerably less moisture. This indicates that any beneficiation
which would change the size distribution, hence the moisture content, would not produce
a product of unsatisfactory moisture content.
Inasmuch as a pasty consistency in talcum powder would be undesirable, samples
which by exposure or otherwise have taken on exce ss moisture must be restored by
proper drying. One problem arising from flotation experiments was the .tendency for the
products to agglomerate after drying. It is possible that this moisture can be removed
by spray drying. The problem of proper drying is to be considered further in the beneficiation phase of the program.

A physical property closely related to moisture content and particle size is the
absorptive power of talc. The hygroscopic property is highly important, inasmuch as it
is a factor in deodorizing, coloring, in the carryirig of perfume or other agents, and in
. the retention of moisture. Because this subject is only partly understood at this time,
it will not be reported on here. The property is not immediately pertinent to the other
me chanical and physical relationship s in this report except through moisture content.

Protected Document--Subject to Protective Order

M E M.O R I A L

I
I
I
I
I
I
I
I

I

Absorptive Power

BAT TEL L E

I
I
I
I
I
I

INSTITUTE

I
I
I
I
I
I
I

JNJAZ55_000000927

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

21
Alkalinity

The 15 samples of Italian talc collected at Cranford were measured for pH on a
Beckman pH meter standardized at neutrality and checked with Beckman buffer solutions
of pH 7 and pH 10. The figures are accurate to about O. 1. The samples were prepared
by mixing 5 grams of talc with 10 cc of distilled water (pH 6.9). The solutions were
agitated and permitted to stand for 2 hours prior to their measurement. The pH of the
Cranford samples ranges from 9.0 to 9.3 (Table 16).

TABLE 16.

pH OF CRANFORD SAMPLES

Date of
Sample

pH

8-10-56
8-20-56
8-28-56
9-6-56
9-12-56
9-19-56
9-27-56
10-4-56
10-12-56
10-18-56
10-29-56
11-6-56
11-15-56
11-30-56
12-22-56

9. 1
9.0
9. 3
9. 1
9. 1
9.2
9.2
9.2
9.2
9. 0
9. 1
9. 3
9.0
9. 1
9. 2

To see if there was any relationship of pH to other physical properties, a Cranford
sample with a pH of 9. 2 was sized and the fractions were measured (Table 17). The
size fractions each measured 9. 2, which showed that within the precision of the instrument there was no difference due to particle size or variation in the concentration of
carbonates. Studies in progress will determine the practicality of removing the dolomite to the degree that the pH will'be lowered. Effective lowering of the pH would
lessen Johnson and Johnson's expense of the acid additive.

BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000928

 I
I
I
I

22
TABLE 17.

pH OF SIZE FRACTIONS OF ITALIAN TALC

Tyler Mesh Size

pH

Uns eparated

9.2

+200
-200+250
-250+270
-270+325
-325+400
-400

9.2
9.2
9.2
9.2
9.2
9.2

Acid Solubility

Acid-solubility measurements were made as a part of the study of the carbonate
component of Italian talc. It has been demonstrated that carbonate comprises the major
amount of the contamination and that its removal decreases abrasiveness and improves
lubricity.
The per cent solubility was first determ.ined gravim.etrically by the m.ethod outlined in Johnson and Johnson's Raw Materials Specifications sheet. The figures obtained were considerably lower than the 6 per cent solubility lim.it perm.itted by the
specifications; however, they were greater than the figures anticipated from. the sIUall
amount of carbonate minerals observed petrographically. Solubility analyses determined gravim.etrically ranged from. 2.10 to 2.81 per cent, showed no relation to any of
the physical property measurements. It was assumed that there was either a greater
am.ount of soluble m.atter in the im.palpable fine fraction, or that there was a constant
large sam.ple los s during handling.
To resolve the problem the carbonate cOIUponent was determ.ined petrographically
to be primarily dolomite, and a volumetric method of analysis was devised to analyze
closely for small concentrations of dolomite in talc (Appendix B). The figures derived
from these methods and computations are pre sented in this report as "equivalent dolomite". It is the measure of the total per cent of dolomite in the sample, not its incidence, which is a function of grain size. The equivalent dolomite analysis is recommended as a substitute for the previously used gravimetric analysis. The method is
adjustable for larger concentrations, and other computations may be substituted when
carbonates other than dolomite are present.
Equivalent dolomite is correlative with petrographicall!y observed contamination
and with related physical properties both in size fractions and in whole powder. A grab
sample of Italian talc from the large bulk sample obtained from Cranford, containing
slightly higher than average contamination, was analyzed for equivalent dolomite. The
analysis showed 1.87 per cent. To check the analyzed percentage against the incidence,
a series of grain counts was made on separate immersions, running 1.8, 1.8, 2.0, 1.8,
1.9, and 1.9 per cent. The average 1.8+ is essentially the same as the 1.87 per cent

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

I
I
I
I
I
·1
I
I
I
I·

I
I
I
I
I
I·
I
I

JNJAZ55_000000929

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

23
determined volumetrically. In other cases a concentration of dolomite in the coarse
sizes cuts down incidence, as a concentration of dolomite in the fines increases incidence. In the consideration of contaminants in regard to flotation, or of the measurement of lubricity or abrasiveness, the actual incidence of the contaminants is the important consideration. Within the range of Italian No. 1 talc, however, the difference
is usually small.
Table 18 shows a comparison of the gravimetric and equivalent dolomite analyses,
the incidence of contaminants, and lubricity, as listed against increasing abrasiveness.
Table 19 shows contamination and equivalent dolomite compared with decreasing lubricity. These show the primary relationship between abrasion and contamination, and the
secondary relationship of abrasiveness to lubricity in whole powders. This was also
demonstrated by Table 9 of this report. Table 20 compares the equivalent dolomite,
contamination, abrasiveness, and lubricity of sized fractions of Italian talc, demonstrating that the fines contain the more abrasive particles.
Inasmuch as the carbonate 'iil the Italian talc constitutes a rare component in all
size fractions, it~ removal by sizing is not practical. Any practical beneficiation process would be concerned with effectively removing the carbonate from the whole powder,
thus improving the slip while eliminating the major abrasive. The effects of flotation
on lubricity, abrasiveness, and contamination are presented in a report(8) on the beneficiation of Italian No. 2 talc.

REFLECTANCE AND WHITENESS

Discussion

The terminology of properties involving the behavior of light is very complex and
for the purposes of this report the discussion will be limited to reflectance, whiteness,
and gloss. Reflectance is the measurement of the return of light off of a s~rface in
ratio to the intensity of the incident light. This may be measured in terms of brightness,
apart from color. Whiteness may be measured in either terms of reflectance over the
whole spectrum as "lightness", or in the sense of the absence of specific colors. Gloss
is the measure of shininess of surface or specular reflection, as distinct from total
reflection.
Gloss, closely related to reflectance and whiteness, will be discussed in a future
report when ample samples are prepared to enable assessment of the factors which control the property. Gloss is a separate measurement from those here reported.
Although one may visualize the differences between whiteness, lightness, brightness, and gloss, one cannot subjectively differentiate one from another with any precision or determine the contribution of a specific property to over-all effect. The important consideration is the total subjective effect, which is quickly noticeable. However, in figuring means of improving the over-all effect we must relate the contribution
of particle size, shape, and specific contamination to both the over-all effect and to
specific properties. For example, fibrous talc is white, but less reflective than platy
talc. Rutile is highly reflective, but not white. The beneficiation studies designed to

BAT TEL L E

Protected Document--Subject to Protective Order

I

MEMORIAL

INSTITUTE

JNJAZ55_000000930

 "'0

o

It
n
It

C-

O
n

o

l:

3

III

::!.

in
l:

.E
III

!:lei
"'0

o
It
!:l-

<'
o

III

a.
~

c...
Z
c...

~

I~
o
o
o

~

TABLE 18.

COMPARISON OF THE GRAVIMETRIC AND VOLUMETRIC ANALYSES FOR DOLOMITE
AND THE RELATIONSHIP OF EQUIVALENT DOLOMITE TO THE INCIDENCE OF
CONTAMINANTS AND LUBRICITY, AS LISTED AGAINST INCREASING ABRASIVENESS

 -- ----------------- - -

a_
It
n
It

C-

o
o

n

l:

3

III

::!.

in
l:

.E
III

!:lei
"'0

o
It
n

~

TABLE 19.

III

o

a.
~

(II

EQUIVALENT DOLOMITE, ABRASIVENESS, AND PER CENT CONTAMINATION
LISTED AGAINST DECREASING LUBRICITY

»
~
~

ITI

r

Date of
Cranford
Sample

Lubricity-Board
Measurement, sec

9-6-56
11-6-56
9-12-56
9-19-56(c)
10-18-56
8-10-56
9-27-56
8-28-56
10-29-56
10-4-56
8-20-56
10-12-56
12-22-56
11-30-56
11-15-56

1. 083
1.053
1. 030
1. 028
1. 025
1.021
1. 017
1. 007
1. 006
o. 982
D. 971
0.968
0.965
o. 952
o. 936

Incidence of
Contaminants(a),
per cent

Equivalent Dolomite
(Volumetric),
pe r cent

Abrasiveness(b),
10- 3 in. / sec

r
1'1

~

ITI
~

o
:0

»
r

Z
f/I
~
~

c
~
ITI

c...
Z
c...

~

I~
o
o
o
o
o

<D

'"
IV

(a) Determined petrographically.
(b) None, or poor correlative relationship.
(c) Petrographically found to contain rare but coarse dolomite.

<1
1

<1
<1
1
1
1
2
1-2
1-2
2
2-3
2
2-3
2

1.6
1.6
1.5
1.7
1.6
1.5
1.6
1.6
1.6
1.6
1.7
1.7
1.7

I. 7
I. 6

1. 87
2. 15
1. 62
1. 84
1. 88
1. 70
1. 90
1.97
2.32
1. 90
1. 91
2.59
2.69
2.32
2.30

N
\.1l

 "'0

o

It
n
It

C-

O
n

o

l:

3

III

::!.

in
l:

.E
III

!:lei
"'0

o
It
!:l-

<'
o

lD

~

>

III

a.

~
~

TABLE 20.

COMPARISON OF CONTAMINATION, EQUIVALENT DOLOMITE, ABRASIVENESS,
AND LUBRICITY IN SIZE FRACTIONS OF ITALIAN TALC

1'1

r

r
1'1

Tyler Mesh Size

Incidence of Contaminants{a),
per cent
Total
Dolomite
Tremolite

Equivalent Dolomite
(Volumetric),
pe r cent

Abrasiveness,
10- 3 in. / sec

Lubricity-Board
Measurement, sec

1.9

2.14

O. 990

O.
1.
2.
2.
2.
2.

1. 30
1. 59
1.72
2.00
2. 33
2.48

0.889
O. 951
0.980
1. 030
1.043
1. 099

I:
1'1

I:

o
:u
>
r

Z

Unseparated

±2

<2

Trace

+200
-200+250
-250+270
-270+ 325
-325+400
-400

< 1
1

<1
1
1-2
2
2
>2

O-trace
O-trace
Trace
Trace
Trace-l
<1

1-2
2
2
>2

l/I
~
~

C
~

rrI

c...
Z
c...

~

I~
o
o
o

'"

(a) Determined petrographically.

6
1
0
0
1
1

N

a-

 I
I

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

27
remove specific particle sizes, shapes, or contaminants, in order to improve the
appearance of the talc, will be best controlled when the reflectance and color properties
can be assigned to particular components of the powder. As in the lubricity and abrasiveness studies, when the causes of variations are determined, it becomes possible to
v:isualize the means of improving the subjective property.
The reflectance properties of talc begin a new category of measurements. The
reflectance properties are distinct from other physical measurements insofar as direct
relationships are concerned, except when particle size, surface area, and purity are
concerned, as the following experiments demonstrate.
It is apparent at this stage of the. investigation that some degree of over-all
appearance can be controlled by the selective removal of particular particles.

The Measurement of Reflectance and Whiteness

The Italian talc is nearly pure white and highly reflective. The work under way on
reflectance and whiteness is designed to devise a means of improving these properties,
particularly in lower grade talc, as the result of interpreting their variations in response to the variations of other physical properties. The program includes determining
the effect on whiteness and reflectance of the removal of specific sizes, shapes, and
contaminants by beneficiation.
To date the measurements include only those made on a Photovolt Photoelectric
Reflection Meter* and on a Gardner Color Meter*~~. The Photovolt instrument measures diffuse reflectance in terms ofllwhiteness ll or luminous apparent reflectance (LAR).
Whiteness in this sense is a matter of lightness without regard to color. A green
tristimulus filter is used in the measurement, a standard procedure which permits
interlaboratory comparisons. The instrument is calibrated against standard enamel and
porcelain plates. The LAR of the Cranford samples is presented in Table 21, showing
a range in measurement of 95. 0 to 97. 5, with no discernible relationship to other properties of the whole powder.
To determine the relationship between LAR and particle size, measurements
were made on size fractions, which demonstrated greater reflectance in the fine s
(Table 22). This indicates that particle size and surface area are important factors.
To test if particle shape is also a factor, measurements were made on the products of
cyclone classification. These measurements showed that the underflow (platy talc) has
a greater reflectance than the overflow (fine acicular talc). A third test made on flotation products demonstrated that purity of sample is also a factor, the float product
producing a higher reading than, the starting sample. The data related to shape and
purity will be included in a report on the beneficiation of talc.
The Gardner Color Meter, among other applications, measures properties designated as Rd and tb. The Rd measurement is one of reflectivity in the sense of brightness, apart from color. The higher the Rd value obtained, the greater the brightness.
The tb measurement is one of color based on yellowness, but corresponding to whiteness
in near white materials. The lower the tb value the greater is the whiteness.
·Model 610, Photovolt Corporation, New York. New York.
"Gardner Instrument Company. Bethesda, Maryland. This is similar to the instrument used by Johnson and Johnson's Research
.
Laboratory.
BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000934

 I
28

Measurements made on the Cranford samples showed a range q.f 91. 30 to 93. 25
for Rd and 1.55 to 1.95 for +b, with no correlation as yet established with other physical
properties (Table 21).
To test if particle size has any effect on Rd and +b, measurements were made on
a series of size fractions, showing that brightness increased in the finer fractions and
that whiteness increased with fineness except for the minus 400-mesh fraction (Table 22),
In order to interp'ret the aberrant figure the minus 400-mesh fraction will have to be
subdivided and further +b values obtained. It is expected that the concentration of extremely fine acicular particles in the minus 400-mesh fraction accounts for the decr~ase
in the +b measurement. It appears, since 1.60 is the value obtained on the whole powder, that whiteness is lower in the extreme. particle sizes, both coarse and fine.
To test the effect of purity of sample on Rd and +b, measurements were made on
beneficiated products, showing that the r~moval of contamination measurably improves
Rd and +b. Measurements to be made on cyclone products will demonstrate the effect
of particle shape on these properties. These studies will be presented in a forthcoming
report on the beneficiation,of .talc.
Further work is recommended in the matter of improving the sheen of talc. Further investigation is required in the tracing of the specific properties of reflectance to
specific particle s prior to visualizing beneficiation for the improvement of sheen. It is
hoped to be able to adapt a Glossmeter for use on powdered talc in order to be able to
correlate properties and plan beneficiation for the improvement of specular reflectance.
When sheen can conclusively be traced to specific physical propertie s of the powder, I,'
then beneficiation for its improvement can be visualized.

THE DUST COMPONENT

;

For the purpose of this report dust may be defined as that fraction of the talcum
which remains air borne when the powder is shaken from its container. The dust may
be collected for examination by passing a moistened glass slide through the dust cloud
which remains suspended in the air when talc is shaken from a container, or by similarly sampling the suspended material after an open container is struck on the bottom
onto a table or similar surface. Such action produces two classes of matter, a cloud
comprising the bulk of the talc which quickly settles, and a fine portion which does not.
Material so collected has been analyzed petrographically and has been found to be composed primarily of platy talc, essentially free of contaminants or acicular particles.
This talc represents the finer sizes of platelets - the maximum diameter being about
15 J1 in the larger particles. This roughly corresponds to the theoretical <5 -J1 sphere
fraction(2) not including any amount of nonplaty grains or coarser platelets;
The nature of the dust component has been established and it appear s likely that
it is amenable to beneficiation. Work is at present under way devising a means of comparatively measuring the dust component of talc sample s, and to devise
means for its
removal should it be practically separable from the whole powder.

a

BAT TEL L E

Protected Document··Subject to Protective Order

MEMORIAL

INSTITUTE

I

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000935

 I
I
I
I
I
I
I
I
I

29

TABLE 21.

Rd, +b, AND LAR MEASUREMENTS ON THE. CRANFORD SAMPLES

Sample Date

Rd

+b

LAR

12-22-56
9-12-56
8-10-56
10-18-56
11-15-56
9-6-56
8-28-56
9-19-56
9-27-56
8-20-56
11-30-56
11-6-56
10-29-56
10-12-56
10-4-56

93.25
93. 15
93.00
92.55
92.45
92.35
92.20
92. 15
92.10
92.05
91.80
91.75
91.55
91.35
91.30

1. 75
1. 80
1.90
1. 75
1. 75
1. 75
1. 65
1.55
1. 60
1. 90
1. 75
1.75
1.55
1. 60
1.95

97.5
97.0
97.0
97.5
97.0
95.0
96.0
96.0
97.0
95.5
96.0
96.0
97.5
96.0
96.0

I
I
I
I
I
I
I
I
I
I
I
I
I

TABLE 22.

Rd, +b, AND LAR AS RELATED TO THE PARTICLE SIZE OF ITALIAN
TALC

Tyler Mesh Size

Rd

+b

LAR

Unseparated

91.40

1. 60

96.0

85.60 .
89. 15
90.36
90.50
91. 50
92.55

1. 90

91. 0
92.0
92.5
93.0
96.0
96.0

+200
-200+250
-250+270
-270+325
-325+400
-400

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

1. 45
1. 35
1. 30

1. 20
1. 50

INSTITUTE

JNJAZ55_000000936

 I
30
APPRAISAL OF PHYSICAL..,PROPERTY MEASlJREMENTS IN THE
EVALUA TION OF ORE'S AND BENEFICIATION PRODUC TS

The foregoing studies, and those previously reported(l), have established the
relationships between many qf the physic::~l properties of talc and subjective evaluation.
Many of the devices employed were relpful in establishing the interrelationships of physical properties, have served their purpose, and their use is not requisite to evaluate
the acceptability of talc, inasmuch as the interr'elationsh,ip' of properties permits such
an evaluation to be made on the basis' of a ~inimum of ~easurements.
The subjective tests do not measure specific properties and thus are only of compara.tive value in deciding what is th,e specific problem in' a nonacceptable talc, or how
it may be made acceptable by beneficiation. Such tests, however, must remain the final
analysis of acceptability of beneficiation products or in the selection of naturai highgrade tales.
The subjective tests are both,a matter of touch and visual comparison. Tested by
touch, individual consid~ration may be given to slip and abrasiveness. , Quickly noted in
nonacceptable tales or improper grinds of <;>therwise acceptable talcs are dry floury
feelings, pastiness, the rolling of the powder', poor spread which leaves portions unlubricated, and coarse or sharp grit. Visually it may be quickly noted if the powder is
colored or off-white, without sheen, spJ;'eads unevenly, C"ontains coarse brilliant particles, or contains a high compoI}ent of ex~remely fine dus~.
To measure improvement'in'talc, to maintain quality control, or to visualize
proper beneficiation for the improvement of a talc, 'it is necessary to measure specific
physical properties of the powder as a whole, to know the size and shape of the talc
particle s, and the nature of the contaminants.
Following the subjective appraisal, of foremost importance is petrographic examination. Such a study esta~lishes the platy or nonplaty nature of a talc, identifies the
contaminants, and should establish the general size distribution, incidence, degree of
subdivision, habit of aggregation, and crystallographic varietie s, of the talc, carbonates, amphiboles, and accessor.y mineral compo~ents.
In order to beneficiate for the, improvement of slip or the elimination of grit, it is
necessary to know the size distribution,not'only of the crystallographic types 'of talc
present, but also of the different impurities. Size-distribution procedures yield products which may be studied petrographically. These include screening, in the coarser
fractions, and sedimentation in the fine s. The measurements assigned sedimentation
products in usual procedures should be checked petrographically inasmuch as talc
platelets behave in the manner of theoretical sphere s of much smaller dimensions. A
practical method of comparing pow~er~, so lon'g as the theoretical measurements do not
become mistaken for actual diameters~ is the Andreason sedimentation technique, previously reported( 2). When this method is employed with supporting petrography it
should be a satisfactory device for evaluating beneficiation products.
Without proper size and mine.ra~ knm,yledge of a sample of talc, beneficiation
procedures cannot be developed.' Control'over the physical properties of a talc of known
and fairly constant composition could be keptby the use of refinements of the experimental lubricity and abrasion-m'easuring devices. However, a knowledge of the Inineralogy and size distribution is recommended for, any talc.
BAT TEL L E

Protected Document--Subject to Protective Order

ME~ORIAL

NSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

1
I

'I
JNJAZ55_000000937

 I
I
I
I
I
I
I
I
I
I
I
I
I
I

I
I
I
I
I
I
I
I

31
'The measurement of surface area, specific surface, porosity, and average diam ...
eter will be considered further in regard to compactibility and ullage, and the absorptive
power of talc; however, these are not necessary to consider as prerequisite to beneficiation studies for the improvement of the physical properties of talc. The lubricity board
and abrasion machine were built to measure small differences in heretofore purely subjective propertie s and to relate them to established physical measurements. With
proper mineralogical and size-distribution knowledge, these properties will be reflected
in the other physical measurements.
The following presents the measurements which should be attained in beneficiation
products in order to produce material equivalent in quality to Italian No. 1 talc. Improvement of these properties will, of course, produce superior powder, when not improved at the expense of other physical propertie s. Beneficiation studie s on Italian
No. 2 talc have produced powder considerably superior to grade No. 1 Italian talc in
slip, purity, and the absence of grit.
The following are the recommended requirements for beneficiation products to be
the equivalent of Italian No. I talc. The items considered important at this stage of the
investigation are marked with an asterisk.
Mineralogy':'
Platy talc, 90 per cent or more
Nonplaty talc, less than 10 per cent
Carbonates, less than 2 to 3 per cent
Amphibole s, Ie s s than 1 pe r cent
Accessory minerals, trace only
Opaques, none.
Size

Distribution~~

(1)

Whole powder:

Gr·eater
Greater
Grea.ter
Greater

than
than
than
than

150
200
325
400

mesh,
mesh,
mesh,
mesh,

none
less than 1 per cent
less than 10 per cent
less than 20 per cent.

The powder should have a size-distribution curve over its general
range similar to that shown by Andreason sedimentation measuremend 2) of theoretical particle s. Many of the particles finer than
the theoretical 5- J.l. sphere s are unde sirable, repre senting fine
acicular grains and dust. Fines, however, should not be removed
to the extent that the bulk density is raised beyond present
specifi'cations.
(2)

Contaminants:

,Greater than 250 mesh, less than I per cent
-250 to +400 mesh, not more than 2 per cent
Finer than 400 mesh, less than 3 per cent.

Note: There is reason to believe that the grind of Italian No. 1 talc is finer
than optimum for the production of a superior beneficiated talc. Possibly
talc 100 per cent minus 100 mesh would be fine enough. The principal reason
for a minus ZOO-mesh grind for the currently used product may be to reduce
the grit to a size where the platelets mask it. With beneficiated talc this would
not be necessary and there would be less fines to discard.
BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000938

 32
Lubricity Measurement
Greater than O. 93 second, preferably greater than 1 second.
Porosity
Approximately O. 45 to O. 50.
Average Particle Diameter
Approximately 2.4 to 3.3 jJ.. This measurement is made on the
Fisher Subsieve Sizer, the figures are of theoretical particles
but are not to be compared with those from 'the Andr~ason
measurements.
Bulk Density':'
22 to 24 lb/ cu, ft.
Specific Surface, Theoretical
Greater than 6600 cm 2/ g; preferable measurements lie in the
8000-cm 2 /g range.
Abrasion- Machine Measurement
Talc pellets - less than 2.7 x 10- 3 in. / sec, preferably less than
2 x 10- 3 in. / sec.
Carbonate pellets - less than 1 x 10- 3 in. /sec.
Moisture Content':'
Less than 0.08 per cent, preferably 0.05 per cent or less.

pH
Less than 9. 4, preferably closer to 7 in order to lower the
expense of the acid additive.
Acid

Solubility'~

Gravimetric, less than 3 per cent
Volumetric, less than 2 per cent.
LAR
95.0 or greater.
Rd~'

91.0 or greater.

MEMORIAL.

Protected Document--Subject to Protective Order

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000939

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

33
+b* Less than 2. O. (An additional color m.easurem.ent, -a, should be taken
when tales with a yellow-green tint are studied. )
The above measurements concern purity, slip, and grit, and the measurement of
acceptability of beneficiation products. Yet to be reported on are preferred measurements on the Glossmeter, preferred limits of the dust component, absorptiveness, and
compactibility. Although specific problems may arise when other than Italian talc is
considered, the above measurements should generally suffice formo st raw tales in the
measurement of improvement by beneficiation or of acceptability in regard to Johnson
and Johnson's present requirements.

FUTURE WORK

Future work related to' the physical properties of talc includes studies of the absorptive power to talc, measurements of gloss as distinct from whiteness and reflectance, mea~urements of compactibility, the dust component, studies on the effects of
different methods of drying processed talc on its physical properties, and further evaluation of the physical properties and mineralogy of beneficiation products.
Because of immediate pressure on other phases of work for Johnson and Johnson
most of the above studies will be held in abeyance.

REFERENCES

(1) Smith, W. L., "Studies of the Physical Properties of Talc, Their Measurement,
and Comparison", Battelle report to iohnson and Johnson (October 15, 1957).
(2) Macdonald, R. D. , letter report to Johnson and Johnson on the Andreason Sedimentation Pro cedure (April 1, 1958).
(3)

Smith, W. L., and Snider, R. H., "Investigation of the Salgada and Casa Nova Talc
Deposits of Brazil", Battelle report to Johnson and Johnson (May 28, 1957).

(4)

Smith, W. L., letter report to Johnson and Johnson on the talc deposits of Madoc,
Canada (July 25, 1957).

(5)

Brown, W. E. , letter report to Johnson and Johnson on the flotation amenability of
Italian and other tales (January 24, 1958).

(6)

Report in process describing results of classification of talc by cycloning.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMO'RIAL

INSTITUTE

JNJAZ55_000000940

 I
34
(7)

Sc1ar, C. B., Snider, R. H. , Macdonald, R. D. , and Tangel, O. F., "An Investigation of Selected Talc Deposits of the United States ll , Battelle report to Johnson
and Johnson .(February 29, 1956).

( 8.)

Brown, W. E., Smith, W. L., and Macdonald, R. D., liThe Physical Concentration
of Talc Ore s - Flotation " , Battelle report to Johnson and Johnson (May 23, 1958).

The original notes on the laboratory work described in this report are in Battelle
Laboratory Record Books No. 13034, pages 78 through 96; No: 14187, pages 44 through
100; No. 14431, pages 7 through 100; and No. 14677, pages 1 through 8. The work was
done in the period from October 21, 1957, to May 5, 1958.

W LS:djo / gpi/bah

SA T TEL L E

Protected Document--Subject to Protective Order

M E M 0

R I A L'

INS T , T U T.E

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

APPENDIX A

DESCRIPTION OF ABRASION MACHINE AND TECHNIQUE OF OPERATION

8 A T TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000942

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

A-I
APPENDIX A

DESCRIPTION OF ABRASION MACHINE AND TECHNIQUE OF OPERATION

The experimental device described as the abrasion machine in this report (Figures
1 and 2) consists of a 1/20-hp 220-volt 60-cycle 1725-rpm three-phase Westinghouse
electric motor, mounted vertically at operating height, fitted with a steel lap. The steel
lap has a diameter of 5 inches and is designed so a lap cloth may be held in place by a
rubber belt. A Buehler Microcloth was selected as a standard lapcloth. The lap is
housed in a steel bowlS inches deep and 9-1/2 inches in diameter. A spout extends from
the bottom of the bowl to carry the tested slurry into a beaker. A plastic shield is fitted
over the top of the bowl to prevent spatter. A hole in the shield permits observation of
the operation and access to the sample holder and slurry feed tube.
The slurry feed tube and a cylindrical 1/2-inch-diamete r sample holder are fitted
into a metal strip which is fastened in place over the lap. The sample holder is fixed 2
inches from the center of the lap. Directly behind the sample holder the slurry feed tube
is fixed in a similar position so that the clockwise rotation of the lap brings the slurry
which is to be measured under the standard talc pellet. The talc test pellet is held
onto the lap by a 16. I-gram weight which prevents the pellet from skipping or floating on
the rotating lap. The slurry feed tube is connected by a rubber tube with an adjustable
clamp to a separatory funnel held in a ring stand. The funnel is a 500-ml open-top separatory funnel equipped with a stopcock, and serves as the reservoir for the slurry which
is to be measured. The time of operation is kept on a Kodak electric timer.
The pellets are made of minus 400-mesh Italian talc pressed in a F. S. Carver
Laboratory Press under 50,000 psi. The pellets are 1/2 inch in diameter, and the
5. 2-gram samples used make a pellet about 7/10 inch long.
In addition to the standard talc pellets used in the measurement of total abrasive
particles, carbonate pellets were used to measure abrasion by the harder contaminants
alone. The carbonate pellets were made by fusing three parts by weight of sodium carbonate to one part of sodium borate into a melt. The fused melt was then crushed and
pressed similarly as the talc· pellets, under 15,000 psi. Because of swelling during
drying the carbonate pellets must be measured wet, unlike the talc pellets. Also, because of slight solubility, the slurry to be te sted must be a mixture with alcohol instead
of water. The carbonate pellets are less satisfactory than the talc pellets, however,
they se rved a specific expe rimental purpose.
In operation, the slurry, composed of 3 grams of the talc to be measured, in
350 ml of distilled water, passes onto the rotating lap at a rate controlled by a clamp on
the feed tube. The slurry is carried under the standard pellet where the abrasive components wear the pellet at a rate approximating abrasion of the pellet by some 1800 ft/
min of surface composed of the sample slurry. Pure samples of talc were found to effect
little abrasion, while contaminated talc was found to quickly wear away the pellet. The
amount of abrasion loss as measured on a Starrett micrometer caliper is divided by the
number of seconds of abrasion to provide figures representing the degree of abrasion.
Should similar experiments be repeated, the following are important considerations.
In any se~ies of tests the operation time of the machine should be essentially the same.
The talc slurry should be kept in suspension by agitation. A new lap cloth should be used
BAT T EL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000943

 I
I

A-2
as soon as any wear is noticed. All the talc pellets used in a series of tests should be
pressed at the same time. Talc pellets should be dried overnight before measuring, to
prevent any swelling effects of absorption of water. When any reruns are required on the
abrasion machirie, the slurry feed should be adjusted to reproduce former readings before comparative data are sought. To make proper comparative measurements the abrasion machine should be operating so as to make replicate tests showing a difference of not
~~re than O. 1 x lO- 3 in. / sec of abrasion.

8 A T TEL L E

Protected Document--Subject to Protective Order

M EM OR I A L'

INSTITUTE

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000944

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

APPENDIX B

DETERMINATION OF EQUIVALENT DOLOMITE CONTENT
IN ITALIAN TALC BY VOLUMETRIC ANALYSIS

by
W. E. Brown

BATTELLE

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUT~

JNJAZ55_000000945

 : 'ffbf 0
\

f

t..

'- /1

#~~/(--e ;

cr/"
~
'~
#'" )

A'

A

1/1~ ". . ',t.:.,
. I)

, I , '...'

!(/ )~"

V

./

/. (. t ,j)"
r t . "-0 /

t~

~

.'

(j

)( jJ~

\

I:

7

'7'
<?

/r

X

X~9~

X 'f'f
i I

X

/,0 0

..;'

I

X !y-t-t
!'

~

,

,,:;:,

/

/?
'<"":":

,

0,

'

-

' f

'.7

/'

~

' ,~

?'f

~

1

' o'f

=

/~

/UA-~

:.
='::

j
Protected Document--Subject to Protective Order

JNJAZ55_000000946

 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

B-1
APPENDIX B

DETERMINATION OF EQUIVALENT DOLOMITE CONTENT
IN ITALIAN TALC BY VOLUMETRIC ANALYSIS
'by·
W. E. Brown

(1)

Prepare a solution of approximately O. 2N sodium hydroxide and determine exact
normality.

(2)

Prepare a solution of approximately O. 2N hydrochloric acid and determine exact
normality.

(3)

Weigh out for analysis a 5. OOO-gram sample of talc and put in a 250-cc beaker.

(4)

Add 25 ml of distilled water to the talc sample and stir with a glass rod to thoroughly
wet the talc.

(5)

Add 50 cc of the HCI solution prepared in Step (2).

(6)

Heat sample, containing water and HCl, for 45 minutes at 105 C.

(7)

Raise temperature to boiling for approximately 1/2 minute to expel H2C03.
care so that the sample does not boil over,

(8)

Cool to room temperature.

(9)

Add 4 drops of methyl orange indicator to the cooled sample and stir.

5.5"

J.. ::~

~~ C.

(10)

Titrate the sample with the NaOH [from Step (1)] to a yellow end point,
determines the amount of unused acid.

(11)

Calculate the per cent dolomite.
Given:

Normality of NaOH
Normality of HCl

Use

This

An example of the calculations is as follows:

= O. 2055
= 0.2120

[from Step (1)]
[from Step (2)J

Each milliliter of HCI contains 36. 5 x O. 2120 = O. 0077 gram of pure HCl
100
ml of NaOH neutralizes 0.97 ml of HCI

/- J- (;,

1 gram of HCl neutralizes 1.26 grams of dolomite

/-/5
/~

37

47.6 ml of NaOH was required to titrate a 5-gram sample which had been
dige sted with 50 ml of HCl.

BAT TEL L E

Protected Document--Subject to Protective Order

MEMORIAL

INSTITUTE

JNJAZ55_000000947

 B-2
47.6 x 0.97
50.00 - 46. 17
3.83 x 0.0077
0.0295 x 1. 26

::
::
::
::

46.17 ml of unused HCI
3.83 ml HCl consumed by dolomite
0.0295 gram HCl consumed by dolomite
0.0372 gram dolomite dissolved by Hci

0.0372 x 100
5.000

:: 0.74 per cent dolomite.

Note: In order to test the accuracy of this method of analysis, some relatively
pure dolomite (taken from a mineral specimen) was analyzed. The weight of
the sample analyzed was 0.0300 gram. The foregoing analytical method showed
the sample to contain O. 0306 gram of dolomite. Another check test was made
by analyzing a sample of Italian talc for per cent of C02, .and converting the
C02 content to the theoretical amount in dolomite. The. C02 analysis indica.ted
that the dolomite content was 2. 26 per cent. By volumetric analysis the
dolomite content was calculated to be 2. 18 per cent.

BAT TEL L E

Protected Document--Subject to Protective Order

M E M O"~ I A L

I ,N S T I T 'U T E

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

JNJAZ55_000000948