John Almon.

The American journal of science and arts online

. (page 41 of 102)
Online LibraryJohn AlmonThe American journal of science and arts → online text (page 41 of 102)
Font size
QR-code for this ebook


admit of precise definition. This species must have a new or
unappropriated name, and we propose for it that of Kaolinite,
in allusion to the material which furnishes it most commonly-
and abundantly.



Digitized by



Google



852 Messrs. Johnson and Blake on Kaotinite and PhokrUe.

The cbemical compcNsition of this mineral was first dedaced
by Forchhammer from the analysis of a number of kaolins. It
is represented by the formula ^gisXleft, or by «8iSl2fi. The
per cent proportions vary considerably according to the atomic
weights employed in the calculation. In the taole that follows,
p. 358, are given the percentages reckoned on the atomic weights
adopted by Gmelin (Handbook, English ed.), by Bammelsberg
{Handbucn d. Mineralchemie), and by Fresenius (Quantitative
Analysis, 4th ed.).

Of the substances which have come under our notice, having
the above composition, the most striking is the so-called nacrite,
from the Einigkeit mine at Brand, near Freiberg, Saxony. It
is described by Breithaupt (Berg. u. Hiit. Zeit, No. 40, 1866) as
occurring " in snow-white or yellowish six-sided tabular crystals
in fan-shaped or reniform aggregates, and having pearly luster
passing into adamantine. Sp. gr. 2 '68." The analysis of this
mineral made by Richard Miiller appeared in Dana's 9th Sup-
plement, and is quoted below.

DesCloizeaux, in the Supplement to his Manuel de Min^ralo-
gie, p. 549, remarks concerning this mineral as follows: "There
has been recently discovered in Saxony a pholerite, at first called
nacrite^ which occurs in large macled hexagonal plates. These
plates are composed of six triangular sectors, whose boundaries,
though quite vague, nevertheless give indications of composition
parallel to the faces of a right rhombic prism approximating the
angles 120° and 60°. They cleave easily in the direction of the
base of this prism ; their interior structure is fibrous, and their
surfaces are slightly undulated. Notwithstanding the plates are
transparent when sufficiently thin, their action on a polarized
beam of parallel rays is very irregular. In convergent light
there are seen iir each sector the hyperbolas which indicate two
diverging optical axes whose plane is normal to the side situated
upon the hexagonal contour and is consequently parallel to the
principal diagonal of the base of the fundamental prism. The
bisectrix is negative and evidently normal to the plane of cleav-
age. The dispersion of the axes is feeble: at 46° from the
plane of polarization it is shown by the symmetrical distribu-
tion of the colors about the two hyperbolas, and the separation
of the axes is greater for the red rays than for the violet," &c.

Our observations, made on a specimen in the cabinet of Pro-
fessor Brush, are as follows. The crystals occur in hemispheri-
cal groups of about 4 mm. in diameter. These groups have a
radiated structure, as is evident from their cleaving into quite
thin wedge-shaped laminae. The laminae themselves appear to
have a rs^iate structure at right angles to that indicated by the
•cleavage, for, when viewed by polarized light, dark shades of
color branch out irregularly firom near the center of the thin



Digitized by



Google



Messrs. Johnson and Blake on Kaoliniie and Pholerite. 35S

edge of the wedge, and these rotate with considerable uniformity
as the plane of polarization is changed. When magnified, the
surfaces of the laminae are seen to be striated in. three directions.
These strias intersect at angles of 60^ and 120°. The micro-
scopic structure under polarized light gives evidence that the
ultimate plates of these groups are twin or compound crystals.
In a section ground thin, parallel to the cleavage direction, crys-
tals were seen superposed, the outline of one corresponding to
the striae of another, while, optically, they did not correspond.
The portions of crystals thus distinguished by polarized light,
were often «longat^ three or four diameters, and this elongation
had the same relation to the plane of polarization as observed
in the mineral from Summit Hill to be described presently.
Sections of the groups often give an approximately hexagonal
outline. The plates of this mineral are flexible, non-elastic, and
have a soft, soapy feel.

The white pearly luster of the " nacrite" appears to be due to
strata of air included between the separate crystalline plates
composing a mass. It, as well as the crystals presently to be
noticed from Summit Hill, exhibits the colors of pearl. This
can be sewi under the microscope by reflected light. If perpen-
dicular illumination be not used, an oblique position of a plate
on the slide is most £Eivorable to reflect the lignt into the instru-
ment and bring out the color. This iridescence may be due
either to the fine striae upon the crystal or to the colors of thin
plates of eleavage. As a comparative test, quite thin blown
^lass, when crushed into a mass, was found to give the pearly
fuster perfectly, without at the same time exhibiting color ; while
the thinnest glass in the same condition showed bK>th in a high
degree.

A second substance in possession of Prof. Brash was received
firom Prof W. T. Roepper, of Bethlehem, Pa., and was found in
a cavity in a coal seam at Summit Hill, Carbon Co., Pa. It
bore the label Pholerite. It is a brown scaly powder, which on
digestion in hydrochloric acid gives up oxya of iron to that sol-
vent and becomes nearly white. It has a
pearly luster and soapy feel. Magnified
fifty diameters, the substance appears made
up, for the most part, of well aefined crys-
talline plates. The average size of the
plates is *003 of an inch, the largest are
•005 of an inch in breadth ; they are in
general extremely thin. They have, as
nearly as the mode of measurement em-
ployed would show, the angles of perfect
hexagons, 120° (see figure). The method KaoUnha—scaitaoodiwM.^
used was to draw them upon paper under the camera luddai



Digitized by



Google



954 Messrs. Johnson and Blake on Kaoliniie and Pholerite,

with the aid of a straight edge, to a scale of 650 diameters, and
to measure the angles of the drawings with the hand goniometer.
Many of these tables are elongated in a direction parallel to one
of the sides of the hexagon, sometimes to two diameters. They
are striated, and the principal striae concur with this elongation.
Besides these extremely thin and isolated hexagonal plates,
the substance contains prismatic aggregates of similar plates.
These aggregates have all degrees of thickness, amounting in
some instances to *003 of an inch. Some of them are obviously
homogeneous crystals, composed of closely parallel laminae per-
fectly resembling in their aspect prisms of mica. When viewed
laterally they are often quite transparent and have deep trans-
verse striae, which indicate perfect basal cleavage. In other
cases the structure of these prisms is less compact and symmet-
rical ; the plates being loosely combined and somewhat separated
from each other on one side of the prism.

The longer tables, when seen in polarized light, cease to show
a difference of shade on the field, or of tint wim the use of selen*
ite, when the plane of polarization of the analyzer is parallel
with or at right angles to the axis of elongation. The same oc-
curred with crystals on edge, when the plane of polarization
was perpendicular or parallel to the cleavage plane.

The thicker plates, when not in the positions just mentioned,
liave very evident effect on the polarized beam. This indicates
A considerable separation of the optical axes. Like the laminae
of '' nacrite," the crystals of this substance exhibit, when prop
erly illuminated, the colors of pearl. When ignited, the sub-
stance is seen to increase in bulk, and the microscope shows this
to be the result of the exfoliation of the crystals due to the ex-
pulsion of their water of combination. This mineral differs
from the so-called " nacrite" in not being macled. Some frag-
ments from the exterior of the groups of nacrite crystals resem-
ble this "pholerite" closely, showing evident hexagonal outlines
and striae with the angles 120^ and 60^. We found the specific
gravity of the purified mineral to be 2*69. An analysis made
on 442 milligrams by fusion with carbonate of soda gave the
following results :

Silica, - ... - 46*93

Alumina, and trace of oxyd of iron, - 30*81

Water, - - - - . 1402

99-76

The substance described as pholerite by Dr. F. A. Genth (this
Jour., [2], xxviii, 251) is of similar character and occurrence.
It was found in coal mines at Tamaqua, Pa., in scales of a vel-
lowish white color, which became white on treatment with dilate
hydrochloric acid, and at Pottsville, Pa», in snow-white scales of



Digitized by



Google



Messrs. Johnson and Bktke on Kaolinite and Pholerite. 355

a pearly luster. Dr. Gtenth remarks that under the microscope
the scales appear to be clinorhombic. His analysis of this sud-
stance, after purification by hydrochloric acid, is given below.

The first mention of a crystalline substance with the compo-
sition of Forchhammer's kaolin that we have been able to find
is by Wdhler, who describes, under the name steinmark, a pale
yellow coherent mass which is converted by dilute hydrochloric
acid, with solution of a little oxyd of iron, into a wliite shining
powder. (Ann. d. Oh. u. Ph., Ixxx, 122.) With help of a lens,
Wohler found it to consist of "shining lamina, which, when
magnified 200 diameters, were seen to be transparent and to
consist in part of rhomboidal tables. Before treatment with
dilate hydrochloric acid the mass had an earthy fracture which
assumed luster by rubbing, an unctuous feel, and adhered
strongly to the tongue. Sp. gr. 2'6." The locality was Schneck-
enstein, Saxony. The analysis by Prof W. S. Clark, now of
Amherst College, is given below.

In response to our application, Prof. Clark has kindly fevored
lis with a fragment of this substance. We observed that it re-
quires to be acted on with hot concentrated hydrochloric acid
for some time before falling to a white powder. Microscopic
examination of the substance thus purified confirmed our antici-
pation of its close physical resemblance to the minerals already
noticed. It consists of plates and bundles of plates, the largest
being '001 of an inch or less in breadth, and when sufficiently
magnified has a great similarity to the kaolinite from Summit
HilL The angles of the plates, as well as of the strisB which
they exhibit^ approximate 120"^. Under a high power the striae
are seen to be formed by the edges of superposea and conform-
able plates ; some loosely aggregated bundles resembled those to
be noticed presently, as occurring in the kaolinite from near
Richmond, V a. [See note on a subsequent page.]

In 1859, Knop analyzed a mineral of the same composition
from Zeisigwald near Chemnitz, consisting of microscopic sharp
rhombic plates. (Jahresbericht der Chem., 1859, p. 789.)

Stolba has also published an analysis (see below) of a sub-
stance occurring in the coal mines of Schlan, Bohemia, in the
form of brilliant white scales, which is obviously kaolinite.
(Jour, fur prakt. Ch., xciv, 116.)

In his Manuel de Min^ralogie, DesCloizeaux, in describing
pholerite, remarks, p. 190 : " a varietv from Lodfeve, in some-
what contorted scales, exhibits under the polarizing microscope,
indications of two quite divergent systems of axes, of which the
negative bisectrix is almost normal to the plane of the laminas;
the interior structure otherwise appears hignlj irregular." This
** variety " is the mineral analyzed by Pisani from the same lo-
cality (Comptes RenduB, liii, 1072, also Dana's 10th Supplement),



Digitized by



Google



356 Messrs, Johnson and Blake on KaoUnite and Phokrite,

and the words of BesCloizeaux, above quoted, appear to be the
first recorded optical observations on kaolinite. risani's analysis
is ^ven in the subjoined table.

Kaolin is described by nearly all writers as an opake amor-
phous substance. Some have, mentioned it to contain minute
transparent plates, but'have supposed them to be sheets of mica
or other admixture. We have examined microscopically twenty
specimens of kaolin, pipe- and fire-clay. Most of these are of
unknown origin. In them all is found a greater or less propor-
tion of transparent plates, and in the most of them these plates
are abundant, evidently constituting the bulk of the substance.
The kaolin from Diendorf (Bodenmais), Bavaria, is perhaps the
most finely divided of all the white clays we have studied.
When dusted dry upon a glass slide it appears to consist chiefly
of masses of a white substance that are opake or nearly so in
transmitted light, but, when fully illuminated above and below,
have the translucent aspect of snow in the lump. Interspersed
among these masses may be seen extremely minute transparent
plates of irregular rounded outline. When brought into water
the masses are almost entirely resolved into similar transparent

Elates, most of which are not more than 0001 of an inch in
readtk This description applies to all the finer plastic clays.
Even the dark-colored Stourbridge clay is made up m large part
of transparent laminsB, as is a compact sedimentary orownish-gray
pipe-clay from Table mountain, Tuolumne Co., Cal. The same
IS true of the blue fire-clay from Mt Savage, Md., the white
clays of Brandon, Vt., Perth Amboy, N. J., Beading, Pa., Ches-
ter Co., Pa., Long Island, and various other white and colored
clays from unknown localities. On several specimens of kaolin,
^especially on one collected at one of the hematite mines at Beek-
mann, N. Y., we have observed pearly glistening surfaces on the
interior of cavities. Viewed in reflected light, by the micro-
scope, these surfaces were seen to be covered with minute scaly
crystals, or crystalline aggregates, which, however, revealed no
regular outlines.

A white, pulverulent substance, having much the appearance
of powdered starch or wheat flour, found near Richmond, Va.,
was recently analyzed by Mr. Burton in the Sheffield Laboratory.
Its composition agrees with Forchhammer's formula (see below),
and under the microscope it is seen to be made up for the most
part of transparent plates '001 of an inch or less in breadth, and
of prismoidai bundles, obviously composed of loosely aggregated

?lates, similar to those found in the kaolinite from Summit HilL
'hese bundles are usually curved, and their length is often sev-
eral times greater than their breadth. The bundles are fan-
shaped in some instances, and the plates are rarely parallel to
each other. The edges of these bundles are frequently presented



Digitized by



Google



Messrs, Johnson and Blcike on Kaolinite and Pholerite. 857

and in this position they have the greatest effect anon polarized
light. They have least influence on the polarizea beam when
the plane of polarization is perpendicular or parallel to the plates
in this position. The separate plates are of broken and irregular
outline. Grains of quartz are intermingled.

In four other specimens of kaolin from unknown (probably
European) localities, similar prismatic bundles were observed*
The Dundles were usually curved and irregular; in some in-
stances their length was iour or five times their breadth. One
of these four kaolins contained hexagonal plates that could be
made out with ease under a one-fourth inch objective. Two
others, when rubbed between the fingers, assumed a distinct
pearly luster ; and after this treatment, by which the prismatic
crystals were broken up, microscopic examination revealed
abundance of hexagonal plates.

Prof. Brush has called our attention to a specimen of fluor
from Zinnwald on which occurs a white powdery substance
that passes for kaolin. It consists entirely of perfectly definite
hexagonal tables averaging *0005 of an inch in diameter, which
are usually thin but sometimes are aggregated into short prisms.

The kaolin, pseudomorphous after prosopite, from Altenberg,
Saxony, the analysis of which by Richter (Pogg., xc, 316) is
given below, though compact in texture, is found by microscopic
examination to be made up also of hexagonal plates and bun-
dles of plates.

The plasticity of clay is a physical character, and appears to
have a close connection with the fineness of the particles. The
kaolinite of Summit Hill, consisting chiefly of crystal-plates av-
eraging 003 of an inch in diameter, is destitute of this quality.
The nearly pure kaolinite from Bichmond, Ya., occurring mostly
in bundles of much smaller dimensions, the largest being but
*001 of an inch in diameter, is scarcely plastic. The four kao-
lins of unknown origin which we have described as also consist-
ing largely of prismoid crystals are scarcely plastic, though
when rubbed between the fingers they become more soapv to
the feel. So too the crystallized kaolinite accompanying fluor
from Zinnwald is a scarcely coherent unplastic substance.

The more finely divided fire-clay from Long Island, is more
" fiit," while the feodenmais porcelain earth and other clays, in
which the bundles are absent and the plates are extremely small,
are highly plastic. So, too, the Summit Hill crystals, when tritu-
rated in an agate mortar, yield a powder which, when breathed
upon, acquires the argillaceous odor, under the microscope per^
feetly resembles the finer kaolins, and in the wet state is nighly
plastic and sticky.

Sommaruga has published analyses of two Passau kaolins^
Am. Joux. Soi.— Sbgovi) Sseibs, Vol. XLIII« No. 139.— Mat, 1867.
46



Digitized by



Google



8M Messrs. Johnson and Blake on KaoUnite and Pkolerite.

employed in the imperial porcelain manQfactore at Vienna, one
of which IB "fat" and the other "short." The composition of
the two is almost identical (see Chem. Centralblatt, 1866, p. 268),
and the different degree of plasticity is thus evidently connected
with their state of division.

It is possible also that the plasticity of a clay is related to the
form ot the plates of kaolinite, perhaps to their thickness, but
this is a subject that requires farther investigation. Our obser-
vations indicate that the impurer sedimentary clays are the most
Elastic. Some of these are perhaps not so fine as "shorter"
aolins. The plasticity may be, therefore, in part due to the
imparities.

In the subjoined table are given the analyses of various crys-
tallized kaolinites which have been previously referred to.

Analyses of crystallized Kaolinite,

Si Si ft ^«'*^



KM>linite(Nfterite),Freibeiv,SazonT»&Mun6r» 4*7-74 S9*4S 14-07

** (Pbolerite), Summit Hill, Pa., S. W. JohnaoQ, 46-93 8981 1402 ....

( ** ),Tamaqua.Fa,F.A.G€nth, 4«'98 89-65 18*69 0-17

(KaoliD), Richmond, Va^ B. S. Burton, 48-66* 86-61 1288 2-96

•• (Stoinmark), SchneekeosteiD, Saxony, W. &

Clark, 46-76 86-69 1842 0-94

" (Kaolin), Zeisigwald, Sax^ A. Enop, 49-91 86-28 14-86f ....

•• ( " VAltenberg, « R.Rlchter, 46-68 89-89 18-70 0-60

** (Pholerite), Lod^re, France, Pieaai, 4700 89-40 14-40 ....

** ( ** ), Schlan, Bohemia, Stolba, 47-98 86-78 1629 ....

Calculation after Gmelin(Si=16, 21=18-7) requires, 4719 89-12 18-69 ....
« " Rammelaberg (3i=14-8, Sls:18-68)

requires, 47-06 89-21 18*74 ....

« FreMiuuB(I>nma8)(Si=14,Sbsl8-76)

requu^, 46*88 89*76 18*90 ....

We find more than thirty analyses of days, kaolins, and
steinmarks, which obviously agree with the formula above
given. Some of these analyses appear to have been made on
the kaolin as it occurs in nature ; others, however, were made
on the washed kaolin as prepared for the porcelain manufacture;
and in still other cases, as in Forchhammer's investigations, the
clay was first exhausted with hydrochloric acid and tne analysis
was performed on the residue, allowance being made for quartz
and substances insoluble in sulphuric acid.

It is obvious then that the basis of many kaolins and clays is
a soft, white, transparent, infusible mineral, which crystallizes
in forms probably belonging to the trimetric system, has a den-
sity of 2'6, when crystallized has usually a pearly luster, is in-
soluble in dilute hydrochloric acid, and in most of its forms is
difficultly decompdised by hot concentrated hydrochloric acid,
but is resolvable by hot oil of vitriol and dissolves completely
in strong solutions of caustic alkalies. In chemical composition

* Jneluding tome quarts. f Bydtflbrenoe.



Digitized by



Google



Messrs. Johnson and Blake an Kaolinite and PholerUe. 869

it agrees with the formula deduced by Forchhamm'er from hid
analvBes of porcelain days, viz., sSl 4Si d£[.

This substance is not the naerite of Yauquelin or Thomson,
which contained at the most but one per cent of water. It is
not the pholerite of Guillemin, as we shall presently see. The
old terms kaolin, steinmark and lithomarge have been so loosely
applied that they do not define it

The massive yellow steinmark from Bochlitz has the compo-
sition of kaolinite, but with a portion of the alumina replaced
by sesquioxyd of iron. Elaproth's analysis (Chemische Ab*
handlungen, vi, 287) is as follows:



Silica,


45*25


Alumina,


36*50


Sesquioxyd of iron, -


2-76


Water,


1400


Potash,


trace.



98*50

Digested in hot concentrated hydrochloric acid it is scarcely
acted upon, but retains its jellow color without falling to pow-
der, as we have observed with a specimen in Professor Brush's
cabinet.

The steinmark from Buchberg, analyzed by Zellner, that from
Rumoelsberg examined by Bammel8l)erg (Bamm. Handbuch, p.
676),T;hat from Saszka analyzed by v. Hauer (Jahresbericht der
Chem., 1856, 860), and the severite of the latter (Bamm., Hand-
bttch, p. 1012), are evidently impure indurated kaolinite.*

Brongniart and Malaguti have mvestigated a large number of
kaolins, and some of the results of their analyses have led to
the adoption of the formula Si Sisfi (or 2Su9i4£[). But of the
81 analyses by B. and M. but four agree to the above formulae
within 1 per cent of silica, but six within 2 per cent, and but
nine witlun 8 per cent. (See Dana's Min., 4th ed., vol. ii, pp.
249-60.)

Furthermore, the data from which this formula has been pro-
posed, were not derived from the original analyses of the clay, but
m>m these analyses "corrected" by deducting from the total
silica (exclusive of quartz), the loss suffered by boiling the kao*

»

* Hallojeite cannot be confounded with kaolinite although it is another hydrate
of the same silicate of alumina that exists in the latter. Its formula Is 4Si sSl
12^ or i&i SA 4£L The specimens of this mineral from Guatequ6 analvced bj
Boussin^ult, and those from Houscha and Anglar examined by ^rthier, lost one-
half their water (8-9 per cent) on drying at 212°, and thus acquired the formula of
kaolinite. It cannot be assumed that this loss was due to hygroscopic water, for
many substanoes when dried at 212^, or below that temperature, loee a part or all
their ciystal water. Thus selenite loses about three-fourths of its water at 212^.
The quantity is npt, however, definite. Halloysite is of much inferior density
(tp. gr.8=:i) to kaolinite, is more easilj decomposable by adds, and is without doubt
a fairly characteriied spedes.



Digitized by



Google



860 Messrs. Johnson and Blake on Kaoliniie and Pholeriie,

lin for one and a half minutes with a five per cent solutioD of
caustic potash, this loss being assumed to be accidental hydrated
silica. This mode of correction is obviously of no value. On
the one hand, the caustic potash might dissolve the kaoliniie
itself from the more finely divided specimens. Berthier and
Bammelsberg have both observed the solution of kaolin in
fitrong potash ley. On the other hand, treatment for so short a
time would scarcely suffice to remove all the free silica torn a
kaolin that contains a large proportion of that substance, if our
andytical experience enable us to judge. Again, the analyses
appear to have been intended in the first place for technical pur-
poses, and were made, not on specimens selected with reference
to their purity, but on the clays in bulk employed in the porce-
lain manufacture. It is plain that they are not adapted to throw
light on the chemical composition of the basis of kaolin. Least
of all do they give evidence of the existence of the compound
5l5i2fi in the generality of clays.

We have been able to find but two analyses of kaolin made
by Forchhammer's method that lead to this formula. On the
other hand, eighteen of Malaguti's uncorrected analyses agree
with Forchhammer's formula, and eight of them as closely as



Online LibraryJohn AlmonThe American journal of science and arts → online text (page 41 of 102)