Joseph William Mellor.

A comprehensive treatise on inorganic and theoretical chemistry (Volume 2) online

. (page 80 of 156)
Online LibraryJoseph William MellorA comprehensive treatise on inorganic and theoretical chemistry (Volume 2) → online text (page 80 of 156)
Font size
QR-code for this ebook


appears bluish-violet ; lithium, dark reddish-brown ; and sodium, yellowish-brown ;
while W. L. Dudley 5 found that the vapour of potassium at its b. p. has a greenish
colour, and at a red heat, bluish- violet ; the vapour of sodium in thin layers appears
colourless, in thick layers purple-red, and at higher temp, yellow; while the
vapour of rubidium is blue with a greenish tinge. According to E. Linnemann.
the oxidation of sodium at ordinary temp, is accompanied by a greenish phos-
phorescence ; 6 and if, when the phosphorescence has ceased, the metal be heated
to 60 or 70, the phosphorescence reappears. An analogous phenomenon occurs
when sodium acts on water in darkness. The bluish-white streak which sodium
makes on paper shows a green phosphorescence which persists longer than the
greyish- white streak made by potassium because sodium is not so rapidly oxidized
as potassium.

Different numbers have been reported for the specific gravities 7 of the alkali
metals, presumably owing to the presence of impurities, as well as to differences of
temp. The best representative values at 20 are as follows :

Lithium. Sodium. Potassium. Rubidium. Caesium.

Sp. gr., solid (20) . . 0*543 0'9723 0'859 1'525 1'903

At. vol 13-1 23-7 45-5 56'0 71-0

Expansion on melting . 1*51 2'57 2'60 1'657 1-393 per cent.

The metals lithium, sodium, and potassium are thus lighter than water. Lithium
is specifically lighter than any other element solid at ordinary temp. ; it floats on



THE ALKALI METALS 453

petroleum. According to G. Vicentini and D. Omodei, solid sodium at its m.p.
97'8 has a sp. gr. 0'9519, and liquid sodium, 0'9287 ; solid potassium at its m.p.
63'5, 0*851, and liquid potassium, 0'8298. Sodium at its b.p. has a sp. gr. 0'7414 ;
and at the temp, of liquid air 1*0066. Potassium is specifically lighter than sodium,
and has nearly twice its at. vol. The at. vol. of csesium is larger than that of any
known element. K. F. Slotte estimated the edge of the molecular cube of liquid
and solid sodium and potassium to be respectively 8' 7 X 10~ 9 cm. and 10'7 X 10~ 9 cm.
It is of historical interest to note that in 1808 H. Davy determined the sp. gr. of
sodium by finding the proportions of oil of sassafras and naphtha to be mixed to
produce a liquid in which the metal remains suspended without sinking or floating.
This, said W. Ostwald, shows that H. Davy discovered the method of finding the
sp. gr. of solids by suspension, a method attributed by J. W. Retgers to A. Dufour.

Hardness. The mineralogist defines hardness as the resistance which a body
offers to the penetration of points or edges, and when one body A scratches another
body B, then A is said to be harder than B. A series of bodies A, B, C, . . ., is
arbitrarily arranged so that A scratches B, B scratches C, . . ., and not conversely.
It is assumed that if a point or edge of A scratches a plane surface of B, then a point
or edge of B will not scratch a plane surface of A. Taking a hint from R. J. Haiiy,
F. Mohs devised an arbitrary scale of hardness : talc, 1 ; gypsum, 2 ; calcite, 3 ;
fluorspar, 4 ; apatite, 5 ; felspar, 6 ; quartz, 7 ; topaz, 8 ; corundum, 9 ; diamond,
10. A hardness represented by the number 6 means that it can be scratched by
bodies above it in the series, and scratch bodies of hardness below it in the series.
H. Behrens made the scale more even and regular by using a series of alloys.
F. Auerbach prepared fourteen samples of Jena glass of different degrees of hardness,
and found that any one variety could be scratched by the other ; and he therefore
showed that if resistance to scratching be accepted as a definition of hardness, then,
of two samples of glass, that is the harder which makes the deeper scratch on the
other, when the test is made under like conditions.

E. Winkler, F. Grashof, H. Hertz, 8 etc., have studied the stresses which are set
up when two elastic isotropic bodies are in contact over a portion of their surface,
when the surfaces of contact are perfectly smooth, and when the press, exerted be-
tween the surfaces is normal to the plane of contact. H. Hertz showed that there is
a definite point in such a surface representing the hardness defined as the strength
of a body relative to the kind of deformation which corresponds to contact with a
circular surface of press. ; and that the hardness of a body may be measured by the
normal press, per unit area which must act at the centre of a circular surface of
press, in order that in some point of the body the stress may first reach the limit
consistent with perfect elasticity. If H be the hardness of a body in contact with
another body of a greater hardness than H, then for a circular " surface of pressure "
of diameter d ; press, p ; radius of curvature of the line p ; and the modulus
of penetration E,



The resistance to penetration defined in this sense is called by F. Auerbach the
absolute hardness of a body, and he has designed an apparatus for measuring this
property.

While hardness may be regarded as the resistance which a body offers to pene-
tration, the resistance is largely determined by the nature of the applied stress.
There is the resistance which a body offers to the abrasion of, say, a sand-blast ; the
resistance it offers to a press. ; the resistance it offers to a cutting tool ; resistance
to deformation, etc. Accordingly, there is a scratching hardness, an abrasion
hardness, a cutting hardness, an elastic hardness, a tensile hardness, etc. The
usual tests of hardness are static in character, but with kinetic tests, the penetrating
body is moving with an arbitrarily assigned speed. As a result, numerous definitions
of hardness have been proposed, and numerous instruments have been designed



454



INORGANIC AND THEORETICAL CHEMISTRY



for measuring hardness e.g. T. Turner's and A. Martens' sclerometers; 9 A. F. Shore's
scleroscope ; J. A. BrinelTs indentation test; W. J. Keep's drill test, etc. In
T. Turner's instrument the load required to make a pointed diamond cut a scratch
of given width is measured ; in A. Martens' instrument, the width of the scratch
with a definite load is measured ; in J. A. Brinell's apparatus, a hard steel ball
(say 10 mm. diam.) is forced into the smooth surface to be tested, and the depth of
the indentation measured; the quotient obtained by dividing the press, in k.grm.
by the spherical area of the cavity in sq. mm. is taken as a measure of the hard-
ness. P. Ludwik used a steel cone in place of J. A. Brinell's ball. According to
N. S. Kurnakoff and S. F. Schemtschuschny, Brinell's hardness number is really
a measure of the plasticity of a metal. A. F. Shore's scleroscope measures the
rebound of a hard body dropped from a given height on to the surface to be tested,
T. A. Jagger and H. C. Boynton used a drill with a diamond point, and W. J. Keep's
hardness-testing machine measures the number of revolutions required to make
a drill with a definite load cut a hole to a definite depth. A. Kiirth, R. P. Devries
and A. F. Shore, and R. Guillery measure hardness in terms of the tensile strength
or the lower elastic limit of a metal, for, as C. Karmarsch showed, the harder a metal
the greater its tensile strength.

Just as the press, of a gas, at a given temp., is proportional to the number of mols.
in unit space, so, with a homogeneous solid, it might be assumed that the resistance
it offers penetration by another body will increase as the number of atoms in unit
volume increases, and accordingly, S. Bottone 10 was led to postulate that hardness
varies inversely as the at. vol. ; and it is found generally that soft metals have a
large at. vol. and hard metals a small at. vol. For example, with S. Bottone's
scale of hardness :



Hard metals.


Metals of medium
hardness.


Soft metals.


Very soft metals.


At. vol.


H.




At. vol.


H.




At. vol.


H.


At. vol.


H.


Ni


67


1410


Ir .


8-6


984


Al .


10-6


821


Na


23-7


400


Co


6-9


1450


Pd.


8-9


1200


Cd .


12-9 760


K 45-4


230


Mn


6-9


1456


Zn.


9-1


1077


Mg.


13-8 726


Rb


55-8


.


Fe


7-2


1375


Pt .


9-1


1107


Sn .


16-1 651


Cs 71-0


.


Cu


7-2


1360


Au.


10-2


979


Th .


17-3 565


Ca


36-6


405


Cr


7-7





Ag.


10-2


990


Pd .


18-1


570


Sr


35-1


" "-""



Allowing for variations in the hardness due to impurities, it therefore appears that
hardness measures the resistance offered by the mols. of a substance to their separa-
tion by the penetration of
another substance. G. A.
Kenngott, A. Schrauf , C. Bene-
dicks, and J. L. C. Schroeder
van der Kolk have also tried
to establish empirical relations
between hardness and other
physical properties of the
elements. It follows that if
hardness varies inversely as
the at. vol. or atom cone.
the curve obtained by plotting
the hardness of the elements




90

Atomic



210



zw



150 It

weights

Fio. 10. Periodic Curve showing the relation between
the Hardness and Atomic Weight of the Elements.



a periodic character like the at. wt.-at vol. curve



against the at. wt. will exhibit
and this was shown to be the



case by J. R. Rydberg, in the curve illustrated in Fig. 10. C. A. Edwards also



THE ALKALI METALS 455

found the hardness of the solid elements to be a periodic function of the at. wt. ;
and that while there is a close connection between the hardness and at. vol. curves
there is an even closer parallelism between the hardness and absolute m.p. curves.
Practically, all the changes in the m.p. curves are reflected by corresponding changes
in the hardness.

The hardness of a metal is greatly influenced by the presence of small amounts
of other elements. N. S. Kurnakoff and S. F. Schemtschuschny have shown that
when two metals unite to form a solid soln. there is a continuous series of mixed
crystals, the curve of hardness is a continuous one which passes through a maximum,
and this point of maximum hardness generally corresponds with that of minimum
electrical conductivity. When the metals form alloys which solidify as a mechanical
mixture of two components the hardness curve is approximately a straight line.
When the metals form only a limited series of mixed crystals the curve of hardness
is a combination of the two forms previously mentioned ; while, lastly, if the two
metals form a definite chemical compound this may be either harder or softer than
the constituents. Again, in passing from either end of a series of solid solns. towards
the centre of the series, it will be found that the hardness, the limjt of elasticity,
and the tensile strength increase, but the ductility (as measured by the extension
and the reduction of area) and the electrical conductivity decrease. The m.p.
usually changes fairly regularly throughout the series. The facts are illustrated
diagrammatically in Fig. 11. A. Kiirth
showed that with the non-ferrous
metals copper, silver, nickel, alu-
minium, zinc, and tin the hardness
decreased as the temp, rose, but with
annealed steels 0*2 to 0'75 per cent,
carbon the hardness decreased as the
temp, rose to 150 ; increased a little
from 150 to 250; and decreased
between 250 and 500. C. A. Edwards

gave 0'07 for Brinell's hardness of

sodium, and 0*04 for that of potassium. 7 r-

/N *i v s\n P -t tUU /o ft L* UfifUUO /is fun

Sodium at 50 is soft, and at
ordinarv tPtrm it mav HP rmPP7Prl FlGk H- Relation between Hardness and
may be squeezed Qther ph ical p rop ertie a of Solid Solutions
with the fingers, at it is very ductile, (Diagrammatic),
and at 20 it is hard. 11 Lithium is

harder than sodium or potassium, but it can be scratched by lead, drawn into wire,
or rolled into thin plates. Potassium is harder than sodium, but it can be scratched
by lithium, lead, calcium, and strontium. Caesium is one of the softest of metals,
rubidium comes next ; at 10 rubidium is as soft as beeswax. N. Slatowratsky and
G. Tammann 12 tried to find if crystals of potassium and sodium soften in the
neighbourhood of their m.p. by heating them under a load near their m.p. Under a
press, of 27'5 kilograms per sq. cm. the depression 2 below the m.p. was 48 units for
potassium and 60 units for sodium ; 5 below the m.p. the depression was respec-
tively 8-3 and 36 ; 10, 7'7 and 24 units ; and 15, 7 and 18 units. C. A. Edwards
gives 0'07 and 0'037 respectively for the Brinell's hardness numbers of sodium and
potassium. According to J. E. Eydberg, 13 the relative hardness of the alkali
metals at ordinary temp, is indicated in the following table, which also includes
T. W. Eichards' values at 20 for the average compressibilities, jS, of the alkali
metals for press, between 100 and 500 megabars. The values of j8 represent the
average fractional changes of vol. caused by one megabar press, between 100 and
500 megabars referred to the volume V Q of the uncompressed substance dv/(v dp).
Mercury, the standard of reference, has j8=0'00000395 megabars at 20.

Lithium. Sodium. Potassium. Rubidium. Caesium.

Relative hardness . . 0'6 0'4 0'5 0'3 0'2

Compressibility, 3 . . '000009 0*0000156 "00003 17 "000040 O'OOOOGi




456 INOKGANIC AND THEORETICAL CHEMISTRY

The compressibility of caesium is very high, and appears to be connected vvith the
high at. vol. of this element. The change of volume 14 of potassium at atm. press.
is 0'02680 c.c. per grm., and of sodium, 0'02787 ; at 1000 kilogrm. per sq. cm. press.,
0*02368 c.c. for potassium and 0*02555 c.c. for sodium ; at 6000 kilogrm. per sq. cm.
press., 0*0134:7 c.c. for potassium and 0'01873 for sodium; at 12,000 kilogrm. per
sq. cm. press., 0'00642 for potassium and 0'01398 for sodium. The surface tension
of molten sodium in an atm. of carbon dioxide and at a temp, slightly above its
m.p. is 25*75 mgrm. per mm. according to G. Quincke, 15 and 27*23 mgrm. per mm.
according to E. B. Hagen. The surface tension of sodium at 90 in an atm. of carbon
dioxide is 293*6 dynes per cm. ; and of potassium, at 62, 411*5 dynes per cm.
The sp. cohesion a 2 =2a/S, where S denotes the sp. gr. of water and a the surface
tension is 64*4 per sq. mm. for sodium and 101 '1 for potassium. M. von Wogau
found the rates Of diffusion of the alkali metals in 01 per cent, mercury amalgams
are : lithium at 8'2, &=0*66 ; sodium at 9'6, &=0'64 ; potassium at 10'5, A;=0'53 ;
rubidium at 7 '3, &=0'46 ; caesium at 7 '3, &=0'45, where k denotes the quantity
of substance (mol. per litre) which passes per sq. cm. between two planes 1 cm.
apart with unit; difference in cone, on the two planes, per day. W. Wenz found
the velocity of sound in the vapour of potassium at 850 to be 652 metres per sec.
The melting points of the metals, reported by different investigators, vary
somewhat owing probably to differences in the purity of the specimens used for
the determination. 16 The best representative values are indicated in Table III.
According to P. W. Bridgman, the effect of press, in kgrm. per sq. cm. (1 atm.
=r033 kgrm. per sq. cm.) on the m.p. of potassium and sodium is as follows :

1 1000 2000 4000 6000 8000 10000 12000

97-6 106-2 114-3 129-1 142'9 155-0 166'6 177-2



Ka{



K



dB 8-6 8-1 7-4 6'7 6'2 6'8 5'3

. . 62-5 78-7 92-4 115'8 135'4 152'2 166'7 179'6

dd . 16-2 13-7 11-7 9-8 8'4 7*2 6-4



According to G. Tammann, the m.p., 0, of potassium is 0=59* 5+0*0146^
+0-0000007^ 2 . The changes in volume of the liquid dvJdT and of the solid, dvJdT
at the m.p. are respectively 0'000299 and 0'000216 for sodium, and 0'000360 and
0*000276 for potassium. P. W. Bridgman found the effect of press., p, in kgrms.
per sq. cm., on the m.p., 0, of lithium to be

p . . 1000 3000 6000 7000 8000

. . 178-4 182-1 188-8 194'6 199'4 3 2016

The results are thus characteristic of a liquid which contracts on freezing ; and
the mean value of the increase in vol. on melting is 0*006. The boiling points of the
elements also vary considerably. 17 The best representative values are indicated in
Table III. Lithium volatilizes at a bright red heat. According to A. Gebhardt,
the vapour pressure of sodium at

380 420 480 520 540 550 560 570
Vap. press. . 1*2 2-0 6'1 12'4 18'5 23'0 80'2 mm.

and, according to F. Krafft and L. Bergfeld, sodium boils in the vacuum of the
cathode light at about 140, and potassium at about 90. G. Bartha gave for the
b.p. of the metals in the cathode light : 575, lithium ; 420, sodium ; 370, potas-
sium ; 355, rubidium ; and 315, caesium. A. Kroner calculated the vapour
pressure of lithium to be

630 731 765 799 867 900 1000

Li . . . 0-05 0-60 1-0 1-7 3'7 5'5 760 mm.



Kroner gave for potassium and caesium

249-5 288-0 327'2



288-0 327'2 336'0 865-5 881'5 3986

Cs . . . 0-31 1-34 3-61 4-49 6'65 15'88 (397;

K . . . 005 0-21 0-73 0'95 1-51 2'84 4'01



THE ALKALI METALS



457



The value for caesium at 397 is by L Hackspill, who also obtained values for
sodium, potassium, caesium, and rubidium, and for the latter he gave



Rb



250
0-06



292
0-98



305
1-46



330
2-66



340
3'29



353
4-25



366'-



The latent heat of fusion 18 per gram is indicated in Table III along with the
best representative values for the m.p. and b.p. E. Griffiths found the latent heat
of fusion of sodium to be 27'1 cals. per gram ; I. litaka gives 2*60. A. Thum gave
32 '81 grm. cals. for lithium. E. Eengade gave for the heat of fusion of sodium
27 '21 ; potassium, 14*67 ; rubidium, 6144 ; and caesium, 3*766. Trouton's ratios
of the atomic heat of fusion to the absolute m.p. are respectively 1*69, 1*70, 1*68,

TABLE III. HEAT CONSTANTS FOR THE ALKALI METALS.





Lithium.


Sodium.


Potassium.


Rubidium.


Caesium.


M.p.


180


97-6


63-5


39


28-5


B.p. . .


+ 1400


877-5


759


696


670


Heat of fusion












per grm.


.


27-21


14-67


6-144


3-766 Cals.


At. ht. of












fusion


0-941


0-9811


0-1728


0-0802


0-0522


Entropy at 25


7-6


12-2


19-7





~



and 1*66. P. W. Bridgman found the latent heat of potassium changed from 5*51
kilogrammetres per gram to 5*81 at 1000 kilogrms. per sq. cm., while the corre-
sponding values for sodium changed from 12 '90 to 12 '46 ; there is a maximum of
6*22 with potassium at 4000 kilogrm. per sq. cm. press., and a minimum of 11*93
with sodium at the same press. ; the value for potassium at 12,000 kilogrms. per sq.
cm. press, is 4*83, and with sodium 12*72 kilogrammetres per grm. The specific
heats 19 are indicated below. The sp. ht. of sodium from



191 to 83
0-2433



-79-5 to 17'
0-2830



and for potassium from



191 to -80 C
0-1568



-78-5 to O c
0-1662



to 20
0-2970



to 22-3 c
0-1876



to 56-5 to 78
0-3071 0-3191



to 97-63
0-3290



to 56-5
0-1922



to 62-04
0-1980



to 78
0-2137



to 100
0-3330



to 100
0-2170



to 157
0-3330



to 157
0-2245



The sp. ht. of rubidium between 20 and 35 is 0*07923, and of caesium from to
26, 0*04817. K. Lammel's value for lithium at the m.p. (193) is 1*3 ; and for
sodium at the m.p. (100) is 0*35. E. D. Eastman and W. H. Kodebush's value
for the atomic heats of sodium at the absolute temp. T is



TK.



64-6
4-52
4-47



71-1

4-77
4-71



84-6
5-08
5-00



156-8
6-02
5-82



181-7
6-15
5-91



234-7
6-43
6-07



293-5
6-79
6-29



similarly for potassium



TK.

G p

C v



68-6
5-76
5-69



76-0
6-78
5-70



87-0
5-96
5-86



101-8
6-06
5-95



119-3
6-23
6-07



199-5
6-72
6-37



286-7
7-10
6-52



For each metal, therefore, the at. ht. C v rises well above the normal value 6.
For potassium at the m.p. L. Protz gave for the atomic heat of the solid C7 P ==7'81,
and C,=7-26 ; and of the liquid, C p =7-96, and C v =7'25 ; similarly, E. Griffiths
gave for solid sodium C p =7*49, and C f ,=6'71 ; and for liquid sodium, (7^=7*43,



458



INORGANIC AND THEORETICAL CHEMISTRY



and Ct;=6-61. P. Giinther measured the sp. lit. of sodium between 186 and
149, and found the results agreed with those calculated by P. Debye's formula
on the assumption that j3v=125. A. Bernini measured the sp. ht. of sodium
and potassium ; and A. Thum that of sodium and of lithium. The latter gave for
sodium 0-1743 at 273, and 0'3369 at 150. E. Rengade gave for the sp. ht. of
solid sodium 0'281 1+0*0002330 and for the liquid 0'330 at 98 ; for solid potassium,
0-1728+0-0001420, and for the liquid 0*1422+0-0006680 ; for solid rubidium,
0-0802 +0-0001530, and for the liquid 0-092+0-0006680 ; and for solid ceesium,
0-0522+0-0001370, and for the liquid 0*0604-0*0000340. J. Dewar found that
between the b.p. of hydrogen and nitrogen about 223, or 50 K.,



Sp. ht.
At. ht.



Lithium.
0-1924
1-35



Sodium.
0-1519
3-50



Potassium.
0-1280
5-01



Rubidium.
0-0711
6-05



Caesium.
0-0513
6-82



R. Lammel found the at. ht. of lithium rises to 9'45 at the m.p. 193 ; and that of
sodium to 8*23 at its m.p. The at. ht. of sodium rises from 6'5 at to 7 "5 at 96
the theoretical limiting value at constant volume is nearly 6, which is lower than
the value calculated from the formula C p C r =a 2 Tv/p. I. litaka gives for the
sp. ht. of solid sodium at its m.p. 0*330, and of the liquid at the same temp., 0'347,
and the respective at. hts. are 7*59 and 7'98. A. Joannis gives for the sp. ht. of
solid potassium 0'166, and of the liquid 0'25. W. Nernst obtained the relation
CzCsQ/Tm, where C 2 and C 3 respectively denote the mol. ht. of liquid and solid
at the m.p., T m the absolute temp, of the m.p., and Q, the latent heat of fusion.
I. litaka found the relation is not good with the metals bismuth, tin, mercury,
zinc, lead, and sodium, and also with sulphur for the quotients are respectively
4-71, 3-15, 2-40, 2-19, 119, 1*61, and 1-08. A. Thum gave for lithium,







i/, j. j-t/j j.




Sp. ht.
At. ht.
At. ht.


273
. 0-5303
. 3-7121

. (0-9086)


200
0-5890
4-1230

(2-5837)


100
0-6809
4-7663
(4-1965)



50
0-7315
5-1205
(4-8755)





0-7854
6-4978
(5-5657)



50

0-8425

5-8971

(6-3441)



150 180

0-9658 1-0055

6-7606 7-0385

(8-4539) (9-3163



0-6
0-4
0-P

















x














x












^


^










^


***










/


^














f

































The numbers in brackets are by R. Lammel. 20 The sp. ht. of lithium is the largest
of that of all the elements. R. Lammel represents the sp. ht. of lithium at

by the expression 0'7951 +0*00206320 +
0-0000025080 2 +0-0000000142070 3 . The observed
values are 0'3693 at 200 ; 0*5997 at 100 ;
0-6964 at -50; 0'7951 at 0; 0'9063 at 50;
1-0407 at 100 ; and 1'3745 at 190, so that the sp.
ht. curve shows a slight turning-point at about 60,
Fig. 12, possibly corresponding with a change from
one allotropic form a-lithium to another form
j8-lithium. The sp. ht. of sodium was found by
E. Griffiths to increase with temp., but its absolute
value at any given temp, depends on the previous
thermal treatment of the metal. At temp, below
about 60, the sp. ht. is greater when cooled by

quenching than when annealed. The sp. ht. of liquid sodium varies with temp, such
that the temp, coeff . is 0'00034 per degree ; I. litaka found very little variation in
the at. ht. of solid and liquid at the m.p. Hence, E. Cohen and G. de Bruin argue that
under ordinary conditions two allotropic modifications a-sodium and j8-sodium
are present in proportions dependent upon its previous thermal treatment. The
transition point not yet determined lies between and 90 possibly it is near
75. Dilatometric measurements show that the high temp. j8-form has the lower
density ; and since the rapidly quenched metal melts more quickly than the slowly
cooled metal, it is inferred that the change from the a- to the jS-form is attended
by the evolution of heat. Similar observations show there are two allotropic



-100 -100 100 200

Temperatures

FIG. 1 2. Specific Heat of Lithium
at different Temperatures.



THE ALKALI METALS



459



forms of potassium a-potassium and /3-potassium with a transition point in the
neighbourhood of 59 '7. W. Holt and W. E. Sims 21 found that potassium is ductile
and soft at 54 '5, but it becomes quite brittle and has a conchoidal fracture at
60-5.

The best representative values of the varied measurements 22 of the coefficients
of cubical expansion of the five alkali metals are :



Lithium.
0-000190



Sodium.
0-000274



Potassium.
0-000283



Rubidium.
0-000338



Csesium.
0-000338



The coeff. of expansion of sodium between and 95 is uniform and proportional
to the temp., and the coefL is greater for the liquid than for the solid. According



Online LibraryJoseph William MellorA comprehensive treatise on inorganic and theoretical chemistry (Volume 2) → online text (page 80 of 156)