ON THE ST~~~T~RE OF CARBON,

I_-

An X-ray invest.igation of some (( amorphons ))
carbons and graphites tias revealed certain new fea-
tures which it is the purpose of this note to cle.strihe.
As a pr$iminary to the wider prohlems of carbon
structure and the dependence of structure on [.he
origin arid treatnient of the material, a tletail(-il qum-
titntive study of a single. carbon was made, in ortlcr
to find out just how much information the diffuse
x-ray diagram could be made to yield.  'I'he niate-
rial was prepared by pyrolysis of IjoI~vit~~~ider~c
chloride at 1.0000, and is more thin 11'3 :/o carbon.
The foliowing results were obtained.

65 7; of the carhon is in 1 he form of highly perfect
graphite-like layers.  The iiiean tliaaicter of these
layers is oily 16 A. Of Llie grapliite-like layers,
about 45 O/;, show no 1iinLu:il orientation and 55 :io
are groiiped in parallel pairs with spacing 3,7 A, tlic
number of ~~:irallel-lay~~r groups con t :iitiing more
than 2 layers per group being w:ry sinall.

'The remaining 35 7; of the carhun is iit a form so
disordewtl ;is to gisc only ii gaz-iikc c~~i~trit~~iti~Ii to
the total X-ray scattc:rirlg.
Applic,ation of tht: 'f:ourier transforin to tile very
extensive low angle s~~~~t.t~!rirl~ reveals a iiit:;tn in tcr-
particulate distance of 25 A.

*
**

The details of thic structure arc., of cowbe, peen-
liar to the carbon inves Ligated, but tlie I csult\ sugges-
ted two properties nhicl-1 it miqht be of interest to
investigate in otlier caibons.  'I'he first, a ~ery for-
tllnate result, war; the sharp st:paration observed
between the ordcred and disorclwed parts of the
siJpUcttire.  It miglit haw been tupected that in a
carbon stiowiiig such a low degree of crystallinity
*here would be present all degrees of partial disorder,
hut this is not so.  *\part from tlie sinall, perfpct,
graphite-like layers, only itigtily disorckrcd material
1s Present.  It thus seeins clear that the proportion
Of ordered and disordered material is an important
feature of the structure of suc~i carljons.  ~iie otlicr
Point of interest is tlie spacing, 3,7 A, ohserved bet-

mien pairs of small pnrallcl grnphi te-like laycrs, the
spacing in true graphite being 3,35 A.
Investigation of a number of ot?ier (( a~~ior~~lio~is B
carhons.sliowec1 that the sharp separation hetween the
ordered and disortlered parls is of general occiirencc.
A11 the X-ray cliagratiis obtaiiwd can be s~~tisf~~ctoril~
interpreted by supposiiig t.he esistcnce only of small,
pcrfrxt, graphile-like layers togcther with son~e
highly disordcreci material.
For carbons of \t.idely dilT't:rc~nt origin there is a
general relatio~istiip between the di:trncter of the
graphite-like layers arid the proportion of amor-
phous material. This is shown in figure 1. For car-

1:ig. 1.

bons wliicli contain a measurable proportion of disor-
dered material the layer diameter is less than 25 A.
(l'he method of X-ray analysis used is directly
applicable only to carbons which are nearly pure.
'l'his explains the absencc, in figrire 1, of carbons
which are inore than 50 9'0 disordered or in which the.

layer cliametcr is less than 12 A.).

574

ROSA4LIND E.

FRANKLIN

*
**

For the same carbons, the inter-layer spacing and
the mean number of layers per parallel group (as indi-
cated by the form of the (002) hand) have been
measured.  It is found that for fe-x (less than 5)layers
per group tlie spacing decreases sharply a\ the niini-
her of layers increases.  For 2 layers it is 0,7 A, for a
mean value of 2-3 layers it is 3,6 A, and for carbons
having 4 to 5 layers per group it is 3,44 to 3,425 A.
With further increase in the number of layeis per
group the change is small.

It will he noted that this value, 3,it.t A, is still
rnarkedly different from 33, the spacing in graphite.
The carbons mentioned so far are all suhstant
siiow only 2-dimensional order.
pliitc-likc layers are grouped parallel to one another
and ~~~~uidi~taiit, but are 110 t othm\ ise mutually
orieii tatect.  'The X-ray (1i'igJ mis of such material\
 how only (0 0 I) crystalline reflections and 2-diinen-
sional (I'i k) bands of the type dew Ibed hy JViairen
(i'hys. Ikv. 59, 6'33, 194 1).  During fur tlicr graph-
tisation -- induced by tfierm:il treatment -- tlier
gr;iilual and continuow dofortiiation of each (It X)
band tending towards thc transformation of tlip 1)and
into a sene5 of (h k () cr~~lC>lline reflections. \tItich
arc at first very   lie hiit I)ecomc iiiore ihnrply
rleftrietl a5 the pr pro( reds. All i arhvii, wftrth

That is,

SfMC117$ !lo! [('S', !hill1 3,4/ \.

\\'ht

tl1e (h h) I,nnds srti In, sllv\\lllg th

ckgr te 01 i-cliruensioual oi der, the rntei-layer spacing
:tyatii cicciea\ei. A11 ( arl)ons shonii~g dcior ilicd
(it ic) ltaricls or diffuse (It X L) reflections liave app-
rriit inter-laycr spavin+ intermediate hiitnwn 3,6 1 .i

PI& 2.

and 3.35 A.  In reality it appears that tltc c'hanqe in
spaciity from 3,44 to 3,35 Ai is disconfiiiimus.  When
riei~~i~ou~ irig layers are niulually orientated as in a
graphite crystal the spacing is 3,35 -i.  When this

orientation is destroyed by a random translation or
rot:ition of a layer in its own plane, then the spacing
is increased to B,44 K (see fig. 2). 'The apparent
intermediate spacings obwvcci are in reality average
values. This is shown by the figure 3. From the

311

1

343 1

13 3 4 5 (i 7 --L.i t; '3

Fraction of I n y vrs [lis, 111 a cetl

Fig. 3.

form of tire tlefonrit:il (Mi) 1t:tntls (or tiifTuse (h k 1)
rdlcclions) it is pi)ssiiblt: to calculate the pr~)I)ortioI~
of layers at which a translntion or rot::tion, cquira-
letiL t,o a heak in tlie crystnllinc ordor, or N mistake ))
occurs (fig. 2). '['his ca1cul:itinii has ixcn carried
out, and tire app:irent in tt.r-.i:\ycr spcing (given liy
the (002) he) nit.:\surcrl for a niinilier of carbons of
dilferent origin 1ieatc.d to tcinperatiires 11et~~1i
2.300 an(\ 3.OOOo.C.  In Figure 3 it is seen that the
apparciil intcr-layer spac,ing is a function of the pro-
portion of (( rriist.akes )) or displaced layers in thi:
structure. For a JIlktake at, each Iayer   that is, in
tlii: absence of any crystQlline order -- the spacing
3,.'lt-i A is o1)tainett.  The spicing ohtaintxi by i\u'ehIl
and fiiley (I'roc. Phys. Soc., 3. 477, 1945) for the
best-cxystnllised natural graphite available to them
is 3.%rt A, anti this ~aIuc has been confirmed during
the course of the prkscnt work.

*
**

Up *to the present no attenipt, has been macle
correlate the structural factors cjescritled above with

the chemical behaviour of the c:trbons, but tliere are
certain obvious lines which it is Iioped to follow in the
near futrtre.  In the case of the (1 amorphous )I carbons
1t w~iifd be of interest to investigate n.hrtlier the
ordered or disordered prt is at tacked preferentially
in combustion or other rcactions. In thf, case of
c:nhons showing some 3-dimensional order it may be
asked whether or nol phes showing the 3, 1% x
spacing are substantially morc nvnilahle to attack
than thosr which have the true graphite spacing of
3-32 .I.

D TSi:l!SS1ON

SUR 1.F: S 1)E US COXM CX 1 CAT IONS PRECI? 11 RN`I

11. Iiiley.

       In an amorphous structure, it is not pcrinissible
to leave valency bonds unsatisfied, in other wortls, it. is not
nlerely a chaotic jnintrlc of carhon atonis. X striicturc is
ilivolved, hiit one wiiicii is riot sufficiently ordered to give
coherent scatlering of X-rays.  Miss 1;ranklin's pape'. siiggcsts
that we must identify the sugsested t.hree-tlinicllsio
cross-linked, aromatic structure with the (liiortitw?il pti
the samplcs ~l~clie~l.

XI. 1l:inghiiIn. . The stuctiec of fnel carlioiis by X-rup
methods matte by Itiley, f'ranliliu ani1 others have irirlicatxct
that ordcrcd crystallites of graphitc form oiily part of the ~Iiolr
structure.  It is :i little tenipting to ideritify thi cf aniorphonx ,)
carhon as the portion which contritiutes most to t he '( reticti-
vjty )n.  Instead of dniirg so, however, I would like Lo tiiTer the
sirgj+xition that the iirost reactive c:ti~iio:i aloin.; arc tiltJ.;C in

        1 tlii ainorphuuc (or otlicr metastaLle) state
aiid the, stsblc gr:iphitic i~hase.  \V!iilst it is Im!iwi th:it ahn\-e
a ccrt:iiti teni~~er:iiure tlic. orclcrecl portion iircre;a.\cs at tllc
(t:ip<:i~~ of tlie disrrrtlere~i, i! is \vi:il-iiigh irnpossibk tr, cr1it:tiri
data 3s to the rcile at \vhich thi> ~~xoc~ss
brcansc one cannot, in praciil.e, readily

nn itistatit of time.  Sotnc data of 1-1. L. IWey in:ilte it pro-
baljle tlioL tlie change :tsrnci:itetl nrith a market1 rire {if tctttpc-
raturc OCC~I~S fairly rupiiily --- in :I ni:ittcr of riiinutos, perli:ipq,
but nnt of days.

In consitleritig hov the more ordered structure is evoivcil
from tlie les ordt~xtl, it appenrs iiecess:iry to suppose that,
migration, either of carbon atoms or of small groups thereof,
can take place. `I' se migrating atonir (or grot:j>s) \sill
surely contril)ute ve  much lo the rcactivity towards gascs.
X would ernphasise  c i~t~~~ort~~nce of this transition stat.c,
bearing in rniricl the distinctiori inade by Crone hetween tlle
i'enctioris inidergotie by :i ftief 1,iirticle xfiich is I~eing 1ie:itcd up,
anti those associated with its burniiig out ;  with technical
fuels the carbon gcnerai~y has not iiecit previoiisiy heatcct to
tempernturcs such 51s are attained in fuel Jmls.

1LlI`C Of a S>lIIlplt! Of IX~h~JIl fl'oiil One SteLld

It would pixhaps he interesting (lhoogh difficult) to study,
from the theoretical point of view, the magnitudes of the
energy barriers separating (say) mnorphous from graphitic
carbon, making difTercnt assumptions as to Ihe size of the
migrating nnit. Tt might be found, for ezcainlile, that the
migration of individual c:irbon atoms is the rnosk probable
mechanism of tiic ch:~tige ; or, alternatively, that several
atoms move $15 a group.
The progress of the cornhnstioii of a sarnple ol conirnercial
earlmti must ]irovide coridilioiis w0ic.h fiiimczr tlie pro~i~ict ion
of these migrating fragments. We are thus led, as in the study
of iiornogencous g:is-pfi:ise rcactions, to think in terms of
the e1enient;rry evciits whic,lr may control (n) the formation
ant1 (6) the iIest~ticli~~ri OF these ~~i~~/rn~~~z~~ reavtion centres.
Viewed from this slnnd {Joint, the i rcactioii vessel )) berouies a
t~~o-~iinetision~r~ motrile siirixce phase iii which tltc simple

1.1:; the action of ordin:iry Iran der IYxils adsoriitit-rn.  One
must reineiirbet. that in such a 1)hase -- aiiproxirn:iting to a
condriisetl phase -- disailru~ion of citergy by three-body colli-
sions would IJC frequent.

g:lS-llhaSe SIJeCiCS (0,. co, etc ... ) al'e SoIIl~\lh~l\. CkJll~elitr~itCd