
Graphitizable
coiled
carbon nanotubes
101
of
the five-membered ring [Fig. 17(d)]. Note also
that only six new
C,
units are shown in Fig. 17(c)
with
respect to Fig. 17(b).
Once the pair of sites is disconnected from the
growing tubule [Fig. 17(d)], an orthogonal
C2
unit
is
inserted below the five membered ring
[4"
in
Fig. 17(e)]. The latter inserted
C2
unit and the remain-
ing two cis-polyacetylene
C2
segments are finally
displaced by the arrival two orthogonal
C2
units
[Y
in Fig. 18(f)].
The carbon atoms linked to the remaining twenty
coordination sites have now finished to rearrange to
a
fivefold symmetry [Fig. 17(f)]. Consecutive inser-
tion of the five orthogonal
C2
units
[lo,
3"
and
5")
displaced from the catalyst by the arrival of five new
Orthogonal
C,
units closes the seven-membered ring
and completes the knee [Fig. 17(g) and (g')]. Further
growth will yield a (5n,5n) tubule. It will proceed
as already explained in Fig.
15.
Figure 17(g') is
a
Schlegel diagram explanation of the (9,O)-(
5,5)
knee,
equivalent
to
Fig. 17(g). In that diagram
it
is possible
to identify the
C,
units introduced at the different
steps [Fig. 17(a)-(g)] by their numbering.
3.2.2.4
Growth
mechanism
of
a
(5n,5n)-
(9~0)
knee,
involving
from
20n
to
24n
coordina-
tion
sites
of
the
catalyst.
The explanation given
here to pass from
a
(9n,0) tubule to
a
(5n,5n)
tubule
can also be used backwards to pass from a (5n,5n)
tubule to a (9n,O) tubule. The progressive steps are
illustrated in Fig.
18.
The starting point of the knee
is again the blockage
of
tubule growth
at
the seven-
membered ring
[
1"
in
Fig. lX(a)J. The later formation
of
the five-membered ring is only
a
consequence of
tubule growth blockage at the seven-membered ring.
Secondly, probably because of the large space created
on
the other side of the tubule by the elastic bending
[Fig. 12(b)] after tubule blockage,
a
cis-butadiene
[2" in Fig. 18(b)] can be inserted instead
of
the usual
orthogonal
6,
units. This cis-butadiene arriving with
four new coordination sites [black points on
Fig. 18(b)] will be the head of
two
cis-polyacetylene
chains
C3"
in
Fig. 18(c)]. (The freshly arrived
C,
segments
of
the cis-polyacetylene chain, not yet
inserted in the tubule, are represented by dotted lines
for the sake of clarity.) These two chains are started
at the five-membered ring [Fig. 18(e)]. The insertion
of
that cis-butadiene also disturbs the other coordina-
tion sites,
so
that
four other cis-polyacetylene chains
[4" and
5"
in Fig. 18(c)] are also inserted into the
growing tubule. The other logical growing steps
of
the (53-(9,O) knee are very close to the mechanism
explained for the
(9n,0)
tubule. The insertion of the
first cis-polyacetylene units
[3"
in
Figs 18(c) and (d)]
and the coordination of the first parallel
C2
units
[6"
in Fig. 18(d)], followed by the insertion
of
the cis-
polyacetylene units [4O in Figs 18(d) and (e)] and the
coordination of parallel
C2
units
[7"
in Fig. 18(e)]
leads to positioning of the five membered ring. The
closure of that ring, followed by the insertion of the
cis-polyacetylene units
[5"
and
8"
in Figs 18(e) and
in Fig. 18(f)] leads to positioning of the seven-
membered ring. The closure of that ring, during
a
normal (9n,O) tubule growing step [Fig. 19(g) and
(g')], completes the knee and the following steps will
be the growth of the
(9n,O)
tubule proceeding as
explained in Fig.
16.
Figure 18(g') is a Schlegel
diagram of the (5,5)-(9,O) knee, equivalent to Fig.
18(g). In that diagram
it
is
possible to identify the
C,
units introduced at the different growing steps
[Fig. lS(a)-(g)] by their numbering.
An
important conclusion obtained from the model
based
on
the variation of the number of coordination
sites at the catalyst surface is that all double bonds
can be localized on the (9n,0)-(5n,5n) knee and
connected nanotubes.
In
fact, in that model, only
single C-C bonds are formed and there
is
no double
bond formation during the nanotubule growth.
All
of the double bonds are already localized on the
C,
units or on the inserted cis-polyacetylene chains.
As
seen from Figs 17 and 18, there is no double bond
on the sides
of
the five membered ring and there is
only one double bond
on
the seven membered ring.
It should also be pointed out that localizing the
double bonds on the (9n,0)-(5n,5n) knee and con-
nected tubules was
a
very difficult task before the
establishment of this model. Once the double bonds
are localized, after replacing the vacant bonds of the
(9n,0)-(
5n,5n)
knee of Fig. 17(g') by hydrogens,
it
is
possible to have a three dimensional view
of
the knee
(Fig. 19). The introduction of the parameters of
Fig. 19 into
a
more sophisticated program can also
be used in order to minimize the energy and simulate
the real knee angle[18,19].
Concerning the multi-shell tubules, the graphitic
layers
of
the growing nanotubule described are sup-
posed to grow on the same catalyst particle, at the
same time
as
the inner layer does. This
is
in agreement
with the observations of Fig.
11,
where it is possible
to see that the diameter range of the young tubes
(1
minute) is about the same as that of the
old
ones
(20
minutes). Moreover, it has also been observed that
during
a
long exposure time (5 hours) of the tubules
to the reaction conditions, only amorphous carbon
is deposited on the outer layer[4,5], and no tubes
with larger diameters were observed.
As
the diameter of the catalyst particle is supposed
to be close to that of the single-shell tubule[20], or
to that of the inner tubule[8], the number of graphitic
layers might depend on the flow rate
of
acetylene at
the catalyst particle. The graphitic layers are supposed
to be formed by the
Cz
units formed
on
the catalyst
particle, exceeding those needed for the growth of the
multi-shell tubule inner layer. This generalisation
to
multi-layer tubules
is
just a hypothesis, since we do
not have any experimental proof yet.
4.
CONCLUSIONS
The building of knees, tori
and
helices
is
described
bv
a
simple formalism.
(f)] and coordination
of
three parallel
C,
units
[So
Relationships are established between the tubules