Home Search Collections Journals About Contact us My IOPscience

An investigation of the structures and molecular dynamics of natural waxes. I. Beeswax

This content has been downloaded from IOPscience. Please scroll down to see the full text.

1988 J. Phys. D: Appl. Phys. 21 1421

(http://iopscience.iop.org/0022-3727/21/9/016)

View the table of contents for this issue, or go to the journal homepage for more

Download details:

IP Address: 194.27.128.8
This content was downloaded on 05/11/2014 at 13:00

Please note that terms and conditions apply.

J. Phys. D: Appl. Phys. 21 (1988) 1421-1428. Printed in the UK

I Basson and E C Reynhardt

Department of Physics, University of South Africa, PO Box 392, Pretoria 0001,
South Africa

Received 1 February 1988

Abstract. A detailed NMR investigation, supported by DSC and x-ray diffraction

measurements, of the combwax of the African bee Apis mellifera adansonii is
reported. Proton spin-lattice relaxation times in the laboratory and rotating frames,
as well as the proton spin-spin relaxation time, have been measured as a function
of temperature (333 K > T > 110 K). The liquid content of the wax has also been
determined as a function of temperature by employing a simple pulse technique.
The motional parameters associated with the reorientations of the methyl groups
and the chains have been isolated and compared with these parameters for similar
motions in Fischer-Tropsch waxes. The T,? results revealed a minimum which is
associated with the dangling motion of cham ends in the amorphous zone of the
wax. The relaxation results strongly suggest that beeswax is branched to a much
higher degree than Fischer-Tropsch waxes, including oxidised waxes. The high-
resolution 13C spectrum of beeswax in the solid state shows that it resembles
oxidised Fischer-Tropsch hard wax closely. The major difference is that a higher
percentage of carbon atoms are involved in ester groups in beeswax. Oxidised
hard wax contains a higher fraction of carbon atoms with double carbon-carbon
bonds. The average chain length in beeswax, determined by ebullioscoDic
methods, is 40 carbon atoms.

1. Introduction dependence of the motional behaviour of the chains in

medium wax, of average chain length = 28 and A N =
Apartfromtheobvious uses of waxes in producing 4.12, differs significantly fromthebehaviour in octa-
candles and polish, they form important materials in cosane (n-C2J. A N is the root mean square deviation
the manufacturing of products such as ceramics, food, fromtheaverage N. The resultswerefoundtobe
rubber. timber. textiles, tobacco, paints and plastics. consistentwithamodel in which chains of unequal
The collection and interpretation of fundamental data lengtharepacked in anirregularstaggeredarrange-
on waxes are therefore a primary objectiveof industries ment. Apart from the dynamics of the crystalline frac-
using waxes. A great deal of research has been carried tion of hard wax (N= 50, A N = 15.7), the narrow com-
out on syntheticFischer-Tropschwaxes.Structural ponent of the proton line, usually associated with the
aspects such as crystallinity and branching of the chain- mobile amorphous zone of the wax, was studied as a
like molecules have been studied by means of spectro- function of temperature. As far as the physical prop-
scopic and diffraction methods and have led to a struc- erties of the wax are concerned, the importance of the
turalmodelof Fischer-Tropschwaxesconsisting of melting process of the wax. which occurs over a wide
crystalline, amorphous (chain ends), rigid amorphous temperature range, was emphasised. A similar study of
(defects due to branching) and mobile amorphous liquid oxidised hard wax (Reynhardt 1985b) revealed that the
phases. In addition the dependence of hardness, con- activationenergiesassociated with methylandchain
gealing point and density on molecular mass have been reorientations are lower than in unoxidised hard wax.
reported (Le Roux 1969a. b, c,1970, LeRoux er a1 It was concluded that the splitting of the chains during
1976). Thetemperaturedependence of thethermal the oxidation processis an important factor contributing
expansion and thermal conductivity of some waxes has towardsthehigher mobility of the chains. The tem-
been measured (Le Roux er a1 1974). Wide-line NMR perature dependenceof the cell dimensions of the latter
investigations of Fischer-Tropschwaxeshave been two waxes has also been reported (Reynhardt 1986).
reported by Lourens and Reynhardt (1979) and Reyn- Althoughthesynthetic waxes havebeenstudied
hardt(1985a,b). It was foundthatthetemperature fairly extensively, the other main group of waxes, the

0022-3727:881091421 08$02.50 0 1988 IOP PublishingLtd 1421

I Basson and E C Reynhardt

natural waxes, has attracted relatively little attention. signal is entirely due to the relaxation of the liquid
The naturalwaxes can be divided into three subgroups, component.Ateachtemperaturethe average of 50
namelyanimalwaxes,vegetable waxes andmineral signals was obtained with a repetition period of 10Tl.
waxes. The amplitude of the free induction decay above the
Witha view toobtaininginformationaboutthe melting point of the wax was assumed to correspond to
structures and moleculardynamics of the naturalwaxes 100% liquid. The percentage liquid at a temperature
and tocomparethese waxes with synthetic Fisher- below the melting point was obtained by expressing the
Tropsch waxes, a detailed investigation employing wide-amplitude of the FID at the temperatureas a percentage
line and high-resolution NMR in the solid state, x-ray of the amplitudeof the liquid signal at the sameposition
powderdiffraction,differentialscanningcalorimetry on the timescale. It was not necessary to make a cor-
(DSc) and ebullioscopy, was initiated. One wax from rection for the effect of the change of the Boltzmann
each of the subgroups was selected. In this paper the factor on thesignal with changing temperature over this
results of the investigation of beeswax are discussed limited temperature range.
while the results of similar studies on carnaxha wax and Thehigh-resolution I3C solid-statespectrum was
montan wax will be presented in parts I1 and 111 of this obtained at room temperature by employing the com-
series of papers. binedtechniques of magic anglespinning,proton
decouplingandcrosspolarisation.Thespectrometer
was operating with a proton decoupling field of 15 G
2. Experimental details and a sample spinning frequency of 2.2 kHz. The FID
was accumulated 82990 times with acontacttime of
A sample of combwax of the African bee Apis mellifera 1 ms and a repetition period of 4 S .
adansonii was supplied by Sasol Technology, Sasolburg, The ebullioscopicexperimententailed thedeter-
South Africa. mination of the boiling point elevation constant K , of
The DSC thermogram was obtained with a DuPont the solvent toluene, calibrating the apparatus with the
1090 instrument. Cooling
and
heatingrates were n-alkane C36and measuring the boiling point elevation
0.17 K S-'. A T as a function of the mass of beeswax present as a
X-raypowderdiffractogramswererecordedona solute in a known mass of toluene.
Seifert MZ IV diffractometer with Cu Ka radiation and
a nickel filter. An accelerating potential of 40 kV and a
filament current of 35 mA were used. The width of 3. Results
the slits was 0.02 mm. The low-temperature equipment
consisted of a PAAR variable-temperaturecamera 3.1. DSC
cooled by a regulated stream of liquid nitrogen. The
Thethermogramfor beeswax,shown in figure 1,
camera was evacuatedandthetemperatureatthe
exhibitsawidemeltingtransition consisting of two
sample regulated by a PAAR TTK temperature
partially resolved peaks. The melting process starts at
regulator.Spectrafor beeswaxand Fischer-Tropsch
approximately 300 K and the two peaks occur at 329
medium wax were recorded at room temperaturein the
and 337 K.
range 55" 3 2 8 2 18". The 28 positions of the (110),
(020) and (200) reflections of the orthorhombicunit cell
were determined as a function of temperature in the
3.2. X-ray diffraction
range 340 K 3 T 3 80 K.
All NMR measurementsweremade on aBruker Figure2showsthepowderx-raydiffractogramsfor
CXP 200 high-power spectrometer. The temperature beeswaxand Fischer-Tropsch medium wax atroom
dependence of the spin-lattice relaxation times T , and
T I P ,the spin-spin relaxationtime T 2 andthe liquid
content of the wax was determined in the temperature 8 )I) 1
range 333 K 3 T 3 110 K. A saturating comb pulse
sequence, followed by a solid echo, was used to measure
T1. T,, values were obtained from the t-dependence of
the free induction decay (FID) following a spin-locking
pulse of magnitude H 1 = 24.2 G and duration t. T 2was
measured by employing the solid echo technique. Inall
the relaxation time experiments information about the
magnetisationcomponent associated with the solid
phase of the wax was emphasised by applying a linear
apodisation to the tail of the free induction decay.
The liquid content of the wax was determined as a 1
function of temperature in the range 290 to 380 K by
employing a single90" pulse and measuring theintensity
of the free induction decay at -100 p , i.e. when the Figure 1. A DSC thermogram for beeswax.

1422
 NMR of beeswax

Medlum wax The length of the b axis is 7.27 t 0.01 A at 80 K and

increases uniformly to 7.40 % 0.01 A at 240 K. These
values yield an expansion of 1.6% for a and 1.8% for b
in the low-temperature region, i.e. prior to the starting
of the melting process. Above 270 K the length of the
b axis increases at
faster
a rate with increasing
temperature.
The length of the c axis was not determined. Due to
the length of this axis (for Fischer-Tropsch hard wax
c 2: 142 A), the observable 001 reflections occur at very
low 28 values (28 < 6"). In this region the influence of
axial divergence becomes considerable and complicated
corrections have to be applied to the data. In the case
of Fischer-Tropsch hard wax, the length of the c axis
25.43 32.80 LO20 47.60
shows very little variation as a function of temperature
(Reynhardt 1986).

Beeswax 3.3. NMR

Twelve peaks were identifiable on the high-resolution

13Cspectrum shown in figure 4. The peaks at -10 and

25.40 32.80 46.20 47.60

28 Ideg)

Figure 2. X-ray powder diffractograms at room

temperature for medium wax C(28) and beeswax.

temperature. The temperature dependenceof the unit

cell dimensions a and b is shown in figure 3. The length
*
of the a axis is 4.880 0.010 A at 80 K and increases
*
almost linearlyto 4.996 0.010 A at 295 K. In the range
300 to 340 K there is a slight decrease in the value of a.

, . . . l . . l l . i l l . l . . 1 . . I I . . L
4
80 120 160 zoo 240 zao 320
J 200 160 120 90
IPP~I
60 30 0 - 40

T (K)
Figure 4. A I3C high-resolution spectrum of solid beeswax.
Figure 3. The length of the a axis (upper curve) and the b The lower trace is equivalent to the upper trace but with a
axis (lower curve) as a function of temperature. reduction in gain.

1423
 l Basson and E C Reynhardt

Table 1. Chemical shifts and relative intensities of the peaks in the high-resolution spectrum
of beeswax. The results for oxidised Fischer-Topsch hard wax are also included.

Relative intensity (Yo)

Chemical Oxidised
shift (ppm) Functional group Beeswax hard wax
14 methyl
22.8 a-methylene
30 interior methylene 92 92.5
43.5 carbons adjacent to carbonyl and acid groups
50.5 carbons adjacent to ester groups
68-80 hydroxyl 1.2 1.2
86-9 7 triple-bonded carbons 0.6 -
125-35 double-bonded carbons 2.5 4.3
164-78 esters acids 2.7 1.2
21 1-23 carbonyl 1.O 0.8

65 ppm are spinning side bands. The positions of the 10’1 l

remaining ten peaks and the approximate areas under
these peaks, expressed as percentages of the total area
under all the peaks, excluding the spinning side bands,
are listed in table 1. These percentages should be con-
sideredasroughestimatessince it is expectedthat
theintensities of thepeaksdependtosomeextent
on operating conditions and the environments and the
degree of mobility of the different carbon atoms. The i l
dependence of relative peak intensities on the contact
1321
1
time was not investigated. 4 5 6 7 8 9
Thepercentage liquid in the wax is shownasa 1/T <-l)

function of temperature in figure 5. Between approxi- Figure 6. The proton spin-lattice relaxation time as a
mately 290 and 320 K, the percentage liquid increases function of 1OOO/T.
slowly,but abovethelattertemperature it increases
dramatically and reaches 100% at approximately 340 K.
The corresponding T1, results are shown as a func-
tion of in figure 7, which exhibitsaminimum of
4.4 ? 0.3 ms at p = 3.7 K”. Thephasetransition
revealed by the T , results is also observed in this case

* 20
40:
l%
283
, l 0
l
i
/
i

l, h
T (K)
Figure 5. The percentage liquid in beeswax as a function
of temperature.

The T 1results are displayed in figure 6 as a function

of p = 10OO/T. A T , minimum of 0.82 ? 0.06 S occurs at
p = 6.4 K-*. Above 250 K T 1decreases with increasing l
35 43 45 5s
temperature. A phasetransition is observedat = 1/T K-’)

3.13 K” ( T = 320 K) where T , decreasesdiscon- Figure 7. The proton spin-lattice relaxation time in the
tinuously from 0.36 to 0.28 S . rotating frame as a function of 100OiT.

1424
 NMR Of beeswax

at p = 3.13 K" where T I , changes discontinuously from except that it has oxygen-containing groups scattered at
9.6 to 200 ms with increasing temperature. random along the chains.
The temperature dependenceof T , is shown in figure
8. T 2decreases with decreasing temperature and reaches 4.1.2. X-ray diffraction and DSC. From figure 2 it is clear
a constant value of 10.5 f 0.5 ps below fi = 5 K". that the diffractogramsof beeswax and Fischer-Tropsch
wax are identical. i t therefore follows that beeswax also
forms orthorhombic crystals with the chains parallel to
the c axis (Reynhardt 1986).
The linear dependence on temperature of the a and
b cell dimensions between 80 and 240 K indicates that
theexpansion is dueto small anharmonicthermal
vibrations. The NMR results, which will be discussed in
8 4.2.1, show that the chains are stationary on the NMR
timescale in this temperature region and that only the
methyl groups execute threefold reorientations. Only
arelativeshortsection of thechain,located in the
amorphous zone of the wax, is involved in this motion
and should not influence the length of the cell dimen-
L 5 6 7 a sions noticeably. Above room temperature the motions
l:? i10-3 K"]
of thechainsandthemeltingprocess influence the
Figure 8. The spin-spin relaxation time as a function of temperaturedependence of thelength of the crys-
1OOOi T. tallographic axes. The temperature dependence of the
lengths of the a and b axes of beeswax is similar to that
of hard wax and oxidised hard wax (Reynhardt 1986).
3.4. Ebullioscopy In all cases da/d T is smaller above room temperature
than at lower temperatures. However, in the case of
The boiling pointelevationconstant of toluene was
beeswax da/d T becomes negative above 230 K. Reyn-
found to be 3.61K kg mol". From the slopeof a graph
hardt (1986) showed that on accountof potential energy
of A T versus g, the mass of beeswaxdissolvedina
calculations and knowledge of the arrangement of the
constant mass of toluene, an average molecularmass of
chains in the unit cell, reorientationsof the chains should
0.591 kg mol" was obtained. It was further assumed
result in db/d T > da/d T , as observedaboveroom
that monoesters(CNHZNOZ) constitute themain fraction
temperature. From a discussion of the NMR results it
of beeswax, resulting in an average chain length of40
will be seen that these motions become fast above room
carbon atoms.
temperature. Both peaksin the thermogramof beeswax
wereattributed by Craig et a1 (1968) asmelting
transitions. However, the T1 and T1, results show con-
4. Discussion
vincingly thataphasetransition occursat thetem-
perature of the first transition. It seems therefore more
4.1. Structure
likely that the DSC peak at 323 K is associated with this
4.1.1. Composition. Extensivestudies of beeswax of solid-solid phase transition and not with the melting
comb origin produced by the European bee Apis mel- process.
lifera have shown thatit has thefollowing composition.
hydrocarbons (14%), monoesters (35%),
diesters 4.1.3. High-resolution 13C NMR. The peaks of the 13C
(14%),triesters(3%), hydroxy monoesters(4%), spectrum were assigned to specific functional groups by
hydroxy polyesters (8%), free acids (12%), acid mono- comparingthemeasuredchemical shiftswithknown
esters (l%),acid polyesters (2%), free alcohols (1%) ones (Johnson and Jankowski 1972). The groups thus
and unidentified material (6%) (Tulloch 1972a, b, 1980, identified are listed in table 1. Although the intensities
Stransky and Streibll971).Gas-liquid chromatographic of individual peaks could not be measured accurately
analyses of combwax of the European and African bees due to overlapping of peaks, it follows from the results
(Tulloch 1980) have shown that their compositions do listed in table 1 that approximately 92% of the carbon
not differ drastically. atomsarepresentasmethylandmethylenegroups.
Amodelforthestructure of aFischer-Tropsch Carbons attached to hydroxyl groups constitute approxi-
wax has been described by Le Roux et a1 (1976). The mately 1.2%. Asmallpercentage of carbonatoms
amorphous zone contains the chain ends in which the (0.6%) form triple bonds with adjacent carbon atoms,
chains form a staggered pattern due to the wide distri- while 2.5% form double bonds. Of the remaining carbon
bution in chain lengthsin this zone. Therigid amorphous atoms, 2.7% occur in esters and acid groups and 1.0%
zone is caused by branching of the long chains. The in carbonyl groups. Also listed in table l are the cor-
crystalline zone resembles an n-alkanewhile the mobile responding I3C results foroxidised hard wax. The results
amorphous zone contains the liquid component of the are very similar except that oxidised hard wax contains
wax. i t is assumed that beeswax has a similar structure, about twice thepercentage(approximately4.3%) of

1425
 I Basson and E C Reynhardt

Table 2. Parameters associated with motions of methyl groups and chains in beeswax.

Motion

Threefold of methyl groups 3.5 7.4 ? 0.5 (2.0 t 0.6) x

Dangling of chain ends 0.55 I0.05 26.5 ? 2.3 *
(7.9 0.6) x 1 0 - I 2
RET Of chains 30 ? 3 (1.1 ? 0.1) X 10-13

double-bonded carbons and about half the percentage roc asvariableparameters, yielded the full curve in
of ester and acid carbons compared with beeswax. figure 6 and the values listed in table 2. The activation
energy for methyl reorientations is 7.4 "+ 0.5 kJ mol"
which is approximately the same as the corresponding
4.2. Molecular dynamics activationenergies in Fischer-Tropschmedium wax
4.2.1. Laboratory frame spin-lattice relaxation. Since (8.4 "+ 0.4 kJ mol-') and hard wax (7.9 & 0.4 kJ mol")
about 70% of beeswax consists of alkanes and esters,it (Reynhardt 1985a).
Theaverage
environments of
is assumed that on average the ratio of methyl protons methyl groups in the low-temperature solid phases of
to the total number of protons is approximately 3x/2N waxes seems to differ only slightly. Most probably the
where x is the average number of methyl groups per height of thebarriertoreorientation is determined
chain of averagelength N carbonatoms.The spin- largely by intra-molecular contributions. The depth of
lattice relaxation rate due to the modulation of intra- the T I minimum yields a value of 3.5 methyl groups per
molecular methyl dipolar interactions caused by three- chain of average length 40 carbon atoms. If on average
fold reorientations of methyl groups is given by two of these groups are positioned at chain ends, the
(O'Reilly and Tsang 1967) remaining 1.5 methyl groups are distributed at random
along the interior of the chain. This result implies that
beeswax is branchedtoamuchhigherdegreethan
Fischer-Tropsch waxes or oxidisedFischer-Tropsch
(1) waxes (Reynhardt 1985a, b). In these cases branching
where r is the inter-proton distance (1.78A) in a methyl is limited to less than 0.5% of the carbon atoms while
group and in the case of beeswax it is approximately 3.8%. Due
tothe limited temperature region in which the RET
exp(E,/RT).
t, = tom (2) motion of the chainsinfluences the T 1results noticeably,
Similarly, a motion of the chains results in a relaxation the activation energy associated with this motion could
notbedeterminedaccurately.However,fromthe
rate given by
steepest sectionof the In T Iversus /3 curve it is estimated
tobe of the order of 30 kJ mol". Like in Fischer-
Tropsch waxes (Lourens and Reynhardt 1979. Reyn-
hardt 1985a, b) and n-alkanes (Ewenet a1 1974) the RET
whereAMz is thereduction in theprotonsecond motion triggers the transition to the high-temperature
moment due to this motion. The subscripts m and c in phase, which is most likely dominated by the melting
equations (1) to (3) refer to methyl and chain motions process. Unfortunately the temperature region of this
respectively. phase is very limited and reliable motional parameters
The observed TI minimum at p = 6.4 K" occurs at could not be determined.
approximately the same temperature asminima associ-
ated with threefold reorientations of methyl groups in 4.2.2. Spin-spin relaxation time. If the T z plateau value
Fischer-Tropsch waxes (Lourens and Reynhardt 1979, of 10.5 ,us is substituted into
Reynhardt1985a,b)andn-alkanes(Andersonand
Slichter1965). The decrease in T I athigher tem-
peratures is associated with the motion of chains. In which is valid for a Gaussian line, a second moment of
Fischer-Tropschmedium wax thismotionhasbeen 25.4 t 1.3 G?is obtained. Thecalculated proton second
identified
as the reorientational-translational (RET) moments for a rigid lattice and a lattice in which only
motion.Thechain-like molecules performthermally threefold methyl reorientations are present were cal-
activated rotational jumps through 180" between two culatedtobe29.1 G'and26.5 G' respectively. The
equilibrium positions, each jump being accompaniedby latter value is in reasonable agreement with the value
a simultaneous translation along the chain axis by one obtained from the T I plateau.
CH2 group (Ewen et a1 1974, Lourens and Reynhardt Theproton spin-spin relaxationrate is given by
1979). This motion was also identified in hard wax and (Fukushima and Roeder 1981)
oxidised hard wax (Reynhardt 1985a, b).
Fitting equations (1) and (3) simultaneously to the
experimental data by treating x , tom, E,, AM*, E, and

1426
 NMR Of beeswax

The T 2plateau is reached with decreasing temperature should therefore affect the TI results fairly seriously in
when m o t -- 1. Substitutingthisvalue of t and T = theregionjust below the phase transition. The acti-
200 K intoequation ( 5 ) , to= (1.l 5 0.1) X S for vation energy of approximately 18 kJ mol" for the RET
the RET motion of the chains is obtained. motion obtained from the fit to the T I data, excluding
the chain dangling motion, is therefore fairly inaccurate.
4.2.3. Rotating frame spin-lattice relaxation. The Since the activation energy of the dangling motion is
rotatingframe spin-latticerelaxation rateforreori- approximately 26 kJ mol", the activation energy for
entations of the chains is given by (O'Reilley and Tsang the RET motion is expected to be at least 30 kJ mol",
1967) in agreement with the estimate made from the steepest
/ r \ section of the In T1versus p curve just below the phase
transition.
if T I PG T1.
A best fit of equation (6) to the TI, data below the
phase transition is shown in a full curve in figure7. The
motional parameters thus obtained are listed in table 2.
Inanattempttoidentifythenature of themotion
characterised by the T,, minimum, the reduction in the
protonsecondmoment A M 2 = 0.55 ? 0.05 G',
obtainedfromthe fit totheexperimentaldata, was
compared with the calculated reductions in the proton
second moment for different motions. Itwas found that
the calculated reduction for any motion involving an
entirechain, is much too largetoaccountforthe
observed reduction. In a further attempt to identify this
motion, T1, measurementswere also madeatroom
temperature for medium wax, hard wax and oxidised
hard wax. All these measured values were found to be
in the range 6 to 24 ms, while a similar measurement
for the pure alkane n-C24 yielded a valueof 1.2 S , which
is between 50 and 200 timeslongerthanthe values
obtained for waxes. It therefore seems that the motion Figure 9. An illustration of the dangling chain-end motion
resulting in the T I , minimuminbeeswax is present in the rigid amorphous zone (A) of waxes. D represents the
in all waxes,butabsent in n-alkanes. The twomost crystalline zone.
prominentaspectsthat distinguisha wax froman n-
alkanearethedistribution of chainlengthsandthe
presence of side chains in waxes. The threefold reori-
entations of sidechainmethylgroupshavealready 5. Conclusions
beenascribedtothelow-temperature T 1 minimum.
However, it is possible that a relatively small number Theaveragechain-like molecule in beeswax closely
of these methyl groups have higher activation energies resembles that of oxidised Fischer-Tropsch hard wax.
associated with this motion, resulting in a shallow T I , The average chain lengths are 40 and 47, respectively.
minimum. However, the measured T1, value at room Beeswax contains half as many C=C bonds and twice
temperature for medium wax,containingonaverage as many carbons involved in ester groups as oxidised
only 2.4 branches per 1000 carbon atoms (Lourens and Fischer-Tropsch wax. Perhaps the most important dif-
Reynhardt 1979), is comparable with the T I , minimum ference is that beeswax is much more branched than
for beeswax. Since one would expect that the T l pmini- oxidised hard wax.
mum would be proportional to the degree of branching, The potential barriers to threefold reorientation of
the T1, minimum for beeswax should have been much methyl groups in all waxes seem to be about the same.
deeper. The possibility that the observed T,, minimum The RET motion of chainsinfluence the T , resultsat
is associatedwiththe
threefoldreorientations of higher temperatures andtriggers the phase transition at
branchedmethylgroupscanthereforebediscarded. approximately 320 K.
The most likely motion associated with this T1, mini- The dangling motion of chainends in the rigid
mum is the dangling or wobbling of the chain ends in amorphous zone has been observed for thefirst time by
the amorphous or chain end zone as illustrated in figure NMR. Thismotionseemsto bepresent in all waxes
9. This type of motion is thermally activated and should and in thepresent casehas an activationenergy of
result in a T I , minimum. It should be noted that this approximately 26 kJ mol".
motion should influence the T1 results in the vicinity of Thethermalexpansionand crystallographic par-
p = 3.5 K-'. The depth of the TI minimum due to this ameters of beeswax and Fischer-Tropsch waxes are
motion is calculated to be approximately 2.2 S and it approximately the same.

1427
I Basson and E C Reynhardt

Acknowledgments Le Roux J H 1969a J. Appl. Chem. 19 39

-1969b J . Appl. Chem. 19 86
Dr J H le Rouxof Sasol Technology (Pty) Ltd is thanked -1969c J. Appl. Chem. 19 230
-1970 J . Appl. Chem. 20 203
formanyhelpfuldiscussions.SasolTechnology(Pty) Le Roux J H, Smith R D F, Turner R and Weidema 0
Ltd is also thanked for the ebullioscopic measurements 1974 J . Appl. Chem. Biotechnol. 24 81
and for financial support. Le Roux J H, Retief J J and Pretorius S 1977 Ext. Abstr.-
Int. Symp. Anal. Chem. Explor. Min. Process. Mater.
lJohannesburg) 1976 (Pretoria: CSIR) pp 164-70
References Lourens J A J and Reynhardt E C 1979 J . Phys. D: Appl.
Phys. 12 1963
Anderson J E and Slichter W P 1965 J . Phys. Chem. O’Reilly D E and Tsang T 1967 Phys. Reu. 157 417
69 3099 Reynhardt E C 1985a J . Phys. D: Appl. Phys. 18 1185
Craig R G , Powers J M and Peyton F A 1968 Anal. -1985b J . Phys. D: Appl. Phys. 18 2519
Calorim. 155 157 -1986 J . Phys. D: Appl. Phys. 19 1925
Ewen B. Fischer E W, Piesczek W and Strobel G 1974 Stransky K and Streibl M 1971 Collect. Czech. Chem.
J. Chem. Phys. 61 5265 Commun. 36 2267
Fukushima E and Roeder S B W 1981 Experimental Pulse Tulloch A P 1972a J. Am. Oil Chem. Soc. 49 609
N M R (Reading, MA: Addison-Wesley) -1972b J. Am. Oil Chem. Soc. 49 696
Johnson L F and Jankowski W C 1972 Carbon-l3 NMR ~ 1980 Bee World 61 47
Spectra (New York: Wiley)

1428