Mechanochemical reaction of DDT with calcium oxide

Article abstract of DOI:10.1021/es950680j

Evidence is presented that, in the mechanochemical destruction of DDT [2,2-bis(4-chlorophenyl)1,1,1-trichloroethane] by ball milling in the presence of calcium oxide, a complex series of reactions occurs along the pathway to a product that appears to be essentially graphitic, though aromatic chloro and hydroxy substituents are retained to some degree. The production of the various intermediates can be understood in terms of processes initiated at both CaO and steel (of the milling device) surfaces. With the exception of DOE [2,2-bis(4-chlorophenyl)1,1-dichloroethene], most of these intermediates attain maximum concentrations corresponding to <1 mol % of the original DDT and have been characterized only by their mass spectra. In the case of dichlorotolane [bis(4-chlorophenyl)ethyne], however, yields are sufficient for it to be isolated chromatographically as a pure, crystalline solid and characterized further by NMR spectroscopy. After 12 h of milling, no organic materials volatile enough to be detected by conventional GC/MS procedures are present, but the black, graphitic residue does retain some chlorine that is only slowly removed by extended milling.

Full text of DOI:10.1021/es950680j

Research
agitation can produce marked chemicalactivation, resulting
not only in reactions between the mill charge components
but also in reactions with the mill surfaces. Remarkably,
this form of activation has not been widely exploited in
other than relatively simple reactions, and a great range of
chemistry remains open to exploration (1-6). We have
recently given a preliminary report on the potential
application of ball-milling mechanochemistry in the de-
struction of toxic wastes (7), specifically organochlorines,
by reaction with various metal powders and one example
ofa metaloxide (CaO), this work being performed in parallel
with studies of the use of ball-milling for the synthesis of
reactive metal alkyls such as Grignard reagents (8). In
seeking to find less expensive reagents than metal powders
for use in organochlorine destruction, we have examined
the behavior of a variety of metal oxides known to be
effective in related reactions (3) and now provide detailed
results on the use of calcium oxide, seemingly one of the
better reagents for the destruction of DDT [2,2-bis(4-
chlorophenyl)-1,1,1-trichloroethane] in particular. There
is a very long history of the use of CaO in the destruction
of organic matter (9), and recent work has demonstrated
several advantageous features of the use of CaO in the
destruction of some simple chlorinated hydrocarbons, of
particular significance being the fact that ultrafine CaO,
such as might be produced by ball-milling, is a very effective
destructive adsorbent (10).
The environmental ubiquity of trace levels of DDT
because ofits formerlyextensive use as an effective pesticide
is well known (11, 12) as are both chemical and biochemical
pathways to its degradation and destruction (13-15). Thus,
in the presence of a strong base like CaO, DDT is expected
to rapidly undergo formal elimination of HCl to produce
DDE [2,2-bis(4-chlorophenyl)-1,1-dichloroethene] and CaCl2.
Any reaction that DDT may undergo with a metal surface
is more obscure, although there is the general expectation
that any halocarbon may undergo at least initial oxidative
addition to a metal (16). A complex suite of reactions might
then be expected to ensue (16), and consequently, we
anticipated the necessity for the use of a sophisticated
analytical technique such as GC/ MS in the characterization
of the mechanochemical reaction of DDT and CaO. It is
possible to speculate broadly on the nature of the overall
reaction between these two materials, but here it is perhaps
only worthy of note that if organic chlorine removal were
considered formally as a series of HCl elimination steps,
the other reaction product would have to be a highly
unsaturated hydrocarbon:
Mechanochemical Reaction of
DDT with Calcium Oxide
A N N E G R E T K . H A L L , †
,
‡
J A C K M . H A R R O W F I E L D , *
‡
R E I N H O L D J . H A R T , A N D
P A U L G . M C C O R M I C K †
Department of Mechanical and Materials Engineering and
Department of Chemistry, University of Western Australia,
Nedlands, Western Australia 6907, Australia
Evidence is presented that, in the mechanochemical
destruction of DDT [2,2-bis(4-chlorophenyl)-
1,1,1-trichloroethane] by ball milling in the presence
of calcium oxide, a complex series of reactions
occurs along the pathway to a product that appears
to be essentially graphitic, though aromatic chloro
and hydroxy substituents are retained to some degree.
The production of the various intermediates can be
understood in terms of processes initiated at both
CaO and steel (of the milling device) surfaces. With
the exception of DDE [2,2-bis(4-chlorophenyl)-
1
,1-dichloroethene], most of these intermediates attain
maximum concentrations corresponding to <1 mol
of the original DDT and have been characterized only
by their mass spectra. In the case of dichlorotolane
bis(4-chlorophenyl)ethyne], however, yields are
%
[
sufficient for it to be isolated chromatographically as
a pure, crystalline solid and characterized further
by NMR spectroscopy. After 12 h of milling, no organic
materials volatile enough to be detected by conven-
tional GC/MS procedures are present, but the black,
graphitic residue does retain some chlorine that is
only slowly removed by extended milling.
Introduction
Chemistry occurring at surfaces under impact has long been
studied and has numerous practical applications. The field
is commonly referred to as “tribochemistry” or “mecha-
nochemistry” (1), the latter term reflecting the fact that
mechanical agitation ofa material within a mill is a standard
procedure. Since high energy impact between surfaces of,
for example, stainless steel balls can produce transient
temperature rises of large magnitude [and even the
generation of triboplasmas (1)], it is unsurprising that such
2
C H Cl + 5CaO f C H + 5CaCl + 5H O
14 9 5 28 8 2 2
Since mechanochemical activation is known to be effective
in initiating alkene polymerization (6), such an intermediate
could give rise to involatile, high molecular weight products.
Although there is indeed evidence that the final product of
DDT milling in the presence of CaO is a high molecular
weight carbonaceous material, we show herein that the
*
Corresponding author telephone: (61) (9) 380 3170; Fax: (61) (9)
3
80 1116; e-mail address: jmh@chem.uwa.edu.au.
†
‡
Department of Mechanical and Materials Engineering.
Department of Chemistry.
S0013-936X(95)00680-8 CCC: $12.00

1996 Am erican Chem ical Society
VOL. 30, NO. 12, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 0 1

chemistry leading to this and the actual nature of the solid
are considerably more complex than the above equation
would imply.
Materials and Methods
Chem icals. Reagent-grade DDT (Aldrich, 98% 4,4-isomer)
and CaO (Ajax Chemicals) were used as received. Some
investigations were made on various samples ofcommercial
DDT preparations (solids and solutions) obtained through
the (Government) Chemistry Centre of Western Australia.
Some preparations of CaO were made by thermal decom-
position (900 °C) of CaCO3 and Ca(OH)2. Pentadeuterio-
chlorobenzene (Cambridge Isotope Laboratories, 99.8% D)
was used to prepare partly deuterated DDT by the method
described by Ginsburg (17). [The deuteration was only
partial due to the use of the normal form of Cl3CCH(OH)2
FIGURE 1. Residual organic chlorine as a function of time of milling,
estimated. (a) from total organochlorines detectable by GCMS; (b)
by difference from measurement of released ionic chloride.
(
TCI Chemicals, Japan) as a reactant and some degree of
exchange occurring during the H2SO4-catalyzed reaction;
1
H NMR spectroscopy indicating that the DDT product
with temperature programming 45-330 °C. Quantitative
determinations of DDT, DDE, and DDD [1,1-dichloro-2,2-
bis(4-chlorophenyl)ethane], both 4,4′- and 2,4′-isomers for
each, were based on calibration with certified reference
standards from the National Physical Laboratory, Ted-
dington, U.K. The amounts of various other volatile
materials detected were estimated semiquantitatively by
assuming, for a given number of chlorine substituents, a
similar detector response to these standards. Both the Wiley
and NIST (NBS 75K) Mass Spectrometry Databases were
used in searches for compound identification.
contained ∼1 aromatic proton and 7 aromatic deuterons.]
Milling Procedure. Using glovebox procedures under
argon, mixtures of CaO and DDT were loaded together with
hardened steelballs into a steelvial, and the vialthen sealed.
Typically, 10 12-mm-diameter grinding balls were used,
and the ratio of ball mass to powder mass was fixed at 10:1.
For all experiments presently reported, the mass ratio of
CaO:DDT was 7:1. The mixtures were mechanically milled
for various times using a SPEX 8000 mixer mill.
Therm ogravim etry. Thermogravimetric (TG) and dif-
ferential thermal analysis (DTA) measurements were per-
formed on a Rigaku instrument.
Spectroscopic Measurem ents. Infrared spectra were
recorded on a Digilab FTS 45 spectrometer, and nuclear
1
13
Chrom atography. Thin-layer and column chromatog-
raphy experiments were conducted under conventional
normal-phase condition on silica. GC/ MS conditions are
described below.
magnetic resonance spectra ( H and C, both solution and
solid state) were recorded on a Bruker AM300 instrument.
Results and Discussion
Elem ental Microanalysis. C, H and organic Cl analyses
were performed by the Australian National University
Microanalytical Service, Canberra. Also, some C and H
analyses were performed by the Chemistry Centre of
Western Australia, Perth.
Total Inorganic Chlorine Titration (TICT). Chloride
released during the destruction process was determined
by potentiometric titration using a Radiometer pH Meter
Several hundred milling experiments have been conducted
to investigate the wide variety of factors that were antici-
pated to be possible influences upon the process of DDT
destruction in the presence of CaO and then to systemati-
cally evaluate those factors that did prove to be significant.
Reported below are essentially observations on the chem-
istry of the destruction process and a partial rationalization
of its nature. Much of this is necessarily qualitative. A
detailed description and analysis of physical factors that
determine the efficiency of destruction will be given in
subsequent publications.
(
PHM92) equipped with an Hg/ Hg2SO4 reference electrode
and a silver billet sensor electrode. For an analysis, between
.25 and 0.5 g of milled sample was weighed under dry
0
argon, then exposed to the normal atmosphere for 12 h to
allow residual CaO to slowly hydrate. A mixture of water
Characterization of Final Reaction Products. After a
period of approximately 12 h milling under ambient
conditions, the CaO/ DDT mixture appeared as a grayish
powder, seemingly unreactive when exposed to the normal
atmosphere, although still displaying the properties ex-
pected (exothermic reaction with both water and acids)
due to the presence ofexcess CaO. Extraction ofthe powder
with hexane/ acetone/ aqueous HCl revealed no significant
levels oforganic materials detectable by GC/ MS. Extraction
with aqueous acid left an amorphous black powder,
graphitic in appearance but analyzing for significant levels
of both hydrogen and chlorine, the latter consistent with
the fact that free chloride determined on the acidic aqueous
extract corresponded to less than 100% of the chlorine
present in the original DDT. In fact, sampling throughout
the period of milling showed that the chlorine measured
as chloride ion was always less than the level expected on
the basis ofthe chlorine content ofresidualorganochlorines
detectable (and quantified) by GC/ MS (Figure 1). Although
(
8 mL) and concentrated HNO3 (2 mL) was then added and
the mixture stirred for 15 min before the insoluble material
was filtered out and washed carefully with water (30 mL).
The combined filtrate plus washings was titrated with
-
1
standard 0.1 mol L AgNO3. Analytical results were
unaffected if the weighed sample was not firstly exposed
to the air but the reaction with the aqueous HNO3 was
somewhat more violent.
GC/MS Analysis for Organic Reactant and Products.
Milled powder (∼100 mg) was mixed with hexane (4.5 mL),
-1
acetone (0.5 mL), and ∼5 molL HCl(0.5 mL) and sonicated
for 15-20 min. The organic layer was separated off, and
1
-µL samples were used for column injection. The chro-
matographic system consisted of a Hewlett-Packard 5890
Series II GC and 5970 mass-selective detector (EI ionization,
mass range 33-800 amu) on a 30-m DB-5 capillary column
(
0.25 mm i.d., 1 µm film), He carrier, and splitless injection
3
4 0 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 12, 1996

FIGURE 2. Proposed structures for intermediates detected during mechanochemical destruction of DDT in the presence of CaO.
consistent with a range of carbon environments and thus
not excluding the presence of chloro and/ or hydroxy
substituents on some aromatic carbons. An attempt was
made to directly detect residual hydrogen in the black
TABLE 1
Microanalytical Data for Insoluble Products of
Extended Milling of DDT with CaO
sample
%C
%H
%Cl
6.3
%O
15.8a
product that could not have arisen from adsorbed solvent
by measuring the (solid-state) H NMR spectrum of the
2
1
2 h product,
washed dilute
HNO3, water,
ether
2 h product,
washed dilute
HCl, water
74.4
3.5
2
4-h milling product from deuterated DDT, but no signal
could be detected. It is presently uncertain whether this
was due to a true lack of bound deuterium or an extreme
signal breadth. As is shown below, there appear to be many
pathways involved in the reaction between DDT and CaO
in a ball mill; there are, therefore, reasons for expecting the
ultimate product to be a complex mixture, but it is possible
to speculate (see ahead) that one significant pathway might
give rise to a product with the limiting empirical formula
C7H3 and the available microanalytical data at least are not
incompatible with an approach to this C:H ratio. Note
that X-ray powder diffraction served only to confirm that
the black product was trulyamorphous, and infrared spectra
showed only broad, featureless absorptions similar to those
of graphite. Thermogravimetric measurements under the
normal atmosphere provided evidence for the occurrence
ofat least two oxidative processes, with one occurring under
conditions closely similar to those for the single event
observed with pure graphite.
9.8, 8.1 10.1, 14.6a
a
1
2
75.4, 73.9 4.7, 3.4
74.0, 77.3 4.0, 3.6
2.8, 1.8 19.2, 17.3a
a
4 h product,
washed dilute
HCl, water
b
C14H10O3
C14H9ClO2
C14H8Cl2O
74.33
68.72
63.90
79.98
68.04
74.15
4.46
3.71
3.06
4.79
3.21
2.67
0
21.22
13.08
6.08
15.22
0
b
b
14.49
26.95
0
28.69
6.25
b
C42H30O6
C42H24Cl6
b
C35H15ClO6c
16.93
a
b
Estim ated by difference. Theoretical figures for som e possible
reaction products (see Figure 6). c Best fit form ulation (see text).
prolonged milling resulted in a slow approach of the
released chloride value to 100%, for practical purposes the
carbon-containing product of milling of DDT with CaO is
best described as a high molecular weight aromatic
compound (possibly a mixture of compounds) with some
chlorine and oxygen substituents. Microanalyses on these
black materials (Table 1) were somewhat variable, perhaps
because of the adsorption of water and iron oxychlorides,
but were indicative of the presence of at least one chlorine
and possibly up to six oxygen atoms for every 35 carbons.
Organic Interm ediates in DDT Destruction. By ex-
traction of the milling product at various reaction times
(typically, every1-2 h), it was possible to detect and quantify
(at least approximately) both residual DDT and a large
number of intermediates through GC/ MS measurements.
In many instances, these intermediates were compounds
well characterized from studies of other aspects of DDT
chemistry (13-15) and expected to be present as a result
1
3
The solid-state C NMR spectrum of the black solid showed
only a broad peak at δ 130, the breadth of the peak being
VOL. 30, NO. 12, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 0 3

FIGURE 3. (a) Time dependence of the concentrations of DDT, DDE,
andDDDduringmechanochemicaldestructionofDDTinthe presence
of CaO. (b) Time dependence of the concentrations of several trace
intermediates observed during the mechanochemical destruction
of DDT in the presence of CaO. [i-DDE, an isomer of 4,4-DDE; DDMU,
FIGURE 4. Processes possibly initiated at a metal surface in contact
with an organochlorine (16).
acetylene, and in fact dichlorotolane has been previously
identified, during studies of thermal decomposition of DDT
and related compounds, as a decomposition product of
DDD [1,1-dichloro-2,2-bis(4-chlorophenyl)ethane] (18).
Much more recently (19), the difficulties in identifying such
compounds on the basis of mass spectral measurements
have been discussed.
1,1-bis(4-chlorophenyl)chloroethene; DCPA, bis(4-chlorophenyl)-
acetylene; DBP, 4,4-dichlorobenzophenone; DDOL, 1-(4-hydroxyphe-
nyl)-2-(4-chlorophenyl)dichloroethene].
of the use of a basic destruction agent. In others, however,
the compounds appeared to be previously unknown or
nearly so, and where their characterization remains based
exclusively on a mass spectrum, we cannot claim firm
establishment of their structures. Under this proviso, basic
structures for the nearly 50 compounds we have detected
as intermediates in DDT destruction by ball milling with
CaO are shown in Figure 2. In general, with the exception
of the six compounds for which standards were available,
it must be said that considerable uncertainty remains with
respect to an attribute such as the position of chloro
substituents on aromatic rings but that otherwise frag-
mentation patterns (see Supplementary Information) are
fully consistent with the suggested structures. An interest-
ing case, however, is that of the acetylene bis(4-chlorophe-
nyl)ethyne [) 4,4′-dichlorotolane or simply tolane, also
referred to as dichlorophenylacetylene (DCPA in Figure 2)],
which was first identified on the basis ofa computer-assisted
mass spectral data base search as (dichloromethylene)-
fluorene. Fortunately, this compound is present in suf-
ficient quantities after several (4-6) h milling for it to be
easily isolated from the reaction mixture extract by column
The only organic materials detectable by extraction/
GC/ MS in quantities exceeding 10 mol % of the initial DDT
were residual DDT and its elimination product, DDE [2,2-
bis(4-chlorophenyl)-1,1-dichloroethene]. After ∼2 h mill-
ing, DDE is the only extractable/ volatile organic present in
significant amounts (Figure 3a), although trace levels of a
variety of the other compounds listed in Figure 2 remain
detectable for at least 10 h (Figure 3b). Given the well-
known ease of base-catalyzed elimination of HCl from DDT
to give DDE (13), it is hardly surprising that major amounts
are formed in the presence of CaO. It is therefore tempting
to describe the destruction of DDT under the present
circumstances as involving simply the formation of DDE
as the initial step. There are, however, several lines of
evidence that contradict this conclusion. One, possibly
weak because experimental error is large, is that the form
of the %DDE vs time plot does not fit that expected for a
simple A f B f C process. More significant are the
observations of the various minor reaction intermediates.
Thus, when DDE is used as a starting material for milling,
no intermediates detectable by extraction/ GC/ MS are
observed to intervene between it and the finalblack powder.
Thus, the relatively abundant intermediates such as dichlo-
rotolane and DDD observed in DDT milling must surely
arise directly from DDT itself. Though the presence of DDD
1
chromatography (SiO2/ hexane). Both H (AA′BB′ aromatic
proton resonances at δ [7.465, 7.460, 7.456, 7.447, 7.443,
1
3
7
[
1
.438, 7.347, 7.343, 7.339, 7.329, 7.326, and 7.321] and
δ 89.1 (C alkyne), 121.4 (C(Ar) bound to C alkyne), 128.7,
32.7 (CH(Ar)), 134.4 (C(Ar) bound to Cl)] NMR spectra
clearly establish its nature as the symmetrically substituted
C
3
4 0 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 12, 1996

FIGURE 5. Processes possibly initiated at a CaO surface in contact with an organochlorine (20, 22).
further reactions must proceed more rapidly on a metal
is perhaps surprising given that CaO is not a reducing
medium, DDD could however wellbe a product ofhydrogen
abstraction by a radical formed subsequent to the oxidative
addition of DDT to a metal surface (Figure 4), and in this
regard it may be significant that, despite the lack of any
marked etching of metal surfaces within the mills, it was
always possible to detect trace amounts of Fe(III) in the
final powder. We are therefore led to conclude that both
the steel and CaO surfaces (Figure 5) must be involved in
the overall DDT destruction. It is quite probable that initial
products of reaction at one surface may undergo further
reaction at the other, and it is also possible that more than
one process may be initiated at a given surface. Thus, the
fact that isomers of DDE could be detected despite the fact
that the original DDT was essentially isomerically pure
suggests that the “base-catalyzed halogen dance” (20) may
well have accompanied reactions such as elimination on
the CaO surface.
An obvious means of establishing the roles of different
surfaces within a mill is to vary the surfaces and although,
for practical reasons, our primary interest lies in the use of
standard stainless steel mills, an experiment in which
ceramic (zirconia) balls were used in place of those of
stainless steel, although still within a steel vial, provided
revealing data. Reaction was greatly slowed, and after 12
h, DDE in particular remained present at a level equivalent
to ∼46% of the initial DDT, suggesting that, although DDE
might be formed from DDT on the basic CaO surface, its
surface. Another significant observation was that the
ceramic balls appeared to give rise to some minor reaction
products not seen in the all-steel system. An oxygen-
containing species, C14H8Cl4O, is one of the most readily
detected, and its formula is consistent with the possibility
of oxidation of an elimination product from DDT, a reaction
-
that might occur if hypochlorite (OCl ) were present (to
form, say, the epoxide of DDE). Hypochlorite is, of course,
the product of the reaction of chlorine (Cl2) with oxide, and
another minor product that occurs in similar amounts to
the C14H8Cl4O species is a compound, C14H7Cl5, that could
be formulated as a chloro-substituted DDE(i.e., as a product
of electrophilic substitution by chlorine on one of the
aromatic rings of DDE). Chlorine is a possible product of
a pathway (Figure 6) that would generate dichlorotolane,
and so the generation of these various compounds could
be interconnected, although since it is known to be possible
to produce dichlorotolane from DDD (18), there may be
more than one pathway involved. If chlorine is produced
during milling, then certainly it must react further since
there is no evidence of significant oxidizing power in the
final product, so that the above results indicate that under
“normal” milling conditions it must be reduced by the metal
surfaces.
The possibility that calcium alkyls were formed as
important reaction intermediates was excluded by the lack
of marked oxygen sensitivity of the reaction mixture and
VOL. 30, NO. 12, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 0 5

FIGURE 6. Some plausible pathways in DDT destruction.
by the fact that quenching/ extraction with hexane/ acetone/
DCl/ D2O did not result in appreciable incorporation of
deuterium into any ofthe relatively abundant intermediates
detected by GC/ MS. Nonetheless, the presence of acid in
the quenching/ extraction mixture did seem to be necessary
to achieve fullrelease ofthe intermediates, possibly because
it prevented adsorption or occlusion of these materials by
solid CaO/ CaCl2, and some of the very minor but rather
persistent intermediates did not appear to be released at
all in the absence of acid. Since these were oxygenated
species, we interpret this behavior as evidence that they
were phenolic compounds, present in the reaction mixture
as calcium phenoxides. A further indication of this was
that they were more efficiently extracted by ether than by
the rather apolar hexane-acetone mixture. Significantly,
phenol and catechol (1,2-dihydroxybenzene), for example,
both undergo only very slow destruction when milled with
CaO under an inert atmosphere.
(i) Reactions at least formally involving the elimination
of HCl and Cl may occur readily. As noted previously,
2
elimination of HCl from DDT is to be expected in the
presence of a base as strong as CaO (and is the cause of
DDE formation). Cl elimination might well be catalyzed
2
by the action ofCa(II) as a Lewis acid, although it has already
been noted that the milled powders display no significant
-
oxidizing capacity, and therefore any Cl2 (or OCl ) formed
initially must be subsequently reduced.
(
ii) The chloro substituents of DDT may be replaced by
hydrogen in the intermediates. This could occur via formal
loss of Cl+ and ultimate protonation of the product
carbanion during quenching, but deuteriation quenching
experiments indicate that such carbanions cannot be
persistent. Thus the reaction is perhaps better explained
as being due to hydrogen abstraction by a radical produced
at the metal surfaces of the mill system.
Overall conclusions that may be drawn on the basis of
the proposed chemical structures of the intermediates
observed in DDT destruction by the present method are
(iii) Phenyl group migration may occur from C1 to C2
of the original DDT molecule. Such Wagner-Meerwein
rearrangements (21) indicate the involvement of an electron-
deficient carbon species at some point.
(
refer also to Figures 4-6):
3
4 0 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 12, 1996

(
iv) Loss of one carbon atom from the DDT skeleton
Microforms Office, American Chemical Society, 1155 16th St.
NW, Washington, DC 20036. Full bibliographic citation
occurs quite commonly. The trichloromethyl (CCl3) group
is the most obvious cleavage unit.
(
journal, title ofarticle, names ofauthors, inclusive pagination,
volume number, and issue number) and prepayment, check
or money order for $51.00 for photocopy ($53.00 foreign) or
(
v) Migration of chloro substituents around the phenyl
rings appears to be possible. This may indicate that other
reactions known to be undergone by molecules in which
the base-catalyzed halogen dance (20) is observable are
also taking place. Of possibly considerable significance
would be the formation of benzyne derivatives. We have
in fact attempted to detect putative benzyne intermediates
during milling of 1,4-dichlorobenzene by the use of furfuryl
alcohol as a trap but have observed no addition products,
a result which may simply mean that these products
themselves are rapidly destroyed by the milling. The
experiment was also complicated by the rapid oligomer-
$
12.00 for microfiche ($13.00 foreign), are required. Canadian
residents should add 7% GST. Supporting information is
available to subscribers electronically via the Internet at http:
/ / pubs.acs.org (WWW) and pubs.acs.org (Gopher).
Literature Cited
(
1) Heinicke, G. Tribochemistry; Hansen, C., Ed.; Verlag: M u¨ nchen,
1984.
2) Shaw, M. C. J. Appl. Mech. 1948, 15, 37 (See also U.S. Patent
(
2
,416,717, Mar 4, 1947).
(
(
3) Yasue, T.; Katsumata, A.; Arai, Y. Gypsum Lime 1978, 157, 3.
4) Lin, I. J.; Nadiv, S. Mater. Sci. Eng. 1979, 39, 193.
(5) Borisov, A.; Petrova, L.; Makhaev, V. Proceedings of the 1st
International Conference on Mechanochemistry; Russian Acad-
emy of Science: Chernogolovka, 1992; pp 60-62 and references
therein.
ization of furfuryl alcohol in the mill.
(
vi) Aromatic nucleophilic substitution of Cl by O2 to
-
give calcium phenoxides is not an important reaction
pathway (although it seems it can be detected).
(6) Kuzuya, M.; Kondo, S.; Noguchi, A. Macromolecules 1991, 24,
4
047.
(
7) Hall, A. K.; Rowlands, S. A.; Hart, R. J.; Ebell, G.; Donecker, P.;
(
vii) The reactivity of the observable intermediates must
Street, R.; McCormick, P. G. Nature 1994, 367, 223.
be such as to produce high molecular weight materials
involving (as shown by their spectroscopic properties)
graphitic aromatic units.
(8) Rowlands, S. A.; McCormick, P. G. Unpublished work.
(9) Mellor, J. W. A Comprehensive Treatise on Inorganic and
Theoretical Chemisty; Longmans, Green and Co.: London, 1923;
Vol. III, Chapter XXIV.
The complexity of the observations we have made on
the ball milling-induced reactions between DDT and CaO
allows considerable speculation with regard to their ra-
tionalization (Figures 4 and 5). Some plausible pathways
to various observed intermediates and to possible end
products are shown in Figure 6. Regardless of the validity
of these speculations, it is most important to note that the
process (23) does result in efficient destruction of DDT and
the generation of innocuous products, specifically calcium
chloride and a highly insoluble and involatile graphitic
carbon that may remain partly chlorinated.
(
10) Koper, O.; Li, Y.-X.; Klabunde, K. J. Chem. Mater. 1993, 5, 500.
(11) Carson, R. Silent Spring; Penguin Books: Harmondsworth, 1965.
(
12) Note that there are also many naturally occurring organohalo-
gens. See Gribble, G. W. J. Chem. Educ. 1994, 71, 907.
13) (a) Corbett, J. R. The Biochemical Mode of Action of Pesticides;
Academic Press: New York, 1974. (b) Hassall, K. A. The Chemistry
of Pesticides; Macmillan Press: London, 1982.
14) Gould, R. F., Ed. The Fate of Organic Pesticides in the Aquatic
Environment; Advances in Chemistry Series 111; American
Chemical Society: Washington, DC, 1972.
(
(
(15) Henschler, D. Angew. Chem, Int. Ed. Engl. 1994, 33, 1920 and
references therein.
16) See, for example, Powell, P. Principles of Organometallic
Chemistry, 2nd ed.; Chapman and Hall: London, 1988; Chapter
(
2
.
Acknowledgments
(17) Ginsburg, J. M. Science 1948, 108, 339.
(
(
18) Jonas, H. Z. Naturforsch. 1952, 7B, 132.
19) Ramans, D. V.; Rama Krishna, N. V. S. Org. Mass Spectrom. 1989,
24, 903.
We thank Dr. Lindsay Byrne for assistance in the measure-
ment of NMR spectra and Dr. John K. McLeod (Research
School of Chemistry, Australian National University) for
comments on our mass spectrometric measurements. This
work was supported by CRA Ltd.
(20) Bunnett, J. F. Acc. Chem. Res. 1972, 5, 139.
(
(
(
21) Saunders, M.; Chandrasekhar, J.; von R. Schleyer, P. In Rear-
rangements in Ground and Excited States; De Mayo, P., Ed.;
Academic Press: New York, 1980; Vol. 1, p 1.
22) Chambers, R. D.; James, S. R. In Comprehensive Organic
Chemistry; Stoddart, J. F., Ed.; Pergamon Press: Oxford, 1979;
Vol. 1, Chapter 3, p 493.
Supporting Information Available
Quantitative analytical data for the time dependence of the
concentrations of various intermediates observed during ball
milling of CaO/ DDT (mass ratio 7:1), data plotted in Figure
23) Australian Patent No. PCT/ AU93/ 00660; U.S. Canadian, and
European patents applied for.
3
(1 page), and total ion chromatograms and mass spectra
for reaction mixtures of CaO/ DDT (7:1) under ball milling
28 pages) will appear following these pages in the microfilm
Received for review September 12, 1995. Accepted June 24,
X
1
996.
(
edition of this volume of the journal. Photocopies of the
supporting information from this paper or microfiche (105
ES950680J
X
×
148 mm, 24× reduction, negatives) may be obtained from
Abstract published in Advance ACS Abstracts, September 15, 1996.
VOL. 30, NO. 12, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 0 7