Environ. Scl. Technol.
1992, 26, 1324-1326
sphere (where anthropogenic emissions are less important)
the NO levels are likely to be somewhat less. Thus, re-
action with NO is likely to be a major fate of the peroxy
radicals generated from HFC-l34a, and so the products
determined here are likely to be major products of the
atmospheric oxidation of HFC-134a.
To further delineate the atmospheric chemistry of
HFC-l34a, additional research is needed to define the role
of reactions 13 and 14 in the atmospheric oxidation of
HFC-l34a, the fate of CF, radicals, and the aqueous phase
chemistry of CF,COF, COF2, and HC(0)F.
Acknowledgments
We thank Ole John Nielsen (Rim National Laboratory,
Denmark) and Steve Japar (Ford Motor Co.) for helpful
discussions.
Registry No. CF3CFH2,811-97-2; CF,CHFO', 140872-97-5;
02,7782-44-7.
Literature Cited
(1) Farman, J. D.; Gardiner, B. G.; Shanklin, J. D. Nature 1985,
315, 207.
(2) Solomon, S. Nature 1990,347,6291 and references therein.
(3) World Meteorological Organization Global Ozone Research
and Monitoring Project, Report No. 20; Scientific assess-
ment of stratospheric ozone, Vol. 1; 1989.
(4) Fisher, D. A.; Hales, C. H.; Wang, W. C.; KO,M. K. W.; Dak
Sze, N. Nature 1990, 344, 529.
(5) Shine, K.; Denvent, R. G.; Wuebbles, D. J.; Morcrette, J.-J.
Assessment for WG1 Plenary, April 27, 1990.
(6) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Molina, M.
J.; Hampson, R. F.; Kurylo, M. J.; Howard, C. J.; Ravish-
ankara, A. R. Jet Propulsion Laboratory Publication 90-1;
1990.
Degradation of Dibenzodioxin during Chlorine Dioxide Bleaching of Kraft Pulp
Alison K. Daube, Mohammed R. Karlm, Donald R. Dlmmel, Thomas J. McDonough, and Sujlt BanerJee*
Institute of Paper Science and Technology, 575 14th Street, NW, Atlanta, Georgia 30318
The degree of sorption of dibenzodioxin (DBD) to pulp
varies with the lignin content of the latter. DBD is rapidly
degraded by chlorine dioxide in aqueous solution but is
much more stable in the presence of semibleached pulp,
suggesting that sorbed DBD is far less reactive than the
freely dissolved material. Degradation products include
the 2-chloro derivative and products inextricably bound
to pulp.
Introduction
Chlorinated organic compounds are formed during the
bleaching of pulp. The component that has been of
greatest interest is 2,3,7,8-tetrachlorodibenzodioxin(TC-
DD). A potential sowce of TCDD is dibenzodioxin (DBD),
which can enter the bleach plant as a contaminant in the
brownstock pulp (1). The reactivity and pulp-water dis-
tribution of DBD and its derivatives under typical
bleaching conditions are, therefore, essential to the un-
derstanding of TCDD formation. Furthermore, as a stable
hydrophobic material, DBD serves as a model for other
substrates that may undergo chlorination. We have
studied the reactivity and distribution of DBD in pulp-
water systems during chlorine dioxide bleaching, and in
this paper we provide a broad overview of its fate.
1324 Envlron. Sci. Technol., Vol. 26, No. 7, 1992 0013-936X/92/0926-1324$03.00/0
(7) Wallington, T. J.; Nielsen, 0. J. Chem. Phys. Lett. 1991,
187, 33.
(8) Alternative Fluorocarbon Environmental Acceptability
Study. World Meteorological Organization Global Ozone
Research and Monitoring Project, Report No. 20; Scientific
assessment of stratospheric ozone, Vol. 2; 1989.
(9) See, for example, Chem. Br. 1990, 26, 217.
(10) Wallington, T. J.; Gierczak, C. A.; Ball, J. C.; Japar, S. M.
Int. J. Chem. Kinet. 1989,21, 1077.
(11) Wallington, T. J.; Andino, J. M.; Lorkovic, I. M.; Kaiser,
E. W.; Marston, G. J. Phys. Chem. 1990, 94, 3644.
(12) Kaiser, E. W. Int. J. Chem. Kinet. 1992, 24, 179.
(13) Morgan, H. W.; Staats, P. A.; Goldstein, J. H. J. Chem.
Phys. 1956,25, 337-342.
(14) Anderson, L. R.; Fox, W. B. J. Am. Chem. Soc. 1967,89,
4313.
(15) Thompson, P. G. J. Am. Chem. Soc. 1967, 89, 4316.
(16) Hirschmann, R. P.; Fox, W. B.; Anderson, L. R. Spectro-
chim. Acta 1969,25A, 811.
(17) Wallington, T. J.; Andino, J. M.; Ball, J. C.; Japar, S. M.
J. Atmos. Chem. 1990, 10, 301.
(18) Wallington, T. J.; Hurley, M. D. Chem. Phys. Lett. 1992,
189, 437.
(19) Wallington, T. J.; Andino, J. M.; Wine, P. H. Chem. Phys.
Lett. 1991, 176, 103.
(20) Troe, J. Ber. Bunsen-Ges. Phys. Chem. 1983, 87, 161.
(21) Gilbert, R. G.; Luther, K.; Troe, J. Ber. Bunsen-Ges. Phys.
Chem. 1983,87, 169.
(22) US.Standard Atmosphere, 1976; NOAA, NASA, USAF:
Washington, DC, 1976.
(23) Singh, H. B.; Kasting, J. F. J . Atmos. Chem. 1988, 7,261.
(24) Madronich, S.; Calvert, J. C. J. Geophys. Res. 1990,95,5697.
(25) Crutzen, P. J. Rev. Geophys. Space Phys. 1979,17,1824.
Received for review November 11,1991. Accepted February 25,
1992.
Materials and Methods
A loblolly pine Kraft pulp of K no. 40.6 (-6% lignin)
and a bleached hardwood pulp of K number <1 (no
measurable lignin) were used. The pulps were disinte-
grated, re-formed into mats by filtration through coarse
sintered glass, and air-dried. CIOz was determined io-
dometrically. Tritiated water was purchased from New
England Nuclear, DBD from MSD Isotopes, and 1-and
2-chlorodibenzodioxin from Ultra Scientific.
Synthesis of Tritiated DBD. Synthetic trials for the
preparation of tritiated DBD were first conducted with
2H,0. Since DBD is thermally stable, electrophilic deu-
teration was attempted at the high temperature (195 "C)
and dilute acid (4% HCl in DzO) conditions (2) used to
label simple aromatics. However, <2% isotope incorpo-
ration occurred after 43 h, indicating that DBD is too
deactivated to be labeled by electrophilic exchange under
these conditions. Nevertheless, the low reactivity estab-
lishes that once prepared by other means, [,H]DBD should
be very resistant to isotope loss.
[,H]DBD was prepared by condensing [3H]catecholwith
o-chloronitrobenzene (Scheme I). The labeled catechol
was prepared by refluxing catechol in 5% HC1 containing
1 pCi of ,HzO for 24 h. A trial run in deuterated media
showed that complete exchange occurs under these con-
@ 1992 American Chemical Society
Scheme I
-.
3u
5% HCI
OH
DMF 1 base
UOD
3H
0
ditions. Following reflux, the solution was diluted with
ethanol and then distilled to near-dryness. The ethanol
addition-distillation sequence was repeated twice to
maximize azeotropic removal of tritiated water and to
reduce carryover of tritium to the next step.
The labeled catechol was refluxed under nitrogen with
excess o-chloronitrobenzene (to promote consumption of
the labeled catechol) for 1 h in alkaline DMF. The
products were extracted into methylene chloride, and the
labeled DBD was separated by preparative thin-layer
chromatography on silica with 5% methanol in methylene
chloride as the eluent. The other labeled components,
[3H]catechol and [3H]Hz0,are strongly retained under
these conditions. The [3H]DBD was isolated from the
TLC plate with methylene chloride. Its identity and purity
was verified by electron impact GC-MS. A specific activity
of 1.8 WCi/mmol was calculated from liquid scintillation
counting and reversed-phase HPLC.
Recovery Studies. DBD was quantitatively recovered
from water-pulp mixtures with methylene chloride ex-
traction at room temperature. However, if the mixture is
heated (60 "C, 30 min) under typical bleaching conditions,
only 40-60% DBD was recovered by methylene chloride
extraction of the pulp mixture. Clearly, a significant
fraction of the DBD volatilized. However, enough material
remained in solution to react with the bleaching agent.
Sorption Studies. Pulp (0.3 g on an oven-dried basis)
was added to 30 mL of water. [3H]DBD (2000 dpm) was
added, and the mixture was stirred vigorously and allowed
to equilibrate overnight in a closed vessel. For the 40-K
pulp, the water and pulp were separated using a table-top
centrifuge. For the fully bleached pulp, the centrifugal
separation was incomplete and filtration through medi-
um-porosity glass was necessary.
The separated pulp was extracted with methylene
chloride, and both the extracts and the aqueous phase were
counted. For the 40-K pulp, there were twice as many
counts in the total pulp phase than there were in the water
phase. A second experiment with the same pulp and the
solute (DBD) at 10 times the previous level gave the same
pulp to water count ratio. The analysis in this case was
by HPLC with diode-array detection. For the fully
bleached pulp, 0.3 g of pulp was equilibrated with 50 mL
of a DBD solution for 16 h. Extraction and HPLC analysis
as described above yielded a pu1p:water ratio of 0.27.
Bleaching Experiments. An aqueous solution of DBD
was prepared by adding 1 mL of a methanolic concentrate
(1.63 mg/mL) of DBD to water (100 mL) in a 500-mL
round-bottom flask. The 40-Kpulp (3 g) was then added,
and the mixture was mechanically stirred and brought to
60 f 5 "C over 15 min. A freshly standardized solution
of CIOz (nominally containing 1 4 % of Cl,) w a introduced
under the surface of the liquid. The equivalent chlorine
multiple for c l o z (percent active ChlOIine/K number, not
including molecular chlorine) was 0.25. The mixture was
then stirred vigorously for 30 min, quenched with 0.1 N
sodium thiosulfate, cooled to room temperature, and ex-
tracted with methylene chloride. A longer contact period
was used for the time-course experiments. Less DBD was
consumed in the rate than in the product study since the
C1, content in the ClO, solution was lower (2.4% vs 4.6%)
in the former.
The concentrated extracts were analyzed by GC-MS as
well as by HPLC. The principal product, 2-chlorodi-
benzodioxin, was identified by mass spectral and HPLC
comparison with authentic materials. Although the elec-
tron impact mass spectra and the GC retention of 1- and
2-chlorodibenzodioxins are similar, the two isomers are
easily resolved by HPLC on a reversed-phase column. The
chlorodibenzodioxinyield was small (5%),and the absence
of other detectable methylene chloride extractables sug-
gests that the remaining products were polar.
Results and Discussion
Distribution of DBD in Pulp-Water Mixtures.
Owing to its high Henry's law constant of 12.29 atm m3
mol-' at 25 "C (3) DBD can potentially volatilize from
water, although its rate of removal may be retarded by
association to pulp. Our recovery studies show that a
significant fraction of DBD volatilizes at 60 "C in 30 min
under our conditions. Volatilization should be less im-
portant in a bleach plant where the surface area to volume
ratio of the pulp-water mixture is much lower.
The DBD pulp-water distribution coefficient
([DBD],,, *[DBD],,,,) for the 40-K pulp is -200. This
value can t e interpreted with reference to sediment-water
systems where sorption occurs principally to the organic
carbon fraction of sediment. Karickhoff has shown (4) that
sediment-water distribution coefficients normalized with
respect to sediment organic carbon (K,) can be related to
the octanol-water partition coefficient (KO,) of the solute
through eq 1. If pulp behaves similarly to sediment, then
log KO, log KO,- 0.21 (1)
eq 1 used with a DBD log KO,of 4.37 (5) and a 45% es-
timate for the organic carbon content of pulp predicts a
distribution coefficient of -7000.
This value falls sharply if it is assumed that sorption
occurs primarily to the lignin fraction of pulp. Garbarini
and Lion (6) studied the sorption of toluene and tri-
chloroethylene to lignin and cellulose particles and ob-
tained distribution coefficients of 120-150 for lignin-water
and only 0-2 for cellulose-water partitioning. The organic
carbon contribution from the lignin in our pulp is only
-4% since lignin is approximately 62% carbon (7). If
sorption to cellulose is neglected, then eq 1 predicts a K ,
value of -600. While this estimate is 3-fold higher than
the measured value, it is reasonable given that the particle
size of our pulp fibers was much larger than that of the
sediments used to develop eq 1. I t should be noted that
the DBD level used was much higher than that expected
in a mill environment. The distribution coefficient will
apply to a mill situation only if a linear isotherm is as-
sumed.
The finding that sorption to bleached pulp is about
8-fold lower than that to 40-K pulp emphasizes the im-
portance of lignin as the primary sorbent. However, the
degree of sorption to the bleached pulp is still appreciable
and suggests that DBD can bind to non-lignin components.
Resin acids and extractives that remain in bleached pulp
are possible candidates, although at 0.1-0.2%, they cannot
account for the full extent of the observed sorption. Also,
pulp retains a considerable amount of water, and it is very
difficult to completely separate the two phases under
Environ. Sci. Technoi., Vol. 26, No. 7, 1992 1325
 0
03
'h*
0
(0
,(; i"id;geha,,~~3b,,
E/,,, , , , , 'idb', , , , ,
0
minutes
Flgure 1. Loss of DBD from water.
equilibrium conditions. Hence, KO,will be biased high
from solution carryover.
Garbarini and Lion (6) used extracted lignin whereas our
results are based on residual lignin incorporated in the pulp
matrix. The K , value from the 40-Kpulp is broadly con-
sistent with Karickhoff's equation and suggests that lignin
in pulp behaves similarly to the chemically extracted
material. This opens the possibility of estimating sorption
to pulp only from a knowledge of solute KO, and the
properties of the pulp.
Reactivity of DBD in Pulp-Water Mixtures. Ad-
dition of a CIOz solution to an aqueous solution of DBD
degrades the substrate as shown in Figure 1. DBD loss
in the control experiment reflects volatilization. The
chlorination half-life of approximately 30 min is of the
same magnitude as the residence time of pulp in a com-
mercial bleaching stage, indicating that dissolved DBD
would be substantially degraded. Results from corre-
sponding experiments conducted in the presence of the
40-Kpulp are illustrated in Figure 2. Clearly, DBD re-
activity is strongly inhibited through sorption, presumably
to lignin. Possibly, the compound is protected by lignin
by virtue of ita reduced access to the bleaching agent. It
seems that DBD degradation will be optimum when the
lignin is substantially depleted and the DBD is released
to the aqueous phase. Although much of the DBD applied
is consumed, a significant fraction survives the bleaching
process. These results are consistent with an earlier ob-
servation (8) that dioxin precursors are not completely
destroyed by C102.
Product distribution was measured from 30-min
bleaching experiments with [3H]DBD. The 40-K pulp and
water were separated after quenching, extracted with
methylene chloride, and analyzed by HPLC for DBD and
liquid scintillation counting for total activity. Of the or-
iginal activity, 16% was found in the pulp extracts and
16% in the aqueous phase. By HPLC, 8% of the initial
DBD appeared in the water and 4% was found in the pulp
extract. The pulp to water DBD ratio is biased toward
water in contrast to results from our sorption experiments
with the 40-Kpulp. This follows from the removal of lignin
during bleaching, which would reduce the sorptive capacity
1326 Environ. Sci. Technol., Vol. 26, No. 7, 1992
8jh
k without bleaching agent
r n r
a,
,;Jo Q
0
0 50 100 150 200 250
minutes
Flgure 2. Loss of DBD from pulp-water suspensions.
of the pulp. A total of 28% of the activity remained in
the pulp as inextricable material as determined by direct
counting of the pulp fibers. Most of the activity applied
could be accounted for after allowance for volatilization.
Methylene chloride extracts of the pulp and the spent
bleaching liquors contained a small amount (5 mol %) of
2-chlorodibenzodioxin.
Conclusions
Sorption of DBD to pulp occurs principally to the lignin
fraction. In laboratory-scale chlorine dioxide bleaching
experiments, 5% was converted to 2-chlorodioxin and 28%
to products inextricable from pulp. The remainder was
unreacted DBD, unidentified products, and material lost
through volatilization. Sorption to (the lignin component
of) pulp greatly reduced the reactivity of DBD.
Registry No. DBD, 262-12-4; CIOz, 10049-04-4; 2-chlorodi-
benaodioxin, 39227-54-8.
Literature Cited
(1) Voss, R. H.; Luthe, C. E.; Fleming, B. I.; Berry, R. M.; Allen,
L. H. Pulp Paper Can. 1988, 12, T401.
( 2 ) Werstiuk, N. H.; Kadai, T. Can. J. Chem. 1973, 51,
1485-1486.
(3) Shiu, W. Y.; Doucette, W.; Gobas, F. A. P. C.; Andren, A.;
Mackay, D. Environ. Sci. Technol. 1988,22, 651-658.
(4) Karickhoff, S. W.; Brown, D. S.; Scott, T. A. Water Res.
1979, 13, 241-248.
(5) Doucette, W. J.;Andren, A. W. Environ. Sci. Technol. 1987,
21, 821-824.
(6) Garbarini, D. R.; Lion, L. W. Environ. Sci. Technol. 1986,
20, 1263-1269.
(7) Lai, Y. Z.; Sarkanen, K. V. In Lignins: Occurrence, For-
mation, Structure and Reactions; Sarkanen, K. V., Ludwig,
C. H., Eds.; Wiley-Interscience: New York, 1971; Chapter
5, p 178.
(8) McDonough, T. J.; Lafleur, L. E.; Brunck, R.; Malcolm, E.
W. International Pulp Bleaching Conference, Stockholm,
June 1991; Vol. 2, p 171.
Received for review December 26, 1991. Revised manuscript
received March 16, 1992. Accepted March 17, 1992.