extraction Q./hard evidence that PVC's a major source

[ttweed@wildrockies.org (Tony Tweedale)]

[bcc to several Gpeace activists]

from EPA's Jan. '99 revised method TO9A for ambient air P-Cl&Br-D-D/F, sec. 3.1:

"The isomer profiles found in ambient air are similar to those found in
combustion sources.  Isomer PCDD/F profiles related to products and
by-products and are quite different in that only a few and characteristic
isomers are detectable, which clearly indicate they are not from combustion
sources."
--

As many on these lists know, there is some scientific question as to the
major source of dioxins.  While penta (PCP) apparantly accounts for the
majority of dioxins created (89.9% in '95, according to estimate by EPA's
'98 draft 'Inventory of Sources of Dioxin'), common sense indicates that it
doesn't travel far, as most ends up in wood and leaching into soil (for the
treated wood that is buried, e.g. tel. poles).  Some would volatilize, and
more would have escaped to the air during manufacture, but "the" rational
hypothesis is that that amount is negligible compared to combustion
sources.  A pretty definitive proof is the above statement, I imagine.

Anyway, the 1st sentence in the quote is something I hadn't realized, in
the PVC/source debate.  IT IS SEEMINGLY A STRONG PROOF THAT WHAT COMES OUT
OF COMBUSTION STACKS IS PRETTY IDENTICAL IN COMPOSITION TO WHAT'S FOUND IN
THE AIR, EVERYWHERE (ie everywhere ambient samples are taken).  Combined
with the now undisputed facts:  1) that PVC is the major source of chlorine
in municipal & medical waste combustors; and 2) that (see Thomas & Spiro's
Table 1 that summarizes all avail. studies, in issue #1 of "95 _Tox & Env
Chem_):  'mass of chlorine in = mass of PCDD/F coming out the stack' [as
the dioxins are formed 'de novo' from free Cl- (chloride) &/or chloride
salts (mostly H::Cl, that all the sources of Cl in the fuel are converted
to in the energetic combustion chamber environment) when the flue gas is
cooling to the ideal C-Cl covalent bond forming T (very roughly, 200-800
deg Centigr.)], this neat fact starts to close the argument, once again, in
favor of enviros.

True,  many industrial boilers do not burn PVC, but rather another source
of chlorine.  Also, from u.v. (sun) photolysis, PCDD/F in the environemnt
migh be changing the congener composition of ambient air's PCDD/F, such
that PCDD/F volatilized from non-combustion sources in the environment is
transformed into a faux combustion source profile--but this would seem very
unlikely.  Also, profile's of soil, sediment, water, on some
surfaces--anything not directly exposed to a fair amnt of sunlight-- don't
change much at all (half-life studies in these environments show that after
10-20+ yrs, most of the change measured is actually loss to other
enviroments (e.g, in wind borne dust).  So while it may be said that
photolysis might confound the argument somewhat, it seems to me (at first
aquaintance) that this is a potent new argument to make regarding PVC as a
major source of dioxins in the environment.

It seems a robust argument, as apparantly ambient air samples everywhere
have congener profiles matching what comes out of combustion stacks
everywhere.  After writing these p'graphs, I now feel I may have vaguelly
heard of this argument before ...in any case it's nice to have an
authoritative source (even if it's a secondary (literature) source).  A
great study would be a comparison of congener profiles in ambient air
surrounding different sources--PVC stacks vs. non-PVC but equally chlorine
rich stacks; natural sources (eg volcanos); etc.  Also, the profiles of
what's deposited on vegetation that goes into our food chain, that is grown
near these various sources...

---


I came across the above quote while looking into the recent post to
dioxin-l re: acetone extraction [ie extacting the analytes from the sample
medium (eg, PUF, polyurethane foam) in preparation to run through the gas
(or liquid?) chromatographer that elutes the analytes at different rates so
their amnt can be quantified in a mass spectrometer] reporting only half
the amnt of PCDD/F present compared to toluene extraction.  Because a group
I work with funded (via a consent decree w/ Smurfit-Stone Container Corp's
local kraft P&P mill) an ambient valley air sampling study, I asked about
poor acetone extraction of the chem grad student who is working on that
study, and he referred me to the above method (EPA's TO9A method).  He
mentioned it uses benzene extraction.

I have some Q.'s:

!) the method says many labs are switching to toluene, and EPA is
considering it, as benzene is a known human carcinogen (and I guess because
the extraction step is a heated one).  Also the post referred to toluene
extraction.  Not that toluene a non-toxic solvent, but I assume it &
benzene for PCDD/F extraction give very similar results across all
congeners?  Brominated ones too?  Is it that acetone is a somewhat water
soluble (I think), straight chain solvent, compared to benzene & toluene,
that accounts for the difference?

2) 50% seems a terrible recovery rate for the extraction step--how many
methods or labs are using acetone?  Might this be due to health concerns?
The post that was a press article or release from Japan was in the context
of food matrix samples--does acetone (or any other poor-PCDD/F extraction
solvent) need to be used because of different sample matrices, or for any
other reason?  How many labs use acetone extraction, and why are they
allowed to get away with it?!

3) I see from the method hexane is part of the benzene extraction--why? Am
I correct assuming it wouldn't be a win-win (safe, yet excelent PCDD/F
extraction) substitute for benzene or toluene, as the "benzene" extraction
step would then use it alone?

4) Finally, a bit off topic, I see the next step in the method is acid-base
washing w/ a bunch of polar (water soluble) acids & caustics.  I assume,
from reading the introductory sections of the method, this may be to remove
interferants (molecules w/ similar weight & structure), such as the common
halogenated diphenyl ethers, whose oxygen makes them more water soluble?
Regardless, does this step potentially cause significant losses of PCDD/F
too, despite PCDD/F not being very H2O soluble at all?   Also I briefly
read that several other solvent-like compounds are used around this time
(eg chlorbenzene), in the method--are they also used to clean-out
interferring compounds?  Do they present any problems in under or over
reporting PCDD/F because of them?  Finally, any general comments about
interference in PCDD/F methods?

Thanks to anyone who takes up all these Q's!


Tony Tweedale

Causality is a concept not subject to empirical demonstration. -David Hume
(1711-'76)

Temperate but endangered planet.  Enjoys weather, northern lights,
continental drift.  Seeks caring relationship with intelligent life form.
      -Friends of the Earth