Determining source of odour

Karen Potje <kpotje [at] cca__qc__ca> writes

>Does anyone have experience in detecting the sources of odours?  Any
>advice? We have talked about buying gas detection tubes which we
>would deploy whenever the odour is noticed, but we have no idea what
>gases we are looking for.  Does anyone know of a substance which can
>be released to produce an odour without causing any harm to
>collections and staff?  Is this a wacky idea?  Could it possibly
>work?

There are various chemicals that you can release that produce strong
but safe odours (e.g., the 'gas smell' used to mark otherwise
odourless natural gas).  However, I don't think this is necessarily
the best approach because

    (a) your odour problem is intermittent, thus you may need
        multiple releases, even at the correct location, to
        replicate the effect, and

    (b) odours would probably be objectionable if they ended up in,
        say, the main body of the museum.

The various gas tracer methods (sulfur hexafluoride, carbon dioxide,
PFT) that could be used would reduce the odor problem, but, except
for PFT, still suffer from objection (a).  Of these, sulfur
hexafluoride offers high sensitivity but the equipment is somewhat
expensive.  Carbon dioxide is a poor tracer gas in many situations.
Carbon monoxide has been used as a tracer gas in low dilutions and
is relatively easy to detect, but most public institutions freak out
when this is suggested.  PFT (perfluorocarbon tracer) offers the
following advantages:

    1.  PFTs are nontoxic, nonreactive, nonflammable,
        environmentally safe (contains no chlorine), and
        commercially available;

    2.  PFT technology is the most sensitive of all non-radioactive
        tracer technologies and concentrations in the range of 10
        parts per quadrillion of air (ppq) can be routinely
        measured;

    3.  The PFTs technology is a multi-tracer technology permitting
        up to six PFTs to be simultaneously deployed, sampled, and
        analyzed with the same instrumentation. This results in a
        lower cost and flexibility in experimental design and data
        interpretation.

    4.  All six PFTs can be analyzed in 15 minutes on a laboratory
        based gas chromatograph.

So, the advantage is that you can use slightly different gasses
released from up to six different locations simultaneously.  The
gasses can be detected in the parts per quadrillion range and there
are negligible natural background levels.  The perfluorocarbons are
available in permeation sources with calibrated release rates: you
put one source at each site of interest (where you think the odor
might be coming from), and use CATS (capillary adsorbent tracer
sampler) tubes in your lab office to sample the incoming gasses.
CATS tubes are small cigarette sized glass tube containing a
carbonaceous adsorbent specific for the PFTs. The sampler can be
used dynamically (flowing a sample through the CATS) or passively
(opening only one end so as to allow the CATS to sample by
diffusion). The passive mode allows a time integrated PFT
concentration to be measured in a simple manner. The CATS are
shipped back to the laboratory for PFT analysis.  Several real-time
PFT analyzers are available, one which detects four different PFTs
per five minutes (used in residential radon studies). However, as
with sulfur hexafluoride, the PFT technology is expensive.  You
would still nedd to keep a log of the appearance of the smell so
that this could be correlated with

Before you get mixed up in tracer gas studies, I would recommend
that you do some further work of your own to see if you can localize
the source. After all, even with PFTs, you can only run six
different gasses simultaneously, and you don't want to go shooting
in the dark.  The indispensable (and cheap) tool here is a smoke
pencil.  This is a little puffer bottle filled with fine dark powder
that is blown into the air from the puffer to indicate the direction
of local air flow (somewhat more sensitive than a wetted finger).
After this, you move to pressure differential studies using a
differential manometer.  There are electronic units which are stable
to 1 pascal and cost in the sub $1,000 dollar range.

With both techniques, the idea is to establish what your office is
depressurised with respect to (i.e., the most likely route for
infiltration).  Since the effect is intermittent, is there any
relationship between wind direction/velocity and appearance of the
odour?  The elevator shaft sounds like a possible source (gasses do
indeed whoosh up and down elevator shafts), but I would expect some
similar odour complaints from users of the elevator if this is
indeed the source.  If, as you say, there is no relationship between
occupancy of the loading dock the appearance of the smell, you need
to look for another source of the burnt diesel smell. One other
source for this smell that I can think of off-hand would be burning
electrical insulation (scary thought), but I assume you have either
eliminated this as a possibility, or that the smell is not
characteristic of this problem.

If you want any more information, the Institute for Research in
Construction (at NRCC-CNRC) would probably be the people to point
you in the right direction: they have considerable experience in
tracer gas studies. Regards,

JP Brown
Environmental Conservator

                                  ***
                  Conservation DistList Instance 12:52
               Distributed: Wednesday, December 16, 1998
                       Message Id: cdl-12-52-006
                                  ***