Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Ionization type smoke detectors.

Bernard said,
... stopping the Alphas would have the
opposite of the intended effect, a "proof" of his explanation.

Brian said,
Uh? Alphas are stopped in large carbon particles - Oh, all right,
greatly slowed?

Having products of combustion (POC) stop the alphas makes ions from the
POC just like making ions when the alphas interact with air. Therefore,
unless the POC stop the alpha particles before the alphas enter the
ionization chamber, the POC don't inhibit ion production. If the alphas
are stopped in the ionization chamber by whatever, there will be plenty
of ions to detect regardless of whether the ions are from air or from
POC. So Bernard's comment is correct.

By the way, in smoke detectors I have examined, the alpha source is
right inside the ionization chamber, so stopping alphas before they
enter the chamber is a moot point.

Concerning the discussion of kerosene heaters and ventless gas
heaters... We need to discuss at least three issues. The worst scenario
is production of CO because it is highly toxic in small amounts. It
preferentially binds to hemoglobin and prevents the hemoglobin from
picking up oxygen in the lungs and delivering it to the rest of the
body, e.g. the brain. The next-worst scenario is buildup of CO2 because
when the body detects elevated CO2 levels it causes physiological
changes to try to compensate for what it thinks is a problem in getting
CO2 out of the body. We can tolerate much higher levels of CO2 than CO,
but not as high as one might think. If CO2 levels in the air get above
about 3% then physiological effects are observed, and somewhere between
5% and 10% a person can become unconscious and eventually die. The
third situation is oxygen depletion in the air. The body is also pretty
sensitive to drops in O2 levels, although the bad effects of a slight
drop in O2 are not as serious as a slight rise in CO2.

In short, a non-vented heater must not make CO. It will make CO2 and
also deplete O2. The CO2 buildup is slightly worse than O2 depletion,
but both are bad. The hope is that the house has enough air leaks that
CO2 rise and O2 drop will not reach dangerous levels.

In another post Bernard asks more questions about ion chambers. There
are many issues here, and I'll only comment briefly. Yes, the
ionization track of a charged particle that comes to rest in an
ionization chamber is both + and - charges, of course. In a gas, both
+ and - charges are mobile. If the "sensing" electronics are on the
anode of the chamber, the - ions are heading toward it, and the + ions
are heading away from it. If these are both moving with the same
velocity, it yields twice the current in the sensing electronics as
would be the case if only the - ions were mobile (or vice-versa). Thus,
to get big signal, you would prefer that both + and - ions are mobile.
Go back to n*q*v. If only the - ions are mobile, then n is only half
the total ions present. The immobile (or very slow) ions do not
contribute to the signal because their velocity is zero or near zero.

This is another point about ion-chamber electronics not often realized.
You don't need electronics on the anode to detect the - ions and another
set of electronics on the cathode to detect the + ions. You can detect
either polarity of charge moving by a single set of electronics on
either the anode or cathode. Positive charge moving toward the
electrode connected to the sense electronics yields the same polarity of
signal as negative charges moving away from that same electrode (and/or
vice-versa). The detection of the ions *is not* because the ions hit
the electrode and then go into the electronics. Once the ions hit the
electrode, they have already been sensed. The ions' image charges have
traveled through the electronics and meet the ions at the electrode,
neutralizing them at that point. The event is over when the ions hit
the electrode.

A "whole-nother" complication arises if the sense electronics are
integrating the current in order to yield the total ionization created
by the ionizing radiation. You now have to consider how long the
current exists, which means you need to know how far the charges being
sensed travel in the ionization chamber. We can write this a (n*q*v)*t
= (n*q*v)*d/v = n*q*d. Thus, the charge sensed depends on d, the
distance the ions moved during the detection period. If the integration
time of the charge-sensitive amplifier is sufficiently long for the ions
to clear out of the chamber, then the track orientation in the chamber
becomes very important. Tracks that make ions in one end of the chamber
can create a much larger integrated signal than tracks created in the
opposite end of the chamber. The effect is small if both + and - ions
are mobile with about the same velocity. This is true for electrons and
holes in a semiconductor ionization chamber. But in gaseous chambers,
and especially liquid chambers, the + ion drift speed is typically lower
(much lower) than the electron drift speed, so the signal measured can
be quite dependent on where the radiation enters the chamber. This can
be fixed by putting grids in the chamber and making sure the radiation
enters between the grids.

Yet another thing Bernard mentioned is when the HV on the chamber is
sufficient to accelerate ions enough (generally electrons) so they
collide with atoms with sufficient energy to make more ions. That
happens in a Geiger-Mueller tube and some other types of ionization
detectors... but not in smoke detectors.

I didn't take the time to explain all this very thoroughly, but perhaps
the major point to be made is that ionization chambers (even in things
like smoke detectors) is more complicated than one initially imagines.

Michael D. Edmiston, Ph.D.
Professor of Chemistry and Physics
Bluffton University
Bluffton, OH 45817