Dew Points and Close Calls
It''s one of those very hot and sticky July days and the air is
heavily laden with moisture. However, the day has been
lightened by news of the rescue of those miners in Somerset,
Pennsylvania. My wife is from a coal-mining town near that
area and recalls as a child standing near the mine entrance when
the news was not so fortunate and miners died. The close call of
the nine miners in Somerset reminded me of my own close calls
such as the big tree falling on the spot on the road to Hana in
Hawaii we had just traversed 15 seconds earlier. In an earlier
column, I told how Breyer''s ice cream had saved my life as a
child in another close call. If you missed that column, my family
was debating whether it would be Breyer''s or Rakestraw''s ice
cream for dessert. I was designated to go and get the ice cream.
Rakestraw''s ice cream factory was just a block up the alley from
our house. However, we chose Breyer''s. Otherwise, I would
have been at Rakestraw''s when the ammonia plant exploded,
killing two little girls. I''ve loved Breyer''s ice cream ever since.
Back to the weather, we say, "It''s not the heat it''s the humidity."
We could just as well substitute "dew point" for "humidity". At
any temperature, the pressure of water vapor over liquid water or
ice has a certain equilibrium value. This "vapor pressure" goes
up with temperature. If we have a hot sticky day and it cools
down enough at night, the air contains too much water vapor (the
pressure of water vapor exceeds the vapor pressure of water at
the cooler temp) and we find dew on the ground in the morning.
Regarding close calls, I had a close call in my scientific career at
Bell Labs. I was almost fired but trying to measure a dew point
saved the day. Last week, I mentioned arsenic as a toxin that, in
the proper dosage, can be used as a therapeutic agent. My
attempt to measure the dew point of arsenic salvaged my career.
My tale illustrates that metals also have dew points and even the
simplest measurement can present unexpected challenges.
When I arrived at Bell Labs in 1952, I was a na ve 24-year-old.
My first project involved working with lithium in germanium,
which preceded silicon as the material for transistors. I had some
experience with a lithium salt, lithium bromide, in my thesis
work at the University of Pittsburgh. At Bell Labs,I found that
lithium metal was a different beast entirely from any lithium salt.
Being na ve, I didn''t tell anyone when I gave up on the project -
handling metallic lithium was more than I had bargained for. As
a result, I had little to show for my first year at Bell Labs. My
department head, Addison White, told me that they were very
disappointed in my work and that I was on probation. Even so, I
was given a token raise of $20 (can''t remember if this was for a
month or a year!). My career with Bell was in grave jeopardy.
Well, when I dropped the lithium project, I had decided to try to
measure the vapor pressure of arsenic over germanium at various
temperatures. (More precisely, for those acquainted with phase
diagrams, the partial pressure of arsenic along the liquidus curve
in the germanium-arsenic system.) This seemed worthwhile
since such measurements could help guide the transistor makers
in controlling the amount of arsenic in their devices. So, how to
measure the pressure? If you want to measure the pressure in a
tire you use a pressure gage. If you want your doctor to measure
your blood pressure, he or she will use a mercury manometer.
Measuring the pressure of arsenic at high temperatures isn''t that
easy. Without going into detail, there are a number of ways that
involve some pretty sophisticated apparatus and calculations.
These methods had been used to determine vapor pressure curves
for virtually all the chemical elements. Being simple-minded, I
decided on the simple dew point technique, invented many years
earlier to measure pressures over metal alloys. All one needs is a
furnace with a little window and heating elements wired to allow
control of two temperature zones.
Zinc was a favorite metal whose vapor pressure had been
measured using the sophisticated methods. Let''s measure the
pressure of zinc over a zinc-germanium alloy. First, we seal off
chunks of zinc and germanium under vacuum in a long glass or
quartz tube. Next we heat the tube up to melt the zinc and form a
zinc-germanium alloy in one end of the tube. Now let''s start
cooling the other end and watch it through our window. As we
cool down, we suddenly see little droplets of zinc forming. It''s
the dew point of zinc. So, we get out our tables and look up the
vapor pressure of zinc at the temperature of the cold end. That''s
the pressure of zinc over the zinc-germanium alloy. No fancy
Let''s use this dew point method for arsenic-germanium alloys.
Arsenic doesn''t form liquid droplets but that''s ok. We''ll just
watch for solid arsenic condensing out at the cold end of our
tube. So, after forming our arsenic-germanium alloy in one end,
we cool down the other and wait for arsenic to appear. We cool
down way below the temperature that we expected to see
something. Finally a whole bunch of arsenic appears. Ok, we
say, the pressure is lower than we expected and, being smart, we
raise the temperature a bit to watch the arsenic evaporate back to
the hot end of the tube. But, even when we raise the temperature
many degrees, the arsenic not only doesn''t evaporate off but
more arsenic keeps condensing out! What gives?
Belatedly, we decide that we''d better read up on arsenic vapor.
We find that our hot, sticky day has its counterpart in arsenic
vapor; there''s something called a "sticking coefficient". Simply
put, even though we''re way below the dew point, the arsenic
atoms in the vapor just don''t stick to the glassy interior of the
cold end of the tube. Virtually all of them just bounce right back
into the vapor. The arsenic vapor "supercools", much like pure
water can supercool before freezing. Finally, at a much lower
temperature, enough arsenic atoms stick to form a nucleus on
which a flood of arsenic atoms condenses to form a solid deposit.
Obviously, the temperature of the cold end is now useless as far
as calculating the arsenic pressure over our alloy. It''s that glassy
surface that did us in.
What to do? There''s more than one way to skin a cat (with
apologies to Dan and Jeanne and other cat lovers). Let''s "invert"
the dew point method and watch the germanium turn liquid as it
forms the arsenic-germanium alloy. Actually, we hadn''t foreseen
this problem and didn''t have another window in the furnace. At
this point I''ll let you off the hook - I hadn''t foreseen the problem
and my furnace wasn''t flexible enough to make the cold end the
hot end. So, I settled on starting with a chunk of solid
germanium at one end, the arsenic at the other. I then just took
the tube out of the furnace periodically to see at what
temperatures the germanium showed signs of "melting" to form
the alloy. This was tedious but I finally got some data that, while
not the greatest, were acceptable.
But there was another problem. When I mentioned that the
arsenic atoms didn''t want to stick, I was misleading you. I read
that arsenic vapor isn''t just atoms, but also arsenic molecules
with 2 and 4 atoms, just like normal air isn''t oxygen and nitrogen
atoms but molecules, in their cases with two atoms in a molecule.
This complicated other calculations using my data but I figured
out how to do them and wrote a technical memorandum.
I thought the work was pretty neat. Ad White must have agreed.
He called me in to say that I was off probation and welcomed me
to the technical staff of Bell Labs. Addison White passed away
at an advanced age a few years ago. Before his death, I saw him
at a celebration of some sort and remarked to him that he had
almost fired me some 40 years earlier. I was quite surprised
when he remembered the incident vividly.
Ironically, I was to spend the last 17 of my 36 years at Bell Labs
working on lithium batteries. Handling lithium metal became
routine and, 30 years after giving up on lithium, I was made a
Distinguished Member of Technical Staff. And I owe it all to
arsenic and my attempts to measure a dew point.
Allen F. Bortrum