Index Home About Blog
Subject: Re: Microgravity experiments: commercially successful?
From: Henry Spencer <> 
Date: Apr 24 1996

In article <4l6grm$> (Paul F. Dietz) writes:
>>Uh, "commercially successful"?  What commercial markets are there for
>And, more to the point, what possible advantage would there be of
>doing it in orbit?  Everything is more expensive there, vacuum in
>terrestrial accelerators is perfectly adequate, and heat dissipation
>is much easier down here.  Any antimatter factory will be
>dissipating 99.99+% of its input energy as heat.

While the power and thermal problems are nuisances, on Earth there are
going to be equal or greater headaches from gravity, air leaks, and 57
different regulatory agencies. :-)  In particular, non-contact handling
techniques work considerably better in microgravity.

Also, power and heat dissipation are no longer straightforward problems
even on Earth when you get up into the gigawatt range.

Early antimatter manufacturing will probably be done on Earth, but bulk
production will almost certainly be done in space. 
Americans proved to be more bureaucratic           |       Henry Spencer
than I ever thought.  --Valery Ryumin, RKK Energia |

From: "Gordon D. Pusch" <>
Subject: Re: (no subject)
Date: Tue, 13 Aug 1996 22:36:43 -0500

The following message is a courtesy copy of an article
that has been posted as well.

In article <3211112E.4A98@OVPR.UGA.EDU> "RICHARD J. LOGAN"
<> writes:

> Exactly how would an anti-matter engine work?  When the products
> annihilate you're going to get high energy gammas that will in all
> liklihood just pass through any fuel you're trying heat.  How would
> this mechanism be used to produce propulsion?

This is a very common misconception; however, it is incorrect --- 
the majority of the energy comes out as pions, not gammas.  
The following data are taken from R.L. Forward's report entitled
``Antiproton-annihilation propulsion.''

Each p/p-bar annihilation yields an average of 1.49 pi+s, 1.49 pi-'s,
and 1.52 pi0's. A p-bar can *also* react with a neutron (discussed
below), so when it annihilates in heavy nuclei the, branching ratios 
are somewhat different (albeit experimentally poorly determined,
so I won't quote them); furthermore, since most of the pi0's are
absorbed internally by the nucleus, their energy will instead emerge
in the form of spalled nucleons and nuclear fragments generated by
pion-induced fissions.

The simplest ``explanation'' for why nucleon/antiproton annihilation
tends to produce pions is simply because they are the lowest-mass
strongly-interacting-particles in existence, and therefore are
energetically the ``easiest'' particles to produce. If you prefer a 
``quark model'' explanation, then it is because nucleon/antiproton
annihilation actually involves quark/antiquark annihilation.  
Protons consist of two 'ups' and a 'down', neutrons consist of 
two 'downs' and an 'up', and antiprotons consist of two 'anti-ups' 
and an 'anti-down'.  Likewise, the pi+ is an 'up' and an 'anti-down',
the pi- is an 'anti-up' and an 'down', and the pi0 is a linear
combination of an 'up/anti-up' and a 'down/anti-down' pair 
(it also happens to be its own antiparticle). 

When a proton and an anti-proton annihilate, only =one= quark and
anti-quark annihilate, leaving two pairs of quarks and antiquarks
behind.  The easiest way to rearrange these remaining quarks and
antiquarks into ``colorless'' (unconfined) particles so they can
escape is to pair them off into pions; additional quark/antiquark
pairs may also be created to balance out the conservation laws, 
or simply because there is enough energy available to do so 
(since these are sometimes 'strange/anti-strange' pairs, one also
occasionally sees K+/K- pairs among the annihilation products.)  
On the average, 1.49 'pi+'s and 1.49 'pi-'s are produced, and 
1.52 'pi0's.  The most ``branching ratios'' are as follows:

               pi+ | pi- | pi0
  34.5%  --->   1,    1,    2 
  21.3%  --->   2,    2,    2 
  18.7%  --->   2,    2,    1
   7.8%  --->   1,    1,    1
   5.8%  --->   2,    2
   1.9%  --->   3,    3
   1.6%  --->   3,    3,    1

All other reactions occur with a probability of 1.3% or less;
most of them involve Kaons or other more massive particles,
which is why they are less likely. 

Notice that =none= of the above reactions yield gammas directly;
the majority of the gammas coming from p/pbar annihilation are 
from pi0 decay. (A *few* annihilation reactions =do= yield gammas
directly, but they are strongly suppressed by parity conservation.)
Some gammas come from electron/positron annihilation as well, but
they represent an energetically negligible fraction of the yield.

It should also be clear, at this point, why a neutron and an antiproton 
can react --- since a neutron consists of two 'downs' and an 'up', 
quark/antiquark annihilation is still possible, albeit with different
(and still poorly determined) branching-ratios than in the above table.

The pion spectra from p/p-bar annihilation peaks at around 135 MeV;
the spectrum has a long, quasi-exponential tail, so the mean energy
of the charged pions is significantly higher --- around 250 MeV. 
The pi0's decay almost immediately to gammas with a peak energy 
of around 80 MeV and a mean energy of around 200 MeV.

The charged pion mean range at the most probable pion energy is a strong
function of material.  In vacuum, it is about 12 m (the decay muons, by
contrast, have a mean range of about a kilometer). In H2 at .027 gm/cc, 
the pions have a mean range of about 7m; in N2 at .125 gm/cc, it is
about 3 m. In tungsten, by contrast, the most probable mean range of
charged pions is only about 5 cm. This suggests that, just as in a 
thermonuclear device, a properly designed ``tamper'' can have a strong
effect on the conversion efficiency of antimatter-annihilation energy
into thermal energy.

In summary, the majority of the energy released in matter/antimatter
annihilation emerges as charged pions, not gammas. Since pions are
strongly absorbed by matter, and can be confined and focused by
magnetic fields, ensuring that annihilation energy is efficiently
coupled to the propellant is not nearly the problem many have 
falsely believed it would be...

Gordon D. Pusch                     |  Internet: <>
Math and C.S. Div., Bldg.203/C254   |  FAX:      (708) 252-5986
Argonne National Laboratory         |  Phone:    (708) 252-3843
9700 South Cass Ave.                |  
Argonne, IL  USA  60439-4844        |

But I don't speak for ANL or the DOE, and they *sure* don't speak for =ME=...

From: (Henry Spencer)
Newsgroups: alt.war.nuclear,,sci.physics
Subject: Re: Anti Matter Triggered Fusion?
Date: Mon, 19 Jul 1999 23:26:23 GMT

In article <>,  <> wrote:
>>specialized applications like space propulsion, where cost matters less
>>and shifting work to the ground pays off big, the idea has potential.
>The idea only has potential if you can store significant amounts of
>antimatter, in a secure fashion.  Currently you've no way of doing it.

Depends on what you mean by "significant".  The particle-physics guys are
already storing small quantities of low-energy antiprotons quite securely.
There are at least rough notions of cooling them further, throwing in some
positrons, and condensing the antihydrogen to a solid.  There are several
ways of handling little bits of solid antihydrogen without touching it.
This was investigated in some detail a while back; the conclusion was that
plenty of engineering development was needed, but no breakthroughs.

Actually producing enough antiprotons is a bigger problem.  The current
production processes are hideously inefficient, and although considerable
refinement is possible -- the equipment isn't optimized -- it's still
going to be a costly operation with very large infrastructure costs.
(Robert Forward's estimate was that an accelerator the size of Fermilab,
but optimized for antimatter production, could make enough in a year to
test-fire an antimatter rocket engine.  Making enough for practical use in
space exploration would require a production facility the size of the
Hanford works.)
The good old days                   |  Henry Spencer
weren't.                            |      (aka

From: (Henry Spencer)
Newsgroups: alt.war.nuclear,,sci.physics
Subject: Re: Anti Matter Triggered Fusion?
Date: Tue, 20 Jul 1999 17:20:34 GMT

In article <>,  <> wrote:
>>Depends on what you mean by "significant".  The particle-physics guys are
>>already storing small quantities of low-energy antiprotons quite securely.
>Translating to: storing such small amounts that the inevitable leakege
>is not an issue.

While the amounts are indeed small, the leakage likewise has to be fairly
small if you're going to store them for hours or days, as is now fairly
routinely done.  (Leakage is a percentage, not an absolute rate, so it
affects small quantities just as badly as big ones.)

>>...There are several
>>ways of handling little bits of solid antihydrogen without touching it.
>There are?  I love it when people say "there are ways" on something
>that was never done:-)  You may legitimately say "there may be", but
>that's all.

Uh, no, there is no particular question about this, because the methods
involved can be tested quite thoroughly without using antihydrogen.

For example, if you toss in a few electrons to get rid of a few of the
positrons, you put an electrostatic charge on the antihydrogen ice, and
electrostatic levitation of much larger objects is done quite routinely
(for ultra-high-quality gyros, for example).

Another example is that solid hydrogen/antihydrogen, like many substances,
is weakly repelled by a magnetic field, and can be levitated by one.  This
has been done for a number of substances, e.g. graphite, although I don't
know whether anyone has tried it with solid hydrogen yet.

Laser levitation likewise is routinely demonstrated, although it would
rely on having the antihydrogen ice optically clear enough that it doesn't
absorb much of the light (since you don't want it heating up).

None of this is exactly rocket science; non-contact handling is clumsy but
its basic feasibility is not in doubt.  Condensing the gas to antihydrogen
ice is actually a rather bigger challenge.

>>This was investigated in some detail a while back; the conclusion was that
>>plenty of engineering development was needed, but no breakthroughs.
>Scaling something up by 20 orders of magnitude is just engineering,

Surprises can happen, but there are no obvious areas where straightforward
improvements of current technology are definitely *not* adequate.

>>(Robert Forward's estimate was that an accelerator the size of Fermilab,
>>but optimized for antimatter production, could make enough in a year to
>>test-fire an antimatter rocket engine.
>Nonsense is nonsence even if it is coming from Robert Forward...
>A good source, nowadays, will give you something like 10^10
>antiprotons per second.  A real good dedicated source could perhaps up
>it by another 2 orders of magnitude.  So, lets say 10^12.  Will take
>20000 years to make a gram of antiprotons.

Why do you assume that you need a gram of antiprotons to test-fire an
antimatter rocket engine?  For solar-system use, you don't build engines
with charged-particle exhaust; you use the antimatter to heat much larger
quantities of normal matter.  Microgram quantities are adequate for test
firing, as I recall, although you'd need milligrams for operational use of
such engines.
The good old days                   |  Henry Spencer
weren't.                            |      (aka

Index Home About Blog