When possible disaster caused by strange matter is considered, stakes are obviously very high. Therefore also alternatives that do not seem so likely should be taken into consideration. There might for example exist particles that none of present theories predict. One could assign a positive probability for an alternative that there exists a particle whose parameters are in certain intervals.
The following alternative exhibits potentially different behaviour for cosmic rays and the LHC, what comes to security. The point is that the particle might catalyze transform of ordinary matter to strange matter.
Suppose there exists a scalar particle R with following properties. Its mass is about 3.5 TeV and electric charge +1 and color-charge three times the color-charge of a quark. Suppose R has lifetime 6.8 * 10^(-9) s (Lifetime doesn't have to be exactly this. One can also consider other values of electric charge.) Behind the long lifetime is that R has certain quantum number that is preserved in strong interaction and there does not exist lighter particles with suitable quantum numbers. It decays through weak interaction, but particle R' that appears with it in the currents of the weak interaction is almost as heavy as R, so that m®-m(R')<m(Z).
Consider the possibility that R is generated in LHC through a collision of energetic gluon and its antiparticle, and this collision creates pair R, anti-R. Suppose they fly away from each other, picking three quark-antiquark pairs from vacuum. Antiquarks form with R a color-singlet state and quarks make another color-singlet with anti-R. A hadron of the form (R,anti-u,anti-d,anti-u) or something like that is thus created. We pick the mentioned alternative for consideration. Here all three antiquarks have the anticolor corresponding to the color of R. If this hadron is slow enough, it can remain fixed with an atom nucleus. If it had speed 0.07 c, it would need 14 lifetimes to travel 2 m. If we reserve one lifetime for what follows, we get suppression factor e^(-15) = ( 3 269 000 )^(-1). For such speed the collision (R,anti-u,anti-d,anti-u) and, say, Cu_63, would have kinetic energy 139,8 MeV while the total binding energy of Cu_63 is 539.8 MeV. Thus the exotic hadron could not brake Cu_63 totally apart. It is hence conceivable that it could remain fixed with some daughter-nucleus of copper.
After (R,anti-u,anti-d,anti-u) has remained fixed with an atomic nucleus, there happen apparently annihilations untill all antiquarks have been annihilated. Then in the middle of the nucleus is a hadron of a form (R,uuuddd), if the particle has reacted with two protons and a neutron, or (R,uuuudd), if the hadron has reacted with three protons or (R,uudddd) if the particle has assimilated two neutrons and a proton. We pick (R,uuuddd) for consideration. Quarks u and d have in this system alltogether eigth possible states, two colors, two flavors, two spin states. Only six of them is in use.
Hadron (R,uuuddd) could cause stronger nuclear force than usual hadron. It might bind nearby hadrons of the nucleus together to quark matter. If so, it would possibly be feasible for this quark matter to transform itself to strange matter. Every step of increased strangeness might make it only more stable, if it were already in a state of quark matter. This piece of strange matter could cause the whole nucleus to turn to strange matter. If nucleus were big enough, this strangelet could remain stable after decay of R. (More precisely, some daughter-strangelet could remain stable.)
On the other hand, if hadron (R,uuuddd) would strive for "octet" it could assimilate to itself one nucleon more resulting to (Rd,uuuudddd). Here one d-quark has the same color as R and the others have the two other colors. It is probably impossible with the present knowledge to know for sure, whether such reaction were feasible or not, due to complications related to QCD. Therefore one should take into account also possibility that such baryon-number increasing reaction could happen. Moreover, if in this structure would be feasible some u-quarks to transform to s-quarks, one could add more quarks to the same state so that eventually one would have a hadron of the form (Ruds,uuuuddddssss), where first R u d s have the same color and the rest have the two other colors. At least for electric charge it would be feasible for some of the u-quarks to change to s-quarks.
If R would come to exist through a collision caused by cosmic rays, it would decay before it had slowed down enough to catalyze the transfrorm of ordinary matter to strange matter. For such catalysis it should probably be so slow that the collision energy were less than the total binding energy, or slower.
It would need over 80 times its lifetime to slow down through series of collisions with earthly hadrons. Notice that e^(-80)= 1.8 * 10^(-35). Therefore the size and estimated age of the Earth would not be enough to get a strange matter disaster, even if it would turn out that the electric charge of strange matter would be negative. Also majority of ordinary stars would be safe from such disaster, if the catalysis would be the only or most probable way to strange matter. Actually, in a typical galaxy one would get few strange stars if any. Here the amount of lifetimes needed depends on the density of the surface. This number 80 is calculated with granite density.
Neutron stars could cause problems also in this alternative. It is somewhat problematic to base safety considerations on neutron stars, because we know so little about them with certainty. Anyhow, consider the possibility that particle R had another way to decay, through strong interaction, such that a small amount of additional energy were needed since decay products were heavier than the original particle. Then, despite the density of the surface of the astronomical object, the particle could get repeatedly enough energy to decay in a series of collisions that would follow its generation through cosmic rays. Thus it could decay almost certainly before it were slow enough to catalyze transfrom of matter to strange matter. For this conclusion the needed additional energy would have to be small enough, for example 200 MeV, if one estimates assuming all collisions are head-on collisions.
Anyhow, in some galaxies collisions of strange stars could occur. This might launch the transformation of such galaxies to strange matter because of the flow of strangelets from collisions, if the charge of strangelets were negative or neutral. This could lead to same phenomenon in nearby galaxies at least. Perhaps this would look like a group of quasars from very far distance. Perhaps some professional physicists or astronomers could figure out whether characteristics of such groups of strange galaxies would match observations of some groups of quasars.
As far as I know, the order of magnitude of amount of cosmic ray collisions whose CM-frame energy is similar to LHC with full power is about one per year per 10 m^2. If the particle reaction creating R were rare enough, this particle could have avoided observation so far. A single heavy particle or a pair of particles could have skipped observation devices, while a shower of light particles caused by energetic collisions is much easier to detect, because it spreads over large area.
I would appreciate, if some professional physicists would study thoroughly the issues in this post. I myself was a mathematician, but at the moment I am without an academic post. I cannot become a professional physicist in a twinkling of an eye. One professor of particle physics admitted that this scenario cannot be trivially disproved. She recommended to contact another professor whose expertise area was closer to this. He didn't answer. Neither did a couple of other physicists. This message did not receive any comments so far in physforum.com.