Having thought about this at great length for many years (being a Trekker, naturally), it's always seemed to me that a direct matter-to-energy transformation would require very large amounts of energy to be input to prevent any loss through the deconstruction/reconstruction process; hence, it would be very inefficient, not to mention inherently difficult to prevent loss or contamination of the converted energy.
Imho, there's a much more efficient and effective way to do it.
If you'll forgive a slight divergence, scientists have postulated a "Grand Unification Theory" of 4 elemental forces that pervade the Universe - gravity, electromagnetic radiation (includes radio, light, microwaves, X-rays, etc.), and the strong and weak nuclear forces.
It's these last two that we'll concern ourselves with.
The strong nuclear force holds atomic nuclei together.
The weak nuclear force holds electrons in their orbits around the atom.
If we counteract these forces, we can easily deconstruct atoms into their component subatomic particles (proton / neutron / electron); this ultimately accomplishes the same end as the aforementioned matter-to-energy transformation, without the inefficiency that I noted earlier.
Once we deconstruct matter at the sub-atomic level, is it necessary to retain the exact same protons, neutrons, and electrons in order to reconstruct it?
I would say "no".
This makes the process of teleportation much simpler, and in fact, will have some interesting ramifications, as we'll see later.
Before we do any deconstructing, though, we have to know what it is that we're working with; thus, the first step in the process is to determine which atom we'll be dealing with.
As of this date, we have slightly less than 120 elements that have been identified. Therefore, we can use a number between 1 and 128 (2^7) to represent each atomic number and still have several left over. After determining the type of atom, the position of each one will be encoded, using an x,y,z coordinate system. Next, we'll need to know if the atom is an isotope, and which other atoms (if any) the current atom is bonded to. The final necessity will be to include some sort of error correction/checksum/etc. to insure that the data remains valid.
In SMETP, the characteristics of each atom are encoded in a TCP/IP-style packet, in what would be the ICMP message field. Doing it this way, a number of inherently desirable qualities can be designed in; for example:
Trust me on this one.
The next step is to deconstruct the atoms.
Since the objective is to separate the component particles, instead of directly transforming them into energy, we need a way to do so without affecting the sub-atomic particles themselves. This is accomplished by 'ungluing' the atoms. Since atoms are held together by the strong and weak nuclear forces, we need to counteract those forces in order to deconstruct the atoms; we can call this a negative or 'anti-force'.
The first order of business is the electrons; to separate them from the nuclei, a 'null field' must be generated, using a localized 'microburst' of weak nuclear anti-force to counteract the existing force. The net result is to disrupt the orbits of the electrons; once done, a 'collector' with a positive charge will gather the electrons and store them until needed.
A similar method can be used to gather the protons; in this case, the microburst will be one of strong nuclear anti-force, in order to dissociate the nucleus itself. Again, a 'collector' will be used to gather the protons, and a third collector will subsequently gather the left-over neutrons (I'm still thinking about how to collect them).
Once we have the atoms processed and deconstructed into their component particles, all we need is a place to put them; remember, since we have the pattern stored, we don't have to retain the exact same particles. Also, once we want to reconstruct the matter, we're going to need a supply of sub-atomic particles to use in the process. Thus, once matter is deconstructed, it makes sense to use what's already in a dissociated state as the basis for reconstruction, regardless of whether the matrix used is the same or different from the original matter that those particular particles came from.
For reconstruction, the process would effectively be reversed; the nucleus would be reassembled first, after which the electrons would be replaced. Once each individual atom was reconstituted, it would be placed into the molecular matrix in its proper location.
Let's look at a regular molecular matrix as an example (yes, it's crude, but it works for what we're doing):
1 - - 2 - - 3 / / / 4 - - 5 - - 6 / / / 7 - - 8 - - 9 10- - 11- - 12 / / / 13- - * - - 15 / / / 16- - 17- - 18 19- - 20- - 21 / / / 22- - 23- - 24 / / / 25- - 26- - 27
During the process of deconstruction, the individual atoms will be processed in the numerical order shown here; reconstruction will reverse the order.
The atom we're interested in at the moment is #14, which is in the center of the matrix. The other 26 atoms are those that would be considered in the 'local area'. Not every atom in the matrix would have an atomic bond with #14, though; we'll just say that for our purposes, bonds exist between it and #5, #11, #13, #15, #17, and #23.
In the SMETP protocol, atom #1 will be considered as the (0,0,0) point; the Y axis uses an absolute value (#10 would be (0,1,0), and #19 would be (0,2,0), for example.), since deconstruction would, by necessity, have to be done from the top down. The main atom would be identified by its own absolute coordinates, with the bonding atoms encoded in the information field by their absolute x,y,z coordinates (#5 = (1,0,1); #11 = (1,1,0); #13 = (0,1,1); #15 = (2,1,1); #17 = (1,1,2); #23 = (1,2,1)). Relative coordinates could also be used, but would probably be more useful in a simple matrix than in a complex one.
So, for each individual atom, one packet would be encoded as follows ( '|' = field separator):
Make sense so far?