Joel M. Williams

In the particle in the box problem, the particle is not in both boxes at the same time as some would have you believe. It is a set definition situation with the two boxes being part of a set that also contains a particle. Set and subset differences are explored. The atomic electron orbitals can be mimicked by roulette wheel probability with the 0 and 00 serving as boundaries. Considering a marble on the floor as part of the roulette wheel is taking things a bit too far. Likewise, attraction between a positive and negative charge at distance does not make the negative charge part of the positive charge’s orbital system.

Atomic electron orbitals are statistically defined as volumes that represent the bulk of the area under a probability curve. This, of course, has led to the teaching that one of the statistical tails allows orbitals to stretch to infinity. Schroedinger would, before his death, join Einstein in disbelief of the implications that resulted when his creation was carried a bit too far.

What one thinks a statistical treatment means and what it actually means are not always the same. The "particle-in-the-box" enigma is a case in point. Quantum theory proposes that the universe is connected to such an extent that everything is tied to everything else and that, by knowing something, one can know all. Knowing that something in the "particle-in-the-box" was determined by "looking" into one half of the divided box. Thus, the 50% chance of knowing where the particle was instantaneously changed to a 100% chance in both halves. The revelation seemed to be at odds with the original statistical application. In essence, however, it was the application of a single statistical treatment to several different phenomena. In other words, the "particle-in-the-box" problem was the application of the same statistics to both a set and its subsets.

The "box-with-the-particle" was actually a subset of a set that included at least one other box: a "box-without-the-particle". The particle was determined to be in the original, "undivided", box by "looking" (when it was put in the box; if, at no other time). The probability of finding the particle in this "box" was P=1. This guaranteed that one of this box’s subsets would actually contain the particle at any given time: P_A=0 or 1 and P_B=1 or 0. Statistics is time-independent, but condition-dependent; i.e., in another instant the probabilities will be the same, but those constituting the "count" ("looks") may or may not be. Statistical designs work well when there are no discontinuities. The restrictions of one subset are not those of another, but they both must fit the requirements of the set.

Each set/subset level may have different statistical responses. For example, the probability of correctly guessing which half of the box contains the particle is always the same, if the choice of which half is always made randomly! Whether the particle shows up randomly or is always in the same half does not matter because that is not the question that this particular statistical treatment is addressing! If one wishes to know the probability of a particle actually being in one half or the other of the box, then one must always selects the same half. Whether the particle is found or not does not change the probability that it will be found. The statistical treatment is not whether the particle is there or not, but rather what fraction of the "looks" it will be there. Statistical probability is not the result of a single "look". The probability is determined by "looking" many times to determine the "belonging" of an event, object, etc. to a given subset. Of course, "looking" has evolved into standardized, non-visual, techniques that usually imply an infinite number of "looks". Then, there are the statistics of the statistics.

The probability of correctly guessing the whereabouts of the "particle-in-the-box" points out an interesting feature about statistical evaluations; namely, that the actual location of the particle is irrelevant. That is, this probability is determined solely by the number of subsets in the next level down. To illustrate, consider Fig. 1. Case 1 shows the typical way statistical treatments are made; i.e., each of the subsets (A & B) contains the same number of subsets in the next level down. Case 2 shows a nonequal allotment at the next level. The exact location of the particle is determined by "looking" into all the subsets of either subset A or subset B. With random selection of pile A or B in each case, the chance of finding the particle in the select subset is exactly the same: P_A=P_B=1/2. Even if Case 2 had an infinite number of subsets in subset B, the results would be the same. In this case, the particle would probably always show up in subset B, but subset B would only be chosen to be "looked" into half the time. The rigging of subset A to have the particle would not change the probability, but it could provide some very valuable additional information; e.g., it would cut down on the number of boxes to be "looked" into to find (or not find) the particle -- something for which scientists usually strive.

Set theory is nonrigorously explored in the association of an electron with a nucleus below.

Normally an electron and nucleus are thought of as an interacting pair. This seems quite reasonable when addressing the electronic structure of an atom. But, it has led to the idea that a given electron is associated with a given nucleus with a finite (even if small) probability of association even at an infinite distance from the nucleus (see normal view in Fig. 2). Thus, everything in the universe is connected to everything else and, therefore everything "belongs" to everything else. In the sense that everything in the universe "belongs" to the universe, this is undoubtedly true. An electron may be attracted to a positive nucleus by an infinitesimally small force at infinite distance which would in this sense interconnect all charged species, but that does not mean it is part of a nucleus’ orbital system!! In a vast universe, the probability is high that a given electron is not close to a given nucleus, but rather far from it (see the universe view in Fig. 2). Of course, all the electrons and nuclei could pair in infinite time, but it is unlikely that any two given, randomly chosen, ones would. The obvious difference in the two views of Fig. 2 are the result of considering two different subsets of a larger set: associated pairs (normal view) and unassociated pairs (universe view). Somehow it seems unlikely that a nucleus could handle all the electrons of the universe. Few nuclei form ions with even a small number of extra electrons -- nor do they give up many without a terrific fight. One view would have all of the positive charges of the universe at the origin. Then, the normal view might be understandable, if not totally correct.

A subset view of an atomic electron is given in Fig. 3. The electron can belong to the nucleus or not; and it can belong in different ways. It can also transition between subsets. In the depiction here, it "wobbles" between subsets; i.e., it can go on to another state or it can return to its previous state depending on the conditions. [Message on the interior of subset boundaries: "So long, it’s been good to ‘no’ you." What is the best chance of seeing that spouse again as it moves off into space? When hell freezes over -- the big crunch. Sign on the exterior of proton’s boundaries: "Wanted: spouse for lasting relationship. Inquire with sign within."] But it will not continue to wobble forever. Wait an instant and the "wobblies" will be in one set or another; although there may be new "wobblies" in the meantime. Remember: the statistics here are time-independent as long as the conditions remain the same. The conditions surrounding the number of electrons that may be wobbling are different than those for the subsets -- like marbles on the peaks and in the valleys. Sets will invariably have different energy levels; hence, different loadings of "marbles".

An assignment of the subsets to electron behavior is made in Fig. 4. The electron commingling and orbiting probability shapes reflect spherical distribution of an orbiting particle. (This spherical shape is not endorsed here, but rather is used for convenience.) The "unattached" probability reflects the included volume if a particle has an equal probability of being found in any unit volume. The probability of a particle in a subset is given by the area under the curve that describes the particle. If there is no particle present, then P_r=0 at all r’s in the subset! Ergo, the particle-not-in-the-box analog. If there is a particle present, then the integral under the curve =1. Some heavy nuclei (xp+yn) display the devouring activity via electron capture. The regurgitation of an electron is usually too vigorous for the electron to go into orbit. Each subset has smooth return at the boundaries as opposed to the discontinuous behavior of current theories. Extensions of the probability curve outside the subsets have no meaning for the subset. This, of course, is what boundary condition statements mean.

Different probabilities describe the unusual events that occur at the boundaries. If one persists in the notion that there is a finite probability of an electron’s atomic orbit existing at some point in distant space, then which probability does one use -- that based on an orbiting electron or that based on a commingling electron (i.e., a neutron)? Does a neutron have an infinite influence on electrons as a proton is suppose to have? Remember that a neutron is only a proton with an electron and "relish". What influence does a neutrally charged particle [from a (+) particle and a (-) particle] exert in even near space? Nothing much. Then, what about infinite space? If the charge interaction is so important, why isn’t a neutron stable besides when in combinations with protons? Also, if the orbiting probability continued to r=0, protons would rapidly disappear; as it is, attempts to find even one proton disappearing under normal conditions have met great difficulty!

A marble outside the roulette wheel has no bearing on the odds of a marble in the wheel "hitting" a given slot in the wheel. Also, a marble on the floor may not have anything to do with a given roulette wheel, although said marble could be used with said wheel. Also, once a wheel has a marble, it has no need for another. This is the quality of "belonging". The odds of a marble being in the wheel or not correspond to a different set of conditions than those describing the hits a marble will make while in the wheel. Again, boundary conditions are not the cutoff points of continuous functions that go beyond the boundaries, but are the return limits of continuous functions that stay inside the boundaries.