October 31, 2011

Femtotech: computing at the femtometer scale using quarks and gluons from Hugo de Garis

Kurzweilai - How the properties of quarks and gluons can be used (in principle) to perform computation at the femtometer (10^-15 meter) scale.

An atom is about 10^-10 m in size.

The next smallest thing in nature is the nucleus, which is about 100,000 times smaller, i.e., 10^-15 m in size — a femtometer, or “fermi.” A nucleus is composed of protons and neutrons (i.e., “nucleons”), which we now know are composed of 3 quarks, which are bound (“glued”) together by massless (photon-like) particles called “gluons.”

Hence if one wanted to start thinking about a possible femtotech, one would probably need to start looking at how quarks and gluons behave, and see if these behaviors might be manipulated in such a way as to create a technology, i.e., computation and engineering (building stuff).

In this essay, I concentrate on the computation side, since my background is in computer science. Before I started ARCing (After Retirement Careering), I was a computer science professor who gave himself zero chance of getting a grant from conservative NSF or military funders in the U.S. to speculate on the possibilities of a femtotech. But now that I’m no longer a “wager,” I’m free to do what I like, and can join the billion strong “army” of ARCers, to pursue my own passions.

So I started studying QCD (quantum chromodynamics), the mathematical physics theory of the strong force, or as it is known in more modern terms, the “color force.”

Bits and logic gates : the heart of computation

If you want to compute at the femto level, how do you do that? What would you need? To me, the essential ingredients of (digital) computing are bits and logic gates.

A bit is a two-state system (e.g., voltage or no voltage, a closed or open switch, etc.) that can be switched from one state to another. It is usual to represent one of these states as “1” and the other as “0,” i.e., as binary digits. A logic gate is a device that can take bits as input and use their states (their 0 or 1 values) to calculate its output.

The three most famous gates, are the NOT gate, the OR gate, and the AND gate. The NOT gate switches a 1 to a 0, and a 0 to a 1. An OR gate outputs a 1 if one or more of its two inputs is a 1, else outputs a 0. An AND gate outputs a 1 only if the first AND second inputs are both 1, else outputs a 0.

There is a famous theorem in theoretical computer science, that says that the set of 3 logic gates {NOT, OR, AND} are “computationally universal,” i.e., using them, you can build any Boolean logic gate to detect any Boolean expression (e.g. (~X & Y) OR (W & Z)).

So if I can find a one to one mapping between these 3 logic gates and phenomena in QCD, I can compute anything in QCD. I would have femtometer-scale computation. That was the big prize I was after.

The color charge on the quarks and the gluons

There are 4 types of force in the physical world, from weakest to strongest: the gravitational force, the weak nuclear force, the electromagnetic force, and the strong nuclear force. (Actually, their relative strengths depend on the temperature at which these forces act. At the extreme temperatures (energies) that occurred just after the big bang and now at the LHC (Large Hadron Collider) in Geneva, their strengths converge to the same value, a phenomenon called “grand unification.”

The equivalent of the photon is called a gluon. A quark emits a gluon, which is then absorbed by another quark, and this creates the interaction between the two quarks.

There is an essential difference between a photon and a gluon. The photon has no charge of its own, whereas a gluon does have a color charge, in fact, each gluon has 2 such charges. It is bi-charged, or bi-colored. This means that gluons can interact with other gluons, forming complex “glueballs.” I will not be using glueballs in this essay, but they might play an important role in femtotech in the future?!

Strictly speaking, there are more than 3 color charges. In fact there are 6, namely red, blue, green, anti-red, anti-blue, and anti-green. A gluon (at least the type of gluon that I will use in this essay) has one of the first three, and one of the second three. So there are 6 such bi-colored gluons, a red, anti-blue; a red, anti-green; a blue, anti-red; a blue, anti-green; a green, anti-red; a green, anti-blue. In this essay I will use only the red, anti-blue and the blue, anti-red gluons, because (using Occam’s razor), they are all that I need.

Colors are conserved in quark-gluon reactions

How does a gluon interact with a quark? What happens? Remarkably, when a gluon and a quark interact, the gluon may change the quark’s color, and in such a way that the colors are conserved. For example, imagine a red, anti-blue gluon (which from now on will be abbreviated to Gr,~b) interacts with a blue-colored quark (abbreviated from now on to Qb). The gluon will cause the quark to change its color from blue to red, i.e., in symbolic terms:

Gr,~b : Qb -> Qr

In other words, the red, anti-blue gluon acts on the blue (color charged) quark, and converts it into a red (color charged) quark.

Note that before the interaction, there were 3 charges: a red, an anti-blue (both on the gluon), and a blue (on the quark). During the interaction, the anti-blue of the gluon and the blue of the quark cancel, leaving only a red, which is now the color (charge) of the outgoing quark. The colors are conserved.

What would happen if a red, anti-blue gluon (Gr,~b) interacted with a red quark Qr? Nothing. Such an interaction is forbidden in nature, because the color charges in this case are not conserved. Before the interaction, we have a red and an anti-blue charge on the gluon, and a red on the quark. If the quark absorbed the gluon and changed its color from red to blue, then the final charge would be just blue. But that doesn’t match the “2 reds and 1 anti-blue charges” before the interaction. The colors are not conserved, so this interaction is QCD forbidden.

This color conservation operates with the emission of a gluon as well. For example, a red quark Qr could emit a red, anti-blue gluon (Gr,~b) and become a blue quark (Qb). This emission can be represented as

Qr -> Qb + Gr,~b

Note that the colors are conserved. The blue and anti-blue cancel each other, leaving a red on both sides. Color conservation is one of the basic natural laws of QCD.

Now, a gluon that is emitted by one quark can be absorbed by another quark, rather like the way a photon can be emitted and absorbed by two electrically charged particles (which is the basis of the study of quantum electrodynamics, QED). By emitting and absorbing gluons, two quarks can interact with each other and influence each other. I will make heavy use of this phenomenon, as will soon become clear.

The “aha moment”

Probably some of you have already had an “aha moment” on how you might implement femtotech-based computing, based only on what I have said above.

Once I had read about the color charges and gluon emission and absorption, I had my “aha moment.” I felt I had found a way to compute at the femtometer scale, using quarks and gluons, at least in principle. For difficulties facing the practical engineering of these ideas, see towards the end of this essay.

The aha moment gave me the following basic ideas.

a) Represent a bit by the color of a quark. A red for 1, and a blue for 0. (I didn’t need to use green.)
b) To change the state (1 to 0, or 0 to 1), change the color of the quark from red to blue, or vice versa.
c) To change the color of a quark, use an appropriately emitted gluon, i.e., one possessing the appropriate bi-coloring.
d) To implement logic gates (and this was the creative challenging part), use a sequence of gluon emission and absorption (of the same gluon).

[the details of the quark and gluon mapping is described at the source article]

Engineering Challenges

Now that all 3 gates have been mapped to quark-gluon interactions in QCD, one has an “in principle” recipe for femtometer scale computation.

However the practical engineering problems remain, especially when considering something called “asymptotic freedom,” which says that quarks interact weakly when close together, but immensely strongly as they separate, rather like a tough rubber band being stretched. The more it is stretched, the greater the potential energy it has. Similarly with the 3 quarks inside a nucleon.

A nucleon is stable (in the nucleus) because it has 3 quarks, one is red, another blue, and the third green. These 3 colors “sum” to “white” (rather like a spinning color wheel of equally sized red, blue and green sectors), which is analogous to the way an atom, with its positively charged nucleus and its negatively charged electrons, sums to neutrality.

The implicit assumption in the models is namely that using isolated quarks and gluons seems unphysical and unrealistic.

But, if the gluons and quarks are close together, the “stretching rubber band” phenomenon does not occur. There may be particles that contain more than 3 quarks, the so called “exotics,” which may have 3N quarks (a multiple of 3 to maintain color neutrality (“whiteness”) by summing an equal number of red, blue, and green color charges).

There may also be “glueballs” that consist only of gluons that interact in highly non linear and hence complex ways.

Another possibility is to heat up the quark/gluon complex so much that one obtains a quark-gluon “plasma.” At a critical temperature, after cooling the plasma, quark-gluon “chains” may start forming, that may interact in ways similar to the way molecules interact within the cell, i.e., by complementary “lock and key” touching.


The above femtometer scale computation models are “in principle” only. To make them practical will probably require new thinking, to ensure that they are compatible with the severe constraints imposed by the principles of QCD, e.g., quark confinement and asymptotic freedom.

Hopefully, this essay will stimulate other researchers to enter this new research field of femtotech. Perhaps the “other side” of technology (the “building stuff” side, in contrast with the computational side) can be implemented with glueballs as well, or with quark/gluon “condensates.”

One thing is clear. If humanity does not make any progress along the lines of femtotech, sooner or later, human beings (or our artificially intelligent successors) will be scratching at the “nanotech walls” that confine us.

One final comment I’m thinking of trying to create an “attotech” (i.e., on the scale of 10^-18 meters) by using the weak-force particles (W and Z particles) that interact not only with quarks, but with the much lighter leptons (e.g., electrons, etc) as well.

Human technology has progressed from millitech, to microtech, to (recently) nanotech, and this essay attempts to start the thinking on femtotech (and attotech).

This downscaling trend provides a potential answer to the famous “Fermi paradox” (if intelligent life is so commonplace in the universe, “where are they?”). If intelligent creatures or machines can continue to “scale down” in their technologies, the answer to Fermi’s question would become “They are all around us, whole civilizations living inside elementary particles, too small for us to detect.”

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks
blog comments powered by Disqus