/*---------------------------------------------------------------------*/ Error correcting code keeps quantum computers on track Quantum computing is a computer scientists's dream. by exploiting the ability of a quantum system such as an array of atoms to be in many different energy states at once, a quantum computer can, in theory, perform vast numnbers of computations at the same time, tackling problems that would overwhelm conventional machiens. But so far this fantasy has remained just that: a fantasy. No one has built a quantum computer, much less programmed one to calculate anything. Ane one bit of reality keeping computer scientists from realizing this fantasy has been the notorious fragility of quantum states, which makes quantum systems vulnerable to errors. Some recent mathematical disvoeries, however, have given computer scientists cause for optimism. to ensure that information remains intact, classical computers rely on error-corrrecting codes, which include duplicate bits that serve as quality control" for the rest of the data. That strategy could not work for quantum computers it seemed because in quantum mechanics it's impossible simply to duplicate a quantum state; read or copy the state, and you will have altered it. But in a paper published last year in Physical Review A mathematician Peter Shor of AT&T Bell Labs has shown how to nudge a quantum system back into line without looking at it directly. Shor's theoretical feat is now triggering a flurry of error - correction schemes based on his method. Such efforts are extremely important if one wants to do extended quantum computations," Says Seth Lloyd of the MIT a quantum computing pioneer. That's because quantum computers are unlikely ever to operate as reliably as their conventional counterparts. Quantum inforation has a tendency to degrade more quickly than classical information. As they interact with their environment, quantum systems drift away from the state they're supposed to be in. The result is an information leak, because these states encode information, albeit not in the usual way. Each bit of informaiton - a qubit - is not simply a binary 0 or a 1; instead it's a combination of the two, with coefficients that are associated with the probability that an observer will find the qubit in a particular state. For example, classically an atom can be in a ground state or in a higher energy state, but quantum mechanically it can be in a superposition of the two. More generally, a system of n qubits - residing , say in the ground or higher energy levels of n atoms in a row- consists of a combination of 2^n classical states, each accompanied by a coefficient representing the probability that the quantum system will collapse into that state when the qubits are measured. Quantum computation actually takes place on these coefficients as, for example, a burst of laser light alters the mixture of states in a row of atoms. But there's no way to peep into the computation process, because observing a qubit forces it to "choose" between its two states, collapsing many threads of the computation into one. The only time the operator of the computer actually examines a qubit is at the end of a computation, to learn the outcome. As a result, a qubit cannot be duplicated, except by repeating the computation that produced it. And that poses a problem for conventional error correction, which is based on redundancy. For example , the simplest way to ensure accurate storage or transmission of a conventional 0-1 bit of information is to create three copies of the bit and then take a "majority vote" among the copies whenever the bit needs to be read. Such a strategy reduces the probability of an error from, say 1 out of a million to a tiny bit less than 3 out of a trillion. More sophisticated codes can do even better, safeguarding long strings of "information" bits with just a few extra, error-correction bits. But the methods all make the sensible assumption that bits can be read and copied with impunity that information doesn't disappear just because you look at it, as it does in a quantum computer. The key to getting around this barrier came last year when Shor showed how to safeguard a single piece of quantum information by encoding it as a combination of states in a 9-qubit system - spreading the information content of one qubit across nine of them. Shor's code is constructed so that the original quantum information remains intact even if an error occurs in one of the 9 qubits. While basedo nthe conventional "majority vote" approach to error correction, the quantum code corrects errors without explicitly counting ballots. In effect the quantum computer just determines which , if any, vote differs from the others, and registers this information in some ancillary qubits. Counterintuitive as it seems, measuring the ancillary qubits- and thereby losing some of the information stored in them - restores the original 9 qubits to their correct state. The existence of quantum error-correcting codes "came as quite a surprise," says Lloyd. "Before Shor came up with this idea, nobody thought it was possible. " Now that the barrier has been broken, however, quantum error-correcting schemes are proliferating. Reserachers at IBM and Los Alamos National Lab have streamlined Shor's method to one that embeds single qubits in 5 qubit states that are impervious to single errors. Meanwhile, Shor and Rob Calderbank, also at AT&T Reserach and independently, Andrew Steane at Oxford University, have shown how to create quantum analogs of other, more powerful, codes that can correct multiple errors in long strings of bits. "All these other schemes," says Lloyd, "are refinements of Shor's Original Scheme." Even with the surge of results, the theory of quantum error correction is pretty much still getting started, says shor. One problem, Lloyd notes, is that error correction,being a computation of it s own, runs its own risk of making mistakes, "In order for quantum computation to work, you've got to have error correction that is insensitive to errors committed during the process of correction, "There are some ideas floating around on how to do this, but nothing written up. Then there's the ultimate problem: figuring out how to put these schemes to work in an actual quantum computation "The problem with all these techniques is they're good for storing quantum bits, but they're not yet good for computing, " notes shor. No one has figured out a way to perform quantum computation directly on information that is distributed over multiple qubits, as these techniques require. Nevertheless, this also looks like a solvable problem, Lloyd says . Quantum computers may be getting closer to reality and this time it may byte not bite. /*---------------------------------------------------------------------*/ A History of Global Metal Pollution Prod : Million metric tons Cu Cu Pb Pb Zn Zn Prod Emit Prod Emit Prod Emit (1000metric tons) 1850-1900 4 0 5 0 5 0 1901-1910 8 50 12 500 7 300 1911-1920 10 150 10 500 9 600 1921-1930 13 100 13 1000 11 700 1931-1940 15 150 13 1500 13 850 1941-1950 20 150 12 1700 15 1000 1951-1960 30 200 20 2500 25 1500 1961-1970 50 480 30 3300 37 2000 1971-1980 80 600 35 4200 50 3200 1981-1990 76 500 33 3500 63 2700 /*---------------------------------------------------------------------*/ Chemists Clean Ceramics and Coat Enzymes in Plastic -- Meeting of American Chemical Soceity (ACS) ways to take toxic chemicals out of ceramics production line, push up catalytic activity of enzyme - plastic hybrids in organic solvents, drive up student interest in chemistry <> Nothing looks cleaner than a smooth, white ceramic, but they can be messy in the making. Aluminum oxide ceramics, found in everything from lightweight ceramic engine parts to bathroom sinks and lasers start out as an inofensive powder of aluminum oxide particles. Knitting the precursors together into ceramics with particular forms commonly requires hazardous organic compounds and strong acids. Millions of liters of solvents and acids must then be safely disposed of yearly. Ceramics researchers long for a way to clean up their act. A. R. Barron, inorganic chemist at Rice University in Houston, et al reported by breakding down the aluminum oxide particles, they produced even smaller subunits that easily dissolve in nontoxic, nonhazardous water. The material can be put back together as either bulk ceramics like those used for engine parts, or thin films that form protective coatings for carbon fibers and other materials used in high-strength composites. Georgia Tech chemist William Rees, does add a note of caution. "I think it has a lot of potential" "But I don't see it changing the way toilet bowls are made this year" Companies have made bulk ceramics by dissolving a powder of aluminum oxide in organic solvents and then adding organic binders to give the material some body. Next they pour their slurry into a mold, evaporate much of the solvent, allow the material to set, then fire the ceramic at temperatures around 1000 degrees C. To make more complex ceramics, ceramic-makers simply mix other metal oxides such as yttrium - oxide with aluminum-oxide powder. Turning ceramic mixtures into coatings and films requires a different technique, called sol-gel processing, which relies on causticacids and other compounds that are expensive. Researchers search for environmentally friendly alternatives have been stymied because other techniques for doing away with solvents required prohibitively expensive compounds. Water, Barron's group knew is cheap and safe, but the aluminum oxide particles don't dissolve in it. They're too big and heavy. So the researchers decided to break them down. They started with a cheap aluminum oxide powder known as boehmite. Each boehmite crystal is made up of sheets of aluminum atoms surrounded by oxygens, which are stitched together with an array of weak hydrogen bonds. The researchers added a mild, noncaustic acid, such as carboxylic acid, which essentially unzips the hydrogen bonds between aluminum-oxide layers and breaks some of the sheets into tiny rodlike fragments known as alumoxanes. Each of these fragments is surroundded by carboxylic acid groups which -advantageously - have long, oxygen rich tails that readily bind to water molecules. As a result, the alumoxanes readily dissolve in water. The solution itself isn't concentrated enough to be turned into a bulk ceramic. So the researchers add additional boehmite powder; because there aren't enough carboxylic acides to go around, most of these newly introduced particles remain intact but in solution. The result is a claylike substance that can be molded and fired. Barron et al, showed they can make a variety of ceramics with other organic compounds known as acetylacetenoates (acacs), which can hold metal atoms like yttrium, manganese, or erbium. In the mix, the metal atoms on the acacs trade places with aluminum atoms in the alumoxanes. Researchers don't completely understand the reasons for the trade, but they're quite willing to take advantage of it to incoporate other metals into the ceramic. "It's so simple, my grandmother could do it. The rice team recently formed a collaboration with TDA a ceramics development company in Wheat Ridge, Colorado, to test the cost of industrial-scale alumoxane production. The group is also pursuing scientific questions, such as exactly how the metal exchange reaction works. In the end, they hope to have a clean resolution to a messy problem. <<<>>> Enzymes are the movers and shakers of the biochemical world, and plastics have become bulwarks of the material one. Researchers would like to bring these two stalwarts together, combinging plastics with enzymes that could keep the inside of a plastic fuel hose clean by breaking down deposits of organic compounds, but such unions have been hindered, because most enzymes are at home only in water. The organic solvents needed to make polymers typically can't dissolve enzymes - and even if the enzymes did dissolve, they would lose their normal activity. In New Orleans self cleaning pipes moved closer to reality. At the meeting, Jonathan Dordick and U of I, Iowa City, along with Doug Clark at U of Calif. Berkely showed that by linking enzymes to molecules called surfactants, they could dissolve the enzymes in solvents and incorporate them in a variety of plastics, from polystyrene to polymethylmethacrylate, the tough plastic in Plexiglas. Alan Russell, a chemist at the University of Pittsburgh, say "the idea is very powerful, since you're not limited to the polymer you can make." And it's even more powerful says russell because once linked to the surfactants, their enzymes retain their activity in organicsolvents, broadening the range of settings in which the enzyme - laden polymers could be used. Other such hybrids have been formed before, he points out. But these typically show strong enzymatic acticvity only in water. When Dordick and his colleagues began this work several years ago, they weren't thinking about enzymes and plastics at all. Instead , says Dordick, "we were working on how to dissolve proteins into organic solvents" in order to study their behavior and perhaps understand why solvetns curtail enzymatic activity,. Two years ago, the reserachers got the proteins into solution by linking them to two part surfactants. One part, the head group, has a negative chanrge that binds to positively charged lysine groups on proteins. The other part, a hydrocarbon tail, flees from water and prefers to associate with similar hydrocarbons in organic solvents. When the tail rises into the solvent, it pulls the bound protein along for the ride. Not only did the enzymes dissolve, they also retained their activity. Enzymes contain numerous chemical groups with positive and negative charges, which would cause neighboring proteins to bind together, altering their conformation and hampering their activity, if these charges weren't blocked, or "passivated" Water is well suited to doing so, because it is highly polar, meaning that water molecules have negatively and positively charged sides. But hydrocarbons in organicsolvents can't fulfill this role because they are not polar. Dordick suggests that the surfactant molecules can fill in for the water by binding to the ppositively charged groups on the proteins. " The protein is therefore able to resist being forced into an incorrect conformation, and thus the activity of the enzyme tends to be higher. While this explanation is still speculative,... Last year Dordick realized that the technique could be adapted to embed enzymes in plastics. The idea was to add plastic building blocks which are also hydrocarbons to the solvent. In the presence of a pair of common polymerizing compounds, the building blocks, known as monomers, would then come together and form long polymer chains which then entangle themselves in a complex network. Dordick thought that creating this network in the enzyme-doped solent would trap the enzymes, which are too large to escape though the tiny pores in the network. Using that strategy, researchers embeeded a well-characterized enzyme knwown as chymotrypsin, which breaks down other proteins, in polymethylmethacrylate. They showed that even twhen trapped within a plastic 75% to 80% of the enzymes remain active. Alternately the researchers have found that they can bind proteins directly into the polymer strands. They chemically bind polymerizable compounds, such as acrylate groups, to the enzyme before adding the surfactants. Once the proteins and surfactants form ion pairs the whole web can be pulled together into a polymer netowrk when the acrylate groups link up. Because the polymerization techniques used by the Iowa researchers form plastics without high temperatures, there is no danger of denaturing the proteins. To shape their plastics into films, the researchers simply carry out their polymerization reactions between a pair of glass sheets and then peel the plastic off. Dordick thinks such films equipped with enzymes known as peroxidases, could be used to line fuel pipes, making them able to clean themselves by breaking down organic compounds such as tars. Another possibility,he adds, is to creat catalytic membranes that can carry out reactions that are critical in drug synthesis. But before these hybrids go to work in industry, the researchers must still show that the enzymes remain active and stable in the polymer matrix for long periods of time. If they can accomplish this, enzyme plastic hybrids may trigger a bit of a reaction of their own.