Wikipedia:
"In antimatter-matter collisions resulting in photon emission, the entire rest mass of the particles is converted to kinetic energy. The energy per unit mass is about 10 orders of magnitude greater than chemical energy, and about 2 orders of magnitude greater than nuclear energy that can be liberated today using nuclear fission or fusion. The reaction of 1 kg of antimatter with 1 kg of matter would produce 1.8×1017 J (180 petajoules) of energy (by the equation E=mc²). This is about 134 times as much energy as is obtained by nuclear fusion of the same mass of hydrogen (fusion of 1H to 4He produces about 7 MeV per nucleon, or 1.3×1015 J for 2 kg of hydrogen). This amount of energy would be released by burning 5.6 billion liters (1.5 billion US gallons) of gasoline (the combustion of one liter of gasoline in oxygen produces 3.2×107 J), or by detonating 43 million tonnes of TNT (at 4.2×106 J/kg).
Not all of that energy can be utilized by any realistic technology, because as much as 50% of energy produced in reactions between nucleons and antinucleons is carried away by neutrinos, so, for all intents and purposes, it can be considered lost.[2]
The scarcity of antimatter means that it is not readily available to be used as fuel, although it could be used in antimatter catalyzed nuclear pulse propulsion. Generating a single antiproton is immensely difficult and requires particle accelerators and vast amounts of energy—millions of times more than is released after it is annihilated with ordinary matter, due to inefficiencies in the process. Known methods of producing antimatter from energy also produce an equal amount of normal matter, so the theoretical limit is that half of the input energy is converted to antimatter. Counterbalancing this, when antimatter annihilates with ordinary matter, energy equal to twice the mass of the antimatter is liberated—so energy storage in the form of antimatter could (in theory) be 100% efficient. Antimatter production is currently very limited, but has been growing at a nearly geometric rate since the discovery of the first antiproton in 1955.[3] The current antimatter production rate is between 1 and 10 nanograms per year, and this is expected to increase to between 3 and 30 nanograms per year by 2015 or 2020 with new superconducting linear accelerator facilities at CERN and Fermilab. Some researchers claim that with current technology, it is possible to obtain antimatter for US$25 million per gram by optimizing the collision and collection parameters (given current electricity generation costs). Antimatter production costs, in mass production, are almost linearly tied in with electricity costs, so economical pure-antimatter thrust applications are unlikely to come online without the advent of such technologies as deuterium-tritium fusion power (assuming that such a power source actually would prove to be cheap). Many experts, however, dispute these claims as being far too optimistic by many orders of magnitude. They point out that in 2004; the annual production of antiprotons at CERN was several picograms at a cost of $20 million. This means to produce 1 gram of antimatter, CERN would need to spend 100 quadrillion dollars and run the antimatter factory for 100 billion years. Storage is another problem, as antiprotons are negatively charged and repel against each other, so that they cannot be concentrated in a small volume. Plasma oscillations in the charged cloud of antiprotons can cause instabilities that drive antiprotons out of the storage trap. For these reasons, to date only a few million antiprotons have been stored simultaneously in a magnetic trap, which corresponds to much less than a femtogram. Antihydrogen atoms or molecules are neutral so in principle they do not suffer the plasma problems of antiprotons described above. But cold antihydrogen is far more difficult to produce than antiprotons, and so far not a single antihydrogen atom has been trapped in a magnetic field.
Several NASA Institute for Advanced Concepts-funded studies are exploring whether the antimatter that occurs naturally in the Van Allen belts of Earth, and ultimately, the belts of gas giants like Jupiter, might be able to be collected with magnetic scoops, at hopefully a lower cost per gram.[4]
Since the energy density is vastly higher than these other forms, the thrust to weight equation used in antimatter rocketry and spacecraft would be very different. In fact, the energy in a few grams of antimatter is enough to transport an unmanned spacecraft to Mars in about a month—the Mars Global Surveyor took eleven months to reach Mars. It is hoped that antimatter could be used as fuel for interplanetary travel or possibly interstellar travel, but it is also feared that if humanity ever gets the capabilities to do so, there could be the construction of antimatter weapons."
