Enhanced Extreme Ultraviolet Emission from Laser-heated Blow-Off Tin Plasmas

Modern computer chips are becoming so small that the features on them are measured in billionths of a meter. To draw these tiny patterns, manufacturers use a special type of light called extreme ultraviolet. This light has a wavelength so short it can only be generated by creating a super-hot, super-dense cloud of gas, known as a plasma, from a drop of liquid tin. The process involves hitting the tin with a powerful laser to turn it into this glowing cloud, which then emits the light needed to print the circuits. However, making this light efficiently is difficult. If the tin is too dense, the light gets trapped inside the cloud and absorbed before it can escape. If the cloud is too sparse, the laser energy passes right through without heating the tin enough to create the light. Finding the perfect balance is the key to making faster, cheaper computer chips.

Researchers at Pacific Northwest National Laboratory and other institutions recently explored a new way to manage this delicate balance. Instead of hitting a solid piece of tin or a single droplet directly with a powerful laser, they tried a two-step approach using a very thin sheet of tin foil, just one micrometer thick. First, they used a weak laser pulse to gently blow a cloud of tin atoms off the front of the foil. This created a low-density, spread-out cloud of gas floating in front of the metal. Then, a split second later, they hit this floating cloud with a much stronger laser pulse. The goal was to see if heating this pre-expanded cloud would produce more of the useful light than heating a solid block of tin.

The team found that this method worked significantly better. By carefully timing the delay between the weak "blow-off" pulse and the strong heating pulse, they were able to increase the amount of useful light emitted by up to 35 percent compared to hitting a solid target. The secret lay in the density of the tin cloud. When the tin is solid or very dense, the laser energy gets absorbed only at the very surface, creating a hot outer layer that blocks the light from the hotter inner parts. This is called self-absorption, and it wastes energy. In their experiment, the weak first pulse created a cloud that was much less dense and spread over a larger area. When the strong laser hit this cloud, the energy could penetrate deeper and heat a much larger volume of tin evenly. This allowed more of the tin atoms to reach the exact temperature and electrical state needed to emit the specific color of light required for chip manufacturing.

To understand exactly what was happening, the researchers used several tools to watch the plasma as it formed. They took high-speed shadow images to see how the cloud expanded, and they used laser interferometry to measure the density of electrons within the cloud. These measurements confirmed that the two-step method created a large, low-density region where the laser energy could be absorbed efficiently. They also measured the speed of the ions, which are the charged atoms flying out of the plasma. They found that in the optimized two-step process, the ions moved slower but in greater numbers, indicating that the energy was being shared more gently and effectively across the cloud rather than blasting a small area. Computer simulations matched these observations, showing that the low-density cloud allowed the laser to heat the entire volume of tin, rather than just a thin skin.

The study also revealed that the timing was critical. If the second laser hit too soon, the cloud was still too dense, and the light was absorbed. If it hit too late, the cloud had expanded so much that the laser energy passed through without heating enough tin. There was a sweet spot, a specific fraction of a second after the first pulse, where the density was perfect. Interestingly, the researchers found that changing the strength of the first weak pulse did not drastically change the quality of the light, as long as the timing of the second pulse was adjusted to match the new expansion speed of the cloud. This suggests the method is robust and could be tuned for different conditions.

While the results were promising, the researchers noted that this specific setup used a type of laser that is different from the massive carbon dioxide lasers currently used in industrial factories. The experiment used a solid-state laser, which is smaller and more efficient, but the conversion efficiency they achieved was still lower than what is currently possible with the industrial machines. However, the work proves that the principle of creating a mass-limited, low-density cloud before heating it is a powerful way to improve light production. By showing that a simple two-pulse sequence on a thin foil can dramatically reduce the self-absorption that usually limits performance, the study offers a clear path for improving how we generate the light that powers the future of electronics. The findings suggest that by controlling the density and shape of the plasma cloud, scientists can make the process of creating these tiny chips more efficient, helping to keep the pace of technological progress moving forward.