Experimental Demonstration of a Balanced Travelling Wave Parametric Amplifier

In the coldest reaches of the universe, where temperatures drop to near absolute zero, scientists listen for the faintest whispers of light and matter. To hear these whispers, they need amplifiers that can boost incredibly weak signals without adding their own noise or distorting the message. Superconducting travelling-wave parametric amplifiers are the tools of choice for this job. They act like a perfectly tuned microphone for the quantum world, capable of catching a wide range of frequencies at once. However, these amplifiers have a tricky side effect. When they boost a signal, they inevitably create a mirror image of that signal at a different frequency. If the original signal is unknown or covers a broad spectrum, this mirror image creates confusion, making it impossible to tell which part of the data is the real signal and which is the ghostly reflection. This problem has long limited how much useful information these powerful devices could deliver.

A team of researchers from the University of Oxford and the Jet Propulsion Laboratory has now demonstrated a way to solve this confusion. They built a new type of amplifier architecture that acts like a traffic director, sorting the real signal from its unwanted mirror image and the powerful pump energy used to drive the system. By arranging two identical amplifiers between special microwave couplers, they created a setup where the desired signal travels one path while the noise and reflections travel another. The result is a device that provides strong, clear amplification across a wide range of frequencies, effectively doubling the amount of usable bandwidth by eliminating the spectral confusion that has plagued previous designs.

The core of this breakthrough lies in how the researchers managed the phases, or timing, of the waves moving through the system. In a standard amplifier, a strong pump wave mixes with a weak input signal to create a third wave called an idler. This idler is the mirror image that causes the problem. In the new balanced design, the researchers split the input signal and the pump wave into two separate paths. They arranged the timing so that the signal waves in both paths were perfectly synchronized, while the pump and idler waves were perfectly out of sync. When these waves reached the end of the journey and were recombined, the synchronized signals added up to become stronger, while the out-of-sync pump and idler waves canceled each other out. This cancellation happened so effectively that the unwanted components were routed to a separate exit port, leaving the main output clean and free of interference.

To test this concept, the team constructed the device using two superconducting amplifiers made from a material called niobium titanium nitride. These chips were placed on a single wafer and connected to a pair of microwave couplers that acted as the splitting and combining junctions. The researchers fed a weak signal into one port and a strong pump wave into another, then measured what came out of the two exit ports. They found that the device successfully amplified the signal by about 20 decibels across a frequency range that nearly doubled in width, spanning from 1.8 to 3.2 gigahertz. More importantly, they observed that the unwanted pump wave was suppressed by more than 20 decibels, and the mirror image, or idler, was rejected by an average of 15 decibels across the operating band.

The experiment also revealed how the physical construction of the device influenced its performance. Because the two amplifiers were not perfectly identical—due to tiny variations in the manufacturing process on the silicon wafer—the cancellation of the unwanted waves was not perfect. The researchers measured a slight difference in the electrical length of the two paths, which caused a small amount of leakage. However, this imperfection had an unexpected benefit: it helped smooth out ripples in the gain, which are small fluctuations in amplification strength that often occur in these devices. The team noted that if they were to build future versions with the two amplifiers placed right next to each other on the wafer, they could minimize these manufacturing differences and push the image rejection even higher, potentially reaching levels where the mirror image is almost completely invisible.

This work establishes a practical path forward for using these amplifiers in real-world applications where the input signal is unknown or varies widely. In fields like astronomy, where scientists use radio telescopes to map the sky, or in quantum computing, where delicate quantum states must be read out without error, the ability to separate the signal from its mirror image is crucial. Previously, scientists had to use complex filtering or sacrifice bandwidth to avoid confusion. This new balanced approach removes that need, offering a cleaner, wider window into the data. The researchers demonstrated that by simply arranging the components correctly, they could achieve high performance without sacrificing the speed or range of the amplifier, paving the way for more sensitive instruments in the future.

The success of this experiment relies on the precise control of wave phases, a concept that can be difficult to visualize without technical training. Imagine two people walking side by side; if they step in perfect time, their combined movement is strong and steady. If one steps forward while the other steps back, their movements cancel out, and they stay in place. The researchers used this principle to make the useful signal step in time while forcing the unwanted noise to step against itself, effectively silencing it. This simple yet powerful mechanism allowed them to separate the amplified signal from the pump and the idler, turning a device that previously produced a confusing mix of frequencies into a clean, high-fidelity receiver.

The findings suggest that balanced architectures are a viable and practical solution for broadband parametric amplification. The team's measurements confirmed that the device could handle the demands of modern, high-speed data collection while maintaining the low noise levels required for sensitive scientific work. While the current prototype showed some limitations due to the slight differences between the two amplifiers, the results proved that the fundamental concept works. The researchers are confident that with improved fabrication techniques, such as placing the amplifiers closer together to ensure they are more identical, the performance can be further refined. This approach offers a promising route to building the next generation of ultra-sensitive detectors for exploring the universe and advancing quantum technologies.