Researchers have developed a precision magnetometer based on a special material that changes optical properties in response to a magnetic field. The device, which is integrated onto a chip, could benefit space missions, navigation and biomedical applications.

High-precision magnetometers are used to measure the strength and direction of magnetic fields for various applications. However, many of today’s magnetometers must operate at extremely low temperatures — close to 0 kelvin — or require relatively large and heavy apparatus, which significantly restricts their practicality.

“Our device operates at room temperature and can be fully integrated onto a chip,” said Paolo Pintus from the University of California Santa Barbara (UCSB) and the University of Cagliari, Italy, co-principal investigator for the study. “The light weight and low power consumption of this magnetometer make it ideal for use on small satellites, where it could enable studies of the magnetic areas around planets or aid in characterizing foreign metallic objects in space.”

In Optica, Optica Publishing Group’s journal for high-impact research, the research team, led by Galan Moody of UCSB, with Caroline A. Ross of MIT also serving as a co-principal investigator, describe their new magnetometer. They show that the device can achieve a sensitivity comparable to that of other high-performance, but less practical, magnetometers. 

“The magnetometer could be useful for magnetic navigation, providing an alternative navigation source in environments where GPS is jammed, spoofed or unavailable such as underwater, in tunnels or during electronic warfare,” said Pintus. “It could also benefit medical imaging methods such as magnetocardiography and magnetoencephalography, which currently depend on highly sensitive magnetometers that require bulky, costly equipment.”

Turning light into magnetic insight
The new magnetometer was developed as part of the U.S. National Science Foundation’s Quantum Sensing Challenges for Transformational Advances in Quantum Systems program. It builds on previous works in which the researchers used magneto-optic materials to develop a magneto-optic modulator and integrated magneto-optic memories for photonic in-memory computing.

For the new device, the researchers used a magneto-optical material called cerium-doped yttrium iron garnet (Ce:YIG), which was provided by Yuya Shoji from the Institute of Science Tokyo. When an external magnetic field is present, light propagating through Ce:YIG experiences a phase shift that can be detected with an optical interferometer. 

Optical interferometers work by splitting light into two paths and then recombining those paths.  By placing the magneto-optic material in one of the paths, the researchers were able to measure whether the light in that path becomes brighter or dimmer, which was then used to determine the strength of the magnetic field.

To make the magnetometer practical, the researchers built it on silicon photonics, a technology that creates tiny optical devices using the same silicon found in microchips. This allowed them to create a device with minimal size, weight and power consumption that can be integrated with other chip-based optical components such as lasers and photodetectors.

“Historically, magneto-optic materials have been used almost exclusively in optical isolators and circulators, a specialized class of devices that enforce unidirectional light propagation,” said Pintus. “By incorporating magneto-optic materials directly onto a photonic integrated circuit, we expand the range of integrated photonic components and introduce functionalities that stem from their unique properties.”

The magnetometer operates with ordinary laser light, but the authors have shown that injecting quantum light can improve its performance. “The idea is similar to what’s already done in large optical interferometers used to detect gravitational waves, like LIGO,” explained Pintus. “By using squeezed light — a special quantum state of light — we can reduce noise and increase the instrument’s sensitivity.”

High sensitivity from a small device
Using a combination of multi-physics simulations and experimental measurements, the researchers showed that the device can detect magnetic fields ranging from a few tens of picotesla to 4 millitesla. For comparison, Earth’s magnetic field is about 100,000 times stronger than the minimum detectable field, yet around 1,000 times weaker than the maximum field the instrument can measure. This sensitivity matches that of high-performance cryogenic magnetometers, without their restrictive temperature, size, weight or power constraints.

“This research effort brought together specialists in modeling and fabrication of integrated optical devices, material science and quantum-level modeling of light–matter interactions,” said Pintus. “The synergy among these disciplines enabled us to demonstrate a high-performance device with capabilities that would not be attainable through any single field alone.” 

Now that the researchers have taken an important step toward demonstrating the feasibility of their approach, they are working to improve performance by exploring alternative magneto-optic materials and integrating quantum elements for even greater sensitivity. They note that transitioning the research into a commercial product would require the challenging task of creating a fully integrated chip-based system that includes other key components, such as an integrated laser and photodetector.

About Optica
Optica is an open-access journal dedicated to the rapid dissemination of high-impact peer-reviewed research across the entire spectrum of optics and photonics. Published monthly by Optica Publishing Group, the Journal provides a forum for pioneering research to be swiftly accessed by the international community, whether that research is theoretical or experimental, fundamental or applied. Optica maintains a distinguished editorial board of more than 60 associate editors from around the world and is overseen by Editor-in-Chief Prem Kumar, Northwestern University, USA.

Research in the lab of UC Santa Barbara materials professor Stephen Wilson is focused on identifying and understanding the fundamental science of such states and identifying, growing, and characterizing materials that can host the kinds of abnormal properties and phenomena that can support quantum functionalities.

In a new paper published in October 22 issue of the journal Nature Materials, Wilson’s lab group reports on an innovative way to use a phenomenon referred to as frustration of long-range order in a material system to engineer unconventional magnetic states that are useful to quantum applications, although, he makes clear, “This is fundamental science aimed at addressing a basic question. It's meant to probe what physics may be possible for future devices.” 

The paper, titled “Interleaved bond frustration in a triangular lattice antiferromagnet," describes how several types of frustration can come into play in this realm. Geometric frustration refers to a state in which the magnetic moments of a material are in a state of fluctuating disorder, or as mentioned above, frustrated.
 
To understand that phenomenon, Wilson says, “You can think of magnetism as being derived from tiny bar magnets sitting at the atomic sites in a crystal lattice. Those bar magnets are what we call magnetic dipole moments, and they can interact and orient themselves relative to one another in specific ways, depending on the details of a material, to minimize their energy or, said another way, to realize their ground state.” That is the lowest energy state for any system, and any system at absolute zero temperature exists in its ground state.

“If those magnetic moments interact in a way that wants them to point antiparallel to one another, we call that antiferromagnetism.”  Wilson continues, “If they want to interact in this antiferromagnetic way, and if they are sitting on atoms forming a square network, then each moment can be antiparallel to its neighbors. The moments are ‘happy,’ and that is the ground state. In a different network, however, such as a triangle, not every moment can point opposite to its neighbors. They compete with one another, or are ‘frustrated,’ because they don’t know which way to point to realize the ground state of the system. The moments seek equilibrium but are frustrated from achieving it by the geometry of the space they occupy. ”

It turns out that a similar type of frustration can occur with other aspects of the electron, its charge, for instance. In particular, if two neighboring ions try to share an electron across a bond, they can form what is called an atomic dimer. Similar to the case of antiferromagnetism, the formation of these dimers can be frustrated in certain lattice geometries, such as triangular lattices or honeycomb networks. What can then result is a frustrated bond network that is highly susceptible to strain, which can act to relieve the frustration of the bond network. Wilson’s paper is about a system of materials where both of these types of frustration were found to coexist, which is extremely rare.  

Wilson describes this advance as “exciting” because it opens a window into functional control over one frustrated system via a perturbation that impacts the other. Over the past six or seven years, researchers have found that they can engineer a frustrated magnetic state by using materials built from triangular networks of lanthanides, a group of elements located at the bottom of the periodic table. “In principle, this triangular lattice network of properly chosen lanthanide moments can cause a special kind of intrinsically quantum disordered state to arise,” Wilson says. “One thing we tried to do in this project was to functionalize that exotic state by embedding it in a crystal lattice that has an additional degree of bond frustration.” This is of interest, because while there are many different “flavors” of quantum disordered magnetism, in principle, Wilson notes, “Some states can host long-range entanglement of spins, which is of interest in the realm of quantum information. Gaining control over those states via applying a strain in the frustrated bond network would be exciting.” 

If you have two highly frustrated layers that are both very sensitive to perturbations, like strain, or, in the magnetic case, a magnetic field, then the question is whether you can couple the two together, because when one is biased and decides to order, it can potentially couple to the second one and alter it. “It’s a way of imparting in things a functionality or response to other things to which it would otherwise not respond,” Wilson explains. “So, in principle, one can engineer large ferroic responses.You can apply a bit of strain, which induces magnetic order, or you can apply a bit of magnetic field and induce changes to the structure.

“Again, in principle,” he continues, “if you can find a quantum disordered ground state that hosts long-range entanglement, the question then becomes whether you can access that entanglement by, for instance, coupling to another layer, such as bond frustration.” 

Wilson also wants to discover whether, through this process, it is possible to realize different types of intertwined order. “Basically, you could have different types of order that get nucleated because of the proximity of these two frustrated lattices,” he says. “That's the big-picture idea.”

When cell phones, laptops, printers, and processors reach the end of their useful lives, they get sent off to be recycled. What’s left turns into e-waste — and, for many in developed nations, that means the issue seems to disappear. But what really happens to the 62 billion kilograms of technology that get discarded each year? UC Santa Barbara third-year computer science PhD student Pranjali Jain wanted to find out. In particular, she wanted to know more about the environmental impact of e-waste, with an eye toward designing electronics that are more sustainable throughout their lifecycle.  

But as she and her advisor, computer science professor and dean of the College of Creative Studies, Tim Sherwood, began looking into this question, they realized that despite the size of the problem, “there is absolutely no data out there about this,” Jain said. “There are no metrics about what components of electronics are the problem — at least not metrics that are going to be understandable to people who design and build computers.”  

Now, Jain’s project has become a multidisciplinary research effort that recently received the National Science Foundation (NSF) award. 

“We're really delighted to have the support from the National Science Foundation. That is transformative for this truly interdisciplinary project,” said Sherwood, who is one of the grant’s principal investigators, along with associate dean for research and director of the Materials Research Laboratory Ram Seshadri and computer science professor Jonathan Balkind. “And I'm really delighted for Pranjali to see that what began as her project is being recognized at this national level.”

This research also aligns closely with the mission of the university’s world-renowned Institute for Energy Efficiency (IEE), which all three principal investigators are already affiliated with.  By uncovering the material and energy costs embedded in computer hardware, the project supports the IEE’s broader goals of promoting sustainability, reducing environmental impact, and designing more efficient, environmentally responsible systems.

“Computing is changing so much about the world,” Sherwood said. And as the world changes, there’s an important question he said researchers need to be asking: “How should we be building computers in the future to get us ready for the world we want to create together?” 

Bringing E-Waste to Light

According to a recent United Nations report, the amount of e-waste generated in 2022 alone would fill 1.55 million semitrucks. Lined up end to end, these trucks would circle the entire equator. And the problem is getting worse: between 2010 and 2022, the amount of e-waste being produced nearly doubled.

Jain notes that, despite the enormous amount of e-waste, very little is known about what’s in it, particularly when it comes to computer hardware. “If you were to ask a computer scientist what materials go into making this laptop, they couldn't really tell you,” she said. “It’s not easily accessible public information. A lot of it is protected intellectual property, and companies themselves often don't really know exactly how much of which different materials go into making, for example, a laptop, because they also source the components from different manufacturers.”

It can also be difficult to make the connection between design and its ultimate destination. “I sit in my office and I write a line of code,” Balkind said. “But down the line, somebody — likely in a developing nation — is going to sit down in front of a fire pit and melt the computer to get out a valuable metal and, in the process, breathe in extremely noxious fumes.”

Starting with Servers

To start shedding some light on design choices that may have environmental repercussions down the line, the researchers first focused on one of the mainstays of computing: servers. Jain obtained two servers that were destined for e-waste — one from the California NanoSystems Institute at UCSB, and the other from the North Hall Data Center. 

Jain and postdoctoral researcher Alex Bologna spent hours in the Materials Research Laboratory crushing the circuit boards by hand into a powder — a process which, Jain said, was satisfying but also very labor intensive. Then, they dissolved the powder in a powerful acid, and ran the samples through an inductively coupled plasma spectroscopy instrument, which allowed the researchers to quantify the metal content of the servers.

To understand more about the potential negative impacts of metals in the servers, they brought in Dingsheng Li, a professor at the University of Nevada who previously spent time as a postdoctoral researcher at UCSB’s Bren School of Environmental Science & Management. They looked at established metrics for the eco-toxicity of metals, which includes impact on both the environment and human health. They also considered how servers are typically disposed of, Jain said, making an assumption that they would be put into a landfill.  

What they found was surprising: while the printed circuit boards of the servers did contain small amounts of metals known to be highly toxic — chromium, barium, nickel, and thallium — most of the toxicity came from metals like copper and aluminum that are thought to be non-toxic, Bologna said, because these common metals made up a greater proportion of the circuit board. “We can minimize the toxicity of servers simply by making them smaller and minimizing the materials used in them,” he said.

They also learned that a basic electrical component of the circuit boards, the capacitors, have an outsized impact on the boards’ toxicity. “This makes them a potential candidate for redesign or replacement for a computer architecture researcher like Pranjali,” said Bologna, who works with Sheshadri, a distinguished professor of materials and chemistry & biochemistry. 

Broadening the Scope

Now, the research team plans to acquire additional e-waste to study and machines to help process it. They also intend to broaden the project’s scope beyond toxic metals. Additional metrics, such as carbon, water, and energy usage, could be incorporated to further sustainable design and production.

Balkind envisions this research perhaps leading to a dashboard that hardware designers could use to look up the downstream impacts of various design decisions. “There are all these decisions you make during your design process, and if you just had a number in front of you, it might nudge you in the right direction, whether that’s toward an improvement in the toxicity of the components, or to their cost in terms of their carbon footprint, or the water and energy used to make them,” he said.

Eventually, consumers could also have a window into the impacts of the products they buy — perhaps via a system similar to a nutrition label, that describes a product’s contents, enabling buyers to make more informed choices, he said.

Balkind hopes that providing metrics around e-waste can empower engineers, computer scientists, and others who design and build new technology. “We want to provide motivating examples, metrics, and statistics to help people understand that a small decision can have a large — and positive — impact.” 

Data used to move along copper wires. Increasingly in recent years, it has been traveling much faster and more efficiently on light waves via lasers and fiber optic cables, with some short copper connections still in use on data-center racks. That set-up works fine for users who are doing everyday tasks like running a Google search, editing video, or completing an inventory report.

It is, however, far less effective for moving a different kind of data package, one that is many orders of magnitude larger. Even with the most up-to-date technology, transferring massive datasets on the petabit (PB) scale (a petabit being one quadrillion bits) — like those used in data analytics, genomics, experimental physics, or in training large language models such as ChatGPT — create bottlenecks at even the most advanced data center. 

Two students advised by UC Santa Barbara computer science assistant professor Jonathan Balkind, an expert in computer architecture, presented a paper on the topic of data transfer at the 2024 International Symposium on Computer Architecture (ISCA), held June 29-July 3 in Buenos Aires, Argentina. In one important point, Balkind and co-first authors, including Guillem López-Paradís, of the Barcelona Supercomputing Center, and UCSB College of Creative Studies undergraduate student Isaac Hair, who collaborated on the paper with multiple co-authors including PhD student Parker Murray, master’s student Rory Zahedi, and undergraduates Sid Kannan, Roman Rabbat, Alex Lopes, and Winston Zuo, explained that moving Meta’s 29-PB machine-learning dataset from node A to node B with 400Gb/s networking would take roughly an entire week. To achieve an optimized one-hour transfer time would require the network to be 161 times faster than it is now, reaching a bandwidth of 64 Tbit/s, which exceeds the capability of today’s top-of-rack switches.

The Balkind team argues for changing data-center architecture to enable “embodied data movement” — that is, physically carrying the data storage media across the data center. Their idea is to build “data-center hyperloops” (DHLs) to physically transport solid-state disks (SSDs) containing large datasets. Hyperloops consist of a pair of rails that use magnetic levitation to support and transfer a payload inside a low-atmosphere chamber from one endpoint to another. The team’s proposal applies that paradigm at the data-center scale to shuttle SSDs between compute nodes and cold storage. Using a DHL system, the team writes, “A data center can save energy while also raising performance, thanks to bulk network bandwidth being freed for other applications.” DHL, they write, “obtains energy reductions up to 376 times above 400Gb/s optical fiber.”

It might seem counterintuitive to imagine that such immense data hauls could move faster if delivered manually —  on a disk carried by a vehicle operating in a near- frictionless vacuum tunnel. Balkind is quite at ease in that counterintuitive realm. Asked a week prior to the conference what kind of reception he was expecting, he said, only half-joking, “I’m hoping for laughs.”

The reception was good, he reported upon returning, but he might have earned a few laughs, because his idea is so far out of the mainstream thinking as it relates to the increasing challenge of moving massive data parcels faster at reduced cost. The proposal, however, which he developed with his co-authors, is hardly a joke. It is conceptual and theoretical at this time, and it raises many as-yet unanswered questions, but, Balkind says, “The raw numbers from our model show that you can move a disk really fast.” Will it actually work? Time will tell, but, he notes, “If the numbers are good, it’s probably worth looking at, right?” 

The idea has its roots in a paraphrased quote presented in Andrew Tanenbaum’s well-known textbook on computer networking, which Balkind recalls as, “Never underestimate the bandwidth of a station wagon filled with hard disks hurtling down the highway.” Swap out the automobile for a very small “cart” measuring perhaps a few inches per side, and a hyperloop for the pneumatic tubes that used to be common in factories and at drive-through bank windows, and you have an updated analog of the data-packed station wagon hurtling down the highway.  

“We're not using copper or light to move data,” Balkind says. “We're saying instead, ‘Here's a train and a container, and we’re going to load it with disks, and then we just fire it across the data center as fast as possible. In terms of the pure mechanical work of moving these disks from A to B, maybe it could be faster and cheaper than whatever we're doing today.” 

Traditionally, the disks to be moved would have been hard disks, “But those are pretty heavy,” Balkind notes. In his proposed system, he would use SSDs, the densest of which can store around eight terabytes (TB) at a mass of six to eight grams. It’s a technology, he says, “that we haven’t really taken advantage of.”

The concept of physically moving disks has an analog in current everyday practice. Right now, Balkind explains, if you want to get a lot of data into a data center such as Amazon Web Services, “Amazon will let you pay to have a truck come to your door. They connect to your server via a cable, you write your data into this truck full of disks, and they deliver your data to the data center. It’s a much better choice than having to use all of the available network bandwidth you have for the next several weeks to get your data over there.

That process, however, “is still not very fast,” Balkind says. “The hard drives on the truck spin, so downloading the data takes time. Not only are SSDs data-dense, but you can get data into and out of them much quicker than you can from traditional hard disks.” Not to mention the time lost in transit as a truck motors down a highway.

At a data center, servers are stacked vertically in racks, and the racks abut each other to form long horizontal rows with aisles between them. Some of these long rows house machine-learning supercomputers, the enabling technology for today's generative-AI boom. Balkind imagines the hyperloop beneath a subfloor running perpendicular to the racks. “We're not moving any part of the server; we're only moving the storage with the data that you need on it,” he explains. “If you want to put data on, you key in a request for the cart. It shows up under your server, and you transfer your data into or out of it. It can be moved around as necessary. 

“It seems that it could be a win,” Balkind continues. “We've only done the early-stage modeling with some simulation elements, and we're not mechanical engineers, so we didn't build anything. And while various assumptions are baked in, we argue that many potential issues can be resolved with careful and creative engineering.”

It is a question of using the system to move only what it makes sense to move that way. “We want to move only things that we have to move, such as very large datasets for machine learning,” Balkind says. “Otherwise, we're burning energy for no good reason. And in terms of speed, we’re up against physics, because as we move the discs faster, we spend quadratically more energy. So we chose the densest storage we could get, a type of SSD known as m.2. They're about three inches long by an inch wide and weigh about a quarter of an ounce. Basically, you just pack a bunch of them together into the cart, and then move the cart. The carts will be in a library some of the time, or they’ll be plugged into a server rack at the bottom or in transit to their next data pickup or delivery destination.”

As with Tannenbaum’s “station wagon filled with hard disks,” perhaps the power of the Balkind team’s somewhat counterintuitive, but theoretically effective, approach to moving data should not be underestimated.