# Gallium vs. Plasma FIB: What’s Actually Happening Inside the Column > If you’ve spent any time around a FIB-SEM, you’ve heard the shorthand: “Ga FIB” and “PFIB,” tossed around like everyone knows the difference. Most people know that they’re different (one is gentler on some samples, one mills faster, one costs more to run) without necessarily knowing why. The answer comes down to how each instrument […] Source: https://covalent.com/blog/gallium-vs-plasma-fib/ Updated: 2026-09-10T16:55:07+00:00 --- ![](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/themes/covalent/assets/images/blog-default-banner.jpg) ![](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/themes/covalent/assets/images/blog-mobile-banner.jpg) # Gallium vs. Plasma FIB: What’s Actually Happening Inside the Column Valerie Brogden • Sep 10, 2026 If you’ve spent any time around a FIB-SEM, you’ve heard the shorthand: “Ga FIB” and “PFIB,” tossed around like everyone knows the difference. Most people know that they’re different (one is gentler on some samples, one mills faster, one costs more to run) without necessarily knowing why. The answer comes down to how each instrument makes its ion beam in the first place, and once you see the physics, the practical trade-offs make a lot more sense. Every focused ion beam instrument needs a source: something that reliably produces a stream of ions that can be focused, accelerated, and steered onto a sample. There are two fundamentally different ways instrument makers have solved that problem. One uses a literal droplet of liquid metal. The other uses a cloud of plasma. Here’s what’s going on inside each. ## **Gallium FIB: a liquid metal pulled into a beam** The classic FIB source is a liquid metal ion source (LMIS): a tungsten needle wrapped in a heated coil, with a small reservoir of liquid metal feeding down the needle. Heat the reservoir just enough to melt the metal, and it wicks down the tungsten tip by capillary action, where an extractor electrode a short distance away applies a strong electric field. At low field strength, that droplet of liquid metal on the needle tip just sits there, held in a rounded shape by its own surface tension. But push the voltage past a critical threshold, and the electric field wins: the droplet stretches into a sharp, cone-shaped point called a Taylor cone, with a radius of curvature at the very apex measured in nanometers. At that scale, the local electric field is intense enough to strip ions directly off the liquid surface in a process called electrospray. This produces an ion beam from a source spot smaller than almost anything else in the microscope. ![Diagram comparing gallium and plasma FIB with ion trajectories and components.](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/09/image-1.jpeg) *Figure 1- Schematic of the Ga LMIS.* Nearly any metal can, in principle, form a Taylor cone and emit ions this way. Lithium, aluminum, gold, indium, bismuth, and dozens of others have all been demonstrated as LMIS materials at one point or another. But gallium became the industry default for a specific combination of reasons. It melts at about 86°F (30°C), barely above room temperature, so the source runs at a gentle, easy-to-maintain temperature instead of the several-hundred-degree heat many metals would need. It has low vapor pressure, so it stays compatible with the high vacuum inside the column instead of contaminating it. A well-made gallium source can run for 500 to 1,500 hours before it needs replacing. Gallium is also a relatively heavy ion, which makes it efficient at sputtering material. It forms an unusually sharp, bright emission tip, and it ionizes cleanly as a single ion rather than clumping into clusters. And practically, gallium’s characteristic X-ray lines sit cleanly apart from the lines of most sample elements, so when trace gallium gets implanted into your sample during milling (which it does!) it’s far less likely to muddy an EDS spectrum than a lot of the alternatives would be. Gallium’s low melting point is also why it makes a good party trick: it’s solid at room temperature but will start to liquefy in your palm, since human body temperature runs a few degrees above its melting point. It’s memorable enough that it’s become a fixture of microscopy classrooms and trade-show booths for a reason. ![](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/09/image.gif) *Figure 2- Photograph of a Covalent employee playing with gallium. It’s safe to touch but it does embrittle metal, so you need to be careful not to let it touch jewelry.* There’s a nice piece of history behind this technology, too. The liquid-metal-in-a-strong-electric-field idea wasn’t invented for microscopy at all. In the early 1960s, researchers were investigating charged liquid-metal droplets as a potential propulsion source for spacecraft. The idea here was to use tiny, efficient thrusters that would eject charged metal ions instead of burning fuel. That electrohydrodynamic ion source research, done in the context of space propulsion, is the direct ancestor of the LMIS technology that ended up inside every gallium FIB column. ## **Plasma FIB: pulling ions from a cloud instead of a point** Plasma FIB (PFIB) sources work on a completely different principle. Instead of a liquid metal tip, a plasma FIB uses an inductively coupled plasma (ICP) source. Often xenon is used in PFIB’s, however it’s possible to use other gas species as well, including oxygen, nitrogen and argon. Gas is fed into a small dielectric chamber wrapped in an RF antenna. Current oscillating through that antenna at radio frequency creates a changing magnetic field, and that changing field induces a current in the gas. The RF frequency is tuned so that it resonates with the free electrons in the gas, heating them significantly while the much heavier ions stay close to room temperature. Those energized electrons collide with neutral gas atoms, knocking more electrons loose and generating more ions, which collide and generate still more. This cascading ionization process fills the chamber with plasma. Ions are then drawn out through an extraction electrode and formed into a beam, the same way the gallium beam is formed downstream of its source, just starting from a very different kind of source in the first place. ![Diagram of plasma FIB system showing gas inlet, plasma chamber, and electrodes.](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/09/image-2.jpeg) *Figure 3- Schematic of an inductively coupled plasma source.* ## **Why the source shape changes everything about the beam** Here’s where the physics actually matters for your sample. A liquid metal source emits from a sharp cone. At low beam currents, that geometry is a huge advantage: you can use a small aperture and collect a tight, well-defined slice of that cone-shaped emission, giving gallium FIB genuinely excellent spot sizes, down to just a few nanometers. But if you need more current because you’re removing a larger volume of material, you have to open the aperture wider to collect enough of that cone. Because the emission is cone-shaped, spot size grows sharply, roughly exponentially, as current goes up. Push a gallium source to the tens of nanoamps, and the spot size balloons along with it. A plasma source doesn’t have that problem, because it isn’t emitting from a single point at all. The PFIB draws ions from a broad, flat plasma cloud, and that emission is inherently more collimated. Spot size still increases as you push more current through a PFIB, but the relationship is closer to linear than exponential. The practical result, as shown in work comparing the two source types1, is that at high beam currents the spot size for the LMIS FIB is very large while the PFIB has a smaller spot size at high currents but a larger spot size than the LMIS FIB at low currents. ![Diagram of Gallium and Plasma FIB techniques in material analysis.](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/09/image-3.jpeg) *Figure 4- Comparison of the shape of the LMIS beam vs the PFIB beam.* ![Graph showing spotsize D50 versus beam current for LMIS-Ga and ICP-Xe.](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/09/image-4.jpeg) *Figure 5- Spot size comparison of Ga-LMIS and Xe-ICP based FIB columns as a function of the ion beam current.* ## **So which one do you actually want?** Neither source is simply “better.” They’re suited to different jobs, because of that spot-size-versus-current curve. Gallium FIB is the right call when precision at low current matters more than speed. The LMIS Ga FIB is the perfect fit for application such as fine TEM lamella preparation, small failure-analysis cross-sections, and any application where nanometer-scale FIB resolution is the priority and the volume of material you need to remove is small (less than ~30µm). Plasma FIB earns its keep when you need to remove a large volume of material and still keep your resolution intact. PFIB is great for large cross-sections, high-throughput 3D serial sectioning and tomography, and bigger failure-analysis sites where a gallium source would force you to choose between an impractically slow mill and a beam too coarse to trust. There’s often a second reason to reach for PFIB beyond speed, too: some materials, including certain compound semiconductors like GaN, are sensitive to gallium implantation itself, and an inert gas beam avoids introducing an element that can alter the very properties you’re trying to characterize. In practice, the right answer depends on the sample and the question you’re asking of it. That is exactly why it’s worth having both tools, and the judgment to know when each one is doing its best work. *Have a sample that’s got you debating gallium vs. plasma FIB? That’s a conversation we have with customers every week! Reach out to the Covalent Metrology team and we’ll help you match the tool to the job.* **Citations** 1. Young, Rue, Randolph, Chandler, Franz, Schampers, Klumpp, and Kwakman, “A Comparison of Xenon Plasma FIB Technology with Conventional Gallium LMIS FIB: Imaging, Milling, and Gas-Assisted Applications,” *Microscopy and Microanalysis* 17(S2), pp. 652–653 (2011), DOI: 10.1017/S1431927611004132. About the Author ![Valerie Brogden](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2025/12/valerie-brogden-150x150.jpg) Valerie Brogden Technical Marketing Valerie Brogden is a technical marketing professional at Covalent with 15 years of electron microscopy experience. Her expertise includes scanning electron microscopy (SEM) and focused ion beam (FIB) workflows. Before joining Covalent, she worked as an applications scientist at Thermo Fisher Scientific and later managed an electron microscopy lab at University of Oregon. She holds advanced degrees in physics and chemistry. Table of Contents Blogs you may like [![Read time](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2025/07/clock.svg) 11min read ![Semiconductor Failure Analysis: Techniques and What They Reveal](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2025/11/blog-default-banner.jpg) Aug 14, 2026 ### Semiconductor Failure Analysis: Techniques and What They Reveal Semiconductor failure analysis investigates why a semiconductor device or package has failed to meet its functional or parametric requirements (such as specified limits for leakage current, threshold voltage, or breakdown voltage). It combines electrical...](https://covalent.com/blog/semiconductor-failure-analysis-techniques-and-workflow/) [![Read time](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2025/07/clock.svg) 9min read ![Amorphous vs Crystalline Materials: Structure, Properties and Characterization](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/07/adobestock-1897924822-scaled.jpeg) Jul 13, 2026 ### Amorphous vs Crystalline Materials: Structure, Properties and Characterization Crystalline materials have both short- and long-range structural order: their atoms, ions, or molecules are arranged in patterns that repeat over extended distances. Amorphous materials retain short-range order, meaning neighboring atoms, ions, or molecules...](https://covalent.com/blog/amorphous-vs-crystalline-materials/) [![Read time](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2025/07/clock.svg) 3min read ![Topgolf Ball microCT Scan Reveals Hidden Electronics](https://spcdn.shortpixel.ai/spio/ret_img,q_cdnize,to_auto,s_webp:avif/covalent.com/wp-content/uploads/2026/07/picture3.png) Jul 06, 2026 ### Topgolf Ball microCT Scan Reveals Hidden Electronics At Covalent, curiosity doesn’t clock out at 5 p.m. Recently, our team headed to Topgolf San Jose for an afternoon of team building. The agenda was simple: enjoy some friendly competition, spend time together...](https://covalent.com/blog/topgolf-ball-microct-scan-hidden-electronics/) Subscribe to Our Newsletter Subscribe to our newsletter for the latest updates, insights, and breakthroughs - delivered straight to your inbox.