The Fusion Report got together with Keith LeChien, the chief technical officer (CTO) of Pacific Fusion, to discuss the work on their Impedance-matched Marx Generator (IMG), and to hear about their announcement at the end of this interview.

What is an Impedance-Matched Marx Generator (IMG), and what are your goals in the productization of your IMG? 

An Impedance-Matched Marx Generator (IMG) is the core building block of Pacific Fusion’s pulser-driven inertial confinement fusion system. It stores electrical energy in capacitors over a short time, say one to 10 seconds, and then releases that energy in about 100 nanoseconds, creating an ultrafast pulse of electrical energy.

What makes an IMG different is efficiency and simplicity. Traditional pulsed-power systems compress energy through multiple stages of intermediate energy storage. An IMG is designed to deliver that energy in a single step, transmitting power forward in a highly efficient way – approaching ~90% at the module level into a matched load.

This matters for fusion energy because if you can deliver more energy to the target with less stored energy upstream, you reduce system size, cost and complexity. 

A colleague and I invented the IMG concept 15 years ago and published it first in 2017. We built the first prototype unit at Lawrence Livermore National Laboratory in 2019, which is a very small version of what we're now doing at Pacific Fusion. 

Our fusion system, called the Demonstration System, is made up of 156 identical IMG modules and is designed to generate high peak currents and very high fusion output. Our goal for the DS is to achieve net facility gain – more fusion energy output than energy stored in the system – by 2030. 

Are the parts for your IMG and pulser “off the shelf” or are they unique?

At its core, the machine is built from simple materials and repeatable parts assembled in the right way. We manufacture some critical components in-house, while sourcing others, such as our capacitors, from commercial vendors.

Our system and strategy are intentionally designed using mass-manufacturable components with no major supply-chain bottlenecks so that in the future, we can rapidly deploy power plants at scale.

What quantitative results do you have of your IMG and pulser testing?

We’re executing on a milestone-based roadmap designed to achieve net facility gain by 2030:

• Our first phase focused on demonstrating that the key components (bricks, stages) of our pulser worked and that they met the specifications required to power the Demonstration System. Each brick is made up of two capacitors and a switch; a stage is made up of 10 bricks. 
• Next, we will build and demonstrate a full pulser module, which will generate ~2 TW of peak power at full capacity. We will then replicate the remaining ~155 identical modules to complete the system and demonstrate net facility gain.
• In parallel, our Fremont Test Center is focused on repetition-rate testing. A commercial power plant ultimately needs to operate at ~1 Hz. At its most irreducible scale, an IMG is two capacitors and a switch, and we’re running those components continuously to validate performance and durability. We have 10 brick testers that can operate 24/7, allowing us to iterate quickly.  

One of the things about inertial confinement I've always liked is that it's a really straightforward concept. The problem has always been that lasers are sexy, but they're not exactly efficient.

There are lots of different ways to skin the fusion cat (so to speak), but we all essentially have the same end goal. You want to deliver a very high Lawson criterion: plasma pressure times its confinement time. The higher that gets, the higher your fusion gain. The challenge is figuring out where to start to get to that end state as quickly as possible. 

That’s why we’re pursuing pulser-driven inertial confinement fusion: It's very efficient, so we can end up delivering more than 10 percent of the stored energy to the target – at least an order of magnitude higher than other methods such as lasers and potentially factors higher in some cases. 

When your goal is net energy production, efficiency matters. The less energy you have to store to achieve the required conditions, the easier it is to produce more energy than the system consumes.

When do you expect to be able to achieve an engineering break-even proof of concept with your system? 

The simplest way to frame it is: can you get more energy out than you put in?

In 2022, the National Ignition Facility demonstrated ignition by producing more fusion energy than was delivered to the fuel target. The next major milestone is going further, producing more energy than the entire system stores. That’s what we call net facility gain.

What sort of net energy or facility gain do you need to achieve to be able to say make it worthwhile as far as a fusion machine goes

This is an important point. Net facility gain is the next major engineering milestone in fusion — it tells you the system works end-to-end. But it’s not enough to build a commercial power plant.

To operate economically, you need target gains well above one. In a pulsed fusion system, the capital cost of the driver — the fixed hardware — scales inversely with target gain. The driver is roughly the same size and cost, but if each shot produces more energy, your cost per unit of energy drops.

Gains of three start to become interesting. Realistically, you need something closer to the five-to-seven range to achieve net facility gain. If you can reach 10, that’s amazing.

Our Demonstration System is ultimately designed to prove we can operate in that regime — because that’s what ultimately enables a commercially viable fusion power system.

Are there other advantages that you see with your approach? I mean, I would imagine target design is different with the way you're doing it than it is say with a laser-based system which is fairly complex. 

Fundamentally, it’s the same class of physics tools. But one of the advantages of a pulser-based fusion approach is that the system is energy-rich, which makes target design less sensitive. The implosions are also slightly slower, which makes targets more tolerant to manufacturing imperfections, giving pulser-driven ICF several practical advantages. 

Tell us about your new announcement that you are putting out on the 4th.
This is exciting and something that we're really proud of. We're building our Demonstration System to be the first net facility gain fusion driver, but it will also be the highest-flux neutron source on the planet. 

That capability opens the door to uniquely powerful collaborations. Today, on March 4, we announced a call for Expressions of Interest, inviting external users to access the Demonstration System. We’re targeting a broad set of areas: fusion energy research, commercial applications, fundamental high-energy-density (HED) science, and national security.

• In fusion energy science, the DS could support ignition-scale physics experiments, target validation, diagnostics development, and materials testing under extreme conditions. 
• In commercial applications, high-flux radiation environments can help test how components will hold up in outer space over time, radiation tolerance, and evaluation of materials exposed to high-energy debris and thermal loads, medical isotope production, and other applications of pulsed power technology.
• In fundamental HED science, access to intense neutron and X-ray sources can reveal new behaviors and open new engineering frontiers. 
• We also expect collaboration with the U.S. government on applications relevant to national security.

Submissions for Expressions of Interest are due March 31. This precedes a formal call for proposals next year. We expect the process to be lightweight and iterative, working with prospective partners to refine concepts and shape early use cases. Our goal is to engage initial collaborators later this year and begin outlining the first wave of experiments, with more details to follow as the program develops. Those interested can learn more at  users.pacificfusion.com.