Marlene Turner inspects a capillary discharge waveguide in the BELLA lab, which serves as the accelerating chamber
© The Regents of the University of California, Lawrence Berkeley National Laboratory
Advancing particle acceleration for the future
An interview with accelerator physicist Marlene Turner
by Paul Huber
March 2023
Math has successfully predicted many things in physics ahead of their actual observations or experimental data. Two examples are Karl Schwarzschild’s
prediction of black holes in 1916 from Einstein’s General Relativity theory and Paul Dirac’s prediction of antimatter in 1928 from one of his
equations which had two opposite solutions. But there have also been many mistakes and experimental guidance, knowledge gained from real-world
experiments, is needed to know if the math is representing reality or wandering off course.
In particle physics, a big part of experimental guidance comes from giant colliders, which collide particles at high energy then detect and analyze the
products of those collisions. They’re largely defined by the maximum collision energy they can achieve. The Tevatron at Fermilab in the US peaked
at 1.96 TeV (1.96 trillion electron volts*). It closed in 2011 and was notable for the discovery of the top quark, a constituent particle of protons.
The Large Hadron Collider (LHC) at CERN near Geneva, Switzerland is currently operating at a peak of 13.6 TeV. But most physicists think there is more
beyond the Standard Model and exploring that realm will take much higher energies. *An electron volt is a unit of energy equal to the work done by an
electron accelerated through a potential difference of 1 volt.
Current collider technology uses radio frequency (RF) cavities to accelerate the particles through circular tubes, about 27 kilometers circumference
in the case of the LHC. One of the reasons for that great length is a factor called “acceleration gradient”, the amount of speed or energy increase
for a given length of RF cavity. A familiar analogy might be a car accelerating from 0 to 100 kilometers per hour in the distance of half a kilometer.
If that’s at maximum power then the car needs more distance to go faster. The acceleration gradient of RF cavities is physically limited by their
shape and material and it’s getting harder to improve as designs are maximized. In the LHC, protons travel around the 27 km ring millions of times
before reaching the desired collision energy. During all those laps they need to be guided by large steering magnets. These magnets provide the
centripetal force to keep the protons in a circular path but they are expensive and consume a lot of electrical power. This all adds up to much
greater costs with energy increases while funding resources are finite.
Now imagine a technology with a superior acceleration gradient that can achieve high collision energies with less size and lower cost. Physicists may
be able to probe deeper into the nature of matter with available resources. This is what Marlene Turner and others are working to develop, and it may
be what’s needed to help answer some questions in physics that we’re currently stuck on.
Following is my discussion with her about the various aspects involved and the current state of progress:
Q1: Did you originally plan to work in this field or were you drawn to it later?
A career in accelerator physics became my number one choice in university. Being a researcher
was a childhood dream, and a fascination with plasma wakefield acceleration started during my
time as a graduate student. So I guess it's fair to say that this was my original plan.
My first interaction with particle accelerators was back in high school when our class went on a
school trip to Hamburg, Germany, and visited the DESY (Deutsches Elektronen Synchrotron)
laboratory and its accelerators. Growing up in a family with a working background, we always
tried to build and repair devices. My dad was trained as a car mechanic and I helped him repair
cars (and my scooter that kept breaking) in my teenage years. So, seeing a high-energy physics
particle accelerator for the first time was amazing. I was fascinated by its complexity and
sophistication. All the cables, all the magnets, all the controls... Everything was unique and
needed to work just right. The challenge of understanding and working on accelerators excited
me.
After actually starting to work on accelerators, I realized that today's high energy frontier
colliders are limited by their acceleration technology. While working amazingly well as tools to
extend our understanding of the universe and the basic laws of physics, they are limited in their
scientific reach by their size and cost. Existing colliders are already very large and very costly
and it is not clear that we can keep building even more expensive and even larger machines. I
understood that if we want to go far beyond, we need new technology. So I adjusted my course
and started working on the research and development of a new way to accelerate charged
particles—plasma wakefield acceleration, which is a path towards next-generation colliders that
are substantially smaller and less expensive to build and operate.
I am excited about this new particle acceleration technology, the fundamental science to be
done while developing it, as well as the discovery science that it is going to enable. I hope it will
allow us to discover—allow us to understand—things that are not understood to date, and I am
excited to see how these findings will impact society.
Showing the proposed Future Circular Collider for comparison.
In addition to acceleration length, ring collider size is also determined by the radius of curvature. The faster the particles are moving,
the stronger the magnetic field you need to bend their path around the ring. For a given speed, a tighter bend needs a stronger magnetic field,
and you reach the limitations of steering magnets that can be practically built and powered. The solution becomes a ring with a larger radius of curvature.
(image credit: Pcharito)
Q2: In 1979, a paper by Tajima and Dawson proposed "harnessing electric fields of high
amplitude plasma density waves driven by intense laser pulses". Was this the beginning
of the idea to your knowledge?
Yes. In their seminal paper in 1979, Tajima and Dawson proposed to use the enormous fields
sustained by plasma waves for the acceleration of electrons. Charged particles can increase
their energy when interacting with electric fields. The particle’s energy gain is a product of the
accelerating field and the length over which the particle interacts with the field.
Before Tajima and Dawson, people had already used very large electric fields from laser pulses
directly for acceleration. However, in a laser pulse, the electric field is transverse to the direction
of propagation. This means that particles can only interact with the high-intensity laser field (and
therefore be accelerated) over a short time or distance, which limits the particles’ energy gain.
Tajima and Dawson realized that particles can interact with fields caused by laser pulses over
much longer distances if the laser pulse energy is first transferred to a plasma, by the laser
pulse driving and exciting a plasma wakefield. The plasma wakefields are longitudinal (in the
direction that the particle accelerates); also, they co-propagate with the accelerating particles.
However, at the time when Tajima and Dawson published their paper, short, intense laser pulses
were not yet available. They were enabled by the idea of Chirped Pulse Amplification, for which
Donna Strickland and Gérard Mourou received the Nobel Prize in 2018.
For completeness, I would like to mention that intense particle bunches can also drive
wakefields. A dense highly-relativistic particle bunch carries a large transverse electric field that
could be used to drive wakefields. In 1985, Chen et al., proposed using electron bunches to
drive plasma wakefields for particle acceleration.
Q3: I’ve seen it estimated that this technology could yield a 1,000x improvement in
acceleration gradient over RF cavities. Do you think that’s realistic?
Though almost unbelievable, these gradients have already been demonstrated by many groups
through different experiments.
Conventional acceleration technology based on radio-frequency (RF) cavities can achieve a
gradient of around 30 MV/m. That means that in 1 m of acceleration distance, a particle can
gain an energy of ~30 MeV. Lots of ongoing R&D work aims to increase this gradient and some
cavities may be able to get up to 100 MV/m meter. However, reaching even higher gradients is
very challenging because of electrical discharge.
RF cavities are made out of metal and are loaded with strong RF power to create electric fields
that accelerate the particles. However, at some point the fields become so strong that electrons
are ripped out of the metallic walls. That causes a spark (like a mini lightning), which destroys
the electric field and in the long term destroys the cavities. This electrical breakdown
fundamentally limits the gradients achievable in RF cavities.
Plasmas, on the other hand, are ionized—you can think of them as already broken down, with
electrons separated from nuclei. Therefore plasmas can sustain much higher field strengths,
e.g. in plasma wakefields, and ionization no longer limits the achievable gradient. Plasma
wakefields are electron charge density oscillations. You can imagine them to be a bit like water
waves. Particles (or surfers in the water wave analogy) can ride and accelerate on the waves.
So what has been demonstrated in the laboratory?
Using laser pulses to drive and excite wakefields, the group that I am part of (the Berkeley Lab
Laser Accelerator Center, or BELLA, at Lawrence Berkeley National Laboratory) has
accelerated electrons from rest to 4.2 GeV in 9 cm (2014) or 7.8 GeV in 20 cm (2019). These
achievements corresponded to n acceleration gradients of ~ 47 and 40 GV/m. This is more than
1000 times stronger acceleration than the ~30 MeV of RF cavities.
Researchers using particle bunches to drive the wakefields, e.g., the FACET group at Stanford
Linear Accelerator Center, have accelerated electrons by 42 GeV in 85 cm of plasma,
corresponding to an acceleration gradient of 50 GV/m.
Q4: The experiments I’ve read about are currently accelerating electrons. Is that because of
the small mass, and is this equally suited to accelerating larger mass protons as they
do with the LHC?
That is right; plasma wakefield acceleration is typically used to accelerate light particles such as
electrons or positrons. From the physics point of view, any charged particle can be accelerated,
including heavier particles such as protons or muons. However, in practice, circular accelerators
based on RF cavities work very well for heavy charged particles and tend to be the primary
choice. Let’s take a step back into fundamentals to look at the reasons.
Circular accelerators have at least one acceleration section, along with many dipole bending
magnets all over the ring to keep particles on their circular trajectories. Particles pass the
acceleration section repeatedly and keep gaining energy on each turn. The acceleration
gradient is not that important because particles pass very often.
There is also energy loss as particles are bent onto a circular trajectory by dipole magnetic
fields. When the direction of a charged particle is changed (e.g. by a dipole magnetic field), the
particle emits radiation (called synchrotron radiation) and consequently loses energy. So, on
each turn, particles gain energy from the acceleration sections, but they also lose energy in the
form of synchrotron radiation. This is a quantum-mechanical process, but a good classical
analogy is turning a corner in a car — the faster you are going, or the harder the turn, the more
energy you lose.
Energy loss to synchrotron radiation scales as the square of the bend radius and the fourth
power of the energy. At some point the energy loss per turn equals the energy gain in the
accelerating section, and that is the maximum energy possible in the circular accelerator.
The amount of energy loss due to synchrotron radiation in each turn also depends on the fourth
power of the particle’s rest mass. Therefore lighter particles radiate much more energy than
heavy particles on the same trajectory.
Let me give you an example, the world's longest accelerator is a circular machine 27 km long —
the Large Hadron Collider at CERN. A previous accelerator in the same tunnel, the Large
Electron-Positron collider (LEP), had been used to accelerate electrons and positrons up to an
energy of about 0.2 TeV per beam. The LHC, which accelerates protons, can reach up to 7 TeV
per beam
Circular accelerators will likely continue to be used for heavy particles such as protons.
However, they have long since reached a practical limit for light particles such as electrons. For
those, one typically uses linear accelerators (which have almost no synchrotron radiation
losses). However, in linear accelerators, particles pass through the acceleration section only
once. There it makes sense to switch to a technology with a very high acceleration gradient.
Accelerating electrons to, let's say, 10 TeV would require 100 km of RF cavities with an
acceleration gradient of 100 MeV/m, but only 1 km of plasma accelerator stages.
Q5: Do you think a better acceleration gradient would mean the end of ring colliders or would
they still have benefits, such as beam recirculation?
Circular and linear accelerators are different tools for different jobs. Rings will keep being
important tools for high energy physics based on heavy particles such as protons. Scientists are
also considering the use of circular machines to accelerate heavy leptons such as muons (the
muon mass is comparable to the proton mass). Muons pose their own set of challenges, as they
are unstable particles that decay. Apart from that, circular machines do offer exciting possibilities
for energy recovery, e.g., to decrease collider power consumption.
The enormous gradients of plasmas may also enable similar energy recovery techniques for
linear accelerators, however, it will take many years of R&D for these technologies to be ready.
Q6: When driven internally by a particle beam or laser pulses, the wake diminishes over a
short distance (centimeters?) as energy dissipates. Is this a serious problem, and if so, is
there a way to overcome that and sustain the wake by driving it externally?
The lifetime of the plasma wave is around picoseconds (10-12 s), and the lifetime of the plasma is
nanoseconds (10-9 s). In a plasma wave, the energy decoheres and dissipates over a couple of
oscillations, which is also on the picosecond scale. At first that seems short (~mm scale at the
speed of light).
However, the time duration of the laser pulses and particle bunches is only tens of
femtoseconds (10-15 s) and the laser pulse and electron bunch travel in the same direction at
approximately the same speed. The electron bunch follows the laser pulse very closely for
hundreds of femtoseconds to picoseconds. To some extent, what happens after that doesn't
really matter.
We are actually working on ideas to purposefully remove all the remaining energy after the
electrons have passed. We want to recuperate and reuse this energy and thus make particle
accelerators greener by decreasing their power consumption.
Q7: Does this type of acceleration impact the issue of luminosity (positively or negatively)
as far as particle bunches and/or ability to focus the beam?
That is a very good question. Let me start with luminosity. Luminosity is a measure of the
collider discovery potential. It describes how many collision events are expected per fraction of
time (say, 50,000 collisions per second). There is a certain probability, called the cross section,
of producing any given new particle. Higher luminosity means higher collision rates and thus
greater likelihood of producing that particle in a reasonable time. Achieving high luminosity is a
major challenge for plasma-based accelerators. The size of the acceleration bubble is much
smaller than conventional RF cavities. It is therefore tricky to pack a similarly large amount of
particles into each wave for acceleration.
However, there are certain tricks to increase luminosity. For example, within limits, we can pack
the particle bunches more densely. Also we can make the beams very very small in both
transverse and longitudinal directions so that at the interaction point it is very likely that two
particles actually collide. The idea sounds simple, but is hard to realize, and we will need to
work out what we call a beam-delivery system, a focusing system that can pack these bunches
tightly after acceleration. Every collider has a beam delivery system, but plasma based colliders
may need an especially sophisticated design.
Another way to increase luminosity is to increase bunch repetition rate. Simply accelerate more
bunches so they collide at a higher rate. Also here, plasmas have an advantage because they
operate at higher frequencies than RF cavities. To summarize, luminosity is a major challenge,
but we are working on ways to increase it to the desired level.
Q8: What do you see as the biggest challenges to overcome in order to advance this
technology to its potential?
There are quite a few challenges on the path to a plasma based collider. These will require
significant R&D efforts over the coming one or two decades.
For an electron-positron collider, the biggest challenge might be the acceleration of positively
charged particles. In RF cavities, particles accelerate in electric fields loaded in vacuum.
However, in plasma, there are both negatively charged free electrons, which sustain the
wakefields, and positively charged ions. The positive ions form a constant background that
focuses negatively charged particles such as electrons during acceleration, but defocuses for
positively charged particles, e.g., positrons. Acceleration of high quality positron bunches is
therefore really hard and we are working on schemes that will allow acceleration without beam
quality degradation.
Another challenge is the acceleration of bunches with high quality at high efficiency. Linear
colliders need insanely high beam quality. Proof-of-principle experiments with plasma
accelerators have demonstrated most parameters, but not all of them together.
Acceleration efficiency is also crucial for colliders. Acceleration efficiency is a measure of how
much wall plug energy is transferred to the particles. For instance, if a particle beam carries a
power of 10 MW and the acceleration efficiency is 10%, 100 MW of wall plug power is required
for acceleration. Typical beam powers are on the order of tens of MW, and 5-10% energy
transfer is typical for linear colliders.
Therefore the power consumption of a next generation collider might be on the order of a few
hundred megawatts, which is really a lot and frankly unacceptable. For reference, the city of
San Francisco has a total power consumption of 700 MW. We are actively researching ideas on
how to make acceleration more efficient and more affordable—for example, by implementing
energy recovery technologies.
Q9: Is it a reasonable assumption that plasma wakefield acceleration will eventually surpass
RF cavities for high energy colliders, and if so, would you have an estimate on when?
I hope that plasma wakefield acceleration will be able to eventually replace RF cavities in
applications where size and cost really matter. Examples can be found across the range of
possibilities. Imagine a linear high energy physics collider that is a kilometer rather than many
tens of kilometers long; or light sources that could fit in a university basement; or medical
treatment accelerators small enough to actually be inserted into the body next to a tumor.
Plasma wakefield technology for applications could become ready over the next ten years. A
particle collider is more on the twenty-year timescale and there are significant outstanding
challenges that will have to be addressed with intense R&D efforts.
On the other hand, RF cavities work really well. My guess is that there will always be some
applications that prefer to keep using them because of their ability to accelerate very high
current beams, or in situations where wakefield drivers (laser pulses or particle beams) are
simply not practical or available.
Q10: You’re currently working at the Berkeley Lab Laser Accelerator Center (BELLA) on a
Petawatt Laser project, correct? Can you describe that a bit?
The BELLA Center is located at Lawrence Berkeley National Laboratory in Berkeley, California
and hosts many lasers, among them three high-power, short pulse laser systems. BELLA is well
known and well established in the advanced accelerator community and the BELLA PW system
is most known for producing 7.8 GeV electron beams over only 20 cm of plasma. To date, this is
still the world record in a laser-driven plasma wakefield accelerator.
Two of the BELLA lasers are on the hundred terawatt level and are used to produce very
high-energy photon beams via Compton scattering (useful for material science and
security
applications) and to develop a plasma based free-electron laser. The most powerful laser is the
BELLA Petawatt, which provides PW laser pulses for R&D on high-energy electron acceleration
in plasma and ion acceleration from solid targets.
The laser systems are used to develop plasma wakefield acceleration technology and to provide
beams for user experiments. The BELLA Center is also developing fiber laser technology
because current Ti:Sapphire laser technology is unable to provide the required repetition rates
for a collider. Fiber lasers provide a new approach to high-power lasers by coherently combining
pulses from many lower-energy optical fiber lasers, and they can work at tens of kilohertz
(whereas BELLA PW has a repetition rate of 1 Hz). We think that fiber lasers are the path to
increased stability and performance as well as higher pulse rate, which is required or desired for
many applications.
Q11: What are your hopes for this technology in the near and long-term future?
My hope is that we can advance the plasma wakefield technology to the point where it can
serve society. It would be exciting to be able to provide affordable, compact and reliable
accelerators for medical, security and scientific applications as well as discovery science at the
energy frontier.
My big, long-term dream is to build a high energy physics collider (preferably based on plasma
wakefield technology) that allows us to go to unprecedented collision energies and to expand
our understanding of the structure of the universe. That is probably the most challenging goal of
all. However, I also am very excited about more approachable goals, e.g., providing the
accelerators for medical applications. Acceleration structures could become so small that they fit
inside the body, which would mean that one could irradiate cancerous cells right where they are
without sending particles through healthy tissues. Apart from that, treatment options could
become significantly more affordable since the accelerators themselves will be much more
compact. That would also make treatment options more accessible
New technologies have a potential to change the world. Think about transistors that enabled
modern computers, the invention of the Internet, or switching from huge TVs based on cathode
ray tubes to compact and elegant flatscreens. We are changing the technology on which particle
acceleration is based and who can imagine all the exciting possibilities of widely available and
affordable accelerators? The unimaginable ways in which our technology will change the world
are what I am most excited about. □
Marlene Turner is a research scientist in the
Accelerator Technology & Applied Physics Division at Lawrence Berkeley National Laboratory, working on high-energy electron acceleration in
laser-driven plasma wakefields and the next generation of particle colliders. She is also a member of the Accelerator Frontier Implementation
Task Force (ITF), which evaluates proposals for new accelerator facilities.
Some resources of interest:
The Berkeley Lab Laser Accelerator (BELLA) Center
Energy consumption, cost considerations could shape future of accelerator R&D (Symmetry Magazine)
The Advanced Proton Driven Plasma Wakefield Acceleration Experiment (AWAKE)
Accelerating: Radiofrequency cavities (CERN)
Beyond Higgs: The Wild Frontier of Particle Physics (World Science Festival video)
Future Circular Collider (CERN)
