The First Wall

The first wall is the wall closest to the burning plasma inside a fusion reactor. This single wall takes the full brunt of the intense heat and particles coming from the plasma head-on, protecting the power-generating heart of the reactor that lies just outside it. On this page we will look, in order from intuition to the research frontier, at the harsh environment the first wall is placed in, why choosing its material is so difficult, and what role it plays as a component of the reactor.

Start with Intuition (High School)

At the center of a fusion reactor, a plasma burns at roughly 100 million degrees. That is hotter than the center of the Sun. The innermost wall of the vessel that confines that flame is the first wall. Just as its name says, it is the “first wall the plasma runs into.”

Imagine holding your hand over a campfire. Even from a distance you feel the gentle warmth, right? The first wall is like pressing not just a palm but an entire wall right up against that fire. And its opponent is a super-high-temperature plasma that is beyond comparison with any campfire. The intensity of the heat the first wall receives, expressed per unit area, reaches anywhere from several hundred kilowatts to several megawatts. Picture the heat of several hundred to several thousand household microwave ovens pouring continuously into one square meter.

The tricky part is that it is not only heat. When fusion reactions occur, particles called neutrons are born, flying at tremendous speed. Because neutrons carry no electric charge, they cannot be bent by magnets; they pass straight through the wall while depositing energy along the way. On top of that, charged particles and fast atoms from the plasma also slam into the wall.

In other words, the first wall suffers a triple affliction all at once: “burned by heat,” “broken from the inside by neutrons,” and “eroded on the surface by particles.” There are not many materials that can withstand such an environment for years. That is precisely why what to use for the first wall is one of the most troubling problems in fusion reactor design.

If we sum up the role of the first wall in a single word, it is a “shield.” By taking damage itself, it protects the device that recovers heat (the blanket) and delicate components such as the superconducting magnets that lie outside it. If the shield is too thin or too weak, the contents get destroyed; but if it is made too sturdy and thick, then the heat cannot be transferred well. Striking this delicate balance is where the skill of first-wall design shows.

Understand the Physics (Undergraduate)

Let us pin down the loads the first wall receives quantitatively. The loads on the first wall can be broadly divided into two: the surface heat load from the heat raining down on the surface, and the neutron wall load brought in by neutrons.

The surface heat load is delivered by electromagnetic radiation from the plasma (bremsstrahlung and synchrotron radiation) and by charged and neutral particles leaking out of the scrape-off layer. On the ITER first wall, this surface heat load is designed to be steadily around $0.25$ to $0.5$ MW/m$^2$. Here MW/m$^2$ is a unit expressing how many megawatts of heat flow in per square meter, and it is the basic yardstick for measuring how harsh the first wall’s environment is. For reference, at the divertor, where heat concentrates more locally, this value becomes larger by orders of magnitude (see Divertor for details).

The neutron wall load, on the other hand, is the energy flux that the fast neutrons born in fusion reactions carry into the wall. In the reaction of deuterium and tritium (the D-T reaction), a single reaction releases $17.6$ MeV of energy, of which $14.1$ MeV is carried away by the neutron. This neutron passes through the first wall and reaches the blanket, where it is converted to heat and used for power generation. The neutron wall load on the ITER first wall is roughly $0.5$ to $0.8$ MW/m$^2$. In future reactors aiming for power generation, this value is expected to rise to $2$ to $5$ MW/m$^2$.

Whether a material can withstand the surface heat load is determined by whether it can carry the heat away into its interior by thermal conduction. The temperature gradient inside the wall in the steady state can be estimated from Fourier’s law.

$$q = -k \frac{dT}{dx}$$

Here $q$ is the heat flux (in units of $\mathrm{MW/m^2}$), $k$ is the thermal conductivity of the material (in units of $\mathrm{W/(m \cdot K)}$), and $dT/dx$ is the temperature gradient. What this equation says is that even when receiving the same heat flux $q$, a material with a larger thermal conductivity $k$ can keep the internal temperature rise smaller. That is why first-wall materials are required to have not only a high melting point but also high thermal conductivity. In fact, in ITER, beneath the surface armor material a chromium-zirconium copper (CuCrZr), a copper alloy with high thermal conductivity, is laid as a heat sink, and cooling water is run through it to carry the heat away.

The impurity contamination perspective is also quantitatively important. When material atoms mix into the plasma, those atoms rob it of energy through radiation and cool the plasma down. This radiation loss becomes stronger the larger the atomic number $Z$. Roughly speaking, because the radiated power of an impurity increases as a power of $Z$, high-atomic-number elements cool the plasma even in tiny amounts. For this reason, a heavy element such as tungsten must be kept below a concentration of about $10^{-5}$ in the plasma. Conversely, a light element is tolerated even if it mixes in somewhat. This difference in the “permissible contamination concentration” greatly influences material selection.

Deepen the Theory (Graduate)

Among the physical processes that determine the lifetime of the first wall, we will dig into sputtering, the erosion of the surface; neutron irradiation damage, the degradation of the bulk; and tritium retention, the uptake of fuel.

Sputtering is the phenomenon in which incident ions transfer momentum to surface atoms and knock those atoms off the surface. The number of atoms released per incident particle is called the sputtering yield. Sputtering has a threshold in the incident energy, and this threshold is determined by the surface binding energy and the mass ratio of the incident particle to the target atom. From the conservation of momentum and energy, the maximum fraction of energy that an incident particle can transfer to a target atom can be written using the mass ratio $M_1$, $M_2$ as follows.

$$\gamma = \frac{4 M_1 M_2}{(M_1 + M_2)^2}$$

Sputtering begins when this $\gamma$ multiplied by the incident energy exceeds the surface binding energy. Beryllium, a light atom, has a small surface binding energy and a low threshold of about $10$ eV, whereas for the heavy tungsten the threshold is a high $200$ to $300$ eV. In other words, if the particle energy at the plasma edge is kept low, tungsten is barely eroded. This is a major reason that makes tungsten attractive as a first-wall material.

Knocked-off atoms redeposit at another nearby location, forming a mixed layer in which multiple elements are blended. Through this repetition of deposition and re-sputtering, the composition of the wall surface changes moment by moment during operation. Predicting not only the properties of the material itself but also such dynamic surface evolution is indispensable in evaluating the lifetime of the first wall.

What governs the degradation on the bulk side is neutron irradiation damage. When a fast neutron knocks a lattice atom out of place, the amount of damage is evaluated in dpa (displacements per atom), which expresses how many times each atom has been knocked out on average. In ITER the lifetime damage to the first wall is on the order of a few dpa, but in a power reactor it is predicted to reach $100$ to $200$ dpa. The point defects and dislocation loops produced by irradiation harden the material while simultaneously making it brittle, and they push up the transition temperature at which it changes from ductile to brittle (the ductile-to-brittle transition temperature). Furthermore, when helium produced by transmutation accumulates at grain boundaries, helium embrittlement, in which the material swells and becomes brittle, also progresses. How to suppress these irradiation effects is a central issue in materials science (for details on the materials side, see Plasma-Facing Materials).

Something not to overlook when discussing the first wall is tritium retention, that is, the accumulation of tritium, the fuel, in the wall. The hydrogen isotopes coming from the plasma adsorb onto the surface of the wall material, diffuse into its interior, and become trapped and accumulate in the defects and redeposited layers created by irradiation. Because tritium is a radioactive substance, a strict safety upper limit is imposed on the amount that accumulates inside the reactor. Choosing materials that do not readily accumulate tritium and preparing the means to recover what has accumulated are decisively important on both the safety and fuel-balance fronts of the reactor. Incidentally, the blanket plays the opposite role of “breeding” tritium, producing fuel just inside the first wall (see Blanket for details).

Putting all this physics together, the conditions demanded of a first-wall material are wide-ranging: high melting point, high thermal conductivity, low sputtering yield, low activation, low tritium accumulation, and robustness against irradiation. There is no perfect material that satisfies all of them, and the history of choices about where to compromise is itself the story of the evolution of first-wall materials.

Research Frontier (PhD)

The choice of first-wall material is converging toward a single direction after a long process of trial and error. Its symbol is the flow from beryllium to tungsten.

ITER originally planned to adopt beryllium as the main surface material of the first wall. Beryllium has the advantages of being very light with an atomic number of 4, so that its radiation loss is small even when it mixes into the plasma; of having a getter effect that captures oxygen and reduces impurities; and of relatively low tritium accumulation. On the other hand, it also carried the weaknesses of a low melting point of about $1287$ degrees C, toxicity, and the resulting handling difficulties. Given this design, tungsten had been chosen from early on for the divertor, where heat concentrates.

Afterward, as research advanced, the trend to unify the first wall too around tungsten grew stronger. ITER has decided to move to a full-tungsten first wall. Tungsten has the highest melting point among metals at about $3422$ degrees C, a low sputtering yield, and less tritium accumulation than beryllium, properties that are decisively important for a power reactor. Behind the shift to all-tungsten lie the safety-management costs that accompany beryllium operation, and the judgment that, foreseeing that future power reactors will certainly be tungsten-based, it is best to concentrate expertise there. The drawbacks are that its high atomic number means its mixing into the plasma must be suppressed to the extreme, and its embrittlement under neutron irradiation. How to control these is a topic under active research today.

A research frontier looking even further ahead is the concept of the liquid metal wall, which replaces the solid wall itself with liquid metal. The idea is to flow lithium, tin, or an alloy of them as a thin film and make this the surface of the first wall. If it is a liquid, then even if the surface is eroded by sputtering, a fresh surface is always replenished, so the very concept of erosion disappears, and the problem of solids becoming brittle from irradiation damage can in principle be avoided as well. It also has the advantage of carrying heat away by convection. In research, both a method of holding the liquid metal in a capillary porous system and a method of flowing it as a film are being examined. On the other hand, many challenges remain toward practical use: the problem that magnetohydrodynamic (MHD) forces act on the conductive liquid flowing in the magnetic field and disturb the flow, the problem of the liquid metal evaporating and mixing into the plasma, and corrosion of the circulation system.

Let me list, with English terms, keywords that appear frequently when reading papers. The first wall is treated as one of the plasma-facing components (PFCs). On the materials side, sputtering yield, redeposition, tritium retention, neutron irradiation damage (dpa), helium embrittlement, and low-activation material appear frequently. Keeping in mind the design philosophy of the full-tungsten wall, and the advanced concepts of the liquid metal wall, capillary porous system, and flowing liquid lithium, will make it easier to follow the latest discussions.

Check Your Understanding

Q1. Why is the first wall called the 'first' wall?

A Because it is the first wall assembled during reactor construction B Because, seen from the plasma, it is the innermost wall that radiation and particles hit first C Because it is positioned first when counted from the outside D Because it handles the first stage of power generation

Q2. What does the unit MW/m2, used for the surface heat load, mean? Roughly how large is it on the ITER first wall?

A It expresses the total power generated across the whole wall, and in ITER it is about 500 MW B It is a unit expressing the number of neutrons passing through the wall per second C It is a unit expressing, in megawatts, the amount of heat flowing in per square meter, and in ITER it is about 0.25 to 0.5 MW/m2 D It is a unit expressing the temperature of the wall, and in ITER it is about 0.5 degrees

Q3. Why must high-atomic-number tungsten be kept to a tiny amount in the plasma?

A Because heavy atoms sink to the bottom of the plasma and hinder the reaction B Because tungsten melts immediately upon touching the plasma C Because high-atomic-number elements rob energy through strong radiation and cool the plasma down D Because tungsten absorbs neutrons and stops fusion

Q4. Give the main reasons the first-wall material moved from beryllium to tungsten.

A Because tungsten has a high melting point, a low sputtering yield, and low tritium accumulation, advantages suited to a power reactor B Because tungsten is cheaper and easier to obtain than beryllium C Because beryllium cools the plasma too much and stops the reaction D Because tungsten has a small atomic number and has little effect even when mixed into the plasma

Q5. How does the liquid metal wall try to solve the problems of the solid wall?

A By making the wall thicker to completely block neutrons B By replacing the surface with flowing liquid metal, so that even when eroded a fresh surface is replenished, and embrittlement from irradiation can also in principle be avoided C By flowing large amounts of cooling water to lower the wall temperature to near room temperature D By rotating the wall to disperse the heat from the plasma

Related Topics

• Plasma-Facing Materials
• Divertor
• Blanket