**6. Selected cases of radiation damage studies in materials relevant for exploration of the Moon**

The previous section provided an overview of computational methods that can be applied to study radiation damage in materials and discussed the advantages of combining such methods into a multiscale approach. This section mainly focuses on the effects of radiation on materials of practical use on the Moon, including several novel and promising materials. We overview the existing radiation damage studies for these novel materials, emphasizing multiscale modeling when available.

*Modeling Radiation Damage in Materials Relevant for Exploration and Settlement on the Moon DOI: http://dx.doi.org/10.5772/intechopen.102808*

## **6.1 Improving the model for solar cell degradation via a multiscale approach**

Generally, degradation of solar cells is modeled via the non-ionizing energy loss (NIEL) approach, the NIEL being the portion of energy loss per unit path length of the projectile converted into displacement damage. According to Akkerman et al. [175] (the definition used in most simulation tools), the NIEL is defined as:

$$\text{NIEL}(E) = \frac{N\_A}{A} \int\_{T\_d}^{E\_{\text{max}}} Q(E\_R) E\_R \left(\frac{d\sigma}{dE\_R}\right)\_E dE\_R = \frac{N\_A}{A} D(E), \tag{4}$$

where *E* is the total kinetic energy of the external particle, *ER* is the portion of kinetic energy which turns into displacement damage, *Q E*ð Þ*<sup>R</sup>* is the partition factor giving the fraction of kinetic energy to be lost to NIEL mechanisms, *NA* is Avogadro's number, *A* is the atomic mass of the lattice atom, *dσ=dER* is the partial differential cross section for creating a recoil atom with energy *ER*, and *D E*ð Þ is the displacement damage function. The integral runs from the minimum energy required to permanently displace an atom to a defect position, that is, the threshold displacement energy *Td*, to *E*max, which is the maximum energy transferred to a recoil atom in a particular interaction. Although the NIEL concept differs from the nuclear stopping power, as it includes also the energy loss to non-ionizing events induced by hadronic interactions, as already mentioned above (see **Table 2**), relevant non-ionizing effects are induced by particles with energies from few to few tenths of MeV, which is the regime of Coulomb interactions.

On the basis of a large set of experimental observations, it is assumed that the degradation of a semiconductor device under irradiation can be linearly correlated with the NIEL [176]. In practice, this means that the number of defects should give a measure of the damage irrespective of their distribution, whether clustered in high density in small regions (as in the case of neutron damage) or homogeneously scattered over a relatively wide volume (as in the case of the low-energy proton or *γ*-ray-induced damage) [177]. Thus, in principle, the damage produced by different particles (with different energies) should be scalable via their NIEL (i.e., the number of displacements), as indeed has been shown in several studies [176, 178–182]. The NIEL scaling is a powerful method for dealing with displacement damage predictions in complex radiation environments, such as on the Moon and in space missions in general. However, deviations from the linearity exist and seem to be associated with the "quality" of the radiation damage at the microscale as induced by different kinds of particles and (or) as influenced by intrinsic defects in the target [183].

Generally, the NIEL is calculated via MC particle transport codes, assuming amorphous target materials, a static *Td* that is constant for each element in all the materials where such element is found (thus, not considering the underlying electronic structure), and a simple linear collision cascade model for the number of final defects [184]. Several quantities in the NIEL formula and, more generally, the overall understanding of radiation damage can be strongly improved via AIMD and TDDFT+MD studies. *Td*, for example, is an important quantity that can significantly affect the NIEL [185, 186] and its accurate estimation can be accessed by AIMD simulations [187, 188]. Recent results for *Td* in semiconductors have shown that the electronic excitations can, in general, reduce their value [122], in line with previous experimental results [189]. The effect of electronic excitations consists in weakening the atomic bonds making it easier to displace an atom from its equilibrium position. On the other hand, the "heating" effect of electronic excitations has the consequence of facilitating the healing of the structure. Even a small change in *Td* affects the calculated NIEL, as can be seen in **Figure 3** showing the NIEL for a

**Figure 3.**

*NIEL for protons and electrons in GaAs for different values of the threshold displacement energy Td calculated with the online SR-NIEL tool [190].*

proton and an electron in GaAs. The NIEL is affected by the choice of *Td* in an energy window of the lunar radiation environment (see **Table 1**). These findings raise an important question about the role of electronic excitations in defect formation that deserves more attention in future works.

Another example of possible improvement in the NIEL model is a more precise calculation of the number of radiation-induced defects and of the "quality" of radiation-induced damage (which type of defects are induced). It has been observed that point-like and clustered defects contribute differently to some degradation parameters [191]. Recent MD studies [192–194] and experimental works [181, 195, 196] have proposed an effective or *adjusted* NIEL to correct the deviations from a linear dependence of degradation parameters on the NIEL and restore a linear relationship. Other MD studies [197] proposed new metrics for counting defects including the effect of a "heat spike", which leads to a much lower rate of final defects as compared to predictions from a simple linear collision cascade model as commonly used in the NIEL calculations based on MC particle transport [184, 198].

On a parallel research stream, multiscale studies in a number of materials combining MD simulations of collision cascades with the electronic stopping from TDDFT offer a more accurate description of both the number and the nature of defects created under realistic conditions. The electronic degrees of freedom and their coupling to the phonons of the target affect the cascade evolution and morphology [170, 171, 173, 174]. This is of relevance for the NIEL which includes a part of energy dissipated to phonons. This fraction depends on the energy of the impinging particle but also on the properties of the material. Some studies have shown that the direction-dependence of the electronic stopping can influence the collision cascades [118]. Other studies have demonstrated that the formation of thermal spikes and therefore of amorphous pockets is sensitive to the electronic specific heat [199] and others that the choice of the model employed for the inclusion of the electronic effects and in particular the overestimation (or underestimation) of electron-phonon coupling can have a significant influence on the number of defects created [171].

#### **6.2 Radiation effects in the next-generation lightweight photovoltaic panels**

As discussed in Section 2, HOIPs have a unique combination of properties particularly interesting for lunar exploration. The general chemical formula for perovskites is ABX3, where A and B are two metal ions with different ionic radii and X is an anion that is six coordinated to the B-site [200]. HOIPs, in particular,

*Modeling Radiation Damage in Materials Relevant for Exploration and Settlement on the Moon DOI: http://dx.doi.org/10.5772/intechopen.102808*

#### **Figure 4.**

*Structure of a HOIP: methylammonium cation (*CH3NH<sup>þ</sup> <sup>3</sup> *) occupies the central A site surrounded by 12 nearest-neighbor iodide ions in corner-sharing PbI6 octahedra [201] (available under the terms of the creative commons CC BY license).*

comprise a negatively charged lead-halide inorganic skeleton where B is a metal cation (Sn2+ or Pb2+), X is a halide anion (I�, Br�, and/or Cl�) and A is a monovalent positively charged organic cation, such as methylammonium (MA<sup>+</sup> = CH3NH3X<sup>+</sup> , where X = I, Br, Cl) or formamidinium (FA<sup>þ</sup> ¼ CH NH ð Þ<sup>2</sup> þ 2 (**Figure 4**).

Despite many advantages, several external factors, such as air, moisture [202], UV light [47, 203], heat, light soaking [204], and partially also radiation [205, 206], induce considerable structural instabilities in HOIPs. An intrinsic instability is also present, caused by a relatively weak cohesion between the organic cation and the inorganic octahedra and predominantly by the low-energy barriers for the migration of halide anions and organic cations, with halide migration being the most prevalent [201, 207–210]. Phase segregation can be induced by large-scale ion migration [211]. However, some of the challenges that HOIPs-based solar cells face on Earth, such as degradation caused by moisture, are not relevant for space applications [212]. Thermal and vacuum stability, high power-conversion efficiency, and radiation resistance are the main challenges in the space context. A sensible choice of the chemical composition, of eventual use in tandem devices [212] (which also helps to reach an efficiency of up to 30%) or incorporation of a functionalized 2D metal-organic frameworks (MOFs) [213], can improve the long-term operational stability of HOIPs.

A relevant collection of DFT studies for HOIPs can be found in Ref. [214]. A recent study based on DFT + compressed sensing-symbolic regression has shown that mitigation of the propensity of halogens to migrate could be achieved by selectively strengthening specific bonds [215]. The study also unveiled the reasons for improved stability given by specific halogens, the origin of the higher stability offered by certain organic cations compared to others, and highlighted in a quantitative and first-principles manner how weak interactions have a significant role in binding the halogens more strongly.

The study of the radiation tolerance of perovskite solar cells is an extremely active field of research. Solar cells based on HOIPs as active layers have been recently sent to space via first campaigns [60, 216]. Several ground-testing experiments have been performed mostly using protons, either with an energy of several tenths of MeV [69, 211, 217] or with an energy of 150 keV, 100 keV, and 50 keV [70, 218, 219], of less relevance for realistic space conditions.

Superior radiation resistance of perovskite solar cells in comparison to commercially available crystalline Si-based cells has been demonstrated [69]. Moreover, experiments have shown that perovskite solar cells have remarkable self-healing

**Figure 5.**

*3D scatter plots of the straggling of 68 MeV protons within the (A) HOIP/CIGS(Cu(In,Ga)Se*2*) and (B) HOIP/SHJ(Si heterojunction) tandem solar cells. The energy loss of the incident 68 MeV protons to recoils is plotted as a function of depth based on SRIM simulations with a total of* <sup>5</sup> 107 *protons. The damage of a real space environment at the orbit of the ISS is shown as a black line. Adapted from [90] (available under the terms of the Creative Commons CC-BY license).*

capabilities (at room temperature) that lower the number of defects caused by proton irradiation [69]. Another experimental study has shown that the proton irradiation effects on the physical properties of HOIPs are strongly dependent on the synthesis method [220] which appeared to affect the strength of specific chemical bonds. In particular, HOIPs, produced by mechano-chemical synthesis, have shown practically no change in their physical properties after irradiation with a high-energy 10 MeV proton beam with doses of up to 10<sup>13</sup> protons/cm<sup>2</sup> .

Recently, multi-junction tandem solar cells (combining HOIPs with previous technologies or technologies investigated in parallel) have also been studied under ion irradiation [217]. Lang et al. [217] carried out SRIM simulations of energy loss of high-energy protons as well as the energy transferred to the recoiling nuclei—a measure of the degradation of PV parameters—in tandem solar cells (**Figure 5**). The study [217] has shown that HOIP/CIGS tandem solar cells possess a high radiation hardness and retain over 85% of their initial performance even after 68 MeV proton irradiation and a dose of 2 <sup>10</sup><sup>12</sup> proton/cm<sup>2</sup> , equivalent to 50 years in space at the International Space Station (ISS) orbit.

First-principles calculations of the atomic knock-on displacement events in HOIPs have shown that such displacements are significant and highly energy-dependent [221]. The work has shown that only certain types of atoms are prone to displacements suggesting that mitigation strategies should be directed toward some chemical species more than others. Overall, further studies are necessary, but existing research proves that HOIPs-based solar cells have a remarkable potential for power generation on missions to low Earth orbit, the Moon, and beyond [62].

#### **6.3 Novel multi-principal-element alloys with enhanced radiation resistance**

#### *6.3.1 Outstanding properties of MPEA for space applications*

Another promising class of novel materials for space applications is multiprincipal element alloys (MPEAs) [222, 223], which combine superior mechanical properties and enhanced radiation resistance [224]. Also known as high-entropy alloys (HEAs) or concentrated solid-solution alloys (CSSAs), MPEAs consist of at least five principal elements with the concentration of each element from 5 to 35% *Modeling Radiation Damage in Materials Relevant for Exploration and Settlement on the Moon DOI: http://dx.doi.org/10.5772/intechopen.102808*

#### **Figure 6.**

*Atomic structure of a body-centered cubic (BCC) AlCoCrCuFeNi HEA. The Al, Fe, Co, Cr, Ni, and Cu atoms are shown in red, magenta, green, blue, cyan, and gray colors, respectively [225] (available under the https://c reativecommons.org/licenses/by-nc-sa/3.0/Creative Commons Attribution License).*

[222]. Despite the complex composition, MPEAs often form single-phase solid solutions (**Figure 6**). The interest of researchers in MPEAs has been growing exponentially in recent years, as they exhibit a paradigm shift in alloy development. MPEAs indeed combine a set of outstanding properties, such as high strength, hardness, fracture toughness, corrosion resistance, strength retention at high temperature [226], good low-temperature performance [227], and recently discovered enhanced radiation resistance, superior to conventional alloys and pure metals [149, 222, 223, 228–233]. Moreover, MPEAs have great potential as 3D printing materials [234]. MPEAs can be printed from a powder, providing manufacturing freedom for lightweight and customizable products of complex geometries for applications in the aerospace, energy, molding, tooling, and other industries, all of the great relevance for the exploration of the Moon.

Recent experiments have shown that MPEAs have a higher resistance to defect formation due to high atomic-level stress and chemical heterogeneity [235]. MPEAs also possess lower void swelling and higher phase stability [236, 237] as compared to conventional alloys. Self-healing capability is another remarkable property of MPEAs [227, 236, 238].

The subclass of lightweight (LW) MPEAs have a great potential for space applications due to their high strength-to-weight ratio [239–241]. The main components of LWMPEAs are low-density elements, such as Al, Mg, Si, and Ti [240]. The latter is of extreme importance for ISRU since 99% of the lunar soil consists of Si, Al, Ca, Fe, Mg, and Ti oxides [5, 242].

Currently, the main focus of computational studies has been on the single-phase random solid-solution (SS) alloys based on transition metals with high densities (Co, Cr, Fe, Ni) for application in radiation environments, in particular in nuclear reactors [148, 149, 232, 236, 243–245]. MD simulations of displacement cascades applied to pure metals and multicomponent alloys [150, 244–248] confirm the experimentally observed reduction of the number of defects and defect clusters in MPEAs compared to pure metals (**Figure 7**).

The electronic stopping power for a proton in binary alloys has recently been calculated using real-time TDDFT [249]. The study has shown that the electronic stopping power of binary alloys is higher than that of pure Ni, suggesting that alloys more effectively stop the incoming particles. Moreover, the inclusion of the electronic stopping into MD simulations of defect formation significantly reduces the final number of surviving defects, as shown in **Figure 8**. The inclusion of both the electron-phonon coupling and the electronic stopping in the 2T-MD model not only reduces the actual number of defects but also notably impacts their final

**Figure 7.** *The number of defects in Ni, NiFe, and NiCoCr from experiments and MD simulations [150].*

#### **Figure 8.**

*Average number of surviving defects in the classical MD cascade, MD cascade including electronic stopping force, and the 2 T-MD cascade at the end of the simulation for 50 keV Ni cascade in Ni, Ni80Fe20, and NiFe [224, 250–252].*

arrangement, namely leading to more isolated point defects and reducing the size of defect clusters in binary and ternary alloys [250–254].

The majority of MD studies focus on binary and ternary MPEAs due to the lack of force fields for alloys with more than three elements. However, some studies exist [233] on defect formation in NiCoFeCr alloy in which fewer defects have been found at the end of the displacement cascade with PKA energies from 10 to 50 keV, as compared with pure Ni. The limitations of the classical MD with force fields and the ways of solving this problem are discussed in the following.

#### *6.3.2 Machine-learning assisted materials discovery*

Classical MD with empirical potentials is the method that proved to work well for large systems and long time scales [139] for the modeling of collision cascades. However, classical interatomic potentials cannot accurately reproduce interactions between the atoms in MPEAs due to their complex structure and lattice distortions leading to internal strain [149, 255, 256]. On the other hand, *ab initio* methods relying on quantum mechanics, such as DFT, can accurately reproduce the interatomic potentials in complex structures but are limited to a small length scale.

Recent developments in machine learning (ML) approaches can provide a solution to this problem. ML-enhanced materials discovery is an emerging and extremely rapidly growing field. The combination of a precise model based on quantum mechanics and ML algorithms have the potential for an efficient and

*Modeling Radiation Damage in Materials Relevant for Exploration and Settlement on the Moon DOI: http://dx.doi.org/10.5772/intechopen.102808*

accurate description of materials properties [257–259]. Much progress has been made in recent years in the development of ML-based interatomic potentials with the input from electronic structure calculations. First applications have shown that accurate potentials can be obtained for many relevant systems [260–265]. ML-assisted calculations have been applied to pure metals, binary, ternary alloys [266, 267], and MPEAs [268–270].

ML and artificial intelligence (AI) may become powerful tools for more accurate multiscale modeling of materials properties. Artificial Neural Networks (ANN) [271] combined with atomistic KMC have already been used to describe the microstructural changes in metals and alloys induced by irradiation [272]. Machinelearned interatomic potentials have been used to study defect formation in refractory MPEAs [273]. The results confirm experimental findings, showing that the 3D migration and increased mobility of defects in MPEAs promote defect recombination leading to more efficient healing. AI, thus, can provide a bridge between different methods, such as DFT, MD, and KMC, and allow for large-scale atomistic simulations of high accuracy, which will accelerate the discovery of new advanced materials.

## **6.4 Radiation resistance of fiber-reinforced polymers and composites for habitats**

Fiber-reinforced polymers (FRPs) are composite materials made of a polymer matrix reinforced with fibers. Typical polymers that are often used include epoxy, vinyl ester, polyester thermosetting plastic, and phenol-formaldehyde resins. Typical fibers include, but are not limited to, glass, carbon, and aramid. In a composite FRP material, the polymer and fiber often have significantly different physical and/ or chemical properties, which remain separate and distinct within the finished structure but are complementary for tailored properties [274]. Because of their low density (lightweight), great moldability, specific strength, stiffness [275], excellent mechanical stability, and good thermal properties, FRPs are being increasingly used as structural materials in aerospace, automotive, marine industries, and civil infrastructures. Hence, FRPs are of great interest for many applications for lunar missions as potential structural materials [276]. Glass fibers (also "fiberglass") can be directly produced from the lunar soil as well as from by-products of metal extraction and can be used to reinforce lunar concrete [277].

The radiation environment on the Moon presents challenges for FRPs with concerns on both the immediate reactions taking place in the materials (short-term effects) and continued post-exposure degradation processes (long-term effects) [276, 278]. In the past decades, many selected FRPs have been ground-tested at different kinds of radiation and particle accelerator facilities for their potential use in space-related radiation environments, including UV-light [279, 280], *γ*-rays [281, 282], electron beams [283, 284] and proton beams [285].

Carbon-fiber composites have been widely used in aerospace industries due to their high-temperature stability and low density along with high strength, as well as superior beam-induced shock absorption [285, 286]. A combined modeling and experimental study of the radiation effect on carbon-fiber-reinforced molybdenumgraphite compound (MoGRCF) [285], including MC simulations of the energy deposited into a realistic structure by a 200-MeV proton beam (**Figure 9**) has show that carbon-fiber-reinforced composites have superior beam-induced shock absorption ability compared to that of graphite.

In the 1980s, the degradation behavior of carbon-fiber-reinforced plastic (CFRP) under electron beam irradiation in various conditions simulating experiments in space has been studied by Sonoda et al. [283]. It has been observed that

#### **Figure 9.**

*MC modeling of the energy deposition for a 200-MeV proton beam interacting with an irradiation target array (MoGRCF) in tandem with the isotope production array downstream [285] (available under the terms of the Creative Commons Attribution 3.0 License).*

there is no change in mechanical properties of CFRP when irradiated by up to a dose of 50 MGy. MC simulations of radiation effects in FRPs have shown that by adding lead nanoparticles it is possible to increase their radiation resistance [287]. According to the study, the addition of 15 wt% of lead nanoparticles to FRPs led to a mass reduction of 64% for the same level of radiation shielding.

An alternative to glass fiber for polymer reinforcement is basalt fiber which offers advantages, such as high specific mechanical and physicochemical properties, biodegradability, non-abrasive qualities, and cost-effectiveness [288]. Arnhof et al. [289] have recently studied mechanical properties of fiber-reinforced geopolymer (FRG) with basalt fiber (i.e., inorganic alumino-silicate polymer) made from lunar regolith simulant as potential shielding and structural material. As basalt fibers can be produced *in situ* at the lunar surface efficiently, they can be used widely to increase the mechanical strength of geopolymers. Overall, geopolymers are advantageous for lunar construction due to their excellent resistance to extreme temperature fluctuations and adequate shielding from radiation [290], as well as enhanced mechanical properties over conventional cement [291].

The additive-manufacturing (AM) techniques for lunar construction from regolith, including FRP materials, and their suitability for ISRU has recently been reviewed in Refs. [292, 293]. The AM techniques for lunar construction include Cement Contour Crafting (CCC), Binder Jetting (BJ), Selective Solar Light Sintering (SSLS) and Selective Laser Sintering/Melting (SLS/SLM) for 3D printing and metal melting, Stereolithography/Digital Light Processing (SLA/DLP), among others. CCC and BJ technologies could be used for outdoor lunar civil engineering. SSLS could be applied to both direct compacting of lunar regolith to ceramic parts and 3D printing. SLA/DLP-based methods could be used for the indoor manufacturing of ceramic instruments, providing higher precision and printing quality and lower defect rate of the printed parts than other AM methods. In the last decade, studies have clearly shown that the 3D-printing technologies will become one of the cornerstones of lunar exploration, providing future astronauts with all the necessary infrastructure [293].

Lunar concrete consisting of mined regolith with the addition of glass fibers (also made *in situ* from regolith containing plenty of silicates) has a high strengthto-weight ratio and can be easily 3D-printed, as tests on lunar regolith simulant have shown [294]. Other studies have shown the promising properties of urea from *Modeling Radiation Damage in Materials Relevant for Exploration and Settlement on the Moon DOI: http://dx.doi.org/10.5772/intechopen.102808*

astronauts' urine as a superplasticizer for lunar geopolymers for 3D-printing applications. The use of urea is expected to reduce the necessary amount of water by about 30% [295].

It is worth mentioning that the 4D printing of a "smart material" with FRPs that responds to radiation-induced damages and aging in a programmable way could be realized in near future [296, 297]. In addition to experiments on the radiation environment in a lab, multiscale computational simulations as mentioned above could be helpful for gaining further insights into the radiation-induced molecular changes occurring in polymers.
