betaFeedback

WorksPhysics and Astronomy

Froude number scaling unifies impact trajectories into granular media across gravitational conditions

Abstract

The interactions of solid objects with granular media is countered by a resistance force that stems from frictional forces between the grains and the media's resistance to inertia imposed by the intruder. Earlier theories of granular intrusion have suggested an additive contribution of these two families of forces and had tremendous success in predicting resistive forces on arbitrary shaped objects. However, it remains unclear how these forces are influenced by gravitational conditions. We examine the role of gravity on surface impact behavior into cohesionless granular media using hundreds of soft-sphere discrete element simulations, we demonstrate that the outcome of impacts remain qualitatively similar under varying gravitational conditions if initial velocities are scaled with the Froude number, suggesting an underlying law. Using theoretical arguments, we provide reasoning for the observed universality and show that there is a hidden dependency in resistive forces into granular media on Froude number. Following the theoretical framework, we show that Froude number scaling precisely collapses impact trajectories across gravitational conditions, setting the foundation for explorations in granular behavior beyond Earth.

Main

Granular materials are ubiquitous on Earth, taking the form of soil, sand, gravel, pebbles, and boulders. They are also prevalent beyond our planet, present on the surfaces of moons, asteroids, and other planets, exhibiting remarkable diversity in size, shape, and dispersity [1, 2, 3, 4, 5, 6]. The interest in exploration and exploitation of non-terrestrial bodies, often strewn with this granular media, has considerably increased during the last decade for scientific investigations and economic reasons [7]. However, any process that involves interaction with granular media will need to overcome the challenges of understanding its wide range of complex mechanical behaviors. This includes history-dependent strength [8], flow anisotropy and non-associated flow rules [9, 10, 11], and nonlinear and non-local behavior [12, 13, 14]. The extent to which a granular media exhibits these characteristics is intimately dependent upon its physical properties as well as the conditions of the surrounding environment. Morphological and environmental conditions such as packing fraction [15], porosity, and cohesional forces [16, 17] influence the rheology and flow properties of grains. Meanwhile a particularly important factor in the response of grains on non-terrestrial surfaces is the fact that they experience a different gravitational force compared to grains Earth. This is expected to strongly influence the behavior and rheology of granular media since their strength is proportional to the confining pressure which in turn scales with gravity. Understanding the role of gravity in granular behavior is key with creating a clearer picture of granular mechanics beyond Earth.

The interaction between solid objects and granular media is a common feature of both natural and human-made processes. This general interaction is referred to as the ‘intrusion problem.’ Excavation [18], animal locomotion [19], vehicular locomotion [20], landing apparatus [21], the study of craters [22], and probing missions on non-terrestrial surfaces [23] are only a few examples. Granular media is complex, though, in that it can exhibit a wide range of behavior depending on its properties and local environment. Packed granular media may behave like a solid body, while agitated granular media may demonstrate fluid-like or even gaseous behavior [24]. All of these states may emerge during a surface intrusion scenario. Of particular interest in this work are the low-velocity oblique impacts, resulting from both natural and human-made causes. Impacts on asteroids and other low-gravity surfaces can create gravitational-bound ejecta and debris that re-impact the surface at lower velocities. The media’s inertial resistance becomes prevalent in resisting surface intrusion during impacts. It remains an open question how the quasi-static strength and inertial resistance of the media evolve when gravitational condition changes. Analyses of grain-grain level data on particles interacting with the intruder can reveal how granular arrangement and its variations resist in the intrusion process [25]. However, a more accessible approach is the study of the impactor’s response and linking its behavior to variations in the environmental conditions. Therefore, we examine how granular media resists the intrusion process and how the trajectory of the intruder evolves by creating normalized behavior maps to parse out the role of gravity.

Parameterization of gravity

Experimental evaluations of intrusion problems under different gravitational conditions are challenging and costly, requiring on-board parabolic flights or drop-towers to override Earth’s gravity. However, these experiments are hindered by factors such as variations in apparent gravity during parabolic flights and size limitations for drop tower experiments. Alternatively, computational methods and theoretical approaches can provide significant insights into the dynamics of granular media in other gravities. For this purpose, we employ a Discrete Element Method (DEM) environment in which the granular medium is idealized as a collection of round cohesionless particles bound by gravity and Hertizan contact forces (See methods for details). To hone in on the effect of gravity, we adopt this framework which allows for changing gravity as an independent parameter. The size, shape, coefficient of friction, material properties, and densities of the impactor and grains are key factors in the surface intrusion problem. In addition, the velocity and launch angle of the projectile will affect the granular response. To isolate the effects of gravity, we keep all of these properties constant except for the initial speed vv of the projectile which is scaled with gravity based on Froude number, Fr=vgr\mathrm{Fr} = \frac{v}{\sqrt{g r}} to preserve the ratio of inertial forces to gravitational forces. All other parameters and morphological properties of the granular media remain unaffected. While Froude number has historical applications in fluid dynamics [26, 27, 28, 29] and gait analysis [30], it applies here because it reports the ratio of inertial to frictional (and therefore gravitational) forces in a system. In parallel to DEM simulations, we use a theoretical model known as the Dynamic Resistive Force Theory (DRFT) [31] to explain our observations and ultimately, present a unified scaling law for low velocity impacts into cohesionless granular media in myriad gravitational conditions.

Impact behavior across gravities

The DEM environment models an array of surface impacts by an impactor of diameter DD in Earth gravity (g=9.81 m/s2g = 9.81\ \mathrm{m/s^2}), Moon gravity (1.62 m/s21.62\ \mathrm{m/s^2}), and asteroid Bennu-like gravity (6.27×10−5 m/s26.27 \times10^{-5}\ \mathrm{m/s^2}). The baseline DEM setup is shown in Fig. 1a. Focusing only on the projectile’s response to a surface impact as a holistic representation of the granular response, three cases may result as initially defined by Wright et al. [32, 33]. The impactor may ricochet and bounce from the impact site, roll out of the initial impact crater whilst remaining in contact with the surface, or fully stop in the crater as shown in Fig. 1b. A velocity range of 1 m/s to 7 m/s for Earth’s gravity translates to a Froude number range of 4 to 25 which is kept constant for all gravitational environments. We vary the launch angle between 20° and 70° resulting in an array of 143 simulation cases for each gravitational condition.

The variations in grain arrangement that develop naturally during settling of grains can result in variations in local packing fraction [25]. In the case of an impact, this condition can significantly influence projectile trajectory, depending on the projectile-to-grain size ratio of the system [33]. In this DEM work, the grains are coarse relative to the impactor with a ratio of about 6.7, meaning we expect trajectory influence due to grain configuration. We demonstrate this variation in the inset of Fig. 1a. To account for these variations in the impact studies and to obtain smoother data for the behavior maps, we perform two sets of impact models in each gravity which differ only in the initial placement of the impactor. We use the same arrangement of the settled bed in the repeated set to minimize the additional variations due to local packing. We also use the same settled granular bed obtained under Earth gravity, but depressurized to use for other gravities to maintain packing fraction between cases.

Impact setup and impact outcomes

Figure 1. (1a) shows conventions for launch velocity and angle. The inset shows variability in path of the projectile when launched from different positions. (1b) shows impact outcomes of ricochets (top), roll-outs (middle), and full-stops (bottom) with colored vectors showing the velocity of the grains and projectile.

We observe remarkable agreement in numerical behavior maps across gravitational conditions as shown in Figure 2a–Figure 2c, which collectively present the outcome of 858 impact simulations. We note that the horizontal intercept of the roll-out boundary region occurs between Froude numbers 5 and 10 and the vertical intercept occurs at or just below 50° for all behavior maps. These maps corroborate experimental observations by Wright et al. [34] that the ricochet behavior concentrates towards the lower right side of the plot and full-stops in the upper left with the roll-outs observed as a transitional zone in-between. We find similar percentage of roll-out to ricochet region for Earth, Moon, and Bennu gravities as 31.5%, 32.9%, and 36.4%, respectively. There is a trend in these measurements with gravity, but this can be attributed to the marginal decrease in packing fraction resulting from the adjustment of Earth’s grain arrangement to accommodate other gravitational conditions.

Behavior maps produced with DEM simulations for three gravitational conditions

Figure 2. (a) Earth (9.81 m/s29.81\ \mathrm{m/s^2}) (b) Moon (1.62 m/s21.62\ \mathrm{m/s^2}) (c) Bennu (6.27×10−5 m/s26.27 \times10^{-5}\ \mathrm{m/s^2}). Behavior maps produced with DEM simulations for three gravitational conditions. Each map is the averaged from four sets of classifications.

The extensive range of gravitational conditions covered in the DEM results highlights the effectiveness of Froude number scaling in unifying the impactor response. This raises the intriguing question of why such a high level of universality exists, considering the inherent complexities of granular media and orders of magnitude change in gravity. We propose an explanation by using a reduced-order theory for intrusion into granular media. Furthermore, this theory offers a robust approach to investigating the intrusion problem from a mean-field perspective, effectively mitigating the influence of the discrete nature of the media that is present in DEM simulations. By eliminating variations stemming from this source of uncertainty, a clearer understanding of the impactor’s response is obtained.

A gravity-normalized theory for intrusion

Inspired by earlier studies on micro-swimming organisms in viscous fluids [35], granular Resistive Force Theory (RFT) was proposed to provide a set of rules to calculate the forces arising from the interaction between solid objects and granular media [36]. In RFT-type models, the granular media is considered a resistant medium that generates forces opposing the motion of intruding objects. This theory, although empirically derived, has demonstrated remarkable effectiveness in predicting resistive forces arising in interactions with granular media in non-inertial regimes [37, 38, 39]. Its mechanical foundation has been elucidated by establishing connections between theory and measurable physical properties of grains both quasi-static, non-inertial regimes [40]. To extend its applicability beyond the quasi-static regime, Dynamic Resistive Force Theory was introduced [31], which is expressed in two dimensions as follows.

τx,z=αx,z(β,γ)H(z)z+λρvn2nx,z(1)\tau_{x,z} = \alpha_{x,z}(\beta,\gamma) H(z) z + \lambda\rho v_n^2 n_{x,z} \tag*{(1)}

Consider a 2D space filled with grains defined by the horizontal direction xx and vertical direction zz, the latter aligned with gravity and pointing downward. When an object of arbitrary geometry ‘intrudes’ into this space, any segment of its perimeter moving into the granular media will experience a resistive traction τx,z\tau_{x,z} where subscripts denote the traction components. Equation (1) includes both the quasi-static and inertial components of the resistance to motion. The quasi-static resistance arises from friction between the grains and depends on the orientation angle β\beta, velocity angle γ\gamma, and depth zz (pictured in Figure 3). This term is determined by empirical resistive force data that form a vector-valued quantity αx,z(β,γ)\alpha_{x,z}(\beta,\gamma) with units of traction per depth (see Supplements for details). A Heaviside function H(z)H(z) ensures segments above the surface do not experience a force. The inertial resistance is calculated using a fitting parameter λ\lambda, bulk density of the granular media ρ\rho, and has a quadratic dependence on the speed vnv_n in the normal direction of the segment defined by nn. By integrating over all infinitesimal areas while excluding areas moving away from the media, the resistive force on the entire object can be obtained.

Conventions for orientation angle $\beta$ and velocity direction angle $\gamma$ of a plate intruding at depth $z$ through a resistive media

Figure 3. Conventions for orientation angle β\beta and velocity direction angle γ\gamma of a plate intruding at depth zz through a resistive media. These quantities provide the basis for the quasi-static formulation of RFT.

While the weight’s dependence on acceleration of gravity is straightforward, additional considerations are required to use Equations (1) in different gravitational conditions. Frictional forces are influenced by pressure, which is influenced by both gravity and depth [41, 42]. While Equation 1 explicitly includes the depth variable zz, the dependence on gravity is implicitly incorporated into αx,z\alpha_{x,z}. Through theoretical and continuum understanding of RFT, we recognize that αx,z\alpha_{x,z} originates from measurable quantities of the granular media and can be reproduced if mass density and coefficient of friction of the grains are known [40]. Thus, to remove gravity dependence from αx,z\alpha_{x,z}, we can envision an Earth gravity-normalized αx,z0\alpha^0_{x,z} defined as αx,z/g\alpha_{x,z}/g.

The inertial term is not dependent on gravity and λ\lambda may be treated as gravity-independent. In addition, preserving packing fraction keeps bulk density constant as gravity changes. In order to make the second term gravity explicit, we rewrite it in terms of Froude number by vn2=grFr2v_n^2 = gr\mathrm{Fr}^2. Combining our new definitions for quasi-static and inertial terms, and dividing the results by Earth gravity, we find a gravity-normalized traction written as

τx,z0=τx,zg=αx,z0(β,γ)H(z)∣z∣+λρrFr2nx,z.(2)\tau^0_{x,z} = \frac{\tau_{x,z}}{g} = \alpha^0_{x,z}(\beta,\gamma)H(z)|z| + \lambda\rho r\mathrm{Fr}^2 n_{x,z}. \tag*{(2)}

This equation shows that, the normalized tractions remain constant in an arbitrary gravitational condition g∗g^* if the Froude number matches between these conditions. Equation (2) coupled with an initial Froude number, an expression for the area in contact with the granular media, and the mass of the impactor are sufficient to form the Equations of Motion (EOM) for the impactor. We can then solve the EOM in an arbitrary gravitational condition to produces an independent theoretical analysis.

Universal behavior and vertical impacts

We can consider a simpler case of intrusion in which the intruder vertically approaches the surface similar to common procedures for development of vertical force laws in Earth gravity [43?, 44]. The purpose of this exercise is to demonstrate how Equation (2) can be used to study impact trajectories in different gravities. Placing a flat plate initially at the surface, and imposing an initial Froude number and driving it downwards into the media, we can write the EOM for Earth and non-Earth accelerations of gravity (gg and g∗g^*, respectively) as

(αz0gz+λρz˙2)A−mg=mz¨(3)(\alpha^0_z g z + \lambda\rho\dot{z}^{2})A - mg = m\ddot{z} \tag*{(3)}
(αz0g∗z∗+λρz˙∗2)A−mg∗=mz¨∗(4)(\alpha^0_z g^* z^* + \lambda\rho\dot{z}^{*2})A - mg^* = m\ddot{z}^* \tag*{(4)}

where zz and z∗z^* represent the position the plate as a function of time tt in Earth and non-Earth conditions and dot notation denotes time derivatives. While the expressions zz and z∗z^* cannot be obtained exactly due to the nonlinearities in the EOM, we can consider an unknown function zz which is the solution to Equation (3). Using a change of variable for time as t=t∗g∗/gt = t^*\sqrt{g^*/g}, Equation (4) takes the same form as Equation (3) with t∗t^* being the new measure of time (see Supplementary Materials for details). With the initial velocity scaled as v0∗=v0g∗/gv_0^* = v_0\sqrt{g^*/g} so that the initial Froude number is the same, we can show that z(t)=z∗(t∗)z(t) = z^*(t^*). This means both plates will pass the same points in vertical space but the plate in lower gravity will require more time to cover that same trajectory. This finding from DRFT is consistent with similar conclusions found using other phenomenological models for granular behavior [41, 45, 46].

Universal behavior and oblique impacts

The EOMs become increasingly complex for the full problem of a round impactor in oblique impact since the area in contact depends on depth and angles β\beta and γ\gamma evolve due to the rotation of the projectile. However, they can be solved numerically once inputs for the DRFT Equation (2) are set to closely match the discrete model as detailed in the supplementary materials. We generate trajectories and behavior maps based on the theoretical equations and use the same bounds for Froude number and launch angle but at smaller increments.

The theoretical solution reveals that the trajectory for a pair of Froude number and launch angle is identical between gravities in oblique impact. This is showcased in Figure 4a for a 45° launch at a Froude number of 24.87 which is expected to produce a ricochet based on our DEM results in Figure 2. This equates to 7.00 m/s initial speed for Earth and 2.84 m/s, and 1.77 cm/s for Moon, and Bennu-like gravities, respectively. The emergence of the same trajectory results in a perfect collapse of theoretical behavior maps shown in Figure 4b which is in agreement with DEM maps of Figure 2. This agreement is achieved by the force law of Equation (2) and rectifies the need to use different pre-maximum and post-maximum force law as done in [32].

Ricochet trajectories for Bennu, Moon, and Earth conditions and a DRFT behavior map

Figure 4. (4a) shows ricochet events modeled at the same Froude number and launch angle in Bennu, Moon, and Earth conditions. (4b) presents a behavior map constructed with the DRFT model. Black represents full-stop, orange represents roll-outs, and white represents ricochets.

There are three distinguishable stages in the trajectories shown in Figure 4a; (i) the projectile motion prior to contacting the resistive media, (ii) the period of interaction with the media, and (iii) the post-impact projectile motion. Considering a round impactor launched with initial velocity v0v_0 and a launch angle of θ\theta in a two dimensional

space defined by xx and zz, with zz pointing down in the direction of gravity, the EOM for the center of mass in Earth gravity is x=v0cos⁡(θ)tx=v_{0}\cos(\theta)t and z=v0sin⁡(θ)t+1/2gt2z=v_{0}\sin(\theta)t+1/2gt^{2} in the xz plane. If acceleration of gravity changes to g∗g^{*} but the launch angle θ\theta is kept the same, a new initial velocity v0∗=v0g∗/gv_{0}^{*}=v_{0}\sqrt{g^{*}/g} should be considered to keep the initial Froude number constant. In the other gravity g∗g^{*}, the EOM becomes x∗=v0∗cos⁡(θ)tx^{*}=v_{0}^{*}\cos(\theta)t and z∗=v0∗sin⁡(θ)t+1/2g∗t2z^{*}=v_{0}^{*}\sin(\theta)t+1/2g^{*}t^{2}. If we use a linear scaling of time as t∗=g/g∗tt^{*}=\sqrt{g/g^{*}}t similar to the vertical intrusion case in the EOM for g∗g^{*} environment, we find that the EOM in g∗g^{*} condition reduce to Earth’s EOM and then produce the equalities x∗(t∗)=x(t)x^{*}(t^{*})=x(t) and z∗(t∗)=z(t)z^{*}(t^{*})=z(t). This results in similar findings to the vertical intrusion case where the spatial response coincides but the temporal domain follows a scaling of t∗=tg/g∗t^{*}=t\sqrt{g/g^{*}}. Therefore, the universality of the trajectory is proven for stage (i). Note that Froude number is not constant during the stage (i) due to gravitational acceleration, but the evolution of Froude number remains identical between gravities.

Based on our arguments for stage (i), the same point of the impactor will make the first contact across gravitational conditions to initiate stage (ii). This will happen at different times tt and t∗t^{*} as illustrated in Figure 5. The gravity-normalized traction τx,z0\tau_{x,z}^{0} experienced by that point PP will also be the same according to Equation (2). If we consider a consecutive frame in time different by dtdt in Earth’s gravity and dt∗dt^{*} in non-Earth, the principle of impulse and momentum between these two frames in Earth’s gravity states that

Illustration of two projectiles in different gravities

Figure 5. an illustration of two projectiles in different gravities gg and g∗g^{*}. The frame at t1/t1∗t_{1}/t_{1}^{*} is the first point of contact whereas t2/t2∗t_{2}/t_{2}^{*} occur further into the intrusion. This figure accompanies the theoretical discussion that shows how Froude number evolves identically across gravitation conditions.

dt∗∫Sτx,z0g∗dA=(Fr[t∗+dt∗]u⃗2∗−Fr[t∗]u⃗1∗)g∗r(5)dt^{*}\int_{S}\tau_{x,z}^{0}g^{*}dA=\left(\mathrm{Fr}[t^{*}+dt^{*}]\vec{u}_{2}^{*}-\mathrm{Fr}[t^{*}]\vec{u}_{1}^{*}\right)\sqrt{g^{*}r} \tag*{(5)}

and for Earth gravity as

dt∫Sτx,z0g dA=(Fr[t+dt]u⃗2−Fr[t]u⃗1)gr(6)dt\int_{S}\tau_{x,z}^{0}g\,dA=\left(\mathrm{Fr}[t+dt]\vec{u}_{2}-\mathrm{Fr}[t]\vec{u}_{1}\right)\sqrt{gr} \tag*{(6)}

Considering the same scaling in time as t∗=tg/g∗t^{*}=t\sqrt{g/g^{*}} and substituting it in Equation (5), we obtain

dt∫Sτx,zgdA=(Fr[t∗+dt∗]u⃗2∗−Fr[t∗]u⃗1)gr.(7)\mathrm{d}t\int_{S}\tau_{x,z}g\mathrm{d}A=\left(F r_{[t^{*}+\mathrm{d}t^{*}]}\vec{u}_{2}^{*}-F r_{[t^{*}]}\vec{u}_{1}\right)\sqrt{gr}. \tag*{(7)}

We know that Fr[t∗]u⃗1=Fr[t]u⃗1F r_{[t^{*}]}\vec{u}_{1}=F r_{[t]}\vec{u}_{1} based on the arguments of stage (i) and the left hand side of Equations (7) and (6) are the same. Therefore, we can deduce that Fr[t∗+dt∗]u⃗2∗=Fr[t+dt]u⃗2F r_{[t^{*}+\mathrm{d}t^{*}]}\vec{u}_{2}^{*}=F r_{[t+\mathrm{d}t]}\vec{u}_{2}. This means that Fr[t∗+dt∗]=Fr[t+dt]F r_{[t^{*}+\mathrm{d}t^{*}]}=F r_{[t+\mathrm{d}t]} and u⃗2∗=u⃗2\vec{u}_{2}^{*}=\vec{u}_{2}, and furthermore that the surface SS must also evolve the same way between the two time frames. Any subsequent time frame is subject to the same argument. As a result, the evolution of Froude number, direction of velocity, and the contact surface area will be the same throughout stage (ii), across gravitational conditions but over different time scales. This scaling also applies to the angular velocity of the impactor that develops due to the interaction with the media. Therefore, not only will the trajectories match, but the rotations will also match between gravities. Since the state of motion at the departure of stage (ii) will be the same across gravities, it then follows via the arguments for stage (i) that trajectories throughout stage (iii) must also coincide.

Discussion

The collective results from DEM simulations accompanied by the theoretical explanation provide a robust case and firm reasoning for the effectiveness of Froude number scaling in impacts and interactions with cohesionless granular media across gravitational conditions. While DEM results show a near-complete collapse of behavior maps, the theoretical explanation uses a non-discrete fundamental description of granular intrusion which is wholly independent from the DEM approach and free from the variabilities inherent in a discrete model. The theoretical arguments unveil how and why Froude number scaling works, and how it spatially collapses all stages of a projectile’s trajectory across gravitational conditions. Furthermore, it is shown that the temporal domains across gravities can be related to each other with a simple scale factor derived from Froude number.

Our findings of unified behavior stem from assumptions that the media is cohesionless and that the compaction level of the media at different gravities is the same as Earth’s compaction, the latter assisting in the isolation of gravity as a sole source of influence. Further considerations should be made in the presence of cohesion and various compactions as these conditions may be more representative of the physical conditions on non-terrestrial surfaces [16, 47]. Nevertheless, our study contributes to the fundamental understanding of gravity’s role in granular impacts and paves the way for exploring this phenomenon in three dimensions and in the presence of more complex and representative granular characteristics such as cohesion, complex grain geometries and size distributions, or variable packing fractions.

Methods

LAMMPS discrete simulation

LAMMPS is an open-source molecular dynamics solver with a broad range of applications from large-scale to nano-scale. For granular media, it is a platform for discrete so-called soft-sphere discrete element method simulations where each grain is modeled as a sphere and interactions with its neighbors are solved by applying force proportional to the level of overclosure between them.

The models in this work use a ‘hertz/material’ definition for solving interactions. For two interacting particles, identified by ii and jj, the normal elastic component of the force (FneF_{ne}) between them is computed in terms of the effective elastic modulus EeffE_{eff}, the effective particle radius ReffR_{eff}, and the overlap between the two particles δij\delta_{ij} by the expression

Fne=43EeffReff1/2δij3/2n(8)F_{ne} = \frac{4}{3} E_{eff} R_{eff}^{1/2} \delta_{ij}^{3/2} \mathbf{n} \tag*{(8)}

A damping coefficient based on work by Tsuji et al. [48] is computed as a function of the coefficient of restitution. Finally, for tangential response, a Coulombic friction model is imposed that is based on a user-input friction coefficient. The details of the values selected for these properties as well as the construction of the impact model are discussed in the next section.

Numerical implementation of DRFT

To solve an impact with idealized conditions of DRFT, Eqn. (2) must be solved. An exact solution is not possible, as the velocity-squared dependence in the inertial component creates a non-linear differential equation. A numerical solution can be achieved, though, by discretizing the circular projectile into a number of linear segments and incrementally solving for its path through the resistive medium. These increments are governed by a fixed time step and for each step the projectile’s overall resistive forces, gravitational forces, and inertia are balanced.

The spatial discretization of the projectile’s surface and the temporal discretization of the solution invite potential sources of error to the model. Circle discretization was tested between 10 and 100 segments, yet model response was observed to saturate at greater above 20 segments. 20 segments therefore became the model’s baseline, though it is notable that changing the number of segments only ever minorly influenced results as the DRFT solution further discretizes segments if they are only partially submerged in the granular surface. Timesteps between 1e-3s and 1e-7s were tested, and the model was observed to saturate above 1e-5s which was then also set as baseline. Overall wall time of the model is extremely brief and with the baseline setup solving for an impact that takes place over 0.1s, the wall time was less than 1s.

Supplementary information

The included Supplementary Information is a single PDF file that provides further details needed for the reproducibility of the results of this work.

Acknowledgments

We acknowledge the financial support from NASA grant number 80NSSC21K0143 and the support from the University of Rochester’s intramural grant. We are also thankful to Mokin Lee for his contributions to this study.

Declarations

The authors have no conflicts of interest to declare.

Declarations

The authors have no conflicts of interest to declare.

Author contributions

H.A. designed and supervised the project. P.M. and E.T. performed and analyzed the simulations. P.M. performed the theoretical analyses and implementations under guidance of E.W., P.S., R.G., A.Q., and H.A. All authors contributed to the discussions and results of the manuscript.

This figure "BennuEB_proto.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "Bennu_maps.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "DRFT_ex.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "DRFT_map.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "DRFT_map_stretch.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "E1scatter_proto.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "E2scatter_proto.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "EarthEB_proto.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "Earth_map_v2.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "Earth_map_v2_stretch.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "Earth_map_v3.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "Earth_maps.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "MoonEB_proto.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

Figure 6 (PDF p. 30).

This figure "Moon_maps.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "RFP_3mm_CP.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "ab_fig_mod.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

This figure "alpha_values.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

Schematic of particle motion and surface deformation

Figure 1 (PDF p. 35).

Diagram showing a red incident arrow at angle $\theta$ striking a granular surface, with particle-flow panels and an inset

Figure 7 (PDF p. 36).

This figure "bg_convention.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

Comparison of Bennu-g, Moon-g, and Earth-g

Figure 8 (PDF p. 38).

Figure 1 (PDF p. 39).

This figure "comparison_v1.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

Figure 1 (PDF p. 41).

This figure "convention_main.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

Schematic of a filament interacting with a surface at angle $\theta$ and velocity $V$, with three panels showing successive configurations

Figure 9 (PDF p. 43).

This figure "drft_imp_main.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

This figure "sample5.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

Unnumbered figure

Figure 10 (PDF p. 46).

This figure "sample6.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "Notes_230516_160910.jpg" is available in "jpg" format from: http://arxiv.org/ps/2307.10998v1

This figure "example_assembled.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

Three parallel surfaces with particle trajectories and coordinate axes

Figure 1 (PDF p. 52).

Bennu (g = 6.27E-05 m/s²), Moon (g = 1.62 m/s²), and Earth (g = 9.81 m/s²) trajectories

Figure 11 (PDF p. 53).

This figure "setup.png" is available in "png" format from: http://arxiv.org/ps/2307.10998v1

This figure "setup_v2.jpg" is available in "jpg" format from:

http://arxiv.org/ps/2307.10998v1

This figure "timescaling.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

This figure "universal_maps.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

This figure "variability.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

This figure "variability_v2.png" is available in "png" format from:

http://arxiv.org/ps/2307.10998v1

References

  1. [1]Walsh, K. J. et al. Near-zero cohesion and loose packing of bennu’s near subsurface revealed by spacecraft contact. Science advances 8, eabm6229 (2022).DOI
  2. [2]Sugita, S. et al. The geomorphology, color, and thermal properties of Ryugu: Implications for parent-body processes. Science 364 (2019).DOI
  3. [3]Walsh, K. J. et al. Craters, boulders and regolith of (101955) Bennu indicative of an old and dynamic surface. Nature Geoscience (2019). URL http://dx.doi.org/10.1038/s41561-019-0326-6.
  4. [4]DellaGiustina, D. et al. Properties of rubble-pile asteroid (101955) bennu from osiris-rex imaging and thermal analysis. Nature Astronomy 3, 341–351 (2019).DOI
  5. [5]Rozitis, B. et al. Asteroid (101955) bennu’s weak boulders and thermally anomalous equator. Science Advances 6 (2020). URL https://doi.org/10.1126/sciadv.abc3699.
  6. [6]Michikami, T. & Hagermann, A. Boulder sizes and shapes on asteroids: a comparative study of Eros, Itokawa and Ryugu. Icarus 357, 114282 (2021).DOI
  7. [7]Galache, J., Graps, A. et al. Asime 2016 white paper: Answers to questions from the asteroid miners EPSC2017, 985 (2017).
  8. [8]Schofield, A. N. & Wroth, P. Critical state soil mechanics Vol. 310 (McGraw-hill London, 1968).
  9. [9]Weinhart, T., Hartkamp, R., Thornton, A. R. & Luding, S. Coarse-grained local and objective continuum description of three-dimensional granular flows down an inclined surface. Physics of fluids 25, 070605 (2013).DOI
  10. [10]Sulem, J. & Vardoulakis, I. Bifurcation analysis in geomechanics (CRC Press, 1995).
  11. [11]Nedderman, R. M. et al. Statics and kinematics of granular materials Vol. 352 (Cambridge University Press Cambridge, 1992).DOI
  12. [12]Da Cruz, F., Emam, S., Prochnow, M., Roux, J.-N. & Chevoir, F. Rheophysics of dense granular materials: Discrete simulation of plane shear flows. Physical Review E 72, 021309 (2005).DOI
  13. [13]Jop, P., Forterre, Y. & Pouliquen, O. A constitutive law for dense granular flows. Nature 441, 727–730 (2006).DOI
  14. [14]Henann, D. L. & Kamrin, K. A predictive, size-dependent continuum model for dense granular flows. Proceedings of the National Academy of Sciences 110, 6730–6735 (2013).DOI
  15. [15]Vo, T. T., Nezamabadi, S., Mutabaruka, P., Delenne, J.-Y. & Radjai, F. Additive rheology of complex granular flows. Nature communications 11, 1476 (2020).DOI
  16. [16]Scheeres, D. J., Hartzell, C. M., Sanchez, P. & Swift, M. Scaling forces to asteroid surfaces: The role of cohesion. Icarus 210, 968–984 (2010).DOI
  17. [17]Mandal, S., Nicolas, M. & Pouliquen, O. Insights into the rheology of cohesive granular media. Proceedings of the National Academy of Sciences 117, 8366–8373 (2020).DOI
  18. [18]Dorgan, K. M. The biomechanics of burrowing and boring. Journal of Experimental Biology 218, 176–183 (2015).DOI
  19. [19]Hosoi, A. & Goldman, D. I. Beneath our feet: strategies for locomotion in granular media. Annu. Rev. Fluid Mech 47, 431–453 (2015).DOI
  20. [20]Thoosen, A. et al. Revisiting scaling laws for robotic mobility in granular media. IEEE Robotics and Automation Letters 5, 1319–1325 (2020).
  21. [21]Ulamec, S. et al. Landing on small bodies: From the rosetta lander to mascot and beyond. Acta Astronautica 93, 460–466 (2014).
  22. [22]Neukum, G. & Ivanov, B. A. Crater size distributions and impact probabilities on earth from lunar, terrestrial planeta, and asteroid cratering data. Hazards due to Comets and Asteroids 359–416 (1994).
  23. [23]Walsh, K. J. et al. Near-zero cohesion and loose packing of bennu’s near subsurface revealed by spacecraft contact. Science Advances 8, eabm6229 (2022). URL https://www.science.org/doi/abs/10.1126/sciadv.abm6229.
  24. [24]Jaeger, H. M., Nagel, S. R. & Behringer, R. P. Granular solids, liquids, and gases. Rev. Mod. Phys. 68, 1259–1273 (1996). URL https://link.aps.org/doi/10.1103/RevModPhys.68.1259.DOI
  25. [25]Miklavčič, P. M., Sánchez, P., Wright, E., Quillen, A. C. & Askari, H. Sub-surface granular dynamics in the context of oblique, low-velocity impacts into angular granular media. Icarus 115089 (2022).DOI
  26. [26]Ogilvie, T. F. Nonlinear high-froude-number free-surface problems (1967).DOI
  27. [27]Tahara, Y., Stern, F. & Rosen, B. An interactive approach for calculating ship boundary layers and wakes for nonzero froude number. Journal of Computational Physics 98, 33–53 (1992).DOI
  28. [28]Tahara, Y. & Stern, F. A large-domain approach for calculating ship boundary layers and wakes for nonzero froude number. Journal of Computational Physics 127, 398–411 (1996).
  29. [29]He, J. et al. Froude number, hull shape, and convergence of integral representation of ship waves. European Journal of Mechanics-B/Fluids 78, 216–229 (2019).DOI
  30. [30]Alexander, R. M. Estimates of speeds of dinosaurs. Nature 261, 129–130 (1976).DOI
  31. [31]Agarwal, S. et al. Modeling of the interaction of rigid wheels with dry granular media. Journal of Terramechanics 85, 1–14 (2019).arxiv.org/abs/1901.10667
  32. [32]Wright, E. et al. Ricochets on asteroids: Experimental study of low velocity grazing impacts into granular media. Icarus 351, 113963 (2020). URL https://www.sciencedirect.com/science/article/pii/S001910352030333X.DOI
  33. [33]Wright, E. et al. Ricochets on asteroids ii: Sensitivity of laboratory experiments of low velocity grazing impacts on substrate grain size. Icarus 376, 114868 (2022).DOI
  34. [34]Wright, E. et al. Ricochets on asteroids: Experimental study of low velocity grazing impacts into granular media. Icarus 351, 113963 (2020).DOI
  35. [35]Gray, J. & Hancock, G. The propulsion of sea-urchin spermatozoa. Journal of Experimental Biology 32, 802–814 (1955).DOI
  36. [36]Li, C., Zhang, T. & Goldman, D. I. A terradynamics of legged locomotion on granular media. science 339, 1408–1412 (2013).DOI
  37. [37]Slonaker, J. et al. General scaling relations for locomotion in granular media. Physical Review E 95, 052901 (2017).DOI
  38. [38]Luck, K. S., Campbell, J., Jansen, M. A., Aukes, D. M. & Amor, H. B. From the lab to the desert: Fast prototyping and learning of robot locomotion. arXiv preprint arXiv:1706.01977 (2017).arxiv.org/abs/1706.01977
  39. [39]Thoesen, A., Ramirez, S. & Marvi, H. Screw-generated forces in granular media: Experimental, computational, and analytical comparison. AIChE Journal 65, 894–903 (2019).DOI
  40. [40]Askari, H. & Kamrin, K. Intrusion rheology in grains and other flowable materials. Nature Materials 15, 1274–1279 (2016).DOI
  41. [41]Roth, L. K., Han, E. & Jaeger, H. M. Intrusion into granular media beyond the quasistatic regime. Phys. Rev. Lett. 126, 218001 (2021). URL https://link.aps.org/doi/10.1103/PhysRevLett.126.218001.DOI
  42. [42]Agarwal, S., Goldman, D. I. & Kamrin, K. Mechanistic framework for reduced-order models in soft materials: Application to three-dimensional granular intrusion. Proceedings of the National Academy of Sciences 120, e2214017120 (2023).DOI
  43. [45]Sunday, Cecily et al. The influence of gravity on granular impacts - ii. a gravity-scaled collision model for slow interactions. A&A 658, A118 (2022). URL https://doi.org/10.1051/0004-6361/202142098.
  44. [46]Sunday, C., Murdoch, N., Tardivel, S. & Michel, P. Froude scaling for rovers on small body surfaces  16 (2022). URL https://hal.archives-ouvertes.fr/hal-03835761.
  45. [47]Sánchez, P. & Scheeres, D. J. Cohesive regolith on fast rotating asteroids. Icarus 338, 113443 (2020).DOI
  46. [48]Tsuji, Y., Tanaka, T. & Ishida, T. Lagrangian numerical simulation of plug flow of cohesionless particles in a horizontal pipe. Powder technology 71, 239–250 (1992).DOI

Paper details

vlnc:W_rb7wmu4w
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents