Potential loss of energetic ions including alphas and radio-frequency tail ions due to classical orbit effects and magnetohydrodynamic instabilities (MHD) are central physics issues in the design and experimental physics programme of the SPARC tokamak. The expected loss of fusion alpha power due to ripple-induced transport is computed for the SPARC tokamak design by the ASCOT and SPIRAL orbit-simulation codes, to assess the expected surface heating of plasma-facing components. We find good agreement between the ASCOT and SPIRAL simulation results not only in integrated quantities (fraction of alpha power loss) but also in the spatial, temporal and pitch-angle dependence of the losses. If the toroidal field (TF) coils are well-aligned, the SPARC edge ripple is small (0.15–0.30 %), the computed ripple-induced alpha power loss is small (~0.25 %) and the corresponding peak surface power density is acceptable (244 kW m−2). However, the ripple and ripple-induced losses increase strongly if the TF coils are assumed to suffer increasing magnitudes of misalignment. Surface heat loads may become problematic if the TF coil misalignment approaches the centimetre level. Ripple-induced losses of the energetic ion tail driven by ion cyclotron range of frequency (ICRF) heating are not expected to generate significant wall or limiter heating in the nominal SPARC plasma scenario. Because the expected classical fast-ion losses are small, SPARC will be able to observe and study fast-ion redistribution due to MHD including sawteeth and Alfvén eigenmodes (AEs). SPARC’s parameter space for AE physics even at moderate Q is shown to reasonably overlap that of the demonstration power plant ARC (Sorbom et al., Fusion Engng Des., vol. 100, 2015, p. 378), and thus measurements of AE mode amplitude, spectrum and associated fast-ion transport in SPARC would provide relevant guidance about AE behaviour expected in ARC. Understanding the confinement of energetic alpha particles generated by fusion is a long-standing goal of the worldwide fusion research programme (Kolesnichenko 1980; Fasoli et al. 2007; Gorelenkov, Pinches & Toi 2014). Energetic ions are subject to redistribution and radial loss due to both classical orbit phenomena (first-orbit loss, ripple) and magnetohydrodynamics (MHD) (sawteeth, neoclassical tearing modes (NTMs) and Alfvén eigenmodes (AEs)). These phenomena can potentially impair system performance due to loss of plasma self-heating, but typically the most dangerous consequence is localized heating of plasma-facing components. The use of discrete coils to generate the toroidal magnetic field in a tokamak causes the field strength to vary slightly in the toroidal direction at constant R and Z. This variation, termed toroidal field (TF) ‘ripple’, breaks the axisymmetry of the toroidal field. It has been appreciated since the early days of fusion research that ripple can cause radial transport – and loss – of energetic ions (Artsimovich, Mirnov & Strelkov 1965; Stringer 1972; Connor & Hastie 1973; Goldston, White & Boozer 1981) that could lead to excessive localized heating and failure of plasma-facing components. So-called ‘ripple-trapped’ ions with very small pitch angle that are effectively trapped in the shallow magnetic well between adjacent TF coils as well as the much larger population of banana-trapped ions are susceptible to ripple-induced transport. The magnitude of ripple-induced transport is an increasing function of the ripple amplitude, and the ripple amplitude is a function of the number of TF coils and their size relative to the plasma. For a given plasma size, the ripple-induced fast-ion loss can be reduced by either increasing the number of TF coils or the coil size. Both approaches affect the cost and diagnostic access of tokamak designs: increasing the number of TF coils increases cost and deceases access, while increasing the TF coil size increases cost and increases access. Thus, TF ripple is an essential consideration in the design of all burning plasma experiments. This paper evaluates ripple-induced fast-ion loss in the SPARC tokamak design (Creely et al. 2020). We use two existing numerical codes, ASCOT (Varje et al. 2019) and SPIRAL (Kramer et al. 2013a), to compute the orbits of alpha particles in the rippled magnetic field from birth at 3.5 MeV until they either thermalize or hit the wall. Because the orbit simulations are rather time-consuming, they have been carried out not for the SPARC V2 design but for its immediate predecessor (V1E), which had slightly smaller ripple (0.15 % versus 0.3 % at the outer edge) but identical plasma conditions (toroidal field, plasma current, major radius, minor radius, elongation: BT, Ip, R0, a, κ). First, for a number of TF coil configurations, we compute the ripple magnitude as a function of the number of TF coils and as a function of how well the TF coils are aligned. Then we compute the fraction of lost-alpha power for these configurations to the last closed flux surface (LCFS) and its poloidal and toroidal spatial variation. We then extend the orbit simulations from the LCFS to a candidate SPARC wall shape to obtain an estimate of the alpha particle contribution to the wall heat load. Finally, we estimate the surface heating due to alpha losses that would occur in the SPARC V2 coil design based on the V1E simulations. As in all tokamaks, the ripple magnitude δ(R, Z) in SPARC is small at the plasma centre but rises rapidly toward the edge. Ripple-induced radial transport of energetic ions increases with the ripple magnitude, so the transport is typically small at the plasma center, where most of the alphas are born (73 % are born inside ρpol < 0.6 where the radial coordinate ρpol = √((Ψ − Ψo)/(Ψe − Ψo)), where Ψo, Ψe, Ψ are the poloidal flux values at the magnetic axis, plasma edge and local position, respectively). The edge ripple for the SPARC V1E design is approximately 0.15 %, which is on the low end of the ripple values for other deuterium-tritium tokamak designs (FIRE, ITER). The TF ripple in a tokamak arises from the fact that the toroidal field is generated by a discrete set of coils, so that the magnetic field is slightly stronger in the plane of a TF coil than half-way between adjacent TF coils. The ripple magnitude δ(R, Z) in a tokamak with N TF coils is defined by δ(R, Z) = (Bmax − Bmin)/(Bmax + Bmin), (2.1) where the maxima and minima are the extrema of the field values as one proceeds along the toroidal direction φ at constant R and Z. Note that the ripple term has N-fold symmetry if the TF coils are perfectly aligned. The ripple δ(R, Z) and 3-D magnetic field topology BR,φ,Z(R, φ, Z) were computed numerically by approximating each TF coil as a set of discrete wires that were equally spaced within the defined shape of the TF coil. The number of wires is configurable, but in practice two wires toroidally and two wires radially was determined to be sufficient. Each wire in turn was represented by a set of ~80 straight wire segments. The magnitude of each wire segment’s magnetic field was computed from the Biot–Savart law and the contribution of all wire segments from all of the TF coils was summed to obtain the net magnetic field at each field point. Figure 1 shows the ripple contours for the nominal SPARC V1E TF design, assuming 18 perfectly aligned TF coils. The ripple near the plasma centre is very low (0.0016 %) and rises to a modest 0.15 % at the outer plasma edge (R = R0 + a) at the midplane. These values are so low that it raises the question of whether the ripple and associated alpha losses in a ‘real-world’ SPARC with imperfectly manufactured coils or imperfectly positioned coils could be dominated by those imperfections rather than by the discreteness of the TF coil set. Four types of TF coil displacements are considered here: up/down, in/out, toroidal offset and vertical tilt, where vertical tilt is a rotation of the coil about an axis along the major-radius direction at the midplane. For each case, candidate displacements (in metres) are sampled from a normal distribution with a specified standard deviation σ, but candidate displacements larger than 1.5σ are excluded. For the toroidal and vertical-tilt calculations, the displacement in metres is converted into a displacement angle by dividing it by the outer major radius of the TF coil or the vertical height of the TF coil, respectively. When the TF coil set is assumed to be perfectly aligned, the toroidal magnetic field is periodic in the toroidal direction. However, when various TF coil misalignments are allowed, then the toroidal magnetic field is no longer periodic; effectively, TF coil misalignments will add some low-N perturbation to the toroidal magnetic field in addition to the high-N (N = 18) generated by the discreteness of the TF coils. For configurations involving misaligned TF coils, we adopt the definition that the ripple is equal to the average of the ripple evaluated for each of the individual 18 TF sectors. Displacements with σ = 1.43 cm show that toroidal offsets and in/out displacements are the most significant. Up/down displacements generate little ripple and vertical tilts generate negligible ripple. The dominant ripple-induced loss mechanisms are ripple trapping and stochastic banana diffusion; collisional ripple diffusion is typically a weaker loss mechanism. Ripple trapping: In the absence of TF ripple, the B × ∇B vertical drift experienced by an ion cancels at the top and bottom of its orbit. However, if the ion has only a very small parallel velocity, |v∥/v| < √(δ) ≈ 0.04 for δ = 0.15 %, the ion cannot climb out of the local magnetic ‘well’ between adjacent TF coils and so it drifts vertically out of the plasma. Rotational transform reduces the effective ripple depth, and the wells disappear entirely for α* = ϵ|sin θ|/(Nqδ) > 1, where ϵ = r/a, θ is the poloidal angle, N is the number of TF coils, and q is the local safety factor. Stochastic ripple diffusion: In the presence of ripple, banana-trapped ions experience an additional radial excursion from a flux surface Δr that, for circular plasma, is given by: Δr = [Nπ / |sin θ|]^(1/2) (q/ϵ)^(3/2) ρδ sin(Nφb), (3.1) where ρ is the gyroradius. If Δr is sufficiently large, successive banana tips decorrelate and experience rapid radial diffusion. The threshold condition is given approximately by: δ > δs = (ϵ / (Nπq))^(3/2) (1 / (2ρq')), (3.2) where q' = dq/dr. Numerical ion orbit-simulation codes including ASCOT and SPIRAL were used to model full alpha gyro-orbits in the 3-D magnetic fields. ASCOT (version 5) and SPIRAL both incorporate collisions (slowing down and pitch-angle scattering) and simulate full orbits to calculate lost-alpha fractions and surface power densities on wall and limiter components. Kinetic plasma profiles and magnetic equilibrium were extracted from predictive TRANSP simulations of the full-performance SPARC DT scenario during the burn flat top (R0 = 1.85 m, a = 0.57 m, κsep = 1.97, δsep = 0.54, BT = 12.2 T, Ip = 8.7 MA, ne0 ≈ 4.3 × 10^20 m−3, Te0 ≈ 22.2 keV, Ti0 ≈ 19.8 keV, Zeff = 1.5, PRF = 11 MW, Pfus ≈ 110 MW, Pα ≈ 22 MW, and Q ≈ 9). Approximately 20,000 alpha markers with birth energy E = 3.5 MeV were followed. Simulations demonstrate that ripple-induced alpha loss to the LCFS is ~0.25% for the nominal 18-coil configuration, which is sub-dominant to first-orbit loss (2.8%). The alpha loss increases approximately linearly with the displacement standard deviation σ. Scans reducing the number of TF coils from 18 to 12 show consistent results between ASCOT and SPIRAL, with total loss rising to 9.27% (6.28% ripple-induced) for 12 coils. Toroidal peaking for 18 coils is very modest (~1.3–1.4 peak-to-minimum), whereas 12 coils produce substantial toroidal localization (peak-to-minimum of ~6). Poloidally, losses are heavily concentrated in the upper outboard region (θ > 0°) due to the upward B × ∇B drift. Evaluating wall/limiter power deposition indicates that recessing RF antenna surfaces by 1 cm behind adjacent limiters reduces alpha power deposition on antennas by a factor of ~40. For the nominal case, peak limiter heat load is modest (244 kW m−2) but increases with coil misalignment. Misalignment standard deviation σ > 0.7–1.8 cm could lead to problematic heat loads. ICRF-driven fast-ion tails (He-3 minority) have effective temperatures (~215 keV) well below the alpha birth energy (3.5 MeV) and are concentrated in the core (ρ < 0.5), where ripple is negligible. Thus, RF tail losses are expected to be negligible in standard SPARC operations. With low classical fast-ion losses, SPARC provides an excellent platform for investigating alpha-driven collective phenomena, including Alfvén eigenmodes (AEs) and sawtooth redistribution at high Q (Q ~ 9–11) where self-heating is dominant. The normalized alpha beta ratio βα/β scales as ~T^(5/2). At Q = 9, SPARC's βα/β is within 20% of ITER and ARC. The most unstable toroidal mode number nmax scales with R0*sqrt(n)/q^2, resulting in nmax values in SPARC comparable to inductive ARC scenarios (within 20–25%), making SPARC experimental AE observations highly relevant to ARC. The SPARC tokamak exhibits low ripple-induced alpha and RF tail fast-ion losses under nominal TF coil alignments. Coil misalignment tolerances should be maintained within the sub-centimeter range (σ < 0.7–1.8 cm) to prevent excessive localized limiter heating. SPARC will operate in a self-heated burning plasma regime suitable for studying burning plasma fast-ion transport and Alfvén eigenmode dynamics relevant to next-step devices like ARC. ASCOT and SPIRAL both model full gyromotion orbits (GO) and include neoclassical slowing down and pitch-angle scattering. ASCOT employs a volume-preserving algorithm (VPA) or Leap-Frog method and Landau collision operator (Langevin equation) with energy diffusion, while SPIRAL utilizes an adaptive Adams method and deterministic slowing down with stochastic pitch scattering. Both codes show excellent cross-code agreement for fast-ion tracking in SPARC.