Fact-checked by Grok 2 weeks ago

Diffusivity

Diffusivity, also known as the diffusion coefficient, is a fundamental physical property that quantifies the rate at which particles such as atoms, molecules, or ions spread through a medium via random thermal motion, driven by concentration gradients.^[1]^[2] It serves as the proportionality constant in Fick's first law of diffusion, where the diffusive flux J is given by J = -D \nabla C, with D representing diffusivity and \nabla C the concentration gradient.^[1]^[2] In the SI system, diffusivity has units of square meters per second (m²/s), reflecting its nature as an area traversed per unit time.^[1]^[2] Diffusivity manifests in various forms depending on the transport process, including mass diffusivity (for molecular diffusion in gases, liquids, or solids), thermal diffusivity (for heat conduction, defined as \alpha = k / (\rho c_p), where k is thermal conductivity, \rho is density, and c_p is specific heat capacity), and others like electrical or ionic diffusivity in specific contexts.^[3]^[1] Mass diffusivity, in particular, is crucial for understanding phenomena such as gas permeation, solute transport in solutions, and atomic migration in materials.^[3]^[2] Fick's second law, \partial C / \partial t = D \nabla^2 C, extends this to describe how concentration evolves over time in diffusive systems.^[2]^[1] The value of diffusivity is influenced by temperature (often following an Arrhenius relation D \propto e^{-E_a / (RT)}, where E_a is activation energy), pressure, medium viscosity, particle size, and structural factors like porosity or tortuosity in heterogeneous materials.^[1]^[3]^[2] In gases, it increases with temperature and decreases with pressure; in liquids, the Stokes-Einstein equation D = kT / (6\pi \mu r) (with k as Boltzmann's constant, T temperature, \mu viscosity, and r particle radius) provides a key model.^[1] Diffusivity plays a pivotal role in fields like materials science, chemical engineering, environmental modeling, and biology, enabling predictions of processes from alloy annealing to drug delivery.^[3]^[1]

Fundamentals

Definition

Diffusivity, denoted as the diffusion coefficient D, is a fundamental transport property that quantifies the rate at which particles or molecules spread through a medium due to random motion. It serves as the constant of proportionality between the diffusive flux of a substance and its concentration gradient, characterizing how quickly a species disperses under non-equilibrium conditions.^[4] At its core, diffusion arises as a spontaneous process driven by the random thermal motion of particles, often referred to as Brownian motion, which occurs without the influence of external forces such as pressure or electric fields. This microscopic randomness leads to a net macroscopic transport from regions of higher concentration to lower ones, tending toward equilibrium. The diffusion coefficient D encapsulates the mobility of particles within the medium, with typical values in liquids on the order of $10^{-9} to $10^{-10} m²/s, reflecting the scale of this spreading over time.^[1] The concept of diffusivity was formalized in 1855 by German physiologist Adolf Fick, who drew an analogy between particle diffusion and Jean-Baptiste Fourier's earlier law of heat conduction, establishing a proportional relationship that underpins modern descriptions like Fick's laws. This historical framing highlighted diffusivity's role as a material-specific parameter, typically expressed in SI units of square meters per second (m²/s), which conveys its dimensional nature as an area swept per unit time. A relatable illustration of diffusivity in action is the gradual dispersion of a drop of ink introduced into a container of still water, where the initially concentrated dye molecules undergo random collisions and thermal agitation, visibly spreading to uniform color over time as a result of their Brownian trajectories.^[5]

Units and Dimensions

Diffusivity, denoted as D, possesses dimensions of length squared per unit time, expressed as [D] = \mathrm{L}^2 / \mathrm{T}, where \mathrm{L} represents length and \mathrm{T} represents time.^[6] This dimensional form arises from its role in transport phenomena, directly linking it to analogous coefficients such as kinematic viscosity, which shares the same dimensions but pertains to momentum diffusion, and thermal diffusivity, which governs heat propagation.^[7] In the International System of Units (SI), diffusivity is quantified in square meters per second (m²/s).^[1] Representative values illustrate its scale across phases: approximately $10^{-9} m²/s for molecular diffusion in liquids like water at room temperature, and orders of magnitude lower in solids, such as $10^{-12} m²/s for self-diffusion in metals at elevated temperatures near their melting points.^[8] For practical applications in experimental contexts, diffusivity is often converted to smaller units like square centimeters per second (cm²/s), where 1 m²/s = 10⁴ cm²/s, or square micrometers per second (μm²/s) for microscale analyses.^[7] Thermal diffusivity \alpha, defined as \alpha = \frac{k}{\rho c_p}—with k as thermal conductivity, \rho as density, and c_p as specific heat capacity at constant pressure—also carries dimensions of L²/T and SI units of m²/s.^[9] While both properties describe the rate at which perturbations propagate through a medium, mass diffusivity fundamentally arises from random molecular motions driving concentration gradients, whereas thermal diffusivity stems from phonon or electron-mediated heat conduction, highlighting their distinct underlying physical mechanisms despite dimensional similarity.^[9]

Theoretical Framework

Fick's Laws

Fick's laws, proposed by Adolf Fick in 1855, mathematically describe the transport of matter through diffusion as a process driven by concentration gradients, analogous to heat conduction described by Fourier's law.^[10] These laws form the cornerstone of diffusion theory, linking the diffusive flux to spatial variations in concentration and enabling the prediction of concentration profiles over time.^[11] The diffusivity D, a material-specific property, serves as the proportionality constant in these relations, quantifying the intrinsic rate of diffusive spreading.^[12] Fick's first law expresses the diffusive flux \mathbf{J} (the amount of substance crossing a unit area per unit time) as proportional to the negative gradient of concentration c, in vector form:

\mathbf{J} = -D \nabla c

where \nabla c is the concentration gradient and D is the diffusion coefficient.^[11] This equation indicates that matter flows from regions of higher to lower concentration, with the flux direction opposite to the gradient. In one dimension, it simplifies to

J_x = -D \frac{dc}{dx},

facilitating analysis in planar geometries.^[13] The law arises from a phenomenological observation but can be derived microscopically from random walk models, where net displacement results from probabilistic particle motions biased by concentration differences, assuming linear response near equilibrium.^[11] Fick's second law governs the time evolution of concentration in non-steady-state diffusion and is derived by combining the first law with the continuity equation, which enforces mass conservation: \frac{\partial c}{\partial t} + \nabla \cdot \mathbf{J} = 0. Substituting \mathbf{J} = -D \nabla c yields

\frac{\partial c}{\partial t} = \nabla \cdot (D \nabla c).

For constant D, this reduces to the diffusion equation

\frac{\partial c}{\partial t} = D \nabla^2 c,

a parabolic partial differential equation describing how concentration spreads temporally.^[11] In steady-state conditions (\frac{\partial c}{\partial t} = 0), the equation simplifies to \nabla \cdot (D \nabla c) = 0, implying constant flux and often linear concentration profiles in one dimension. Non-steady-state cases capture transient diffusion, such as the initial rapid spreading followed by slower equilibration.^[12] These laws rest on key assumptions, including an isotropic medium where diffusion properties are direction-independent, a constant diffusion coefficient D independent of concentration or position, and the neglect of convective flows or chemical reactions that could alter the transport.^[12] Solutions to the second law require specifying initial and boundary conditions; for an infinite domain, such as solute release from an instantaneous point source, the concentration evolves as a Gaussian profile c(x,t) \propto \frac{1}{\sqrt{4\pi D t}} \exp\left(-\frac{x^2}{4 D t}\right). In finite domains, like diffusion between impermeable boundaries with fixed surface concentrations, conditions enforce no flux at walls (\mathbf{J} \cdot \mathbf{n} = 0) or specified values at interfaces, leading to series solutions via separation of variables.^[11]

Diffusion Coefficient Models

Diffusion coefficient models provide theoretical frameworks for estimating the value of the diffusion coefficient D in various physical systems, serving as inputs to Fick's laws of diffusion. These models derive D from fundamental physical parameters such as temperature, viscosity, activation energies, and molecular interactions, tailored to specific phases like liquids, gases, and solids. They enable predictions of diffusive behavior without relying solely on experimental measurements, though validation against data remains essential. In liquids, the Einstein relation connects the diffusion coefficient of spherical particles undergoing Brownian motion to the medium's viscosity via Stokes' law. For a particle of radius r in a fluid of viscosity \eta at temperature T, the relation is given by

D = \frac{k T}{6 \pi \eta r},

where k is Boltzmann's constant. This expression arises from balancing the frictional drag force on the particle with the random thermal forces driving its motion, assuming low Reynolds number flow. The model applies to colloidal suspensions and molecular solutes in dilute solutions, predicting that D increases with temperature and decreases with particle size or solvent viscosity.^[14] For activated diffusion processes, prevalent in solids and viscous liquids where particles must overcome energy barriers to move, the Arrhenius form describes the temperature dependence of D:

D = D_0 \exp\left(-\frac{E_a}{R T}\right),

with D_0 as the pre-exponential factor representing the frequency of attempts to jump barriers and entropic contributions, E_a the activation energy, R the gas constant, and T the absolute temperature. The exponential term captures the Boltzmann factor for surmounting the energy barrier, while D_0 typically ranges from $10^{-5} to $10^{-3} m²/s in solids, depending on the lattice vibration frequencies and jump distances.^[15]^[16] This model is foundational for interpreting diffusion in crystalline materials, where E_a reflects vacancy formation and migration energies.^[15] In gas-phase systems, particularly binary mixtures, the Chapman-Enskog theory derives the self- and mutual-diffusion coefficients from the Boltzmann equation, accounting for molecular collisions. For binary gases, the diffusion coefficient D_{12} scales approximately as

D_{12} \propto \frac{T^{3/2}}{P} \cdot \frac{1}{\sqrt{\mu}},

where T is temperature, P pressure, and \mu the reduced mass of the species. This first-order approximation incorporates collision integrals that depend on the intermolecular potential, often modeled as Lennard-Jones, yielding D values on the order of 10⁻⁵ m²/s at standard conditions for common gases like air. The theory predicts inverse pressure dependence due to collision frequency and a strong positive temperature exponent from increased mean free path. Higher-order corrections refine accuracy for dense or polyatomic gases.^[17] For diffusion in crystalline solids, lattice models employ a random walk approximation, treating atomic jumps between lattice sites as uncorrelated events. In three dimensions, the diffusion coefficient is

D = \frac{a^2 \Gamma}{6},

where a is the lattice spacing (jump distance) and \Gamma the jump frequency, often \Gamma = \nu \exp(-E_m / kT) with \nu a vibrational frequency (~10¹³ s⁻¹) and E_m the migration barrier. This isotropic model assumes a simple cubic lattice with equal jumps in all directions, linking macroscopic diffusion to microscopic hopping rates via the mean square displacement \langle r^2 \rangle = 6 D t. It underpins vacancy-mediated diffusion in metals and semiconductors, with extensions for correlated jumps in ordered alloys.^[18]

Types of Diffusivity

Self-Diffusivity

Self-diffusivity, denoted as D^*, refers to the diffusion coefficient that quantifies the random, thermally activated migration of atoms or molecules of the same species within a homogeneous pure substance, occurring without any imposed concentration gradients.^[19] This intrinsic process arises from thermal fluctuations driving atomic jumps in a lattice or molecular motions in a fluid, independent of external chemical potential differences.^[20] In solids, self-diffusivity typically follows an Arrhenius temperature dependence, D^* = D_0 \exp(-Q/RT), where D_0 is the pre-exponential factor, Q is the activation energy, R is the gas constant, and T is temperature. To measure self-diffusivity experimentally, radioactive isotopes are commonly employed as tracers to label and track the movement of identical species in the host material, allowing precise determination of atomic displacement over time.^[21] For instance, in metals, a thin layer of the radioactive isotope is deposited on the surface, and after annealing at elevated temperatures, the penetration profile is analyzed using sectioning and radioactivity counting to derive D^* from the Gaussian distribution of the tracer.^[22] This isotope method ensures that the tracked atoms are chemically identical to the matrix, isolating the self-diffusion mechanism.^[23] In ionic solids, self-diffusivity is related to ionic mobility \mu through the Nernst-Einstein equation, D^* = \frac{[k](/page/K)T}{[q](/page/Q)} \mu, where k is Boltzmann's constant, T is temperature, and q is the ion charge, linking random diffusion to directed drift under an electric field.^[24] This relation holds under conditions of low defect concentrations and negligible correlations between jumps, providing a fundamental connection between transport properties in electrolytes and solids.^[25] Representative examples include the self-diffusivity of hydrogen in metals, where values around $10^{-12} m²/s at room temperature highlight its exceptionally high mobility compared to other interstitials, facilitating rapid permeation in applications like hydrogen storage.^[26] Similarly, self-diffusion of carbon in iron exhibits much lower rates at room temperature, on the order of $10^{-20} m²/s or less in ferrite, underscoring the role of lattice structure in controlling atomic jumps. These cases illustrate how self-diffusivity governs fundamental processes like annealing and phase transformations in pure materials.^[27]

Interdiffusivity

Interdiffusivity, denoted as \tilde{D}, characterizes the mutual diffusion between two distinct species in a binary mixture, primarily driven by gradients in chemical potential across the interface rather than simple concentration differences.^[28] This coefficient extends Fick's laws to multicomponent systems by accounting for the thermodynamic forces that govern atomic exchange in non-ideal solutions.^[29] In practical terms, interdiffusivity quantifies the rate at which atoms of one species penetrate and replace those of another during contact, influencing material homogeneity and phase stability. A key relation for interdiffusivity in binary substitutional alloys is provided by Darken's equation, which links it to the self-diffusivities of the individual components:

\tilde{D} = N_A D_B^* + N_B D_A^*

Here, N_A and N_B represent the mole fractions of species A and B, respectively, while D_A^* and D_B^* are their respective self-diffusivities, measured under conditions of uniform composition. This equation assumes a random-walk mechanism mediated by vacancies and highlights how interdiffusivity emerges from the weighted average of intrinsic mobilities, adjusted for composition. Self-diffusivity serves as a foundational component in such interdiffusion models, enabling predictions of collective transport in alloys.^[30] The Kirkendall effect exemplifies the consequences of unequal interdiffusivities in binary systems, where markers initially placed at the diffusion interface shift toward the side with slower-diffusing species due to unbalanced vacancy fluxes.^[31] First observed in brass (Cu-Zn) diffusion couples, this phenomenon demonstrates that atomic diffusion is not lattice-conservative, as faster zinc out-diffusion creates excess vacancies that condense into voids on the zinc-rich side.^[32] The effect underscores vacancy-mediated mechanisms and has implications for defect formation in alloys. In semiconductor processing, interdiffusivity governs dopant redistribution, such as boron diffusion into silicon during thermal annealing, which is essential for precise p-n junction formation but can lead to unwanted profile broadening if not controlled.^[33] In metallurgy, interdiffusion drives phase formation in diffusion couples, for instance, between low-carbon steel and aluminum, resulting in layered intermetallic compounds like FeAl that enhance joining strength but risk brittleness. These examples illustrate interdiffusivity's role in tailoring material properties through controlled atomic mixing.

Tracer Diffusivity

Tracer diffusivity, denoted as D_t or D^*, refers to the diffusion coefficient of a trace amount of a labeled species, such as an isotope, within a host medium where the concentration of the tracer is sufficiently dilute to avoid significant interactions among the tracers themselves.^[34] This measures the random atomic or molecular motion driven by thermal fluctuations, providing fundamental insights into the kinetic processes governing atomic jumps in solids, liquids, or gases. In the dilute limit, tracer diffusivity for a solute approximates the self-diffusivity of the host atoms when the solute and host have similar sizes and interactions, though it is perturbed by solute-host frictional effects that alter the local environment.^[34] Theoretically, tracer diffusivity is linked to the friction experienced by the tracer particle through the Einstein-Smoluchowski relation:

D_t = \frac{k_B T}{\zeta},

where k_B is Boltzmann's constant, T is the absolute temperature, and \zeta is the drag or friction coefficient arising from interactions between the solute tracer and the surrounding host medium.^[35] This relation underscores that D_t quantifies the balance between thermal energy driving diffusion and dissipative forces impeding motion, applicable to tracer scenarios where viscous drag dominates in the host lattice or fluid.^[35] In applications to alloys, tracer diffusivity is commonly measured using radioactive isotopes to probe vacancy-mediated diffusion mechanisms, revealing how point defects facilitate atomic exchange. For instance, the isotope ^{59}Fe has been employed as a tracer in austenitic steels, such as Fe-17 wt% Cr-12 wt% Ni alloys, to determine volume and grain-boundary diffusivities, with activation energies (e.g., Q_{\ce{Fe}} \approx 280 kJ/mol for volume diffusion) indicating vacancy jump frequencies and solute-vacancy binding effects.^[36] These measurements help elucidate diffusion pathways in complex alloys, distinguishing lattice diffusion from enhanced boundary paths and informing models of creep or phase transformations.^[36] Unlike chemical diffusivity, which incorporates thermodynamic driving forces from concentration gradients and activity corrections, tracer diffusivity isolates pure kinetic mobility by neglecting the thermodynamic factor \Gamma = \frac{\partial \ln a}{\partial \ln c} (where a is activity and c is concentration), focusing solely on uncorrelated random walks of the labeled species.^[37] This distinction allows tracer studies to serve as a baseline for understanding mobility without compositional influences, as in the Darken relation linking tracer to interdiffusion coefficients.^[37]

Influencing Factors

Temperature Dependence

The diffusion coefficient in most systems exhibits a pronounced temperature dependence, commonly analyzed through Arrhenius behavior where plots of the natural logarithm of diffusivity (ln D) versus the reciprocal of absolute temperature (1/T) yield straight lines. This linearity facilitates data fitting to determine the pre-exponential factor D_0 (intercept) and activation energy E_a (negative slope times the gas constant R), providing insights into the underlying mechanisms.^[38] In gases, the dependence is relatively weak, with D scaling approximately as T^{3/2} from kinetic theory due to increased molecular velocities, corresponding to low effective activation energies on the order of a few kJ/mol.^[39] By contrast, liquids and solids show stronger variations, with E_a values ranging from tens to hundreds of kJ/mol, reflecting the need to overcome molecular or atomic barriers. The physical basis for this temperature sensitivity varies by phase. In liquids, the free volume theory explains enhanced diffusivity as temperature rises, positing that thermal expansion creates transient voids or "free volume" that allow molecules to redistribute and jump more frequently, with the probability of sufficient free volume formation increasing exponentially with temperature.^[40] This leads to moderate activation energies, typically 10-30 kJ/mol for simple liquids. In solids, diffusion relies on discrete atomic jumps via vacancy mechanisms—where atoms exchange positions with neighboring lattice vacancies—or interstitial paths, both requiring thermal activation to surmount high energy barriers associated with lattice strain and defect formation. Vacancy concentrations themselves follow Arrhenius statistics, amplifying the overall exponential temperature response and yielding E_a values often exceeding 100 kJ/mol.^[41] A notable phenomenon across material classes is the compensation effect, where variations in D_0 and E_a are correlated such that materials with higher activation energies exhibit proportionally larger pre-exponential factors. This manifests as a linear relationship between ln D_0 and E_a in Arrhenius plots compiled from diverse systems, attributed to shared entropic and enthalpic contributions in the transition state for diffusion. For instance, in metallic alloys, this correlation links self-diffusion parameters to melting points, enabling predictions of diffusivity trends without phase-specific details.^[42] Quantitative trends highlight phase-specific scales: in polymers, activation energies for penetrant diffusion (e.g., organic molecules in polystyrene) range from 40-100 kJ/mol, often resulting in diffusivity doubling approximately every 10-20°C near ambient conditions due to the polymer chain dynamics facilitating segmental motion. In metals above the melting point, liquid-state diffusivities show exponential increases with temperature but with reduced E_a (typically 20-50 kJ/mol for self-diffusion in liquid Al or Cu), contrasting the steeper solid-state rises and enabling rapid atomic mixing in high-temperature processing.^[43]^[44]

Concentration and Structural Effects

The diffusion coefficient in metallic solid solutions typically exhibits a concentration dependence that is minimal at extreme compositions but becomes significant at intermediate solute levels, often due to lattice distortions and solute-solute interactions. In binary alloys, this can lead to upward trends where diffusivity increases with solute concentration owing to enhanced vacancy-solute binding, or downward trends from clustering that impedes atomic jumps. For example, in the Cu-Ni system, the tracer diffusion coefficient of Cu exceeds that of Ni by approximately 30% at 89.9 at.% Cu over temperatures from 923°C to 1049°C, reflecting compositional effects on the lattice parameter and vacancy formation.^[45] Similarly, interdiffusivity in Mg alloys shows strong concentration dependence, with values extracted via Boltzmann-Matano analysis revealing variations tied to phase stability and thermodynamic factors.^[46] Structural features of the medium profoundly influence diffusivity, particularly through anisotropy in crystalline materials versus isotropy in liquids and amorphous phases. In crystals with hexagonal close-packed (hcp) structures, such as magnesium and titanium, self-diffusion exhibits directional anisotropy due to the lower symmetry, with migration energy barriers differing between in-basal-plane and out-of-basal-plane paths; for instance, diffusion in Mg is faster within the basal plane, while in Zn it proceeds more readily along the c-axis normal to the basal plane.^[47] This contrasts with face-centered cubic (fcc) metals like nickel, where hydrogen diffusivity varies by orientation, being highest along the 〈111〉 direction and lowest along 〈100〉, with ratios up to 2.5 between these axes at 300 K, attributed to elastic anisotropy and self-stress effects.^[48] In liquids, random atomic arrangements yield isotropic diffusivity, unaffected by crystallographic directions. Phase transitions in alloys can cause abrupt changes in diffusivity as the crystal structure or composition alters, often resulting in jumps at phase boundaries due to shifts in vacancy concentrations or lattice parameters. In eutectic systems, for example, crossing from one phase to another during solidification or annealing leads to enhanced diffusivity in the lower-melting phase, facilitating rapid solute redistribution. Such effects are evident in diffusion phase transitions, where the kinetics of phase formation involve concentration gradients that amplify or suppress atomic mobility across interfaces. Microstructural disorder, such as in amorphous glasses, generally reduces diffusivity compared to ordered crystals by increasing activation energies through disrupted pathways, though specific cases show enhanced rates due to excess free volume. In amorphous silicon, self-diffusivity is higher than in crystalline silicon by up to five orders of magnitude at around 700°C, linked to higher diffusion entropy from structural flexibility despite a comparable activation energy.^[49] Conversely, in polycrystals, grain boundaries act as fast diffusion conduits, with coefficients 10⁴ to 10⁶ times greater than bulk values owing to their open, disordered structure that lowers jump barriers; for instance, in oxides like NiO, this enhancement arises from higher atomic jump frequencies at boundaries.^[50]

Measurement Methods

Experimental Techniques

Interdiffusion experiments typically involve preparing diffusion couples by joining two materials with different initial compositions, annealing them at elevated temperatures to allow atomic intermixing, and then analyzing the resulting concentration profiles across the interface. These profiles are obtained through techniques such as electron probe microanalysis (EPMA) or secondary ion mass spectrometry (SIMS), which provide detailed elemental distributions as a function of position. The Boltzmann-Matano analysis is then applied to these profiles to determine the concentration-dependent interdiffusion coefficient \tilde{D}(c), transforming the data into a form that integrates the flux equation derived from Fick's first law under the assumption of a constant Matano interface.^[51]^[52] This method, originally formalized by Matano in 1933 building on Boltzmann's earlier work, enables the extraction of \tilde{D} values varying with composition, particularly useful for binary systems where diffusivity is not constant.^[53] Tracer methods are widely used to measure tracer diffusivity in solids, involving the deposition of a thin layer of radioactive isotopes onto a sample surface, followed by annealing to promote diffusion. After annealing, the sample is serially sectioned—often by mechanical grinding or sputtering—and the radioactivity in each section is counted using techniques like scintillation or gamma-ray detection to construct a penetration profile of the tracer concentration versus depth. This profile, when fitted to solutions of Fick's second law for thin-film sources, yields the tracer diffusion coefficient D^*, providing insights into single-component mobility without interference from chemical gradients.^[54]^[55] Such approaches, refined since the mid-20th century, are particularly effective for metals and alloys, with examples including the diffusion of chromium tracers in nickel-based systems.^[56] Non-destructive techniques offer advantages for sensitive or repeated measurements, such as nuclear magnetic resonance (NMR) spin-echo methods for liquids. In pulsed gradient spin-echo (PGSE) NMR, magnetic field gradients are applied to encode molecular displacement, and the attenuation of the spin-echo signal intensity follows the Stejskal-Tanner equation, allowing direct calculation of self-diffusion coefficients from the slope of signal versus gradient strength squared. This method excels in liquids and solutions, providing high-resolution data on molecular mobility without sample alteration.^[57] For thin films, Rutherford backscattering spectrometry (RBS) serves as a non-destructive profiling tool, where high-energy helium ions are scattered off atomic nuclei to generate energy spectra that reveal elemental depth distributions. These spectra are analyzed to obtain concentration profiles, from which diffusion coefficients are derived by modeling intermixing in layered structures, as demonstrated in studies of copper diffusion in indium sulfide films.^[58] For gases, the Loschmidt cell method is a standard technique for measuring binary diffusion coefficients. It consists of two chambers separated by a porous frit or partition, initially filled with different pure gases. Diffusion is initiated by removing the partition, and the evolving concentration profiles are monitored using optical interferometry or gas chromatography, with the diffusivity extracted by fitting the data to analytical solutions of Fick's laws.^[59] In liquids, beyond NMR, the diaphragm cell method enables steady-state measurements of mutual diffusion coefficients. This involves two compartments separated by a porous diaphragm, one containing pure solvent and the other a solution; the rate at which the concentration difference equalizes is measured (e.g., via sampling and analysis), yielding D from the exponential decay characteristic time.^[60] The Taylor dispersion technique provides a complementary dynamic method, injecting a solute pulse into laminar solvent flow within a capillary tube and analyzing the peak broadening at the detector using the Taylor-Aris equation to determine the diffusion coefficient, ideal for rapid assessments in dilute systems.^[61] Accuracy in these experimental techniques is influenced by several error sources, including surface oxidation during annealing in diffusion couples, which can alter initial boundary conditions and introduce artifacts in concentration profiles. Other challenges encompass inhomogeneities in sample preparation, imprecise sectioning depths in tracer methods, and gradient non-idealities in NMR. Typical precision for diffusivity measurements ranges from 5-10%, with tracer sectioning achieving reproducibility within a few percent under controlled conditions, while advanced SIMS-based profiling can reach 0.5-1% signal precision for deeper analyses.^[55]^[62]^[34]

Computational Approaches

Computational approaches to diffusivity prediction rely on simulations that model atomic or molecular motions at the microscopic level, providing insights into diffusion mechanisms without requiring physical experiments. These techniques span classical stochastic methods, quantum mechanical computations, and data-driven models, often integrating diffusion coefficient models like the Arrhenius equation for validation. By solving the underlying equations of motion or probability distributions, they enable the estimation of self-, tracer, or interdiffusivity in diverse systems, from crystalline solids to amorphous materials. Molecular dynamics (MD) simulations represent a cornerstone of these approaches, evolving atomic positions over time using classical force fields to capture realistic trajectories. Diffusivity is derived from the mean squared displacement (MSD) of particles, quantified by the Einstein relation:

D = \lim_{t \to \infty} \frac{\langle \mathbf{r}^2(t) \rangle}{6t}

in three dimensions, where \langle \mathbf{r}^2(t) \rangle denotes the ensemble-averaged squared displacement from initial positions after time t. This method excels in atomic-scale simulations of liquids, solids, and interfaces, revealing details such as correlated jumps in vacancy-mediated diffusion or the impact of defects on transport. Best practices emphasize selecting appropriate time windows for linear MSD fitting to ensure accuracy, particularly in systems with anomalous diffusion behaviors. Monte Carlo methods, especially kinetic Monte Carlo (kMC), address longer timescales by discretizing diffusion onto a lattice and simulating stochastic events like atomic hops. Jump probabilities are computed from transition rates via transition state theory, k = \nu \exp(-E_a / k_B T), where \nu is the attempt frequency, E_a the activation energy, k_B Boltzmann's constant, and T temperature; these probabilities guide random selections to evolve the system according to the master equation. In lattice diffusion scenarios, such as self-diffusion in metals, the approach incorporates nearest-neighbor interactions to adjust rates, enabling efficient modeling of rare events over experimentally inaccessible durations. This technique is particularly valuable for predicting interdiffusivity in alloys by sampling configurational spaces.^[63]^[64] Ab initio methods, grounded in density functional theory (DFT), provide parameter-free calculations of key energetic barriers driving diffusion in solids. Vacancy formation energies, essential for estimating equilibrium defect concentrations, are computed using supercell approximations with generalized gradient approximations like PBE, often achieving convergence within 50 meV. Migration barriers, and thus activation energies E_a, are obtained via the climbing-image nudged elastic band technique, which identifies minimum-energy paths for atomic jumps; these are combined with formation energies to yield diffusivities using expressions like D = f a^2 \nu \exp(-(E_f + E_m)/k_B T), where f is a correlation factor, a the jump distance, and E_f, E_m formation and migration energies. High-throughput DFT frameworks have generated databases for dilute solutes in hosts like Al and Ni, demonstrating RMS errors below 0.2 eV against experiments and facilitating rapid screening of diffusion properties.^[65] Machine learning surrogates accelerate diffusivity predictions by training on databases of DFT-derived data, bypassing the high cost of individual simulations. Neural networks and other models, such as random forests or gradient boosting, map inputs like elemental compositions, atomic radii, and electronic structure descriptors to outputs like activation energies or full diffusion coefficients in alloys. For instance, frameworks trained on vacancy-mediated diffusion datasets in conventional alloys achieve prediction accuracies with errors under 10% for self-diffusivity, enabling exploration of high-entropy systems. These approaches, often incorporating explainability techniques to highlight features like packing efficiency, support alloy design by forecasting tracer and interdiffusivity trends.^[66]

Applications

In Materials Science

In materials science, diffusivity governs key processes in engineering materials, influencing both processing and performance. During alloy design, controlled diffusion is essential for homogenization in casting, where as-cast microsegregation—arising from solute partitioning during solidification—is eliminated through heat treatments that promote atomic migration to achieve uniform composition. This diffusion-driven homogenization reduces coring and improves mechanical properties, as demonstrated in computational models for complex alloys like nickel-based superalloys. In precipitation-hardening steels, such as martensitic stainless steels, diffusivity controls the nucleation and growth of strengthening precipitates during aging; for instance, the diffusion of titanium and molybdenum enables the formation of fine intermetallic phases that enhance strength without sacrificing toughness.^[67]^[68] Interdiffusivity plays a pivotal role in multicomponent alloys, dictating element redistribution during high-temperature processing. In semiconductor fabrication, precise control of dopant diffusivity is critical for device performance; phosphorus diffusion into p-type silicon wafers creates n-type regions with tailored concentration profiles, enabling the formation of p-n junctions essential for diodes and transistors. Experimental profiles show that phosphorus exhibits concentration-dependent diffusivity, influenced by vacancy mechanisms, allowing junction depths on the order of micrometers to be achieved through thermal annealing. The diffusivity of oxygen in oxide scales significantly affects corrosion resistance and oxidation kinetics in metals. Protective oxide layers, such as alumina or chromia on high-temperature alloys, rely on low oxygen diffusivity to limit further ingress and scale growth; Wagner's theory quantifies this by linking parabolic oxidation rates to ambipolar diffusion of oxygen ions and electrons through the scale. For example, in nickel-based alloys, low oxygen diffusion coefficients in the oxide layer enable self-limiting scale thicknesses that protect against catastrophic oxidation. Measurement techniques like marker experiments validate these diffusivities, confirming inward oxygen transport as the dominant mechanism in many systems.^[69] A classic illustration of diffusivity's impact on material integrity is the formation of Kirkendall voids in Cu-Zn brass alloys, where unequal interdiffusion rates—zinc diffusing faster than copper—generate excess vacancies that coalesce into voids at the interface, leading to porosity and potential mechanical failure. This phenomenon, first observed in diffusion couples annealed at 780°C, demonstrated that atomic fluxes are species-dependent, challenging earlier assumptions of coupled diffusion and highlighting risks in brazing or welding applications. The voids can compromise ductility and fatigue life, underscoring the need for alloy composition adjustments to balance diffusion rates.^[70]

In Biological and Chemical Systems

In biological systems, diffusivity plays a critical role in nutrient transport, particularly for oxygen, which diffuses through tissues at rates around 1.5–2.0 × 10^{-9} m²/s, enabling supply to cells but often becoming a limiting factor in dense structures like biofilms.^[71] In biofilms, oxygen penetration is restricted to approximately 50–90 μm from the surface due to consumption by surface bacteria, creating hypoxic zones that slow metabolic rates and influence microbial community dynamics.^[72] This diffusion limitation shapes biofilm physiology, favoring anaerobic processes deeper within the matrix and impacting applications in wastewater treatment and infection control.^[73] Drug delivery in biological contexts relies on diffusivity across cell membranes, where passive transport is governed by the drug's partition coefficient, a measure of its lipophilicity that determines equilibrium distribution between aqueous and lipid phases.^[74] For instance, drugs with higher partition coefficients, such as certain lipophilic antibiotics, exhibit enhanced membrane permeability, facilitating targeted delivery but also raising concerns about off-target accumulation.^[75] These models, often extending Fick's laws to account for reactive boundaries, predict diffusion rates that are essential for optimizing pharmacokinetics in therapeutic design.^[76] Reaction-diffusion systems in biology, such as those driving morphogenesis, highlight how diffusivity influences pattern formation; in Turing mechanisms, differences in diffusion coefficients between activator and inhibitor species determine the wavelength of instabilities, typically on the order of cell sizes during embryonic development. For example, in protein folding kinetics, the process is modeled as diffusion along a free-energy landscape, where coordinate-dependent diffusivity modulates folding rates, with slower diffusion in compact states extending timescales to microseconds.^[77] In chemical systems like groundwater, pollutant diffusivity contributes to contaminant spread, particularly through back-diffusion from low-permeability aquitards, prolonging remediation efforts over years.^[78]

References

[1]
Diffusion Coefficient Definition - COMSOL
Jan 14, 2015 · The diffusion coefficient is most simply understood as the magnitude of the molar flux through a surface per unit concentration gradient out-of-plane.<|control11|><|separator|>
[2]
[PDF] Lecture 5: Diffusion Coefficient (Diffusivity)
Diffusion coefficient, also called Diffusivity, is an important parameter indicative of the diffusion mobility. Diffusion coefficient is not only encountered in ...
[3]
Diffusivity - an overview | ScienceDirect Topics
Diffusivity is defined as the ability of a medium to diffuse a fluid within its texture, represented as the mass transfer rate in various food-processing ...
[4]
Diffusivity - an overview | ScienceDirect Topics
Mar 1, 2025 · Diffusivity (D) is defined as the capacity of a substance to diffuse within a given medium, quantified in units of m² s⁻¹.
[5]
Diffusion | McGraw Hill's AccessScience
German physiologist Adolf Fick discovered the mathematical laws of diffusion in 1855. ... analogous to Fourier's law for heat conduction in a temperature ...
[6]
https://www.physics.umd.edu/courses/Phys374/fall05/exams/sol_ex1mu/Ex1muP1-S.htm
[7]
Dimensional Analysis of the Diffusion Equation - UMD Physics
(a) In the diffusion equation, in order for the dimensions to work out, the dimensions of D have to be [L2/T]. (Note that the dimensions of f do not matter.Missing: diffusivity | Show results with:diffusivity
[8]
[PDF] THE MATHEMATICS OF DIFFUSION - eng . lbl . gov
... unit of quantity, e.g. gram or gram molecules, then it is clear from (1.1) that Dis independent of this unit and has dimensions (length)2. (time)"*, e.g. cm2 ...
[9]
Diffusion and Partitioning - SERC (Carleton)
Mar 13, 2012 · Typical diffusion values (from high to low):. a) Gas: D ~ 10-5 m2/s in air at 300 K. b) Aqueous solution: D ~ 10-9 m2/s in water at 300 K. c ...<|control11|><|separator|>
[10]
[PDF] Chapter 5: Diffusion in Solids
Jan 28, 2011 · self-diffusion: D=3x10-21 m2/s interdiffusion: D=2.4x10-12 m2/s. Factors That Influence Diffusion. Chapter 5 - 16. Diffusion and Temperature. • ...
[11]
Thermal Diffusivity | Concept & Overview - Thermtest
Nov 14, 2019 · Experimental Methods of Finding the Thermal Diffusivity ... k = c m d ( T 3 − T 4 ) A ( T 1 − T 2 ) k = cmd \frac{(T3-T4)}{A(T1-T2)} k=cmdA(T1−T2) ...
[12]
https://cushman.host.dartmouth.edu/courses/engs43/Chapter2.pdf
[13]
[PDF] Transport Properties: Diffusion, Fick's Laws, Random Walks
Feb 16, 2009 · Here we derive Fick's First Law of Diffusion (not really a law; just a mass balance). Now, we need to find the local density at the planes, J−λ ...
[14]
[PDF] Chapter 2: Diffusion
In this chapter, we explore some consequences of the existence of a diffusive flux by concentrating on cases in which there is no mean flow in the system and ...
[15]
[PDF] Lecture 3: Diffusion: Fick's first law
Understand why diffusion can “surprisingly” occur against the concentration gradient? • Learn how to deduce the Fick's first law, and understand the meaning of ...
[16]
[PDF] the brownian movement - DAMTP
(of Lyons)-had been convinced by direct observation that the so-called Brownian motion is caused by the irregular thermal movements the molecules of the liquid.
[17]
Enthalpy-entropy compensation of atomic diffusion originates ... - NIH
Aug 7, 2020 · Experimental data accumulated over more than 120 years show not only that diffusion coefficients of impurities ordinarily obey the Arrhenius law ...
[18]
[PDF] Gaseous Diffusion Coefficients - TR Marrero
The transport properties appear finally in the. Chapman-Enskog theory as solutions of infinite sets of simultaneous algebraic equations, and the transport ...
[19]
Brownian motion, random walks, and the diffusion equation
In this chapter, we begin our discussion of diffusion in solids by developing the diffusion equation. This equation provides the mathematical description of how ...Missing: 3D formula
[20]
In-situ Measurement of Self-Atom Diffusion in Solids Using ... - Nature
Dec 4, 2018 · Self-diffusion can be defined as the thermally activated random migration of atoms in solids. It is a central matter transport process and its ...
[21]
The Self-Diffusion of Copper | Phys. Rev.
The self-diffusion of copper has been measured with a radioactive isotope prepared by neutron bombardment of zinc. The isotope is electroplated in a thin ...
[22]
Self-diffusion in molybdenum - Taylor & Francis Online
Aug 20, 2006 · The radioactive isotope 99Mo has been used to study self-diffusion in polycrystalline and single crystal specimens of molybdenum in the ...
[23]
[PDF] THE ISOTOPE EFFECT
In the self-diffusion isotope effect measurement the tracers are of the same chemical species as the solvent ions, and the correlation factor for diffusion fc.
[24]
A generalization of the nernst-Einstein equation for self-diffusion in ...
We show that the Nernst-Einstein equation can be generalized for a high defect concentration solids solid to relate the mobility or conductivity to the self- ...
[25]
Self-diffusion in molten sodium chloride: a test of the applicability of ...
The results can be represented by the expressions DNa = 8x 10-4 exp (-4000/RT) and DCl = 33 x 10-4 exp (-8500/.RT). These values, when inserted in the Nernst- ...
[26]
Hydrogen in Metals - ScienceDirect.com
For metals, H diffusivity may reach values as known for ions in aqueous solutions. Thus, thermodynamic equilibrium is reached within comparably short times even ...
[27]
[PDF] Lecture 3: Introduction to Diffusion
9). Thermodynamics of diffusion. Fick's first law is empirical in that it assumes a proportionality between the diffusion flux and the concentration gradient.
[28]
Interdiffusion - an overview | ScienceDirect Topics
Interdiffusion is a process of diffusional exchange of atoms across two materials that are in contact. This is driven by the chemical potential gradient across ...
[29]
Thermodynamic interdiffusion coefficient in binary systems with ...
1. A new form of diffusion coefficient termed as thermodynamic interdiffusion coefficient has been introduced. It is defined as: D ̃ (μ) =(N B D A(μ) +N A D B(μ) ...
[30]
L. S. Darken, Transactions of AIME, Vol. 175, 1948, pp. 184-201.
ABSTRACT: The well-known Darken equation has been widely accepted in analyzing interdiffusion problems since 1948. The diffusion researchers have never ...
[31]
The Discovery and Acceptance of the Kirkendall Effect
This article reports how Kirkendall discovered the Kirkendall Effect in a short research career in which he produced only three papers.
[32]
Chemical interdiffusion in binary systems; interface barriers and ...
Dec 20, 2011 · It allows computing the material velocity in the reacting system in which reactions at several moving interfaces occur. All reactions lead to ...
[33]
Mechanisms of boron diffusion in silicon and germanium
Jan 16, 2013 · In this paper, we review a large amount of experimental work and present our current understanding of the B diffusion mechanism.
[34]
Overview of SIMS-Based Experimental Studies of Tracer Diffusion in ...
Sep 23, 2014 · Tracer diffusivities provide the most fundamental information on diffusion in materials, and are the foundation of robust diffusion ...
[35]
Application of Molecular Dynamics Simulations in Molecular ... - NIH
The diffusion coefficient D is related to friction coefficient ξ by Einstein-Smoluchowski equation (Eq. 8). Friction coefficient ξ depends on the sizes and ...
[36]
Tracer diffusion of59Fe and51Cr in Fe-17 Wt Pet Cr-12 Wt Pet Ni ...
The volume and grain-boundary diffusion of 59 Fe and 51 Cr have been studied in an austenitic iron alloy containing 17 wt pct Cr and 12 wt pct Ni.
[37]
Effect of the thermodynamic factor on the intrinsic and tracer ...
Feb 15, 2023 · It was shown that the tracer diffusivity, D A * , behaves as 1 / Γ and that the intrinsic diffusivity, D A , does not depend on Γ . In this ...Missing: distinction | Show results with:distinction
[38]
[PDF] Introduction to Temperature Dependent Behavior, Loading, and Wear
Arrhenius Behavior. • RRRRRRRR = AARR−QQ11/RRRR ... What is diffusivity of H in Fe at 600°C? Cussler, E. L. (1997). Diffusion: Mass ...
[39]
DIFFUSION COEFFICIENT - Thermopedia
As has been noted, diffusion in liquids encounters greater resistance and the diffusion coefficients for liquids lower than 104 to 105 times. One of the ...
[40]
Experimental tests of free-volume tracer diffusion in water and other ...
Mar 12, 2014 · In this work, we wish to investigate the temperature dependence of the tracer diffusion of a solute in a number of different solvents, listed in ...
[41]
[PDF] how do atoms move through solids? Diffusion mechanisms Vacancy ...
Temperature dependence of the diffusion coefficient, follows the Arrhenius dependence. z is coordination number. (number of atoms adjacent to the vacancy).
[42]
Diffusion-melting correlations and the compensation effect in atomic ...
Nov 15, 1994 · ... compensation effect (CEF) associated with the linear dependence ln D 0 =aΔE+C between the Arrhenius activation energy ΔE and preexponential ...Missing: Ea | Show results with:Ea
[43]
(PDF) Diffusion Coefficients and Activation Energies ... - ResearchGate
Oct 15, 2025 · As expected, the activation energies of diffusion EA show a strong dependency from the molecular volume of the investigated substances. At the ...
[44]
Self diffusion in liquid aluminium - IOPscience
Aluminium self-diffusion coefficients follow an Arrhenius law with an activation energy of 280±70 meV. The Sutherland-Einstein equation relating viscosity ...<|control11|><|separator|>
[45]
Concentration Dependence of Diffusion Coefficients In Metallic Solid ...
Dec 7, 2017 · The concentration dependence is slight in both extremes of concentration and strong at intermediate concentrations. The diffusion coefficient ...
[46]
A short review on diffusion coefficients in magnesium alloys and ...
Boltzmann–Matano analysis [74] is a common method for extracting concentration-dependent inter-diffusion coefficients, while some errors are caused due to the ...
[47]
First-principles study of self-diffusion in hcp Mg and Zn - ScienceDirect
Calculated results show that diffusion is faster in the basal plane in Mg and along the normal of the basal plane in Zn, in accordance with the available ...
[48]
Anisotropy of hydrogen diffusion in nickel single crystals - Nature
Mar 22, 2017 · In this paper, we investigate the anisotropy of hydrogen diffusion in pure nickel single crystals using both an experimental approach and a thermodynamic ...
[49]
Self-Diffusion in Amorphous Silicon | Phys. Rev. Lett.
Jan 13, 2016 · In general, diffusion in amorphous materials is expected to yield a lower activation energy compared to its crystalline counterpart, chiefly ...Abstract · Article Text · ACKNOWLEDGMENTS
[50]
Grain Boundary Diffusion - an overview | ScienceDirect Topics
Grain boundary diffusion is defined as the process of solute migration along the grain boundaries of polycrystalline materials, occurring more rapidly than ...
[51]
[PDF] Improving the Calculation of Interdiffusion Coefficients
The Boltzmann-Matano analysis of a dif- fusion profile, which yields the interdiffusion coefficient as a function of composition, has been extensively u ~ e d .
[52]
[PDF] Lecture 7: How to determine the binary interdiffusion coefficient in ...
This is Boltzmann - Matano equation. By measuring c(x) experimentally, we will be able to give ( )c x c. ∂. ∂. (we no longer needs to know c t.Missing: diffusivity | Show results with:diffusivity
[53]
[PDF] Generalization of Matano's Method: Interdiffusion in Solutions with ...
May 12, 2015 · In his method Matano adapted the solution given by Boltzmann, but derived the concentration dependent diffusivity from the data of the ...
[54]
Radioactive tracer technique for molecular diffusion coefficients in ...
H.Y. Sohn. Experimental techniques in fluid-solid reactions. 2020, 479-514. https://doi.org/10.1016/B978-0-12-816466-2.00011-8.Missing: sectioning annealing seminal
[55]
[PDF] Diffusion in Copper and Copper Alloys. Part I. Volume and Surface ...
Oct 29, 2009 · The tracer-sectioning method can yield results which are reproducible to within a few percent. In practice, however, thi!'! accuracy i!'! not ...Missing: seminal papers
[56]
[PDF] Tracer diffusion of Cr in Ni and Ni–22Cr studied by SIMS - HAL
The methods used by most investigators to measure tracer diffusivity involve machining (sectioning or grinding) to generate a depth-concentration profile.<|separator|>
[57]
NMR SPIN-ECHO SELF-DIFFUSION MEASUREMENTS ON FLUIDS ...
N.M.R. SPIN-ECHO SELF-DIFFUSION MEASUREMENTS ON FLUIDS ... Investigations of vesicle gels by pulsed and modulated gradient NMR diffusion techniques.Missing: diffusivity | Show results with:diffusivity
[58]
Copper diffusion in thin In2S3 layers investigated by Rutherford ...
Copper diffusion in thin In2S3 layers was investigated by measuring copper concentration profiles with Rutherford Backscattering Spectroscopy (RBS).
[59]
Chapter 6. Diffusion Couple Technique: A Research Tool in ...
A number of error sources that may appear when the diffusion couple technique is used as "a research tool" are discussed in detail. ResearchGate Logo.
[60]
Understanding Molecular Simulation - ScienceDirect.com
Understanding Molecular Simulation. From Algorithms to Applications. Book • Second Edition • 2002. Authors: Daan Frenkel and Berend Smit. Understanding ...
[61]
[PDF] Introduction to the Kinetic Monte Carlo Method
Monte Carlo refers to a broad class of algorithms that solve problems through the use of random numbers. They first emerged in the late 1940's and 1950's.
[62]
https://www.researchgate.net/publication/316345160_Chapter_6_Diffusion_Couple_Technique_A_Research_Tool_in_Materials_Science
[63]
https://www.sciencedirect.com/book/9780122673511/understanding-molecular-simulation
[64]
A Computationally Based Approach to Homogenizing Advanced ...
Feb 26, 2011 · We have developed a computationally based approach to optimizing the homogenization heat treatment of complex alloys.
[65]
[PDF] Martensitic Stainless Steels, Precipitation Hardening (code 1810)
Martensitic stainless steel (code 1810) is a precipitation-strengthened alloy with high strength, toughness, and corrosion resistance, but is sensitive to ...
[66]
Evaluating Wagner Oxidation Criteria for Protective Al2O3 Scale ...
Overall, the Wagner model shows high sensitivity to diffusivity and oxygen solubility parameters. Further research is needed to identify optimal methods for ...
[67]
[PDF] The Discovery and Acceptance of the Kirkendall Effect
In 1947, Ernest Kirkendall reported the results of experiments on the interdiffusion between copper and zinc in brass and observed the movement of the interface ...
[68]
A comparative study of oxygen diffusion in tissue engineering scaffolds
Jul 14, 2014 · Examples are the oxygen diffusivities in muscle D = 1.7 × 10−9 m2/s, lung D = 2.21 × 10−9 m2/s, brain D = 1.62 × 10−9 m2/s or blood D = 2.10 × ...
[69]
Contributions of Antibiotic Penetration, Oxygen Limitation, and Low ...
The results demonstrated that oxygen penetrated 50 to 90 μm into the biofilm from the air interface. This oxic zone correlated to the region of the biofilm ...
[70]
Diffusion in Biofilms - PMC - NIH
This article discusses the profound influence of the physics of the diffusion process on the chemistry and biology of the biofilm mode of growth.
[71]
Predicting a Drug's Membrane Permeability: A Computational Model ...
Apr 28, 2017 · (21) While the partition coefficient is used to calculate properties such as membrane permeability and water solubility, it also has importance ...
[72]
Drug Distribution. Part 1. Models to Predict Membrane Partitioning
Dec 15, 2016 · Two structure-based models to predict partitioning into microsomal membranes are presented. An orientation-based model was developed using a membrane template.
[73]
Dispatch Protein folding: Does diffusion determine the folding rate?
A major question about protein folding is whether the coming together by diffusion of different segments of the polypeptide chain is rate-determining.
[74]
Coordinate-dependent diffusion in protein folding - PNAS
In such a diffusion model, the rate of folding is determined by the height of the free-energy barrier and a kinetic prefactor that depends on the diffusion ...
[75]
Controls on the persistence of aqueous-phase groundwater ...
Jun 20, 2020 · Back-diffusion is the phenomenon wherein a solute, which may be groundwater contaminant, diffuses from a low-permeability unit (aquitard) to a ...