Mechanical Behavior · Strengthening mechanisms

Strengthening mechanisms: why alloys get stronger

Every way to make a metal stronger puts something in the way of moving dislocations (line defects in the crystal). Start with the grain size slider in the first lab, then add particles, solute atoms and dislocations, and watch the total at the bottom grow.

Base metal

Why do smaller grains make a metal stronger? (Hall-Petch)

Grain map. Each colour is one grain. In the middle grain, a source sends dislocations along a slip plane; they pile up at the boundary.
Yield strength against d-1/2. Straight lines are Hall-Petch. Top axis: grain size. Shaded: grains below about 20 nm, where the law breaks.
Pile-up animation

σy = σ0 + k d-1/2

σ0 (friction stress)
k (Hall-Petch slope)
boundary term k d-1/2
yield strength σy
dislocations in the pile-up
stress at the pile-up head
atoms in grain boundaries
Try it: drag the grain size from 100 µm down to 1 µm. The pile-up shrinks from dozens of dislocations to a handful and the yield strength rises. Keep going below 20 nm: fewer than one dislocation fits, and the point enters the shaded region where measured strength stops rising.

How do precipitates block dislocations? Cutting or Orowan looping

Looking down on the slip plane. Blue: the dislocation line. Small particles are sheared (cut). Large ones make the line bow round them and leave a loop behind.
Strength against particle radius (log axis). The weaker path wins, so the dark curve is the lower of the two. Top axis: aging time at the chosen coarsening rate.
Animation
Loops and shear steps
spacing between particles λ
cutting: M Δτcut
Orowan: M ΔτOr
strength added Δσp
peak: radius and time
passes (loops or shears)
Try it: watch one pass with the small particles, then drag the aging time up. The radius grows, the line switches from cutting through the particles to looping round them, and the loops start to pile up around each particle. The strength is highest at the crossover: peak-aged. For coarsening itself, see the coarsening lab.

How do solute atoms strengthen a metal?

A line meeting solute atoms (grey: host atoms, coral: solute atoms, ringed: holding the line now). The line stops at a solute atom and bows until the pull is larger than the atom can hold.
Strength added against concentration. Fleischer grows as c1/2, Labusch as c2/3. Bold: the law used in the total below.
Law used in the total
Animation
Fleischer: M G ε3/2 c1/2 / 700
Labusch: M G ε4/3 c2/3 / 550
strength added Δσss
solute atoms holding the line
mean spacing between them
Try it: set c to 0: the line runs straight through. Raise c to 10%: more solute atoms hold the line and the pinning points get closer. Then raise ε to 1: each solute atom holds harder, so the line bows much deeper before it breaks free.

Why does a metal get stronger as you deform it? (Taylor hardening)

Dislocations in a 100 nm thick foil, 4 µm wide, as a TEM would show them. Line count = density × volume ÷ mean segment length.
Taylor stress against density (log-log). Slope 1/2 because strength follows √ρ.
Picture

Δσρ = M α G b √ρ

mean spacing 1/√ρ
lines in the picture
shear stress α G b √ρ
strength added Δσρ
Try it: move ρ from 1012 (annealed) to 1014 m-2 (cold worked). Density rises 100 times, the spacing between lines falls 10 times, and the stress rises 10 times. Switch to weak-beam to see thin bright lines, as in a weak-beam dark-field image.

How do the contributions add up?

Both sums of the numbers set in the four labs above, for the chosen base metal. The two bars differ only in how the precipitate and dislocation terms combine.
Sum rule
friction stress σ0
grain boundaries
solid solution
precipitates
dislocations
yield strength (chosen rule)
hardness, about 3 σy
Try it: set the precipitates near their peak and raise ρ to 1014 m-2. When the two terms are similar, the root-sum-square bar is clearly shorter than the linear one. Make one of them tiny and the two bars meet. The hardness estimate is what an indentation test would read.

What to remember

One idea, four waysGrain boundaries, solute atoms, particles and other dislocations all stop moving dislocations. Strength is the stress needed to push them past.
Hall-Petch has a floorσy = σ0 + k d-1/2 works while a grain can hold a pile-up. Below about 10 to 20 nm it cannot, and strength stops rising.
Peak aging is a crossoverSmall particles are cut and get stronger as they grow. Large ones are looped and get weaker as they spread apart. The peak is where the two meet.
Adding is not simpleDifferent obstacles of similar strength add as a root-sum-square, not linearly. The linear sum is an upper estimate.
More detail: the equations, the numbers and where the models stop

Hall-Petch

σy = σ0 + k d-1/2. Values for recrystallized pure metals at room temperature, from Hansen (2004), Table 1: Cu k = 0.14, Ni k = 0.16, Al k = 0.04 MPa m1/2, with σ0 about 20 MPa for all three. Mild steel (lower yield stress): σ0 = 70 MPa, k = 0.74 MPa m1/2 (Smith and Hashemi 2006; the same k as Petch 1953). Real values change with purity, strain and test method (Cordero, Knight and Schuh 2016). Here d is the equivalent circle diameter of a grain.

The pile-up count uses the edge dislocation result for a pile-up of length L = d/2: n = π(1-ν) L τeff / (G b), with τeff = k d-1/2/M the shear stress above friction. Then n τeff, the stress at the head, does not depend on d. That is the Hall-Petch condition: the neighbour yields when the head stress reaches a fixed value. The dislocation spacing in the picture is schematic.

Below about 10 to 20 nm, n falls below one, boundaries hold a large share of the atoms (about 3δ/d, with boundary width δ = 0.5 nm) and grain boundary sliding takes over. Simulations of Cu put the strongest grain size at 10 to 15 nm (Schiøtz and Jacobsen 2003). The shaded region is where the line should not be used.

Precipitates

Cutting: weak pair coupling of ordered particles, leading term, Δτcut = (γ/2b) [3π2γ f r / (32 Γ)]1/2, with line tension Γ = G b2/2 (Ardell 1985). It rises as (f r)1/2. Particles that are not ordered resist cutting through misfit, modulus or new surface; γ stands in for all of them here.

Looping: Orowan-Ashby, ΔτOr = 0.4 G b ln(2r̄/b) / (π √(1-ν) λ), with mean planar radius r̄ = √(2/3) r and spacing λ = 2r̄ (√(π/4f) - 1), as used by Seidman, Marquis and Dunand (2002). It falls about as 1/λ.

Aging: the radius grows by coarsening, r3 - r03 = K t (Lifshitz, Slyozov and Wagner), with r0 = 0.5 nm and K = 10 nm3/h. That K is chosen only to put the peak in the range of hours; the real K depends on temperature, diffusivity and interface energy. f is held fixed, so this is the coarsening stage after precipitation is complete. M = 3.06 (FCC) or 2.75 (BCC) turns shear stress into tensile stress.

Solid solution

Fleischer (1963): strong, sparse obstacles, Δτ = G ε3/2 c1/2/700. Labusch (1970): many weak obstacles acting together, Δτ = G ε4/3 c2/3/550. ε combines the size misfit (how much a solute atom stretches the lattice) and the modulus misfit (how much stiffer or softer it is). The prefactors 700 and 550 are the fitted constants given in textbooks (Courtney 2000); treat the result as good to a factor of about 2. c is the atom fraction.

The moving picture is a line with tension that stops at solute atoms and breaks away when the pull on the atom exceeds its strength, which grows with ε. The push on the line is set a little above the Friedel critical stress for the current c and ε. That is the Fleischer picture. Each grey dot is one atom and the line moves in steps of one atom spacing b.

Work hardening

Taylor (1934): τ = α G b √ρ, α between about 0.2 and 0.5. Annealed metals have ρ of 1010 to 1012 m-2; heavily cold-worked metals 1014 to 1016 m-2 (Hull and Bacon 2011). G and b: Hansen (2004), Table 1, for Cu (45 GPa), Al (26 GPa) and Ni (79 GPa); G = 83 GPa for steel (Callister, Table 6.1); b = 0.248 nm for ferrite. These are shear moduli of a polycrystal with random grains. The dislocation and creep labs use a different quantity, the modulus for a single dislocation in the real crystal (Frost and Ashby), which is a little lower: 42.1 GPa for Cu, 25.4 for Al, 78.9 for Ni.

Adding up

Linear: σy = σ0 + σgb + σss + σp + σρ. Root-sum-square: the precipitate and dislocation terms, which act as obstacles of similar strength on the same slip plane, combine as √(σp2 + σρ2) (Kocks, Argon and Ashby 1975). Real alloys fall between the two. σ0 from a Hall-Petch fit already holds some solute and dislocation strength, so the sum can count a little twice. Hardness about 3 σy is Tabor's rule for metals that do not harden much.

On this site: Dislocations and Burgers vectors · Schmid factor and slip systems (where M comes from) · Reading a tensile curve · Creep and deformation maps · Fatigue and S-N curves · Nucleation and coarsening lab · EBSD: KAM and GND density · What indentation tells you beyond hardness · Weak-beam dark field · Two-beam defect simulator · Grain boundaries and the CSL

Questions people ask

What is the Hall-Petch relationship?

It says yield strength rises as grains get smaller: σy = σ0 + k d-1/2. Grain boundaries stop dislocations. Smaller grains hold shorter pile-ups, which push less hard on the next grain, so a larger applied stress is needed.

Why does Hall-Petch break down for nanocrystalline metals?

Below about 10 to 20 nm a grain is too small to hold even one dislocation in a pile-up, and a large share of the atoms sit in boundaries. Deformation moves to the boundaries themselves (sliding and rotation), so strength stops rising and can fall.

What is Orowan looping?

When particles are too strong to cut, the dislocation bows between them until the bows meet behind each particle. The line moves on and leaves a loop around every particle. The stress needed is about G b divided by the particle spacing.

Why does over-aging make an alloy weaker?

With longer aging the particles grow and their number falls, so they sit further apart. Past the peak they are looped, not cut, and the looping stress falls as the spacing grows.

What is solid solution strengthening?

Solute atoms that are larger, smaller, stiffer or softer than the host atoms distort the lattice around them. A dislocation line is held where it meets them. More solute and a larger misfit give more strength, roughly as c1/2 (Fleischer) or c2/3 (Labusch).

Why does cold working make a metal stronger?

Plastic deformation multiplies dislocations, from about 1012 to 1015 m-2 or more. Moving dislocations must cut through the others, and the stress to do so grows as the square root of the density (Taylor).

How do you add strengthening contributions together?

Friction, grain boundary and solid solution terms are usually added. Obstacles of similar strength on the same slip plane, such as precipitates and forest dislocations, add closer to the square root of the sum of squares. Measure the total and compare.

How do you measure dislocation density?

Count line length per volume in a TEM foil of known thickness, or use X-ray peak broadening. EBSD gives the geometrically necessary part from lattice curvature; see the KAM and GND page.

References

Show the 14 references
  1. E. O. Hall, The deformation and ageing of mild steel: III, discussion of results, Proceedings of the Physical Society B 64, 747-753 (1951). doi:10.1088/0370-1301/64/9/303
  2. N. J. Petch, The cleavage strength of polycrystals, Journal of the Iron and Steel Institute 174, 25-28 (1953).
  3. N. Hansen, Hall-Petch relation and boundary strengthening, Scripta Materialia 51, 801-806 (2004), Table 1. doi:10.1016/j.scriptamat.2004.06.002
  4. Z. C. Cordero, B. E. Knight and C. A. Schuh, Six decades of the Hall-Petch effect: a survey of grain-size strengthening studies on pure metals, International Materials Reviews 61, 495-512 (2016). doi:10.1080/09506608.2016.1191808
  5. W. F. Smith and J. Hashemi, Foundations of Materials Science and Engineering, 4th ed., McGraw-Hill (2006), p. 242 (Hall-Petch constants).
  6. J. Schiøtz and K. W. Jacobsen, A maximum in the strength of nanocrystalline copper, Science 301, 1357-1359 (2003). doi:10.1126/science.1086636
  7. J. D. Eshelby, F. C. Frank and F. R. N. Nabarro, The equilibrium of linear arrays of dislocations, Philosophical Magazine 42, 351-364 (1951). doi:10.1080/14786445108561060
  8. A. J. Ardell, Precipitation hardening, Metallurgical Transactions A 16, 2131-2165 (1985). doi:10.1007/BF02670416
  9. D. N. Seidman, E. A. Marquis and D. C. Dunand, Precipitation strengthening at ambient and elevated temperatures of heat-treatable Al(Sc) alloys, Acta Materialia 50, 4021-4035 (2002). doi:10.1016/S1359-6454(02)00201-X
  10. R. L. Fleischer, Substitutional solution hardening, Acta Metallurgica 11, 203-209 (1963). doi:10.1016/0001-6160(63)90213-X
  11. R. Labusch, A statistical theory of solid solution hardening, physica status solidi (b) 41, 659-669 (1970). doi:10.1002/pssb.19700410221
  12. G. I. Taylor, The mechanism of plastic deformation of crystals, part I, Proceedings of the Royal Society A 145, 362-387 (1934). doi:10.1098/rspa.1934.0106
  13. U. F. Kocks, A. S. Argon and M. F. Ashby, Thermodynamics and kinetics of slip, Progress in Materials Science 19, 1-291 (1975) (superposition of obstacle strengths).
  14. D. Hull and D. J. Bacon, Introduction to Dislocations, 5th ed., Elsevier (2011), chapters 9 and 10; T. H. Courtney, Mechanical Behavior of Materials, 2nd ed., McGraw-Hill (2000), chapters 4 and 5; W. D. Callister and D. G. Rethwisch, Materials Science and Engineering: An Introduction, Wiley, Table 6.1 (elastic constants).
Cite this page: Tripathy, Manisha. “Strengthening Mechanisms in Metals.” untethered atom, 2026, https://untetheredatom.com/mechanical-behavior/strengthening-mechanisms.
BibTeX 
@misc{tripathy2026strengthening,
  author = {Tripathy, Manisha},
  title  = {Strengthening Mechanisms in Metals},
  year   = {2026},
  howpublished = {\url{https://untetheredatom.com/mechanical-behavior/strengthening-mechanisms}},
  note   = {Interactive web tool}
}
Last updated 24 September 2026.