betaFeedback

WorksPhysics and Astronomy

The shape memory effect and minimal surfaces

Abstract

Martensitic transformations, viewed as continuous transformations between triply periodic minimal surfaces (TPMS), as originally proposed by Hyde and Andersson [Z. Kristallogr. 174, 225 (1986)], is extended to include paths between the initial and final phases. Bravais lattices correspond to particular TPMS whose lattice points are flat points, where the Gaussian curvature vanishes. Reversible transformations, which correspond to shape memory materials, occur only if lattice points remain at flat points on a TPMS throughout a continuous deformation. For the shape memory material NiTi, density-functional theory (DFT) yields irreversible and reversible paths with and without energy barriers, respectively. Although there are TPMS for face-centered gamma-Fe) and body-centered (alpha-Fe) cubic lattices, gamma to alpha deformation paths are not reversible, in agreement with non-vanishing energy barriers obtained from DFT.

Martensitic transformations are diffusionless solid-to-solid transformations triggered by the rapid collective movement of atoms across distances smaller than a lattice spacing [1, 2, 3]. This transformation is abrupt, showing a structural discontinuity at a particular temperature, and is often accompanied by changes in physical properties that can be exploited for applications. Materials exhibiting martensitic transformations include metals and alloys [2], ceramics [4], and biological systems [5].

The shape memory effect (SME) [3, 6], where a material returns to its original shape when heated after a plastic deformation, accompanies some martensitic transformations. Shape memory materials have been known since the 1930s [7], but studies of this effect began in earnest only in the 1960s, with the discovery of NiTi-based materials [8, 9]. Ni-Ti alloys are the most widely used shape memory and superelastic alloys, combining a pronounced SME with superelasticity, corrosion resistance, biocompatibility, and superior engineering properties.

Here, we study martensitic transformations from a geometric perspective with no restriction on point-group-subgroup relations between the two end phases to make our model appropriate for both reconstructive and non-reconstructive transformations. The martensitic transformation is interpreted as a continuous deformation between minimal surfaces, and includes information about the relevant symmetry and topological properties of the two end states. We build on the work of Hyde and Andersson [10] to develop a theory of martensitic transformations that encompasses not just the end-points but also the path between the initial and transformed phases. Transformation paths whose lattice points remain at flat points on surfaces in a space of triply periodic minimal surfaces (TPMS) are expected to be shape memory materials. Our approach provides a purely geometrical distinction between reversible and irreversible transformations.

The local shape near a point of a three-dimensional smooth surface is determined by the principal curvatures k1k_1 and k2k_2 measuring the maximum and minimum bending near that point. The geometry of a surface is characterized by the mean curvature M=−12(k1+k2)M=-\frac{1}{2}(k_1+k_2) and the Gaussian curvature K=k1k2K=k_1k_2 [11, 12]. A surface with K=0K=0 everywhere is flat, e.g. a plane. A surface with M=0M=0 everywhere is locally area-minimizing and is called minimal. Soap films, the catenoid (the surface of revolution of a catenary), and the plane are standard examples of minimal surfaces. Minimal surfaces with repeating units in three independent directions are called TPMS. Minimal surfaces can have points where K=0K=0; these are called flat points, and will play a central role in what follows. Flat points on the minimal surfaces considered here are isolated.

TPMS can have space groups, point groups, just as ordinary crystals [13]. The first TPMS were discovered by Schwarz [14], including the primitive (PP), diamond (DD), and hexagonal (HH) surfaces. Schoen [15, 16] discovered the gyroid (GG), which is an intermediate structure when deforming PP to DD surfaces along the Bonnet path and is locally isometric to both the Schwarz PP and DD surfaces [15, 17]. The rPDrPD family discovered by Schoen [15] and Karcher [17] is a set of TPMS connecting the DD and PP surface other than by the Bonnet transformation. Additional structural details of these surfaces are shown in the Supplementary Information. Since the point group of a lattice should be a subgroup, both proper and improper, of that of a surface [18], we require the point group of the crystal to be a subgroup (proper and improper) of that of of the TPMS at two end states to ensure the conformity between the surface and the crystal. While for intermediate states, because they are unstable, the group-subgroup relation, which may act as a main role in the geometric explanation to the twinning mechanism[19], between the surface and the lattice is not strictly required.

The austenite (B2) phase of NiTi has the space group Pm3ˉmPm\bar{3}m. After a rapid quench, NiTi transforms to martensite. DFT calculations [20, 21, 22] confirmed that there exist barrierless transformation paths of NiTi from the B2 phase to its martensite phases. We study NiTi structures along transformation paths by comparing the calculated and real ratios a/ca/c and a/ba/b, where aa, bb, and cc are lattice parameters, and unit cell angles α\alpha, β\beta, γ\gamma to ascertain if they can meet the requirements mentioned above. The calculated result is obtained by first constructing a “unit cell” from flat points representing lattice points rather than actual atoms of a multi-atom basis on the corresponding TPMS, then compute angles and ratios between cell edges. The actual lattice parameters can be obtained by a uniform scaling factor once the ratios are given. According to results listed in the Supplementary Information, all initial and final states calculated by DFT of shape-memory NiTi correspond to a TPMS.

DFT calculations [20] also show that energy barriers vary with the paths of the B2→B19′\mathrm{B2}\to\mathrm{B19'} NiTi martensitic transformation, with the energy barrier going to zero when a path meets our geometrical criteria. Paths involving only the monolayer (resp., bilayer) ⟨100⟩{011}\langle100\rangle\{011\} shear/shuffle have an energy barrier up to 37 meV/atom (resp., 22 meV/atom), while the path that admits the intermediate phase 109∘109^\circ–B19′\mathrm{B19'}, allowing both shear and structural relaxation, has no barrier, with the energy per atom decreasing throughout the transformation [20]. Paths such as B2→109∘–B19′→B19′\mathrm{B2}\to109^\circ\text{–}\mathrm{B19'}\to\mathrm{B19'} correspond to a continuous deformation of the initial surface, where each surface along the path a TPMS (formally, in the oPb/oPa family [25] [24]) and is, therefore, reversible. But for paths that allow only the bilayer ⟨100⟩{011}\langle100\rangle\{011\} shear, as shown in Figure 1(c), the coexistence of the B19′\mathrm{B19'} lattice and another lattice with γ=119∘\gamma=119^\circ cannot form with flat points on the same TPMS. In other words, to describe the lattice in Figure 1(c), there is a discontinuous transition surface consisting of both pieces of the minimal surfaces shown in Figure 1(e,Figure 1f). Thus, the surfaces are either non-minimal, or lattice points on a minimal surface depart from the flat points. Hence, according to our model, an energy barrier is obtained.

Deformation from B2 to B19′ NiTi by bilayer shear

Figure 1. Deformation from B2 to B19′ NiTi by bilayer ⟨100⟩{011}\langle100\rangle\{011\} shear. (a) Unit cell of NiTi. (b) View of undeformed B2 NiTi from the [011ˉ][01\bar{1}] direction. B2 deformed by bilayer ⟨100⟩{011}\langle100\rangle\{011\} shear, with (c) γ=98∘\gamma= 98^\circ for B19′ and (d) γ=109∘\gamma= 109^\circ for 109°–B19′. Red dashed lines outline the unit cell and blue lines label the bilayers. (e) and (f) Minimal surfaces corresponding to the B19′ and 109°–B19′ lattices, drawn with the package Mesh [23], with white spheres labelling lattice points (not atoms).

The best known and technologically most important martensite transformations are found in iron and steel [26]. When cooled rapidly, high-temperature austenite γ\gamma–Fe (FCC) transforms to martensite α\alpha–Fe (BCC). Explanations for this transformation include the celebrated Bain tetragonal distortion [27], the Kurdjumow–Sachs (KS) [28] and Nishiyama–Wassermann (NW) [29, 30] shear models, together with the Bogers–Burgers (BB) [31] and the Olson–Cohen (OC) [32] hard-sphere models. Where energies along transformation paths were calculated [33, 34], a barrier of approximately 20 meV/atom was found for the γ→α\gamma\to\alpha transformation of Fe.

The geometries for the Bain, NW and KS transformation paths connecting the FCC and BCC lattices are shown in Figure 2, while the BB (OC) path is shown in Figure 3(a,Figure 3b,Figure 3c). The Bain path Figure 2(a,Figure 2b) is based on a deformation along ⟨001⟩\langle001\rangle, which yields a bodycentered tetragonal (BCT) structure. The BCT unit cell is compressed by about 21% along the zz-direction and expanded by about 12% along the xx- and yy-directions [35]. Despite its simplicity and intuitive appeal, the orientation relationship between the parent (FCC) and daughter (BCC) phases has not been observed in experiments. The NW path Figure 2(c,Figure 2d) begins with a shear that produces a monoclinic unit cell. Along this path, the unit cell changes from orthogonal to monoclinic and back to orthogonal. Computational studies [34, 35] indicate this path is favored over the Bain path, though there is still a barrier for the transformation. The KS shear model Figure 2(e,Figure 2f), similar to the NW model, starts with a shear that produces a monoclinic cell, but has different orientational relationships. The BB (OC) hard-sphere model indicates the FCC-BCC lattice transformation is realized by a shear of type ⟨112⟩{111}\langle112\rangle\{111\} [31, 32].

Atomic arrangements of the Bain, NW, and KS transformation paths

Figure 2. Atomic arrangements of (a) the Bain transformation path from the FCC to the BCC phase with an intermediate BCT unit cell (shown shaded), which is isolated in (b), (c) the NW path, with an intermediate body-centered triclinic unit cell (shown shaded), which is isolated in (d), (e) the KS transformation path, with an intermediate body-centered triclinic cell, as isolated in (f). Red and blue spheres indicate atoms in the parent and transformed phases, respectively.

Sketches and ball-and-stick models of lattices on D, G, and P surfaces and along the $rPD$ path

Figure 3. (a,b,c) Sketches of the BB (OC) hard-sphere model showing the stacking of {111}\{111\} planes. (a) Regular tetrahedron in the FCC lattice on the DD surface (d) and (j), and the irregular tetrahedron in (c) residing on either the gyroid (e) or the PP surface (f) and (l). (g,h,i) Ball-and-stick models of lattices formed on DD, GG and PP surfaces, respectively. Spheres label lattice points that coincide with flat points. Yellow spheres are lattice points forming the vertices of tetrahedrons sketched by green lines. (j,k,l) depicts the rPDrPD path, with the area shading purple in (j) and (l) being (d) and (f), respectively. Formation of surfaces shown here are given in the Supplementary Information. (m,n,o) Ball-and-stick models of lattices formed along the rPDrPD path with the FCC lattice (m) on the DD surface and the BCC lattice (o) on the PP surface being two end states. (k) A representative of the rPDrPD-family along the path, with the lattice shown in (n). Panels (d,e,f) and (j,k,l) were drawn with the package Mesh [23].

The Bonnet and rPD-family of transformations are the only known continuous transformations that map the minimal surfaces for the BCC and FCC lattices to each other within the space of TPMS [10, 15, 17, 36]. The BCC lattice can form on Schoen’s gyroid during the Bonnet transformation Figure 3(d,Figure 3e,Figure 3f). This is an intermediate state, but due to the incompatibility of point group symmetry, this lattice cannot be stable on the GG surface. However, we can still expect the appearance of a BCC lattice during the Bonnet transformation. Whereas the rPDrPD path involves a vertical stretching of the surface [Fig. 3(j,k,l)] and therefore involves the evolution of a tetrahedron with three faces of isosceles triangles and one face an equilateral triangle [Fig. 3(m,n,o)]. Because no models mentioned above allow the appearance of a BCC lattice twice to conform to the

Bonnet transformation, or the preservation of a tetrahedron required by the rPDrPD path described above then, according to our model, none of them are reversible. Currently, all γ→α\gamma\to\alpha iron transformation paths [28, 31, 32, 37] and relevant molecular dynamic calculations [38] are developed with the models discussed above, and none of them show geometric conformity to the Bonnet/rPDrPD path, yielding non-vanishing energy barriers [39]. Accordingly, we infer that the γ→α\gamma\to\alpha iron martensitic phase transformation is geometrically irreversible; hence, no SME can take place.

Table I provides energy barriers calculated with DFT of materials undergoing martensitic transformations along several paths. As explained above, NiTi deforms according to, for instance, the ⟨110⟩{011}\langle110\rangle\{011\} shear and Fe deforms along the Bain or NW path do not satisfy the geometric requirements for the existence of a non-vanishing energy barrier, which pre-empts those paths exhibiting the SME. Paths with a vanishing energy barrier conform to the geometric requirements we propose for the SME. More details of the paths in Table I are given in the Supplementary Information.

MaterialInitialPathFinalΔE\Delta E
NiTi [20]B2⟨100⟩{011}\langle100\rangle\{011\} shear (basal shear)B19′∼37\sim37
B2⟨100⟩{011}\langle100\rangle\{011\} shear (bilayer shear)B19′∼22\sim22
B2109∘109^{\circ}–B19′B19′0
NiTi [20, 40]B2B19B19′0
NiTi [41]B2[111] elongationR0
NiTi [20, 21, 22, 42]B2⟨100⟩{011}\langle100\rangle\{011\} (bilayer shear and shuffle)B330
NiTi [43]B2⟨1ˉ10⟩{110}\langle\bar{1}10\rangle\{110\} shearB19∼13\sim13
Fe [33, 34]γ\gamma-FeBain (FCT)α\alpha-Fe∼20\sim20
γ\gamma-FeBainα\alpha-Fe∼45\sim45
Fe [34]γ\gamma-FeNishiyama–Wassermann (FCT)α\alpha-Fe∼37\sim37
Fe [44]γ\gamma-FeBogers–Burgersα\alpha-Fe∼25\sim25

Table I. Calculated transformation paths of NiTi and Fe from the indicated initial and final phases. Energy barriers ΔE\Delta E are in units of meV/atom.

The details of the atomic-scale mechanisms responsible for martensitic transformations are important for a quantitative understanding of the transformation processes, including the calculation of energy barriers. Our approach, based on the transformations between minimal surfaces provides a conceptually simple description based on geometrical and topological principles. As with transformation paths such as proposed by Bain and NW, and according to the group-subgroup relation [45], no attempt is made to model the nucleation process, just to establish a relationship between the initial and final phases and, in our case, the intermediate states as well.

In summary, we have studied the martensitic transformation and the SME from the perspective of the differential geometry of TPMS. We propose that martensitic transformations can be understood as continuous deformations between TPMS with the Bravais lattices of two end phases being at flat points, respectively. If a martensitic transformation continues within a space of TPMS, and all lattice points remain at flat points throughout the progress, we expect the occurrence of the SME. This method is relatively simple, general and less time-consuming in comparison to DFT, and thus can be a shortcut for the discovery of more novel shape memory alloys.

References

4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

References

  1. [1]G. B. Olson and M. Cohen. A perspective on martensitic nucleation. Ann. Rev. Mater. Sci., 11:1–30, 1981.DOI
  2. [2]J. W. Christian. The Theory of Transformations in Metals and Alloys. Pergamon, Amsterdam, 3 edition, 2002.
  3. [3]K. Battacharya. Microstructure of Martensite: Why It Forms, and How It Gives Rise to the Shape-Memory Effect. Oxford University Press, New York, NY, 2003.
  4. [4]P. M. Kelly and L. R. F. Rose. The martensitic transformation in ceramics – its role in transformation toughening. Prog. Mater. Sci., 47:463–557, 2002.
  5. [5]G. B. Olson. New directions in martensite theory. Mater. Sci. Eng. A, 273–275:11–20, 1999.
  6. [6]K. Tadaki, K. Otsuka, and K. Shimizu. Shape memory alloys. Ann. Rev. Mater. Sci., 18:25–45, 1998.
  7. [7]A. Ölander. An electrochemical investigation of solid cadmium-gold alloys. J. Am. Chem. Soc., 54:3819–3833, 1932.DOI
  8. [8]W. J. Buehler, J. V. Gilfrich, and Wiley. Effect of low-temperature phase changes on the mechanical properties of alloys near composition tini. J. Appl. Phys., 34:1475–1477, 1963.
  9. [9]G. B. Kauffman and I. Mayo. The story of Nitinol: The serendipitous discovery of the memory metal and its applications. Chem. Educator, 2:1–21, 1997.
  10. [10]S. T. Hyde and S. Andersson. The martensite transition and differential geometry. Z. Kristallogr., 174:225–236, 1986.
  11. [11]A. Gray. Modern Differential Geometry of Curves and Surfaces with Mathematica. CRC Press, Boca Raton, FL, 2nd edition, 1997.
  12. [12]B. O’Neill. Elementary Differential Geometry. Elsevier, Amsterdam, Rev. 2nd edition, 2006.
  13. [13]E. Koch and W. Fischer. A crystallographic approach to 3-periodic minimal surfaces. In H. T. Davis and J. C. C. Nitsche, editors, Statistical Thermodynamics and Differential Geometry of Microstructured Materials, chapter 3, pages 14–48. Springer–Verlag, New York, 1993.
  14. [14]H. A. Schwarz. Gesammelte Mathematische Abhandlungen. Springer, Berlin, 1890.
  15. [15]A. H. Schoen. Infinite periodic minimal surfaces without self-intersections. Technical Report TN D-5541, NASA, Washington, DC, 1970.
  16. [16]A. H. Schoen. Reflections concerning triply-periodic minimal surfaces. Interface Focus, 2:658–668, 2012.
  17. [17]H. Karcher. The triply periodic minimal surfaces of alan schoen and their constant mean curvature companions. Manuscripta Math., 64(3):291–357, 1989.
  18. [18]P. Kocian, K. Schenk, and G. Chapuis. Differential geometry: a natural tool for describing symmetry operations. Acta Cryst., A65(5):329–341, 2009.
  19. [19]H. Chen. Minimal twin surfaces. Exp. Math., 28(4):404–419, 2019.
  20. [20]N. Hatcher, O. Yu. Kontsevoi, and A. J. Freeman. Role of elastic and shear stabilities in the martensitic transformation path of niti. Phys. Rev. B, 80:144203, 2009.
  21. [21]M. Krcmar, G. D. Samolyuk, and J. R. Morris. Stacking faults and alternate crystal structures for the shape-memory alloy niti. Phys. Rev. M, 4:103606, 2020.
  22. [22]J. R. Morris, Y. Ye, M. Krcmar, and C. L. Fu. The role of phase stability in ductile, ordered B2 intermetallics. Acta. Cryst. A, 980:610, 2006.
  23. [23]M. Weber. Minimal surfaces, 2018. This work is licensed under the Creative Commons Attribution A. Fogden and S. T. Hyde. Parametrization of triply periodic minimal surfaces. ii. regular class solutions. Acta. Cryst. A, 48:575–591, 1992.
  24. [25]H. Chen and M. Weber. An orthorhombic deformation family of schwarz’ h surfaces. Trans. Amer. Math. Soc, 374:2057–2078, 2021.
  25. [26]E. Pereloma and V. Edmonds, editors. Phase transformations in steels: Diffusionless transformations, high strength steels, modelling and advanced analytical techniques., volume 2. Elsevier, 1st edition, 2012.
  26. [27]E. C. Bain and N. Y. Dunkirk. The nature of martensite. Trans. AIME, 70:25–47, 1924.
  27. [28]G. Kurdjumow and G. Sachs. über den mechanismus der stahlhärtung. Zeit. f. Phys., 64:325–343, 1930.
  28. [29]Z. Nishiyama. Martensitic Transformation. Academic Press, New York, NY, 1978.
  29. [30]G. Wassermann. Einfluß der α-γ-umwandlung eines irreversiblen nickelstahls auf kristallorientierung und zugfestigkeit. Archiv für das Eisenhüttenwesen, 6:347–351, 1933.
  30. [31]A. J. Bogers and W. G. Burgers. Partial dislocations on the {110} planes in the b.c.c. lattice and the transition of the f.c.c. into the b.c.c. lattice. Acta Metall., 12:255–261, 1964.
  31. [32]G. Olson and M. Cohen. A mechanism for the strain-induced nucleation of martensitic transformations. J. Less Common Met., 28:107–118, 1972.
  32. [33]S. V. Okatov, A. R. Kuznetsov, Yu. N. Gornostyrev, V. N. Urtsev, and M. I. Katsnelson. Effect of magnetic state on the γ-α transition in iron: first-principles calculations of the bain transformation path. Phys. Rev. B, 79:084111, 2009.
  33. [34]L. H. Zhang, M. J. Cheng, X. H. Shi, J. W. Shuai, and Z. Z. Zhu. Bain and nishiyama–wassermann transition path separation in the martensitic transitions of fe. RSC Adv., 11:3043–3048, 2021.
  34. [35]L. Sandoval, H. M. Urbaßek, and P. Entel. The bain versus nishiyama–wassermann path in the martensitic transformation of fe. New J. Phys., 11:103027, 2009.
  35. [36]A. Fogden. Parametrization of triply periodic minimal surfaces. iii. general algorithm and specific examples for the irregular class. Acta. Cryst. Sect. A, 49(3):409–421, 1993.
  36. [37]C. Cayron. Continuous atomic displacements and lattice distortion during fcc–bcc martensitic transformation. Acta Mater., 96:189–202, 2015.
  37. [38]S. Karewar, J. Sietsma, and M.J. Santofimia. Effect of pre-existing defects in the parent fcc phase on atomistic mechanisms during the martensitic transformation in pure fe: A molecular dynamics study. Acta Mater., 142:71–81, 2018.
  38. [39]J. Meiser and H. M. Urbas̆sek. α ↔ γ phase transformation in iron: comparative study of the influence of the interatomic interaction potential. Modelling Simul. Mater. Sci. Eng., 28:055011, 2020.
  39. [40]K. G. Vishnu and A. Strachan. Phase stability and transformations in niti from density functional theory calculations. Acta Mater., 58(3):745–752, 2010.
  40. [41]J. G. Niu and W. T. Geng. Anti-precursor effect of Fe on martensitic transformation in TiNi alloys. Acta Mater., 104:18–24, 2016.
  41. [42]N. Hatcher, O. Yu. Kontsevoi, and A. J. Freeman. Martensitic transformation path of niti. Phys. Rev. B, 79:020202 (R), 2009.
  42. [43]S. Kibey, H. Sehitoglu, and D. D. Johnson. Energy landscape for martensitic phase transformation in shape memory NiTi. MRS Online Proceedings Library, 57:1624–1629, 2009.
  43. [44]T. L. Wang, J. L. Du, S. Z. Wei, and F. Liu. Ab-initio investigation for the microscopic thermodynamics and kinetics of martensitic transformation. Prog. Nat. Sci. Mater., 31:121, 2021.
  44. [45]K. Bhattacharya, S. Conti, G. Zanzotto, and J. Zimmer. Crystal symmetry and the reversibility of martensitic transformations. Nature, 428:55–59, 2004.

Paper details

vlnc:W_g4a5e5jg
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