Microstructure Evolution and Mechanical Properties of Austenite Stainless Steel with Gradient Twinned Structure by Surface Mechanical Attrition Treatment

nanomaterials

Article Microstructure Evolution and Mechanical Properties of Austenite Stainless Steel with Gradient Twinned Structure by Surface Mechanical Attrition Treatment

Aiying Chen 1,* , Chen Wang 1, Jungan Jiang 1, Haihui Ruan 2 and Jian Lu 3

1 School of Material Science & Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China; [email protected] (C.W.); [email protected] (J.J.) 2 Department of Mechanical Engineering, The Hong Kong Polytechnic University, Hong Kong 810005, China; [email protected] 3 Department of Mechanical and Biomedical Engineering, City University of Hong Kong, Hong Kong 999077, China; [email protected] * Correspondence: [email protected]; Tel.: +86-21-55270632

Abstract: Gradient structures in engineering materials produce an impressive synergy of strength and plasticity, thereafter, have recently attracted extensive attention in the material families. Gradient structured stainless steels (SS) were prepared by surface mechanical attrition treatment (SMAT) with different impacting velocities. The microstructures of the treated samples are characterized by gradient twin fraction and phase constituents. Quantitative relations of gradient microstructure with

impacting time and mechanical properties are analyzed according to the observations of SEM, TEM, XRD, and tests of mechanical property. The processed SSs exhibited to be simultaneously stiff, strong,

Citation: Chen, A.; Wang, C.; Jiang, and ductile, which can be attributed to the co-operation of the different spatial distributions of multi- J.; Ruan, H.; Lu, J. Microstructure scaled structures. The formation of gradient twinned structure is resolved and the strengthening by Evolution and Mechanical Properties gradient structure is explored. of Austenite Stainless Steel with Gradient Twinned Structure by Keywords: austenitic stainless steel; surface mechanical attrition treatment; multi-scaled twin; Surface Mechanical Attrition mechanical property; strengthening Treatment. Nanomaterials 2021, 11, 1624. https://doi.org/10.3390/ nano11061624 1. Introduction Academic Editor: Yang-Tse Cheng Austenitic stainless steels (SS) have been widely developed in various engineering applications because of their excellent workability, corrosion resistance, superior low- Received: 3 June 2021 temperature toughness, etc. [1,2]. However, the low strength of the austenitic SSs is a Accepted: 15 June 2021 Published: 21 June 2021 dominant drawback in the ﬁelds of commercial reactor, vessel internals, and coolant piping [3–5]. Conventionally, austenitic SSs cannot be hardened by heat treatment, but

Publisher’s Note: MDPI stays neutral by cold working, which results in a significant increase of dislocation density, and strain- with regard to jurisdictional claims in induced martensite transformation in some types of austenitic SSs [6,7]. The transformation published maps and institutional affil- strengthening combining with refinement of grain sizes can also be achieve by repeated hot- iations. worked, cold-worked and low-temperature treatment [8]. However, these strengthening methods exhibit several disadvantages, such as low cold formability, deterioration of high-temperature stability, decrease of corrosion behavior, etc., [9,10]. Therefore, the high- performance SSs with new strengthening mechanisms need to be developed. A recently emerged concept of coherent twin boundary (TB) is incorporated into Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. the design for strengthening SSs [11,12]. TBs are special internal boundaries, in which This article is an open access article strengthening is based on dislocation-TB interactions and the toughening is originated distributed under the terms and from the slips of partial dislocations along the coherent TBs [13–15]. Thereby, high-density conditions of the Creative Commons twins with a finer spacing, especially down to the nanometer scale, can lead to a significant Attribution (CC BY) license (https:// enhancement of flow strength and improvement of ductility. However, high density of creativecommons.org/licenses/by/ refined twins in engineering materials is not easily formed [16–18]. At present, high- 4.0/). density nanotwins are obtained under the extreme conditions of the model materials,

Nanomaterials 2021, 11, 1624. https://doi.org/10.3390/nano11061624 https://www.mdpi.com/journal/nanomaterials Nanomaterials 2021, 11, 1624 2 of 11

such as the copper foil with a thickness of several tens of micrometers [19,20]. In order to develop an advanced structural steel with high strength and large ductility, an austenite SS composed of multi-scaled twin bundles and nanocrystalline is produced in one-step by an advanced technology of surface mechanical attrition treatment (SMAT) with high- impacting frequency [21,22]. The SMAT set-up was first introduced by K. Lu and J. Lu in 1999, which was based on the vibration of spherical balls using high-power ultrasound [23]. The randomly flying balls were blasted on the surface propelling by an extremely high sound pressure as the driving force, then nanocrystalline (NC) layer was produced on a variety of metallic materials, together with a gradient grain size distribution [24–26]. In order to control and regulate the microstructure, a number of parameters are involved in SMAT, including ball size, impact velocity, ball density, impact kinetic energy, and so on [21,27]. Large number of research works show that the hardness and tensile strength of SMATed materials significantly increase, and strongly depend on the kinetic energy of balls and shots [21,24,25]. The high-frequency SMAT generates very high-impacting velocity and thereafter resulting in large local strain rate, which is the critical factor to induce twinned structure [21,27–29]. In this work, microstructures evolution and mechanical properties of the SMATed SS by different impacting velocities are studied. Quantitative relationships between impacting velocity with the grain size, martensite transformation, and twinned structure are provided. The strengthening of the gradient nanotwinned structure is proposed. These results are very important for better understanding of the deformation behavior of gradient twinned structure.

2. Materials and Methods The chemical compositions of austenite 304 SS is 19.1Cr, 9.9Ni, 0.04C, 0.15Si, 1.40Mn, 0.005S, 0.026P and balance Fe (all in mass %). The 304 SS sheets (70 mm × 50 mm × 1 mm) were treated by SMAT with a high frequency of 20 kHz. The impact velocity was calculated about 7 m s−1 for austenite SS balls and 10 m s−1 for bearing steel balls [27]. The different impacting velocities originate from the different material properties of balls, where bearing steel is harder than austenitic stainless steel, thus, induces a larger impacting velocity. The 304 SS sheets were treated by different durations with a number of 100 balls on both sides of the 304 SS sheets. When treating the samples, the SMAT was carried out in a symmetrical time interval of 0.5 s from front to back sides, thereafter, a symmetrical microstructure is obtained. The detailed parameters are provided in Table1. The residual strain, deﬁned by

4 ν ε = ln o (1) 3 νf

was used to measure the macroscopic plastic strain of the materials after SMAT, where ν0 and νf are the initial and ﬁnal volume of the treated samples, respectively.

Table 1. Processing parameters of SMAT.

Vibrating Impact Velocity Ball Diameter of Number of Time Treated Side Frequency (kHz) (m s−1) Material Ball (mm) Ball (pcs) Interval (s) 20 7 Stainless steel 3 100 0.5 Front and back sides 20 10 Bearing steel 3 100 0.5 Front and back sides

The samples for the tensile tests were cut into dog-bone-shaped specimens with a gauge length of 30 mm and a cross section of 6 mm × 1 mm. Tensile tests were performed at room temperature at a strain rate of 6.7 × 10−4 s −1 using a MTS Alliance RT/50 Materials Testing System (MTS, Eden Prairie, MN, USA). Four specimens were used to obtain consistent stress–strain curves. Vickers microhardness was measured by applying a load of 50 g for 15 s and taking the average of ﬁve separate measurements on each depth. A Philips Xpert X-ray diffractometer (XRD, Philips, Eindhoven, The Netherlands) with Cu Kα radiation was used to determine the phase constitution and estimate the phase content. Nanomaterials 2021, 11, x FOR PEER REVIEW 3 of 11

Testing System (MTS, Eden Prairie, MN, USA). Four specimens were used to obtain consistent stress–strain curves. Vickers microhardness was measured by applying a load of 50 g for 15 s and taking the average of five separate measurements on each depth. A Nanomaterials 2021, 11, 1624 Philips Xpert X-ray diffractometer (XRD, Philips, Eindhoven, The Netherlands) with3 of Cu 11 Kα radiation was used to determine the phase constitution and estimate the phase content. Scanning electron microscopy (SEM) observations were performed with a HITACHI S- 4200 field emission scanning electron microscope (Hitachi Ltd., Tokyo, Japan). Transmis- Scanning electron microscopy (SEM) observations were performed with a HITACHI S-4200 sion electron microscopy (TEM) observations were made with a JEM 2010 transmission ﬁeld emission scanning electron microscope (Hitachi Ltd., Tokyo, Japan). Transmission electron microscope (JEOL, Tokyo, Japan) with an operating voltage of 200 kV. The TEM electron microscopy (TEM) observations were made with a JEM 2010 transmission electron foils were ion-thinned at low temperature. The statistical distribution of the grain size was microscope (JEOL, Tokyo, Japan) with an operating voltage of 200 kV. The TEM foils were estimated from TEM images, and twin fraction of the ultra-fine twins with spacing (λ1 µm) ion-thinned at low temperature. The statistical distribution of the grain size was estimated lower than 1 µm was determined by calculating the area fraction of grains containing from TEM images, and twin fraction of the ultra-ﬁne twins with spacing (λ1µm) lower than 1twinsµm was from determined SEM observations. by calculating the area fraction of grains containing twins from SEM observations. 3. Results 3.3.1. Results Mechanical Properties 3.1. MechanicalFigure 1 displays Properties the hardness variation as a function of treatment time of SMATed 304 SSFigure samples1 displays by SS theballs. hardness The hardness variation distributions as a function of all of treatmentthe treated time samples of SMATed present 304concave-shaped SS samples by curves. SS balls. Three The characteristics hardness distributions are pointed of allout. the One treated is the samples enhanced present hard- concave-shapedness at center for curves. all the Threetreated characteristics samples, indicating are pointed a bulk out. strengthening. One is the enhanced Second hard-is the nesssaturation at center hardness. for all theAlthough treated the samples, maximal indicating hardness a is bulk high strengthening. up to 476 HV Secondat the surface is the saturationfor the SMATed hardness. specimen Although withthe 20 min, maximal but th hardnesse other specimens is high up with to 476 12 HV and at 15 the min surface have forthe thevery SMATed close value. specimen Third with is the 20 min,gradient but thechange other of specimens the hardness, with where 12 and the 15 minhardness have theamplitude very close from value. the surface Third isto the center gradient is obvi changeously ofsteeper the hardness, with the whereincrease the of hardness treating amplitudetime. from the surface to center is obviously steeper with the increase of treating time.

Figure 1.1. Dependence of Vichers microhardness distributiondistribution with depth of SMATed stainlessstainless steelssteels byby SSSS balls.balls.

The engineeringengineering stress–strainstress–strain curves curves of of the th SMATede SMATed samples samples by by SS ballsSS balls (in Figure(in Figure2a) show2a) show a very a very obvious obvious enhancement enhancement of of the the strength strength with with treating treating time, time, where where thethe yieldyield stressstress ( σ0.2)) ranges ranges from from 540 540 to to 910 910 MPa, MPa, much much higher higher than than that that of the of original the original 304 SS 304 steel SS σ steel(267 (267MP). MP). The Theultimate ultimate tensile tensile stress stress (σb) (increasesb) increases from from 684 684MPa MPa (0 min) (0 min) to to810 810 MPa MPa (1 ε (1min) min) and and 980 980MPa MPa (20 min). (20 min). Moreover, Moreover, the total the elongations total elongations to failure to ( failureεb) of the ( bSMATed) of the SMATedspecimens specimens with treating with times treating of 1–20 times min of also 1–20 maintain min also high maintain levels highfrom levels61% to from 32%. 61%The togood 32%. combination The good combination of high strength of high and strength large ductility and large is owing ductility to the is owing higher to strain-hard- the higher strain-hardeningening capacity, as capacity, shown asin shownthe true in stress–s the truetrain stress–strain curves of curves Figure of 2b. Figure The 2strain-harden-b. The strain- hardeninging exponent, exponent, n, can be n, canestimated, be estimated, according according to the Ludwik–Hollomon to the Ludwik–Hollomon equation, equation,

n σ =σ=σσY + +KεKε (2)(2)

where σ and ε are the true stress and strain, σY is the initial stress, and K is the strength constant. As to the SMATed samples by SS balls, the n reaches the maximum of 0.78 for the one treated by 10 min, and then falls back to initial level of 0.73 for the one treated by 20 min. More interestingly, the true stress of all the SMATed samples reach a close saturation value of 1273 MPa, as illustrated in Figure2b. Nanomaterials 2021, 11, x FOR PEER REVIEW 4 of 11 Nanomaterials 2021, 11, x FOR PEER REVIEW 4 of 11

where σ and ε are the true stress and strain, σY is the initial stress, and K is the strength where σ and ε are the true stress and strain, σY is the initial stress, and K is the strength constant. As to the SMATed samples by SS balls, the n reaches the maximum of 0.78 for constant. As to the SMATed samples by SS balls, the n reaches the maximum of 0.78 for the one treated by 10 min, and then falls back to initial level of 0.73 for the one treated by the one treated by 10 min, and then falls back to initial level of 0.73 for the one treated by 20 min. More interestingly, the true stress of all the SMATed samples reach a close satu- Nanomaterials 2021, 11, 1624 20 min. More interestingly, the true stress of all the SMATed samples reach a close4 satu- of 11 ration value of 1273 MPa, as illustrated in Figure 2b. ration value of 1273 MPa, as illustrated in Figure 2b.

Figure 2. Tensile curves of 304 SS before and after SMAT by SS balls. (a) Engineering stress–strain FigureFigure 2.2. TensileTensile curvescurves ofof 304 SS before and after SMAT by SS balls. ( (aa)) Engineering Engineering stress–strain stress–strain curves; (b) true stress–strain curves; the insets in (a) are the SMATed samples before and after the curves;curves; ((b)) truetrue stress–strain curves; the the insets insets in in ( (aa)) are are the the SMATed SMATed samples samples before before and and after after the the tensile test. tensiletensile test.test. When using bearing steel balls with a higher impact velocity to treat the sample, a WhenWhen usingusing bearingbearing steelsteel ballsballs withwith aa higherhigher impactimpact velocityvelocity toto treattreat thethe sample,sample, aa rapid strengthening is found, shown in Figure 3a. Specifically, the SMATed sample rapidrapid strengtheningstrengthening is is found, found, shown shown in Figurein Figure3a. Speciﬁcally, 3a. Specifically, the SMATed the SMATed sample treatedsample treated by 1 min exhibits a 701 MPa, while that is 540 MPa for the stainless steel balls. bytreated 1 min by exhibits 1 min aexhibits 701 MPa, a 701 while MPa, that while is 540 that MPa is for 540 the MPa stainless for the steel stainless balls. Comparedsteel balls. Compared with the SMATed sample treated by stainless steel balls with 10 min, the sam- withCompared the SMATed with the sample SMATed treated sample by stainless treated steel by stainless balls with steel 10 min,balls thewith sample 10 min, by the bearing sample by bearing steel balls presents higher yield strength (950 MPa) and ultimate strength steelple by balls bearing presents steel higherballs presents yield strength higher yiel (950d MPa)strength and (950 ultimate MPa) and strength ultimate (1070 strength MPa), (1070 MPa), together with a close elongation to fracture (37%). Correspondingly, the together(1070 MPa), with together a close elongation with a close to fracture elongation (37%). to Correspondingly,fracture (37%). Correspondingly, the strain-hardening the strain-hardening exponent, n, also is larger than the samples with the same treating time exponent,strain-hardening n, also exponent, is larger thann, also the is sampleslarger than with the the samples same with treating the same time astreating shown time in as shown in Figure 3b. These results indicate that the strength increases while the elonga- Figureas shown3b . in These Figure results 3b. These indicate results that indicate the strength that the increases strength while increases the elongation while the areelonga- not tion are not sacrificed. Figure 4a gives the change tendency of strength with treatment sacriﬁced.tion are not Figure sacrificed.4a gives Figure the change4a gives tendency the change of strength tendency with of strength treatment with time, treatment where time, where the yield strength gradually increases and tends to a saturation value of 910 thetime, yield where strength the yield gradually strength increases gradually and increa tendsses to and a saturationtends to a valuesaturation of 910 value (stainless of 910 (stainless steel balls) and 950 MPa (bearing steel balls). These ultimate strengths are 980 steel(stainless balls) steel and balls) 950 MPa and (bearing 950 MPa steel (bearing balls). steel These balls). ultimate These strengths ultimate arestrengths 980 (stainless are 980 (stainless steel balls) and 1070 MPa (bearing steel balls), respectively. Additionally, the steel(stainless balls) steel and balls) 1070 MPaand 1070 (bearing MPa steel (bearing balls), steel respectively. balls), respectively. Additionally, Additionally, the strengths, the strengths, including yield and ultimate strengths, both present a linear relationship with includingstrengths, yield including and ultimate yield and strengths, ultimate both strength presents, both a linear present relationship a linear with relationship elongation with to elongation to fracture within the treatment time of 15 min for stainless steel balls and 10 fractureelongation within to fracture the treatment within time the treatment of 15 min fortime stainless of 15 min steel for balls stainless and 10 steel min balls for bearing and 10 minsteel for balls, bearing and then steel reach balls, saturation, and then reac as seenh saturation, from Figure as seen4b. from Figure 4b. min for bearing steel balls, and then reach saturation, as seen from Figure 4b.

Figure 3. Tensile curvescurves ofof SMATedSMATed 304 304 SS SS by by bearing bearing steel steel balls. balls. (a ()a Engineering) Engineering stress–strain stress–strain curves; Figure 3. Tensile curves of SMATed 304 SS by bearing steel balls. (a) Engineering stress–strain curves;(b) true ( stress–strainb) true stress–strain curves. curves. curves; (b) true stress–strain curves.

NanomaterialsNanomaterials2021 2021, 11, ,11 1624, x FOR PEER REVIEW 5 of 11 5 of 11 Nanomaterials 2021, 11, x FOR PEER REVIEW 5 of 11

FigureFigure 4. 4. RelationshipRelationship of ofstrength strength with with treatment treatment time time(a) and (a )elongation and elongation to fracture to fracture (b). (b). Figure 4. Relationship of strength with treatment time (a) and elongation to fracture (b). Figure 55 showsshows the the fracture fracture morphologies morphologies of the of SMATed the SMATed samples samples treated treatedby stain- by stain- lessless steel balls balls with with 10Figure 10 min, min, 5 which shows which exhibits the exhibits fracture a mix amorphologies mixfracture, fracture, characterized of characterizedthe SMATed by tearing samples by lips, tearing treated lips, by stain- microcracks,microcracks, and andless dimples. dimples. steel balls The The withfacture facture 10 surfacemin, surface which is mostly isexhibits mostly composed a composedmix fracture,of the ductility of characterized the ductility fracture by fracture tearing lips, characteristiccharacteristic after aftermicrocracks, tensile tensile test testand (in (indimples.Figure Figure 5a), The5 buta), facture butthe dimples the surface dimples are is mostlyof aremuch of composed smaller much smaller size, of the and ductility size, and fracture characteristic after tensile test (in Figure 5a), but the dimples are of much smaller size, and exhibitexhibit a a graded graded increase increase from from surface surface to center. to center. Tearing Tearing bandsbands are found are at found the surface at the surface exhibit a graded increase from surface to center. Tearing bands are found at the surface withwith a thickness of of 5 5µm,µm, and and then then shallow shallow and andparabolic parabolic dimples dimples are observed are observed at the at the with a thickness of 5 µm, and then shallow and parabolic dimples are observed at the subsurface, as given in Figure 5b. At the subsurface of 150 µmµ from the surface, large subsurface, assubsurface, given in Figure as given5b. Atin Figure the subsurface 5b. At the of subsurface 150 m fromof 150 the µm surface, from the large surface, mi- large crocracksmicrocracks are are formed, formed, as as shownshown in in Figure Figure 5a,5a, where where deep deep shearing shearing edges edges and fine and shear- ﬁne shearing ing bands are clearlymicrocracks found, are as shownformed, in as the shown magnification in Figure 5a,of Figurewhere deep5c. Some shearing deep edges dim- and fine shear- bands are clearly found, as shown in the magniﬁcation of Figure5c. Some deep dimples ples with microcracksing bands are areobserved clearly in found, the subs as urfaceshown and in the interior, magnification as indicated of Figure by arrows 5c. Some deep dim- with microcracks are observed in the subsurface and interior, as indicated by arrows in in Figure 5c,d. ples with microcracks are observed in the subsurface and interior, as indicated by arrows Figure5c,d. in Figure 5c,d.

Figure 5. SEM observations of fracture morphology of SMATed sample by stainless steel ball with 10 min. (a) Global Figure 5. SEM observations of fracture morphology of SMATed sample by stainless steel ball with morphology; (b–dFigure) magnifications 5. SEM observations of the corresponding of fracture A,morp B, Chology zones of in SMATed (a), showing sample the bymorphologies stainless steel of surfaceball with NC 10 min. (a) Global layer, subsurface,morphology; and center10 zones. min.(b–d ) (magnificationsa) Global morphology; of the corresponding (b–d) magniﬁcations A, B, C zones ofin the(a), showing corresponding the morphologies A, B, C zones of surface in ( aNC), layer, subsurface,showing and the center morphologies zones. of surface NC layer, subsurface, and center zones. 3.2. Microstructure

The mechanical behaviors of the SMATed samples are controlled by the microstructures. Figure6a shows the XRD patterns of the as-received 304 SS and SMATed samples by stainless steel balls. The as-received 304 SS is primary composed of γ-austenite (fcc) phase, Nanomaterials 2021, 11, x FOR PEER REVIEW 6 of 11

3.2. Microstructure

Nanomaterials 2021, 11, 1624 The mechanical behaviors of the SMATed samples are controlled by the microstruc-6 of 11 tures. Figure 6a shows the XRD patterns of the as-received 304 SS and SMATed samples by stainless steel balls. The as-received 304 SS is primary composed of γ-austenite (fcc) phase, and a slightly α’-martensite phase. However, α’-martensite transformation occurs duringand a slightly SMAT, αand’-martensite the content phase. of α’-martensite However, phaseα’-martensite increases transformation with the increase occurs of treat- during mentSMAT, time and until the to content a saturation of α’-martensite level of 35%, phase as shown increases in Figure with 6b. the Similarly, increase ofthe treatment SMATed time samplesuntil to aby saturation bearing steel level balls of 35%, also asexhibit shown the in same Figure tendency,6b. Similarly, while the contents SMATed of samples α’- martensiteby bearing phase steel ballsare relatively also exhibit smaller the same than tendency,these SMATed while samples the contents by stainless of α’-martensite steel phaseballs. are relatively smaller than these SMATed samples by stainless steel balls.

Figure 6. 6. XRDXRD patterns patterns (a ()a )and and phase phase content content (b () bat) atthe the surfaces surfaces of the of the SMATed SMATed samples samples by stain- by stainless lesssteel steel balls. balls.

TheThe morphologies morphologies of of the the as-received as-received 304 304 SS SSand and SMATed SMATed samples samples by stainless by stainless steel steel balls are are shown shown in in Figure Figure 7,7 ,which which are are selected selected from from the the distance distance of 50 of 50µm µclosem close to the to the surface. The The as-received as-received 304 304 SS SS shows shows equiaxed equiaxed grains grains with with an an average average grain grain size size of of 50 50 µm (inµm Figure(in Figure7a). 7a). But But many many shear shear bands bands occur occur in in the the samples samples afterafter SMAT,SMAT, asas indicated by byarrows arrows in Figurein Figure7b. 7b. When When the the treatment treatment time time increases, increases, denser denserand and ﬁnerfiner shearshear bands are areobserved, observed, as shownas shown in Figurein Figure7c,d. 7c,d. Some Some shear shear bands bands cut cut through through the the whole whole grains grains inin one parallelone parallel direction, direction, and and some some intersect intersect with with each each other other inin anan angleangle of 70°, 70◦, as as displayed displayed by bysolid solid arrows arrows and and dot dot line line arrowsarrows in Figure 7c,d7c,d respectively. respectively. It Itshould should be benoted noted that that the the Nanomaterials 2021, 11, x FOR PEER REVIEW 7 of 11 multi-scaled shear shear bands bands are are formed formed in inthe the bu bulklk material material in the in theSMATed SMATed specimens, specimens, but but the area area fraction fraction gradually gradually decreases. decreases.

Figure 7. 7. MorphologiesMorphologies of of SMATed SMATed samples samples by by stainless stainless steel steel balls balls from from the the distance distance of 50 of µm 50 µ tom to the thesurface. surface. (a) ( 0a) min; 0 min; (b )(b 3) min;3 min; (c )(c 10) 10 min; min; (d ()d 15) 15 min. min.

The detailed microstructures of the SMATed samples are analyzed by TEM observations, as given in Figure 8. Typical surface microstructure of the SMATed sample by stainless steel balls with treatment time of 10 min is composed of nanocrystalline (NC)/ultra- fined grains (UFG) with an average grain size of 180 nm, as shown in Figure 8a, and presents random crystallographic orientation, as indicated by the diffraction rings of the selected-area electron diffraction (SAED) pattern in Figure 8a. High density of dislocations is observed in the NC/UFG, and dislocation cells also appear, as indicated in Figure 8a. In the deeper zone, multi-scaled twins with single and intersect slip systems are observed. The typical morphologies of twins are given in Figure 8b,c, which are obtained from the zones at 50 µm and 300 µm depths from the surface, respectively. The deformed twins are characterized by finer single twins with higher density in the vicinity of surface, and gradually transit to coarser intersected twins in deeper zone. The dominant spacing (λ) of the single twin is lower than 100 nm (in Figure 8b). Intersected twins present an angle of about 70°, as shown in Figure 8c. High density dislocations are also observed at TBs and interior of twins, as marked by dot arrows in Figure 8c. These results suggest that the high-density shear bands observed in SEM images are the single or intersect twins.

Figure 8. Typical microstructures of the SMATed samples by stainless steel balls with 10 min. (a–c) Bright-field TEM images at depths of 0, 50, and 300 µm, respectively.

Nanomaterials 2021, 11, x FOR PEER REVIEW 7 of 11

Nanomaterials 2021, 11, 1624 7 of 11 Figure 7. Morphologies of SMATed samples by stainless steel balls from the distance of 50 µm to the surface. (a) 0 min; (b) 3 min; (c) 10 min; (d) 15 min.

The detailed microstructures of the SMAT SMATeded samples are analyzed by TEM observations, asas givengiven inin FigureFigure8 .8. Typical Typical surface surface microstructure microstructure of of the the SMATed SMATed sample sample by stainlessby stain- steelless steel balls balls with with treatment treatment time time of 10 of min 10 ismi composedn is composed of nanocrystalline of nanocrystalline (NC)/ultra-ﬁned (NC)/ultra- grainsfined grains (UFG) (UFG) with an with average an average grain size grain of 180sizenm, of 180 as shownnm, as inshown Figure in8 a,Figure and presents 8a, and ran-pre- domsents crystallographicrandom crystallographic orientation, orientation, as indicated as indicated by the diffraction by the diffraction rings of the rings selected-area of the se- electronlected-area diffraction electron (SAED)diffraction pattern (SAED) in Figure pattern8a. in High Figure density 8a. High of dislocations density of isdislocations observed inis observed the NC/UFG, in the and NC/UFG, dislocation and dislocation cells also appear, cells also as indicatedappear, as in indicated Figure8a. in In Figure the deeper 8a. In zone,the deeper multi-scaled zone, multi-scaled twins with singletwins with and intersectsingle and slip intersect systems slip are systems observed. are The observed. typical morphologiesThe typical morphologies of twins are givenof twins in Figure are given8b,c, in which Figure are 8b,c, obtained which from are theobtained zones atfrom 50 µthem andzones 300 at µ50m µm depths and 300 from µm the depths surface, from respectively. the surface, The respectively. deformed The twins deformed are characterized twins are bycharacterized ﬁner single by twins finer with single higher twins density with higher in the density vicinity in of the surface, vicinity and of gradually surface, and transit grad- to coarserually transit intersected to coarser twins intersected in deeper twins zone. in The deeper dominant zone. The spacing dominant (λ) of thespacing single (λ twin) of the is ◦ lowersingle thantwin 100is lower nm (in than Figure 100 8nmb). (in Intersected Figure 8b). twins Intersected present antwins angle present of about an 70angle, as of shown about in70°, Figure as shown8c. High in Figure density 8c. dislocationsHigh density are disloc alsoations observed are also at TBs observed and interior at TBs ofand twins, interior as markedof twins, by as dotmarked arrows by indot Figure arrows8c. in These Figure results 8c. These suggest results that thesuggest high-density that the high-density shear bands observedshear bands in SEMobserved images in SEM are the images single are or the intersect single twins. or intersect twins.

Figure 8.8. TypicalTypical microstructures microstructures of of the the SMATed SMATed samples samples by stainless by stainless steel steel balls withballs 10with min. 10 ( amin.–c) Bright-field (a–c) Bright-field TEM images TEM atimages depths at ofdepths 0, 50, of and 0, 30050, andµm, 300 respectively. µm, respectively. 4. Discussion 4.1. Effect of Impacting Velocity on Graded Structure According to the results of hardness distribution, it can be reasonably thought that the microstructure of SMATed specimens exhibit a gradient change, involving NC/UFG grains at the surface, multi-scaled twins and dislocation substructures interior, as observed in Figure8. The gradient structure is caused by the attenuated strain rate of SMAT [ 21,27]. During impacting deformation, three competitive mechanisms control the microstructure of the austenitic SS, basically, slip, twinning, and transformation [30,31], which are all relative to the impacting strain rate. Two impacting velocities are used in this experiment. Correspondingly, the higher impact velocity provides greater impacting force and larger strain rate, thereafter, increases the probability of deformation twins. Comparing the microstructure of the two types of samples by the low- and high-impacting velocities, the martensite transformation is inhibited in the higher-impacting velocity, exhibiting a relatively smaller content as shown in Figure6b. The grain sizes at the surfaces of the two samples exhibit a gradual decrease with impacting time, and the high-impacting velocity results in a rapid refinement, as shown in Figure9a. The grain sizes reach a saturation size of 138 nm by the high-impacting velocity with 10 min and 150 nm by the low-impacting velocity of 20 min, showing a close grain size. On the other hand, a large number of stacking faults and deformed twins are formed inside the austenite matrix, especially at the depth of 50–150 µm depth from the surface for all the SMATed samples, as shown in Figure 9b,c. The twin fraction increases significantly with increasing impacting velocity, especially in the center zone of the samples. These results suggest that the twinning deformation is sensitive to the impacting velocity. Nanomaterials 2021, 11, x FOR PEER REVIEW 8 of 11

4. Discussion 4.1. Effect of Impacting Velocity on Graded Structure According to the results of hardness distribution, it can be reasonably thought that the microstructure of SMATed specimens exhibit a gradient change, involving NC/UFG grains at the surface, multi-scaled twins and dislocation substructures interior, as observed in Figure 8. The gradient structure is caused by the attenuated strain rate of SMAT [21,27]. During impacting deformation, three competitive mechanisms control the microstructure of the austenitic SS, basically, slip, twinning, and transformation [30,31], which are all relative to the impacting strain rate. Two impacting velocities are used in this experiment. Correspondingly, the higher impact velocity provides greater impacting force and larger strain rate, thereafter, increases the probability of deformation twins. Compar- ing the microstructure of the two types of samples by the low- and high-impacting velocities, the martensite transformation is inhibited in the higher-impacting velocity, exhibiting a relatively smaller content as shown in Figure 6b. The grain sizes at the surfaces of the two samples exhibit a gradual decrease with impacting time, and the high-impacting velocity results in a rapid refinement, as shown in Figure 9a. The grain sizes reach a saturation size of 138 nm by the high-impacting velocity with 10 min and 150 nm by the low- impacting velocity of 20 min, showing a close grain size. On the other hand, a large number of stacking faults and deformed twins are formed inside the austenite matrix, especially at the depth of 50–150 µm depth from the surface for all the SMATed samples, as shown in Figure 9b,c. The twin fraction increases significantly with increasing impacting Nanomaterials 2021, 11, 1624 velocity, especially in the center zone of the samples. These results suggest that the 8twin- of 11 ning deformation is sensitive to the impacting velocity.

FigureFigure 9.9. GrainGrain sizesize distributiondistribution atat thethe surfacesurface ((aa),), twintwin fractionfraction withwith λλ11µμm lowerlower than than 1 1 µmµm of the SMATed samplessamples byby stainlessstainless steelsteel ballsballs ((bb)) andand bearingbearing steelsteel ballsballs ((cc).).

Since the multi-scaled twins traverse the entireentire thicknessthickness ofof thethe sheets,sheets, thethe strainstrain should also penetrate the bulk materials. Th Thee residual strains of the samples after SMAT are shown in Figure 10,10, where where the the plastic plastic strain strain increases increases with with treating treating time, time, and and attains attains a asaturation saturation strain strain of 0.17 of 0.17 for forthe thesample sample by st byainless stainless steel steelballs ballswith 15 with min 15 and min 0.20 and for 0.20 the forone theby onebearing by bearingsteel balls steel with balls 10 withmin. The 10 min. occurrence The occurrence of residual of strain residual means strain that means bulk thatvolume bulk is volumereduced is after reduced SMAT, after and SMAT, the higher and theimpact higher velocity impact generates velocity larger generates strain. larger The Nanomaterials 2021, 11, x FOR PEER REVIEWstrain.saturation The strain saturation implies strain that implies the impacting that the energy impacting or impacting energy or time impacting cannot timefurther cannot9 gen-of 11 further generate plasticity of the sheet due to the strengthening of the treated materials, erate plasticity of the sheet due to the strengthening of the treated materials, and thereaf- and thereafter, an upper limit of grain size and twin fraction is reached. ter, an upper limit of grain size and twin fraction is reached.

Figure 10. Dependence of residual strain of SMATed samplessamples withwith treatmenttreatment time.time.

4.2. Synergetic Reinforcement of the Graded Structure The superior mechanical properties can be attributedattributed toto thethe complexcomplex microstructuremicrostructure induced by the SMAT process, such as the gradient distribution of twin fraction (Figure9b,c) induced by the SMAT process, such as the gradient distribution of twin fraction (Figure and martensite phase (Figure6b). The outer NC layer, especially the hard α0 phase, provides 9b,c) and martensite phase (Figure 6b). The outer NC layer, especially the hard α′ phase, superior stiffness and hardness, and exhibits a limited plastic deformation by grain rotation provides superior stiffness and hardness, and exhibits a limited plastic deformation by and grain boundary slide [28]. The high-density twins lead to a signiﬁcant enhancement grain rotation and grain boundary slide [28]. The high-density twins lead to a significant in ﬂow strength and an improvement of ductility by the dislocation accumulation at enhancement in flow strength and an improvement of ductility by the dislocation accu- TBs [18,24]. The junctions between the various structures are clearly “functionally graded,” mulation at TBs [18,24]. The junctions between the various structures are clearly “func- that is, they possess a gradual spatial change in properties, which can promote an effective tionally graded,” that is, they possess a gradual spatial change in properties, which can transitional region for stress redistribution [32–34]. The hardness variation of SMATed promote an effective transitional region for stress redistribution [32–34]. The hardness sample by stainless steel with 10 min exhibits obviously gradient change during different variation of SMATed sample by stainless steel with 10 min exhibits obviously gradient stages of tensile deformation, indicating an important smoothing role of the gradient change during different stages of tensile deformation, indicating an important smoothing structure in deformation, as shown in Figure 11. Notably, the NC/UFG surface also exhibits anrole increased of the gradient hardness structure during in the deformation, tensile deformation, as shown which in Figure means 11. theNotably, strengthening the NC/UFG still operates.surface also During exhibits the deformationan increased process, hardness the during multiple-scale the tensile twins deformation, strengthen which by secondary means twinningthe strengthening and intersected still operates. twinning, During and changethe deformation the orientation process, of thethe subgrainsmultiple-scale to transfer twins thestrengthen deformation by secondary to the adjacent twinning grains and or subcrystalsintersected [twinning,31,33,34]. Throughand change relieving the orientation the stress of the subgrains to transfer the deformation to the adjacent grains or subcrystals [31,33,34]. Through relieving the stress concentration by the graded structure, the extension of deformation is achieved, which endows the SMATed samples high plasticity (Figures 2a and 3a). More importantly, the refinement by the twin boundaries, such as the intersected twinning or multiple twinning, can significantly enhance the hardness and strength by the dislocation pile-up at TBs, similar to the Hall-Petch rule [18]. Correspondingly, the hardness gradually increases during tensile deformation, until their hardness is close to the surface NC/UFG layer, as observed in Figure 11. Hence, graded distribution of multi- scaled twins is necessary to obtain both high strength and high plasticity by multiple deformation behaviors to release stress concentration compared to the uniform material.

Figure 11. Hardness distribution with the dependence of tensile strain of the SMATed sample by stainless steel balls with 10 min.

Nanomaterials 2021, 11, x FOR PEER REVIEW 9 of 11

Figure 10. Dependence of residual strain of SMATed samples with treatment time.

4.2. Synergetic Reinforcement of the Graded Structure The superior mechanical properties can be attributed to the complex microstructure induced by the SMAT process, such as the gradient distribution of twin fraction (Figure 9b,c) and martensite phase (Figure 6b). The outer NC layer, especially the hard α′ phase, provides superior stiffness and hardness, and exhibits a limited plastic deformation by grain rotation and grain boundary slide [28]. The high-density twins lead to a significant enhancement in flow strength and an improvement of ductility by the dislocation accumulation at TBs [18,24]. The junctions between the various structures are clearly “functionally graded,” that is, they possess a gradual spatial change in properties, which can promote an effective transitional region for stress redistribution [32–34]. The hardness variation of SMATed sample by stainless steel with 10 min exhibits obviously gradient change during different stages of tensile deformation, indicating an important smoothing role of the gradient structure in deformation, as shown in Figure 11. Notably, the NC/UFG Nanomaterials 2021, 11, 1624 surface also exhibits an increased hardness during the tensile deformation, which means9 of 11 the strengthening still operates. During the deformation process, the multiple-scale twins strengthen by secondary twinning and intersected twinning, and change the orientation of concentrationthe subgrains to by transfer the graded the deformation structure, the to extensionthe adjacent of grains deformation or subcrystals is achieved, [31,33,34]. which Throughendows relieving the SMATed the stress samples concentration high plasticity by the (Figures graded2a structure, and3a). More the extension importantly, of de- the formationreﬁnement is achieved, by the twin which boundaries, endows suchthe SMATed as the intersected samples high twinning plasticity or multiple (Figures twinning, 2a and 3a).can More signiﬁcantly importantly, enhance the refinement the hardness by andthe twin strength boundaries, by the dislocation such as the pile-up intersected at TBs, twinningsimilar toor themultiple Hall-Petch twinning, rule can [18]. signific Correspondingly,antly enhance the the hardness hardness gradually and strength increases by theduring dislocation tensile pile-up deformation, at TBs, until similar their to hardness the Hall-Petch is close rule to the [18]. surface Correspondingly, NC/UFG layer, the as hardnessobserved gradually in Figure increases 11. Hence, during graded tensile distribution deformation, of multi-scaled until their hardness twins is necessaryis close to to theobtain surface both NC/UFG high strength layer, as and observed high plasticity in Figure by 11. multiple Hence, deformation graded distribution behaviors of to multi- release scaledstress twins concentration is necessary compared to obtain to both the uniform high strength material. and high plasticity by multiple deformation behaviors to release stress concentration compared to the uniform material.

FigureFigure 11. 11. HardnessHardness distribution distribution with with the thedependence dependence of tensile of tensile strain strain of the of SMATed the SMATed sample sample by by stainlessstainless steel steel balls balls with with 10 10min. min. 5. Conclusions Effect of different impacting velocities during SMAT on microstructure and mechanical properties of 304 stainless steels were investigated. The following conclusions can be obtained. (1) The microstructures of SMATed sample are composed of NC/UFG surface and multi- scaled twins interior in a gradient distribution, characterized by graded twin fraction and martensite phase. The high-impacting velocity stimulates the formation of ultra- ﬁned twins and inhibits the martensite transformation, where the twin fraction is high up to 72% for the sample by bearing steel balls with 10 min. (2) The tensile strength of the SMATed sample increases with the treatment time, and reaches a saturation level. The high-impacting velocity results in a simultaneous increase of strength and ductility compared with the low-impacting velocity. (3) The impressive combination of strength and ductility originates from the juxtaposi- tion of multiple reinforcing microstructure. The gradient structures of twin fraction result in a graded change of hardness, playing an important smoothing role during tensile deformation.

Author Contributions: Conceptualization, J.L. and H.R.; methodology, A.C. and J.L.; formal analysis, A.C. and. J.J.; investigation, A.C.; resources, A.C. and C.W.; data curation, C.W., J.J.; writing—original draft preparation, A.C.; writing—review and editing, A.C., J.J. and H.R.; supervision, J.L.; project administration, A.C.; funding acquisition, A.C. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded by NSFC, grant number 51771121, and Science and Technology Committee of Shanghai Municipality, grant number 20ZR1437500, and Innovation Program of Shanghai Municipal Education Commission, grant number 2019-01-07-00-07-E00015. Data Availability Statement: The data presented in this study are available on reasonable requests from the corresponding author. Conﬂicts of Interest: The authors declare no conﬂict of interest. Nanomaterials 2021, 11, 1624 10 of 11

References 1. Wu, Z.; Zhang, K.; Zhou, C.; Zhou, Z.; Zhang, W.; Bao, F.; Zheng, J.; Zhang, L. Warm deformation enhances strength and inhibits hydrogen induced fatigue crack growth in metastable 304 and 316 austenitic stainless steels. Mater. Sci. Eng. A 2021, 818, 141415. [CrossRef] 2. Schino, A.D. Manufacturing and application of stainless steels. Metals 2020, 10, 327. [CrossRef] 3. Jia, S.; Tan, Q.; Ye, J.; Zhu, Z.; Jiang, Z. Experiments on dynamic mechanical properties of austenitic stainless steel S30408 and S31608. J. Constr. Steel Res. 2021, 179, 106556. [CrossRef] 4. Panov, D.; Kudryavtsev, E.; Chernichenko, R.; Smirnov, A.; Stepanov, N.; Simonov, Y.; Zherebtsov, S.; Salishchev, G. Mechanisms of the reverse martensite-to-austenite transformation in a metastable austenitic stainless steel. Metals 2021, 11, 599. [CrossRef] 5. Hu, C.Y.; Wan, X.L.; Zhang, Y.J.; Deng, X.T.; Wang, Z.D.; Misra, R.D.K. The synergistic effect of grain boundary and grain orientation on micro-mechanical properties of austenitic stainless steel. J. Mech. Behav. Biomed. Mater. 2021, 118.[CrossRef] [PubMed] 6. Cui, F.; Pan, Q.S.; Tao, N.R.; Lu, L. Enhanced high-cycle fatigue resistance of 304 austenitic stainless steel with nanotwinned grains. Int. J. Fatigue 2021, 143.[CrossRef] 7. Kaoumi, D.; Liu, J. Deformation induced martensitic transformation in 304 austenitic stainless steel: In-situ vs. ex-situ transmission electron microscopy characterization. Mater. Sci. Eng. A 2018, 715, 73–82. [CrossRef] 8. Järvenpää, A.; Jaskari, M.; Kisko, A.; Karjalainen, P. Process and properties of reversion-treated austenitic stainless steels. Metals 2020, 10, 281. [CrossRef] 9. Laleh, M.; Hughes, A.E.; Tan, M.Y.; Rohrer, G.S.; Primig, S.; Haghdadi, N. Grain boundary character distribution in an additively manufactured austenitic stainless steel. Scr. Mater. 2021, 192, 115–119. [CrossRef] 10. Silva, M.B.R.; Roche, V.; Blanco, T.M.; Viet, N.H.; Balancin, O.; Cabrera, J.M.; Jorge, A.M. Effect of processing conditions on the microstructure, mechanical properties, and corrosion behavior of two austenitic stainless steels for bioimplant applications. Metals 2020, 10, 1311. [CrossRef] 11. Tan, J.; Zhang, Z.; Zheng, H.; Wang, X.; Gao, J.; Wu, X.; Han, E.H.; Yang, S.; Huang, P. Corrosion fatigue model of austenitic stainless steels used in pressurized water reactor nuclear power plants. J. Nucl. Mater. 2020, 541, 152407. [CrossRef] 12. Calmunger, M.; Chai, G.; Eriksson, R.; Johansson, S.; Moverare, J.J. Characterization of austenitic stainless steels deformed at elevated temperature. Metall. Mater. Trans. A Phys. Metall. Mater. Sci. 2017, 48, 4525–4538. [CrossRef] 13. Kang, C.; Tang, J.; Luo, H.; Ye, K.; Ma, Y.; Shi, D. Study on the deformation mechanism of the inflection point in tensile curve of nano/ultrafine-grained austenite stainless steel rolled at different temperature. Mater. Sci. Eng. A 2020, 786, 139367. [CrossRef] 14. Rai, P.K.; Shekhar, S.; Yagi, K.; Ameyama, K.; Mondal, K. Fretting wear mechanism for harmonic, non-harmonic and conventional 316L stainless steels. Wear 2019, 424–425, 23–32. [CrossRef] 15. Song, Y.L.; Li, C.S.; Li, B.Z.; Han, Y.H. Grain character and mechanical properties of Fe-21Cr-15Ni-6Mn-Nb non-magnetic stainless steel after solution treatment. Mater. Sci. Eng. A 2019, 742, 662–671. [CrossRef] 16. Liu, T.; Xia, S.; Shoji, T.; Bai, Q.; Zhou, B.; Lu, Y. The topology of three-dimensional grain boundary network and its influence on stress corrosion crack propagation characteristics in austenitic stainless steel in a simulated BWR environment. Corros. Sci. 2017, 129, 161–168. [CrossRef] 17. Chen, A.Y.; Hu, W.F.; Wang, D.; Zhu, Y.K.; Wang, P.; Yang, J.H.; Wang, X.Y.; Gu, J.F.; Lu, J. Improving the intergranular corrosion resistance of austenitic stainless steel by high density twinned structure. Scr. Mater. 2017, 130, 264–268. [CrossRef] 18. Matsushita, A.; Mine, Y.; Takashima, K. Enhanced resistance to fatigue crack propagation in metastable austenitic stainless steel by nanotwin bundles. Scr. Mater. 2021, 201, 113976. [CrossRef] 19. Lu, L.; Chen, X.; Huang, X.; Lu, K. Revealing the maximum strength in nanotwinned copper. Science 2009, 323, 607–610. [CrossRef] 20. Zhang, B.B.; Tao, N.R.; Lu, K. A high strength and high electrical conductivity bulk Cu-Ag alloy strengthened with nanotwins. Scr. Mater. 2017, 129, 39–43. [CrossRef] 21. Chen, A.Y.; Ruan, H.H.; Wang, J.; Chan, H.L.; Wang, Q.; Li, Q.; Lu, J. The influence of strain rate on the microstructure transition of 304 stainless steel. Acta Mater. 2011, 59, 3697–3709. [CrossRef] 22. Chen, A.; Liu, J.; Wang, H.; Lu, J.; Wang, Y.M. Gradient twinned 304 stainless steels for high strength and high ductility. Mater. Sci. Eng. A 2016, 667, 179–188. [CrossRef] 23. LU, K.; Lu, J. Surface nanocrystallization (SNC) of metallic materials-presentation of the concept behind a new approach. J. Mater. Sci. Technol. 1999, 3, 193–197. 24. Bahl, S.; Suwas, S.; Ungàr, T.; Chatterjee, K. Elucidating microstructural evolution and strengthening mechanisms in nanocrystalline surface induced by surface mechanical attrition treatment of stainless steel. Acta Mater. 2017, 122, 138–151. [CrossRef] 25. Chen, A.Y.; Jia, Y.Q.; Pan, D.; Li, X.; Lu, J. Reinforcement of laser-welded stainless steels by surface mechanical attrition treatment. Mater. Sci. Eng. A 2013, 571, 161–166. [CrossRef] 26. Benafia, S.; Retraint, D.; Yapi Brou, S.; Panicaud, B.; Grosseau Poussard, J.L. Influence of Surface Mechanical Attrition Treatment on the oxidation behaviour of 316L stainless steel. Corros. Sci. 2018, 136, 188–200. [CrossRef] 27. Chan, H.L.; Ruan, H.H.; Chen, A.Y.; Lu, J. Optimization of the strain rate to achieve exceptional mechanical properties of 304 stainless steel using high speed ultrasonic surface mechanical attrition treatment. Acta Mater. 2010, 58, 5086–5096. [CrossRef] 28. Liu, S.; Gao, S.Y.; Zhou, Y.F.; Xing, X.L.; Hou, X.R.; Yang, Y.L.; Yang, Q.X. A research on the microstructure evolution of austenite stainless steel by surface mechanical attrition treatment. Mater. Sci. Eng. A 2014, 617, 127–138. [CrossRef] Nanomaterials 2021, 11, 1624 11 of 11

29. Zhu, L.; Wen, C.; Gao, C.; Guo, X.; Chen, Z.; Lu, J. Static and dynamic mechanical behaviors of gradient-nanotwinned stainless steel with a composite structure: Experiments and modeling. Int. J. Plast. 2019, 114, 272–278. [CrossRef] 30. He, Q.; Wang, Y.F.; Wang, M.S.; Guo, F.J.; Wen, Y.; Huang, C.X. Improving strength-ductility synergy in 301 stainless steel by combining gradient structure and TRIP effect. Mater. Sci. Eng. A 2020, 780.[CrossRef] 31. Chen, A.Y.; Zhu, L.L.; Sun, L.G.; Liu, J.B.; Wang, H.T.; Wang, X.Y.; Yang, J.H.; Lu, J. Scale law of complex deformation transitions of nanotwins in stainless steel. Nat. Commun. 2019, 10.[CrossRef][PubMed] 32. Wu, X.L.; Jiang, P.; Chen, L.; Zhang, J.F.; Yuan, F.P.; Zhu, Y.T. Synergetic strengthening by gradient structure. Mater. Res. Lett. 2014, 2, 185–191. [CrossRef] 33. Yang, M.; Pan, Y.; Yuan, F.; Zhu, Y.; Wu, X. Back stress strengthening and strain hardening in gradient structure. Mater. Res. Lett. 2016, 4, 145–151. [CrossRef] 34. Zhu, Y.; Wu, X. Perspective on hetero-deformation induced (HDI) hardening and back stress. Mater. Res. Lett. 2019, 7, 393–398. [CrossRef]