Energetic TWIP - Department of Materials Science and Metallurgy

Energe&c TWIPDavid DyeKhandaker Rahman, Ananthi Sankaran, Vassili Vorontsov Adventures in the Physical Metallurgy of Steels, 25 July 2013

The Problem?High density, high velocity penetrators defeat armour by;“Shear Plugging” - locally imparting so much energy that thematerial comes close to melting, and burrowing through, orBehind armour effects such as spall fragments caused by shockwaves on the back side

Armour systemsDisturber - acts to rotate / break and spread out theprojectile, maximising its areaAbsorber - absorbs all of the energy impartedSpall liner - catches spalled fragmentsAdapted from Thomas EL, Opportunities in protection materials science and technology for future army applications.The National Academies Press, 2012.

Absorbing energyChapteFailure occurs onnecking, when thematerial runs out ofhardening.160014001200Then, any instabilityand stress localises,resulting in failure.True Stress (MPa)1000800600Ultimate Stress,ductility400Yield Stress20000 10 20 30 40 50True Strain (%)

Performance diagram for steels1.4. Thesis Structure 510090LowStrengthHighStrengthAdvanced/UltraHigh Strength80TWIP7060*Elongation (%)5040LowcarbonsteelsIFMildIF - HSIS302010CMnTRIPATI-K12 MILDP / CPArmox 370THSLAMIL-DTL-12560MartensiticArmox 440TATI-500 MILATI-600 MILMASTERARM 50000200 400 600 800 1000 1200 1400 1600 1800 2000Tensile Strength (MPa)

How do TWIP steels work? 5 µm(b)TD00120 µm101RD111

Repeated twinning= dynamic Hall-Petch (AR, ε=22%)(a)(b) Twin 1 Twin 2111000111(c)0.5 µm(d)Zone Axis: [011]250 nm50 nm

Do they perform in high rate / blast?

(a)The blast centre20 µm

Centre of the bulge. Conclusions 1(a)(b)(c)0.5 µm(d)5 nmMaterial from centre of blast crater - (a) BF TEM overview, (b) examination of internalstructure of a selected twin

... into the twin(c)0.5 µm 5 nm(d)(e)2 nm(f)1 nmMaterial from centre of blast crater - HREM of the twin imaged (down [110]) in (b),showing numerous intrinsic SFs

Fault density near a twin120 nm

analysis, but we acknowledge that the choice of function should ultimately be placed on aMeasuring grain sizestheoretically sound foundation, which would be a useful topic for further work based, e.g. onrecrystallisation modelling [173].6050(a)CDFWeibull fitRaw EBSD100806050(b)PDF154040Frequency30206040CDF (%)Frequency3020105PDF (x10 -2 )10201000 2 4 6 8 10 12 14 16000 2 4 6 8 10 12 14 16Grain Size (µm)0Figure 5.1: Weibull smoothing procedure where (a) the cumulative distribution function (CDF) of theraw EBSD data is fitted using a Weibull function and (b) smoothed probability density function (PDF)is plotted using a derivative of the fitted Weibull.

CR + anneal(a)(b)(c)00110111120 µm20 µm40 µm(d)(e)(f)1.01m 2m AR 24hr 96hrNo. normalised PDF0.80.60.40.2TDRD70 µm100 µm0.01 10 100Grain size (µm)

Effect of grain size20001500True Stress (MPa)10005000.7µm4.3µm10µm45µm84µm00 10 20 30 40 50True Strain (%)

150nm200nm200nmrmation to 5 % engineering strain showing the relative twin thickness is influenced by the iniin size. (a) 0.7 µm, (b) 4.3 µm, (c) 10 µm, (d) 45 µm and (e) 84 µm grain size material.Finer grains = thinner twins (ε=5%)(a)0.7 µm(b)0.7 µm 0.7 µm(c)(d) 84 µm60nm(e) 84 µm30nm84 (f) µm25nm

Twins in the fine grain…

Thinking about the problem…Flow Curve:= y + m⇥2000Then necking criterion gives1500So: ⇥ f =1U el = m 2 /2Eu = my/mU pl ' 1 2m2ym!True Stress (MPa)10000.7µm4.3µm50010µm45µm84µm00 10 20 30 40 50True Strain (%)σ y(MPa) m (GPa) ε f(%) U pl(GPa)FG 750 2.26 67 0.99AR 480 1.98 76 0.92CG 250 1.7 85 0.81

But Intrusion...Flow Curve:= y + m⇥2000Then impose a max. allowable strain.If failure strain is greater, thenU t' y ⇥ t + 1 2 ⇥2 t mTrue Stress (MPa)150010000.7µm4.3µm50010µm45µm84µm00 10 20 30 40 50True Strain (%)σ y(MPa) m (GPa) ε f(%) U max(GPa) U 30(GPa)Target TWIP 1300+ 2.2+ 41 0.7 0.47FG TWIP 750 2.26 67 0.99 0.32AR TWIP 480 1.98 76 0.92 0.23CG TWIP 250 1.7 85 0.81 0.15Ti-6Al-4V 950 1.0 10 0.13 0.13*Armox 440 1300 1.54 16 0.21 0.21*Goal: 3X strength, +10% hardening & 2X usable energy adsorbtion

Pressurised ESR: Additions ofup to 1 wt% N increaseaustenitic stainless steelstrengths to ~2GPa.ESR originally developed in the1960s in the UK in Sheffield.But, no onshore PESR capability.Adding interstitial N

Adding microalloyingTi, Nb, V additions can increase strength without interferingwith the TWIP mechanism.Both retarding grain growth in hot rolling (μm grains), andproviding nm precipitatesTi additions promote TiN – limit to 0.1wt% - maybe 150 MPaV additions – limit to 0.4 wt% - up to 300 MPaSee Scott et al, Int J Mat Res 102(5):538-549, 2011.

Initial Process Route DevelopmentMicroalloy with 0.5wt. % TiHeat to 1150°C for 15minutes and quenchCold roll to 75%reductionHeat treat at850°C for 1, 2 and5 minutes.Need a recrystallisation heat treatment (T,t) to generate thecorrect grain size AND ~nm precipitate size

+Ti TWIPMelted from master alloys, hot rolled, annealed - 30 µm grain size500 µm

ConclusionsThe TWIP mechanism does operate in the blast regime.Not all of the ductility is useful.Improved, higher strength TWIP alloys that retain the hardeningrate associated with the TWIP mechanism would be desirable.Some thoughts on how to do this have been presented

Backup Slides

Effect on twin thickness (ε=5%)(a)84 µm(b)45 µm(c)10 µm - AR(d)4.3 µm10 µm(e)0.7 µm6 µm3 µmTD600 nm150 nmRD