6.15.2019

Manna Steel: The internet age alloy


So I decided to revisit the Wootz steel topic and did a dozen hours of research and learned a LOT.  Turns out wootz steel isn't as great as modern alloys but there are a couple tricks from it to use like the embedded carbon nano tubes.  The banding of ferrite and cementite is a great idea since one is hard and one is ductile, however in practice it doesn't give as sharp or as strong of a blade as current methods.  The only reason wootz wasn't unusably soft is because it is high in phosphorus.  Phosphorous seriously hardens steel and also gives some corrosion resistance.  So the slow cooling and therefore large band formation was likely done to make the high phosphorus hard steel workable at all.

Manna steel is my attempt to make the highest tech purest steel in existance.  What I do is use the cutting edge modern metallurgy and combine it with the ancient carbon nano tube tech from wootz damascus steel.  These nanotubes should enhance the toughness of the steel.

Currently I plan to make this steel and do 3 tests on it: Rockwell hardness test, homemade weight drop impact test, and tension test.

The carbon nano tubes will have to improve one of these testing values, here at Naturehacker the theory is great, but results are all that really matter.

So here is the recipie and process:

Per 100g:

98.2g high purity iron 99.9%+

1.8g dried tannin impregnated sugarcane chopped/ground finely (this can be ommited entirely in this process and if it is, the resultant material is the mythical "cold iron".  Cold iron was likely iron metorites that fell into the ocean and the fast cooling likely created martenistic structure that is hard as steel)

(Carbon nano structures and silicon carbide whiskers and iron whiskers can also be added at any concentration but we are attempting to form them in the process itself, plus adding them raw would likely cause them to be dissolved in the iron melt anyway.  We are doing process steps to enhance the creation of these nano structures)

Mix well

Thats it!  The real magic in the temperature control.  Why high purity iron?  We want this to be very reproducible and any impurities would interfere with our process and the steel quality.

The sugarcane is chopped/ground with an herb grinder/coffee grinder then dried.  A concentrated black tea is made and the dry sugarcane is added to a pot with the concentrated tea added and it is cooked down until the sugarcane has the tannins absorbed and everything is dry.  The dry sugarcane should contain roughly 50% carbon so we are shooting for .9% carbon to make sure we are at least .76% carbon.  Since we will quench very fast, bieng over .76 carbon won't be a big drawback.  Sugarcane was selected for its fibrous structure like the plants that were added to wootz steel.  Sugar also will help provide carbon for nanotube growth.  Tannins along with iron help catalyze the nanotube formation.  In addition it contains some silicon (3% of the plant) which will lead to about .08% silicon in the steel.  This silicon reduces cementite inclusions from us bieng hypereutectoid and also may graphetize some of the carbon which will lubricate the blade.  Also the silicon may provide a bit of corrosion resistance.

Process:

The mixture is melted in a suitable crucible in an inert gas furnace to 1540c for 4 hours.  This is to allow the steel to melt and become homogenous.

Next it is poured out in a mold the shape and thickness of an object like a knife.

Immediatly the mold and steel is quenched  in a combination of dry ice and antifreeze/water 50/50 mix (DIAF).  This is done to keep grain structure as fine as possible with the fast cooling and also get a full transformation from austenite to martensite.  These cycles between austenite can be done under pressure if possible, at high enough pressure (3-5 Gpa) carbon isn't even needed to achieve high hardness.  Also an electrical field may help induce iron whisker formation.

Next the steel is placed at 740c (inert atmosphere if possible) for 40 mins.  This is to convert the grains back to austenite.

The steel is quickly quenched in the DIAF bath.  This shock causes micro fractures and breaks the grains into smaller grains.

Steel is again placed at 740c for 40 mins to convert the now smaller grains back to austenite.

Steel is again quenched in DIAF for smaller grains.

Steel is again placed at 740c for 40 mins.

Steel is again quenched in DIAF and let sit for 1 hour this time to ensure full conversion to martensite from austenite. 

Now the steel should have very small grain martensite.

If grinding say a basic blade shape (but not beveling), reheat to 740c for 30 mins and let air cool.  Doing that at this step will alow you to grind a softer steel but with the small grains.  Do not attempt to put an edge on the blade at this step, that should be done at the end.  After grinding reheat to 740c for 40 mins and quench in DIAF.  This step can be avoided by making the original casting in the shape you want from the beginning.  It may be possible to grind a rough edge in this step if using an inert gas furnace to prevent carbon loss at the thin edge, but still the results won't be uniform throughout the steel since some parts are thinner than others.

Next temper the steel at 150c for 96 hours then quench in DIAF.  Typical tempering is 190-205c or 260c for 4 hours however lower temperatures for longer always results in more orderly crystal growth.  The hope is increase in toughness while keeping almost all the hardness.  Hardness expected the be Rc 60+ after this process and toughness should be high from small crystals and also carbon nano tubes.  Holloman-Jaffe equation used to calculate the time based on equivalency of 260c for 4 hours.  May be worthwhile heat treating under compression with heated plates on a 20-40 ton press or any tonnage.  Also applying a electrically insulative yet thermally conductive coating on the plates and applying a voltage of any size can help form iron whiskers.

If making a blade grind the edge and finish (polish) at this step bieng sure to keep the knife under 100c (every pass on the sander cool in cold water with detergent before continuing).  Ideally use a wet sanding belt ideally zirconium.

Below is my research:

-------------------

Manna steel: bringing wootz into the internet age advanced metallurgy

Iron looses magnetism 770c.  Letting sit at 912c will help mix the atoms for better diffusion? Or vary bwtween open pack and close packed structure in annealing for diffusion

https://www.futurity.org/melting-iron-858942/

1700c, 1600 inert furnace
https://cmfurnaces.com/furnace/series-1700/

Heat treating graph
https://www.efunda.com/processes/heat_treat/softening/annealing.cfm

The longer the anneal the more diffusion and softer it becomes.  So start with very high carbon and/or hot/cold short steel and super long dendritic anneal will soften.

Phosphorus causes cold short perhaps wootz was a way around the cold shortness
https://www.sciencedirect.com/topics/engineering/phosphorus-content

High phosphorus in wootz
http://materials.iisc.ernet.in/~wootz/heritage/WOOTZ.htm

Widmanstatten steel
http://www.engineeringenotes.com/metallurgy/steel/widmanstatten-structure-of-steels-formation-and-control-metallurgy/25939

Wootz is advanced material
http://materials.iisc.ernet.in/~wootz/heritage/WOOTZ.htm

How to grow grains and etch
http://www.engineeringenotes.com/metallurgy/steel/austenitic-grain-size-of-steel-metallurgy/25901

Fine pearlite harder than course
https://www.quora.com/Why-does-fine-pearlite-has-more-hardness-than-coarse-pearlite

More about the grains
https://www.google.com/url?sa=t&source=web&rct=j&url=https://web.utk.edu/~prack/MSE%2520300/FeC.pdf&ved=2ahUKEwiassbEvOviAhUOTawKHbfsDNwQFjAKegQIAhAB&usg=AOvVaw3R8p48QrnyIZcDJro6_Y3G

So from here I relaize that we want to grow large austenite grains then skip the austenite cementite part because that moves Cementite outaide the grain.  We want cememtite only inside the grain then convert diretly to pearlite.  That way the grain boundaries are non existant.  This can be done with a perfectly eutectic amount of carbon (.76% well dispersed) or by just fast quenching of the austenite.   We will do the latter since not all our carbon will dissolve and some will be nano tubes so we need to overshoot the eutectic point since we are not sure how much of our carbon will actually form cementite.  Since we will overshoot to around 1.5% carbon, we will just assume we need a fast quench to keep the austenite grain structure without cementite around the borders.

The austenite should be as large as possible, even a single crystal would be great. 

Vanadium slows crystal growth so we might want that to make the finest pearlite possible or just grow the perlite at a low temp that still allows diffusion to happen.

Creating bainite and martensite
http://www.virginia.edu/bohr/mse209/chapter10.htm

So basically after this I relaize we want extremely large austenite crystals (so long time in austenite temp range) then we want to cool super rapidly into martensite.  Then we want to temper at 250-650 c for ultra fine pearlite. (lower better for thin bands which are better)

Also I am realizing that having the super fine microstructure will negate the need for the high phosphorus to harden it.  Vanadium also isn't needed if we use low temperatures.  Boron also isn't needed to harden the ferrite phase since that phase is so thin.  Tantalum also won't be needed to form carbides since the iron should take up all the carbon well.  Mabye they could be added in the future at low amounts but for now our heat treatments should be sufficient.  I still do want to keep the silica from the sugarcane to graphetize some of the carbon for a self lubricating finish.

So 1540c to melt in inert atmosphere for 3 hours to make sure carbon nanotubes form.
Then rapid cool to 1100c or lower.  Then set at 1100c for 12 hours (can be tuned to be lowerset temp depending if we can get reliable carbon percentage.  Ultimately we want to be as closr to the low transition temp as possible without going over it.  The closer we get the more uniform the crystal structure will be and therefore stronger).  Then we want a rapid cool to 300c or lower.  Set at 300c for 72 hours to grow ultra fine bands.  These will not be visible to the naked eye.


To start we will use simply pure iron and carbon (graphene for uniform distribution and low ash content).

0.76% carbon so eutectoid steel.

Melt at 1540c (or higher) in inert atmosphere for 3 hours to ensure uniform carbon distribution and no carbon loss from burning.

Super fast cool into martensite

Soak (anneal) at minimum 727c, more like 735c (roughly +10c) to be safe for 72 hours for large as possible and uniform as possible (from relatively low temp) austenite crystal grains.  This converts hopefully uniform martensite to large and uniform austenite crystals.

Super fast cool into martensite

Soak (temper) at 205 to 250c, say 240c for 72 hours

Cooling and martesenite
http://www.gowelding.com/met/carbon.htm

How single crystals are made
https://www.theengineer.co.uk/rolls-royce-single-crystal-turbine-blade/

Single crystals without special cooling several hours just below melting point
https://www.eurekalert.org/pub_releases/2018-10/ifbs-pdm101718.php

Austenite and martensite
https://www.sciencedirect.com/topics/materials-science/austenite

Austenite grain size does matternjin martesenite
https://www.bladesmithsforum.com/index.php?/topic/13846-structure-of-martensite/

Tempering 205c to 250c max to avoid embrittlement
https://en.m.wikipedia.org/wiki/Tempering_(metallurgy)#Physical_processes

Meta crystals in design
https://www.designnews.com/materials-assembly/meta-crystals-make-materials-tougher-and-lighter/168232725860387

Effects of alloying elements on steel
https://www.google.com/url?sa=t&source=web&rct=j&url=https://online.tugraz.at/tug_online/voe_main2.getvolltext%3FpCurrPk%3D32837&ved=2ahUKEwj34Jvc9uviAhXKmq0KHYB2C2MQFjALegQIAhAB&usg=AOvVaw3vrzDu_m5xodch7SuErcpq

Rust free iron pillar of delhi- high phosphorus
https://en.m.wikipedia.org/wiki/Iron_pillar_of_Delhi

Silica reduces cementite inclusions and also cementite increases corrosion
https://www.sciencedirect.com/topics/engineering/cementite

SO start with point steel then introduce silica later on then introduce sugarcane and tannins.

Prior austenite grain size smaller better, niobium helps
https://www.niobelcon.com/NiobelCon/niobium/niobium_in_automotive_flat_steel/auto_martensitic_steel/tough_martensite/

But austenite needs coarse grains to create martensite upon cooling
http://www.engineeringenotes.com/metallurgy/steel/austenitic-grain-size-of-steel-metallurgy/25901

So turns out we don't want to grow the austenite grains much, just enough so we can form martensite at our cooling rate.  Also we want the grains to be high uniformity so it must sit slightly above the austenization temperature for a while.

Grain coarsening temp is over 1000c
https://www.scientific.net/AMR.213.600

How grains form and eutectoid best hardness
https://knifesteelnerds.com/2019/02/25/how-fast-do-you-have-to-quench-hardenability-of-steel/

Liquid nitrogen quenching
https://en.m.wikipedia.org/wiki/Cryogenic_hardening

How to get fine grained steel with additives
https://www.google.com/amp/s/pmpaspeakingofprecision.com/2009/11/24/5-ways-fine-austenitic-grain-size-affects-your-machine-shop/amp/

So you want to have fine grained austenite and quench super fast so you can achieve martensite then temper at 205c ish over a long period.

Smaller grains more impact tough at given hardness
https://www.bladeforums.com/threads/carbon-steel-blades-and-grain-size.1503062/

Quenching leads to small grain size also parameters
https://www.google.com/amp/s/www.researchgate.net/post/Why_quenching_in_steel_leads_to_reduction_in_grain_size_whereas_in_case_of_aluminium_it_increases_the_grain_size/amp

Reducing grain size with annealing and quenching
https://www.bladesmithsforum.com/index.php?/topic/27264-when-exactly-is-grain-size-reduced/

So looks like the best process is quench super fast from as high temp as possible after melting.  Then anneal at 735c for 45 mins.  Quench super fast.  Anneal at 735 for 45 minutes.  Quench super fast.  Temper at 205c for 9 hours.

Super fast quenching should be antifreeze and dry ice.


What is happening is the first quench assures smallest possible starting grain size.  The next anneal creates our starting austenite grains.  Subsequent quenching induces fractures in the grains as the grains convert to martensite.  Then when annealed at above the 727c austenite temp for eutectoid steel these fractures become new smaller grains.  Doing this twice assures we are getting roughly the minimum grain size possible.  The final quench from austenite gives us martensite structure which is based on the final grain size.  This is extremely hard and brittle.  Next we tenper the steel at 150c for 72 hours to achieve increased toughness without sacrificing much hardness.

190-205c temper for 4 hours
https://www.bladeforums.com/threads/temper-too-long.645024/

Cryo treating
https://www.bladeforums.com/threads/deep-cryogenic-tempering.1311938/

Eutectoid types
https://www.bladeforums.com/threads/working-the-three-steel-types.673173/

1095 vs 1083
http://www.knifenetwork.com/forum/showthread.php?t=58969

So I think I am going straight to the big guns and using sugarcane with tannins as a carbon source.  I will shoot for .9% carbon so perhaps calibrating by carbonizing the sugarcane in inert atmosphere and testing carbon content.  Since we will have a super fast quench in dry ice antifreeze, we can go a little hypereutectoid.  Reason for sugarcane and tannins is the silica and potential for nanotube formation.

For testing need hardness tester and charpy impact tester and tensile tester.

Rockwell tester $1000
https://www.ebay.com/itm/HFS-R-Rockwell-Hardness-Tester-Gauge-Load-150-Kgf-Includes-Accessories-/272761552335

Tensile tester around $1300
https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.com%2Fulk%2Fitm%2F291842516122

Make my own drop test impact tester.

Improving toughness process
https://books.google.com/books?id=noe78HLQR4cC&pg=PA168&lpg=PA168&dq=tempering+at+180c&source=bl&ots=oRr1X3m7lS&sig=ACfU3U2yqt1KKMs0MZI39PIclZT2p8djHw&hl=en&sa=X&ved=2ahUKEwie7cjw3e3iAhWywFkKHXlwAOIQ6AEwAXoECAkQAQ#v=onepage&q=tempering%20at%20180c&f=false

We want the equivalent temper of 260c for 4 hours.  Mabye 150c for 72 hours?
http://baronespecialtysteel.com/?page_id=170

Rapid quench after temper then cryo tenper time logarithmic and temperature is linear time temp equation!
https://www.bladeforums.com/threads/tempering-temperature-time-relationship.1520037/

http://mmm.sjtu.edu.cn/userfiles/1/files/2008-96_Canale.pdf

Hardness = T(C+log(t))

1000= 260(C)+260log(4)

1000=260c+156

844=260c

C=3.25

Or (500-156)/260
C=1.32

1000=150*3.25+150log(x)

1000=487.5+150logx

512.5/150=logx
3.41=logx
10^3.41=x
X=2570

Or 500=150×1.32+150logx
2.01=logx

So ya lets say around 96 hours

Heat treating before or after grinding
https://www.bladeforums.com/threads/heat-treat-before-or-after-grinding.514695/

Grind after HT with wet zirconium belt
https://www.bladeforums.com/threads/hamons-and-grinding-after-heat-treat.1187810/

Point is grind bevel after HT to avoid problems
https://www.bladeforums.com/threads/heat-treat-then-start-the-grind.1126340/

60 grit blaze 180 grit gator fine green scotchbrite belt for polishing
https://www.bladeforums.com/threads/heat-treat-then-start-the-grind.1126340/page-2

Heat treatment under presure greatly increases martensite
https://www.tandfonline.com/doi/full/10.1080/21663831.2019.1612792

Can we get iron whiskers to form? Possibly just add silicon carbide whiskers since they dint degrade at steel melting temps
https://www.sciencedirect.com/topics/engineering/ceramic-whisker

Silicon carbide whiskers are about twice as strong as iron whiskers so we will just use those.  Can be added around 5-60% optimuly around 20%?
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=RE32843.PN.&OS=PN/RE32843&RS=PN/RE32843

Silicon cadbide basically dissolves in molten iron so that probably wont work.  Mabye we can get some produced by the sugarcane but im not going to add more.
https://www.washingtonmills.com/silicon-carbide/sic-properties

https://www.cheegoole.com/application-silicon-carbide-steelmaking/

Whisker formation theory
https://ieeexplore.ieee.org/abstract/document/1430806

Electrostatic theory of whisker formation
https://www.google.com/url?sa=t&source=web&rct=j&url=https://arxiv.org/pdf/1401.7689&ved=2ahUKEwihxLeYpe7iAhWmxVkKHfPQDEkQFjAAegQIAhAB&usg=AOvVaw0X7QUHVyNH6eSbbYNifxq5

Smaller austenite grains reduces martensite start temp
https://www.sciencedirect.com/topics/engineering/martensite-start-temperature

Also high pressure reduces martensite start temperature
https://www.tandfonline.com/doi/full/10.1080/21663831.2019.1612792

So get super small grains and use high pressure to achueve martensite in pure iron

Nickel also reduces martensite start temperature
https://page-one.springer.com/pdf/preview/10.1007/BF03377892

So cold iron will be no nickel and sky iron will be 4-30% nickel, optimum around 8, 10, or 11%.

Permeabilities of different iron nickel alloys
https://www.softmagneticalloy.com/soft_magnetic_alloys.html

Iron has much higher saturation than nickel
https://www.jstor.org/stable/77809?seq=1#page_scan_tab_contents

So nickel has a detrimental effect not only on magnetic permeability but also on saturation of magnetics.

Small grain size helps start martensite but slows it down after 30%
https://www.sciencedirect.com/science/article/pii/S0264127519300620

No comments:

Post a Comment

Thank you for your feedback! Sharing your experience and thoughts not only helps other customers but also helps me to improve what I do!