Musical Instruments - The steeldrums

HARDENING STEEL BY NITRIDING
SABINA SCHÄRER and FELIX ROHNER - PANArt AG

1. Introduction
Nitriding procedures are thermo-chemical processes, where the surface of the steel is enriched with nitrogen. Treatments normally occur at temperature of between 500°C and 590°C in a medium which can give off nitrogen. Such media are for example gas, plasma, saltbath or of powdered form. In our case, a gas media is used in the process called gas-nitriding.
Normally nitriding is applied to special tools and elements of machines and cars. It is rare to find gas-nitriding in use with sheet metal, except in a few cases; but it can be found in use, in particular by the spring industries.
PANArt Ltd has performed experiments with different surface treatments [for steel] in search of a stronger material. Nitriding in a saltbath has been used for the Blackpans; here [the active elements] [] are not only nitrogen, but also carbon. The duration of this process is [relatively fast] between 10 to 40 minutes, and the [resultant] surfaces get very hard. Gas-nitriding is a longer duration process, but it can be better controlled. The effect of gas-nitriding is to improve wear resistance, fatigue strength, corrosion protection and the qualities of the steel used in [steeldrum] musical instrument [manufacture].

2. Nitrogen and Nitride Formers
Nitrogen forms a chemical bond with [among others,] the elements aluminium [Al], magnesium [Mg], silicon [Si], titanium [Ti], vanadium [V], chromium [Cr], molybdenum [Mo] and iron [Fe]; to produce nitrides. These elements [when alloyed with; or remain residual traces in iron [Fe] steels] are called nitride formers. Nitrides are hard elements, and [act] similarly to the familiar carbides [compounds of carbon] precipitated into the iron matrix [as associated with case-hardening using carbon].
The radius of atomic nitrogen is 50% of the radius of atomic iron; and about 8% smaller than atomic carbon. Therefore the inclusion of atomic nitrogen into the interstitial position of the iron lattice is possible. [This behaviour is classified in Physical Chemistry as being of the non-stoichiometric compounds of the interstitial carbides [MxC], nitrides [MxN] and hydrides [MxH]; where the respective atoms may occupy some, if not all, of the small spaces between the larger metal atoms.]
For tools and elements of machines and cars; gas-nitriding is normally applied to alloyed steel to get a high hardness (to 1200 Vickers*). For [steeldrum] instruments this hardness is too high; the steel can not easily be deformed, and it breaks. But [for our purposes and requirements, we start with] deepdrawn steel (Fe PO4 / St 14.03), having a very low content of nitride formers; the hardness [we seek to control] will then not increase too much. 
  [ * Vickers Hardness = HV ]

3. The Process
The forming of nitrides is a chemical reaction between nitrogen and iron, [this chemistry is not easily achieved however]. []
To work at normal [atmospheric] pressures, ammonia gas [] is used [to produce the required nitrogen]. [The initial stage of gas-nitriding of iron using ammonia, is virtually the reverse of an ammonia Haber production process where; with iron as a catalyst, undertaken at similar temperatures - but here without the high pressure - the first stage of this reversible reaction now proceeds instead to the alternate equilibrium; to dissociate the ammonia into its constituent gasses - nitrogen and hydrogen.]
[At standard temperatures and pressures (stp), iron is inert to nitrogen; or to its absorption. The diatomic molecule of nitrogen is anyway too large to penetrate the surface; and even if it were a single atom, it would not have enough kinetic energy to penetrate into the fields of the iron lattice matrix.

Fig.1 - The splitting of gaseous Ammonia 2NH3 ----> 3H2 + N2

IN an oven; a current of ammonia gas is passed over the work pieces which are heated slowly to 580°C. The ammonia, splits under catalytic [Fe] interaction on the glowing surface, dissociates into atomic nitrogen and atomic hydrogen (Fig.1). Only the atomic nitrogen penetrates into the metal surface. However, the side reaction of converting atomic nitrogen into molecular nitrogen is faster than the absorption of nitrogen into the hot iron surface. [To proceed however], the partial pressure of the nitrogen-atmosphere on the iron surface must be [kept relatively] high. In order to maintain an advantageous partial pressure of atomic nitrogen; a strong current of ammonia must be continually replaced and circulated around the work.

[At the appropriate temperature, activity begins. The precise chemistry is a little unclear, but the results are demonstrable. Either through direct chemical action with iron, or assisted by secondary catalytic processes, the active monatomic] nitrogen enters the surface matrix of the metal. [The nitrogen then proceeds to diffuse randomly within the lattice, towards the centre of the material. In this state - the iron (alpha)-a-lattice remains in body centred cubic (bcc) form up to 906°C - the nitrogen in the centre of the cubic iron matrix, is classed as the interstitial nitride [Fe₄N]; despite its classification, the nitrogen can migrate and diffuse within the material.]

[Conditions are different around the surface however. The chemistry, still a little unclear; but again either through direct chemical action, or by secondary catalytic processes, the iron atoms at the surface become scavenged and break away. The lattice structure begins to break up and opens. In this area, but only for a very short distance into the material, with the ionic bonding of the (bcc) lattice structure broken; it is believed that the stoichiometric ferrous [Fe₃N₂] and ferric [FeN] nitrides form and collect. Within this area too, some percentage will be of the interstitial [Fe₄N] type.]

[Throughout the process duration, the surface area sees high activity; allowing a steady flow of nitrogen into the body of the material, and experiencing the in and out flux of atoms that are not finally reacted or captured.] Near surface nitrides conglomerate into small crystals, [further stressing the surface lattice.] This is the compound layer. [The layer slowly deepens as the process continues].

The layer of material beneath, and the main body of the material, is called the diffusion layer. []
The duration of the process can be varied between 5 to 100 hours. Afterwards, the oven is slowly cooled down.

Fig.2 - Compound layer of nitrided sheet-metal (9h)

Fig.3 - Diffusion layer of nitrided sheet-metal (9h)

6. The Problems
It is a special case to make [steeldrum] instruments with nitrided sheet steel. [Because of its stiffness,] nitrided sheet steel is not meant to be deformed. [For our purposes] nitriding can only take place after 'sinking'; this is why the process is divided into two steps: In the first, the steel is deepdrawn; in the second, it is nitrided.
As tuning is performed after nitriding, a degree of ductility is needed; further, the steel should not be so hard as to break.
Previously where [drum-head] steel has been deformed by hand; the unequal areas of thickness caused unequal areas of hardness over the work area of the material. [Deepdrawn sheet addressed this problem by allowing a material of uniform thickness to result.]

7. The Parameters
Nitrided sheet steel parameters such as hardness and thickness of both the compound layers, and the diffusion layer, are mainly influenced by the composition of the steel. PANArt has worked with several sheet metals of different composition. This requires the knowledge of the composition of the material. Alloyed Steel ZStE for example, gets very hard with nitriding because of the nitride formers such as silicone [Si] and titanium [Ti] contained in the steel. So the process duration has to be adjusted.

Nowadays we work with common steel (Fe PO1 or Fe PO4), which is used for deepdrawing. This steel has a very low content of nitride formers.

The nitriding process duration is between 7 and 12 hours; left any longer - say 2 hours, and the hardened steel would break with further deformation. So the final parameters are strongly influenced by the duration of the nitriding process.

With the correct selection of steel composition, and setting of the nitriding process duration, the properties of the sheet material can be adjusted to our needs.
It has been a long path, of trial and error, to find the right parameters for steel with which to make [steeldrum] instruments.

8. Stress+Strain Diagram

Fig.4 - Stress+Strain diagram of St 12.03 steel

Fig.5 - Stress+Strain diagram of gas-nitrided St 12.03 steel (9h)

The two stress-strain diagrams (Fig.4 and Fig.5) respectivly show the sheet steel before and after nitriding. The steel is a common deepdraw steel Fe PO1 (St 12.03); and has already been deepdrawn.

The comparison leads to the following observations:

• The tensile strength increases from 300N/mm2 to 500N/mm2 (MPa)
  
• The yield point increases more than twice from 160N/mm2 to 378N/mm2
  
• The modulus of elasticity increases from 190kN/mm2 to 200kN/mm2 (GPa)
  
• The uniform elongation decreases from 26% to 9%

9. Treatments after Nitriding**
After forming the note shells, the instruments are heated to release macro-stresses. Fig.6 shows a stress-strain-diagram of nitrided steel, which has been heated for 4 minutes to 500°C. The E-modulus, the yield point and the tensile strength again increase. This shows that the properties are again improved.
It is not understood why this happens, probably because of a shocking in air, so the nitrides age and brace the iron matrix.

Fig.6 - Stress+Strain diagram of gas-nitrided St 12.03 steel - Heated for 4 minutes at 500°C - Used in Pang Manufacture

11. Conclusions - General
Hardening sheet steel by gas-nitriding offers new possibilities in making sheet material stronger. It is a [relatively] easy treatment with low costs.
The processes can be [generally] divided into two steps; first the [deepdrawn] forming, and second the hardening by gas-nitriding.

The material does not need to be strengthened with work hardening, [as was traditional with 'sinking'. Deepdrawn sheet] - the rawform hemisphere - [addressed part of this problem, and additionally allows a material of uniform thickness to result.]
The tuner becomes independent from the steel producers, and no longer needs to search for special composition steels.
The gas-nitrided sheet steel has a new structure, and the sandwich unites several [desirable] qualities.
The new structure is a sandwich with an hard surfaces and soft core; it is [near] brittle and at the same time ductile.

The new structure has a new combination of characteristics which does justice to the complexity of the instruments [manufacture and tone]. Remember the string? It has a strong core around which is wound another soft material.

Although instruments like steelpans can be built [using] this new material; a consequence of this new material is the development of new note-shapes, and of new sounds.

It has been a challenge to invent new [materials and] instruments and PANArt are sure it will enrich the future [of] music.

12. Acknowledgements
The authors wish to thank Prof. Rufer, HTA Biel CH; B. Stucker, Härterei Duap, Herzogenbuchsee CH and H. Schaub, Härterei Hans Schaub AG, Fällanden, CH.

13. References
1. H. Maeder and J. Bartanus, Diplomarbeit 1999, Werkstoffkunde HTA Biel, CH.
2. Merkblatt Stahl 447, Nitrieren und Nitrocarburieren,1983 Düsseldorf, D.

============
** NOTE: Editorial placement changes:
The original order of position of headings/subjects of Sections numbering 9 (was 10) and 10 (was 9) have been interchanged here.
eEd.
============

Notes on Hardening, Annealing and Yield point Following from the comment of Schärer & Rohner (Section 9) that 'the E-modulus, the yield point and the tensile strength again increase' - linked to their requirements to 'heat treat for 4 minutes to 500°C' (anneal) their work-pieces to 'release macro-stresses' after having imprinted their 'notes' on the 'nitrided hemispherical form'; and where they conclude that 'It is not understood why this happens, probably because of a shocking in air, so the nitrides age and brace the iron matrix'; leads on to the following discussion on some properties of nitrided steel, as observed by others. In general, we are reminded that we are discussing the properties of the bulk-material (the inner diffusion layer) of the sandwiched nitrided sheet, which comprises in the main a solid solution of iron and nitrogen. (The effect of the hardened 'skin' in the stress-strain test, as it comprises approximately 3.3 x 10-3% of the surface area of a 0.9mm thick sample; would be masked by the bulk of the material; and notwithstanding that because of its brittleness, and also the porosity of its structure; would have experiences brittle-surface-fracture - somewhere in the elastic region of the test.) Without quantitative data on the resultant percentage of nitrogen contained in the diffusion layer; and further without intermediate time related stress-strain data, of samples taken between the time of nitriding and the time of the annealing process; we remain at a disadvantage to be able to determine whether time (age) or annealing, or which mixture of both, may be affecting the final results. But what we do know is that, similar to carbon-steels, the properties of nitrided-steels will be affected by the concentration of nitrogen in solid solution. And further that the saturation of nitrogen at higher (process) temperatures, may force super-saturation at lower (ambient) temperatures, to promote characteristics of ageing when the excess nitrogen, in time, comes out of solution. But perhaps more relevant to this discussion, is the following note by D. Tabor in relation to the pinning of dislocations to reduce stress. It would appear that, in annealing their work pieces to reduce work-hardening stresses, Schärer & Rohner have inadvertently encouraged (accelerated) the further migration of nitrogen to preferential interstitial and boundary sites within the material, that would then further increase the overall strength of the diffusion layer. This conclusion is based on the observation that the final stress strain diagram (Fig.6) supplied by Schärer & Rohner, matches to some degree the characteristics described by Tabor in Fig.B below. Two notes are taken from Tabor; Work-hardening and Upper yield point.   Work hardening Dislocations interact with one another. Continued deformation produces interlocking of dislocations and makes it more difficult for them to move. Although the details are still the subject of discussion this is the basic reason for work-hardening as a result of deformation. Upper yield point The edge dislocation is a region of large elastic stress. At LBC in Fig.A-(a) the material is heavily compressed, at MF strongly extended. If impurity atoms are present in the lattice they will tend to concentrate at regions where they can relieve the stress. For example a small impurity atom will tend to take the place of the original atom B. This will lower the intensity of the stress field around the dislocation. It is precisely the presence of the stress field that makes it easy for a dislocation to move. The presence of the impurity at B thus tends to 'pin' the dislocation and a larger stress than normal will be required to make it move. Once the dislocation has moved across to position C or D it is in its normal environment and is now much easier to move. This explains the behaviour for example of certain mild steels where the dislocations are pinned by interstitial carbon or nitrogen. When such materials are tested in tension or compression or shear they show an 'upper yield point' (see Fig.B, point B), then a region of fairly constant lower yield stress (CD) followed by a conventional work-hardening curve. If at some point F the stress is removed and the metal is 'aged' the carbon and nitrogen can diffuse to the dislocations and again pin them. The further deformation of the specimen may then show a second 'upper yield point' (G). D. Tabor; Gasses, Liquids and Solids; Ch. 9 The strength properties of solids, Sec 9.2, Pg 161 - 163; Penguin, 1970     eEd - November 2003

Compiled & Processed by Islands Research for:
The Steelbands (Pan) of Trinidad & Tobago
http://www.seetobago.org/trinidad/pan/archive/r&d/panart/ir_archivecopy_hardening_steel_by_nitriding.htm

[2nd Ed] © 2003: tobagojo@gmail.com - 20031029 - 1m20071228 - 2m20140615 Historic Update: 16 November 2003; Last Update: 29 June 2014 02:11:00 TT Processed by: JdeB - Islands Research

UP