After nearly four decades, Platinum resurfaced in the North American jewelry market in the mid 1980's on the coattails of surging Japanese popularity. Early marketing directed at younger, less conservative consumers suggested that wedding bands would be a base from which to expand Platinum alloys into the North American jewelry markets. At that time, Johnson Matthey's predominant method of manufac¬ture for wedding hands was to stamp ring blanks from cold rolled sheet, followed by various forming and machining operations to produce final product. There were several advantages to this manufacturing technol¬ogy including fine equiaxed grain structure resulting in excellent ductility, hence the ability to expand / reshape inventories of fewer base sizes to a multitude of I.D. sizes, band widths and band thicknesses.

Commercially it was the manufacturing method of choice world wide. However, there were significant issues with manufacturing costs such as labor sensitivi¬ties (numerous manufacturing step procedures) and overall metal utilization, as sheet webbing and center discs outweighed finished product ring washers.
Realizing the potential market pressures for projected industry growth, JM took a critical look at this "Industry Standard Practice", and concluded that a sig¬nificant improvement in manufacturing efficiency would be paramount to our future success. Therefore in 1989 we began exploring alternate manufacturing technologies within the global Johnson Matthey family that would be:

1) Technically equivalent to rings stamped from sheet
2) Substantially more efficient to produce

This paper will share the Johnson Matthey assessment of several alternate manufacturing techniques, circa 1989, that led to the selection of Hot Extrusion of Seamless Tube as our primary method of manufacture resulting in the production of over 1/2 million T.O. of heavy wall platinum tube that the North American jewelry industry has converted into over 2 1/2 million wedding bands from 1991 through 2002.
 

The Manufacturing Methods
I. Stamped Ring Blanks

Production of Platinum ring blanks by stamping rings / washers from sheet was the most popular manufactur¬ing method worldwide in 1989 and the Johnson Matthey method of choice.

The manufacturing process, however, was lengthy and carried a substan¬tial quantity of metal to nearly the end of the manu¬facturing process where the rings / washers were actu¬ally stamped, leaving sheet webbing and center disc portions to be returned to the melt stage.

A typical process for manufacturing "stamped ring blanks" requires as many as 12 manufacturing steps or sequences. There are several advantages to this methodology:

• It was the most widely accepted and mature manu¬facturing process and therefore an industry standard.
• Dimensional control was excellent.

• The resulting microstructure was observed to be uniform fine grained equiaxed after re-crystallization annealing, establishing the desired mechanical properties, forming and machin¬ing characteristics.

However, the disadvantages are equally obvious:

• The method is lengthy and therefore labor intensive.

• There is poor ingot to finish product yield with much of the Platinum alloy weight carried throughout the majority of process steps.

• There is a propensity for micro-structural distortion at the sheared surface both ID and OD which require rectification by mechanical metal removal processes.

• Punch and die sets wear quickly and require signifi¬cant maintenance / eventual replacement.

Therefore, our overall assessment was that although stamping ring blanks was a well established method producing technically acceptable parts, there needed to be substantially improved material utilization and / or a more streamlined manufacturing process.
 

II. Cast To Size Ring Blanks
Potentially the least labor intense manufacturing method for near net shape production is to cast ring blanks as close as possible to finished size. Certainly as pre-finished rings requiring only final machining to desired geometry / designs could be pro¬duced in as few as 5 manufacturing steps. However, among the unique properties and characteristics of platinum and the 950 platinum alloys is high melting point. Although this might be an attribute for extreme temperature applications where strength can be main¬tained at temperatures where many glass / ceramic compositions melt, one must also consider the distress on casting tools such as crucibles, flasks, and molds. It is readily apparent that casting Pt is a significant technical chal¬lenge.

Techniques to improve fill and reduce shrinkage such as insulating smaller cross sec¬tions, raising mold preheat temperatures are impracti¬cal due to the small part geometries and limited tem¬perature capabilities of mold material. Increasing the cross-section and number of gates and risers to accom¬modate fill is counterproductive to metal utilization efficiencies. Finally, raising super heat temperature for pour to improve fill increases shrinkage porosity, and also produces undesirable metal / crucible / and metal / mold reactions that induce and / or trap contaminat¬ing vapors and inclusions.

In addition to the technical concerns centered around high melt temperatures, complete fill and porosity, as-cast structures can also adversely affect mechanical properties, and therefore formability and machinabili¬ty.
Our assessment of cast-to-size wedding ring blanks, although promising in regard to minimal processing steps and near net shape capabilities significantly low¬ering manufacturing costs, was overshadowed by the technical disadvantages of non-uniform cast microstructure, shrinkage porosity and inferior mechanical properties compared to ring blanks stamped from sheet.

III. Deep Draw Ring Blanks
Another mature and widely accepted seamless tube manufacturing method, popular primarily for smaller / thinner wall tube products, is the cupping and forming of a sheet disc to form a tube hollow for subsequent mandrel or plug drawing of seamless tube. However, this manufacturing route, like stamped ring blanks, is long and therefore labor sensitive as well as geometry limited.

Here again, the advantages are maturity / acceptance of practice, and uniform fine-grained microstructures with mechanical properties equal to that of stamped ring blanks. However, the manufacturing process is extensive and therefore labor intensive. Working mar¬gin is not adequately reduced as round deep draw discs cut from rectangular sheet, as well as removal of the cupped end, substantially inhibit yield. In addition, depend¬ing upon equipment capabilities, there can be significant limitations on OD / ID Wall and overall length dimensions.

IV. Machined From Solid Bar Ring Blanks
In the late 1980's, Johnson Matthey began a major capital improvement plan to upgrade our machining capabilities with the acquisition of Computer Numerical Control (CNC) equipment including center hole boring capabilities. With this capability we exam¬ined the potential for center drilling solid rod to form heavy wall tube hollows. Boring as-cast rods was rejected for many of the same difficulties presented by casting ring blanks directly, such as porosity and undesir¬able as-cast micros¬ructures.
 
In addition, on a microscopic basis, casting dendrites continued to be predominant throughout the cross-section. Although OD and ID could be readily machined to produce a heavy wall tube, resulting ring blanks could not be readily sized by tile predominant expanding techniques at the time.

The advantages are straight forward - a significant reduction in operations from melt to finished ring blanks compared to stamping ring blanks from sheet, while achieving the desirable porosity-free, fine-grained microstructures with mechanical properties suitable for machining. The disadvantages are primarily related to working margin and physical limitations that restrict overall tube length.

Drilling long lengths of rod with cantilevered tooling to desired concentricity requirements was found extremely difficult beyond 8 - 10 inches, therefore, longer tubes suffered from increased variation in wall thickness.

V. Powder Metallurgy Ring Blanks
By 1989 Johnson Matthey had considerable experience with PGM powders.  Although most other metal industries forming components by PM process¬ing were producing closely controlled powder particle geometries, we were unable to find any vendors capable of dealing with the high Pt alloy melting points, and consequen¬tial excessive orifice wear of bottom feed crucibles.

However, since the process is straight forward, less labor intense than stamping ring from sheet, and metal efficient due to near net shape manufacture, we felt this technology warranted further evaluations.
There was as much variation from batch to batch of one sponge supplier as there was from suppli¬er to supplier. The sponge frequently contained unde¬sirable moisture content and could range from free flowing to agglomerated lumps. Flame spray and arc sprayed powders were significantly more consistent in regard to shape (generally spherical). However, stan¬dard distribution of particle sizes generally resulted in distributions with less than 10% of the particles in the desired range.

Although, we observed significant advantages in the:

• Reduction of process steps required, and therefore reduced labor sensitivities

• Near-net-shape capability and therefore improved material yield

• Potential for a wider range of alloys

• A very fine-grained microstructure with improved ductility

We were unable to develop a reliable source of consis¬tent and cost effective powder. Further, compromises on the integrity of powder source were unacceptable in regard to product quality and / or too costly to rec¬tify by additional secondary operations. 

Hot Extrusion of Heavy Wall Seamless Tube
With the 1989 purchase of a 400 ton extrusion press intended primarily for the production of high volume Pt alloy rod / wire products, Johnson Matthey was in a unique position to investigate the extrusion of cored Pt alloy billets over a mandrel to produce heavy wall seamless jewelry tube that up until that point had been demonstrated only for karat Au alloys. The man¬ufacturing process was successfully reduced from 16 – 18 steps for stamped ring blanks to 8 – 10 for tube extrusion. In addition, the bulk of working margin was removed from the process after step two, permitting rapid return to the initial melt stage. The advantages include:

• The attainment of porosity free fine-grain microstructure - and the associated desirable mechanical properties.

•Excellent dimensional control

• Material utilization superior to stamping ring blanks from sheet.
The only disadvantages we were able to identify initial¬ly included:

• Shallow OD / ID surface distress from frictional inter¬actions between the die, tube and mandrel rectified by secondary operations such as centerless grinding of OD and ID honing.

•Tool wear at high Pt alloy hot working temperatures.

Overall, however, the assessment as compared to all of the other ring blank manufacturing methods was very positive. This process represented a reduction in pro¬cessing steps and increased material yields by 100% while delivering finished ring blank quality equal to or better than that produced by the stamp from sheet process, thus meeting our original objectives and preparing Johnson Matthey For the demand in com¬mercial ring blanks that was to follow over the next ten years. 

The Final Analysis "Stamped Ring Blanks" versus Extruded Tube
A comparison of the manufacturing yields and labor sensitivities for stamped ring blanks and extrusion of tube was made.

Several key material yield advantages of the extrusion method were highlighted by this analysis. At only 10%, the material yield of the stamping process is well below that or the extrusion method at nearly 40%. Additionally, metal employment is lowered by 2000 grams per melt; representing one extra extrusion melt for every six sheet melts for stamped ring blanks. Looking at the incremental yield for each method, it is seen that nearly half of the working margin scrap is returned to melt by the mid-point of the extrusion process; whereas for the sheet method, nearly the entire working margin is carried through the process to the stamping stage.

At first glance it would appear that these improve¬ments in material yield come at the expense of labor hours, as evidenced by the 20% increase in labor time per melt for the extrusion method relative to the stamping method. However, when the overall material yield is factored in, it is evident that for a given size ring blank, the extrusion method will yield approxi¬mately three times the number of blanks that can be stamped from one sheet melt. Therefore, with the necessity of processing three melts by the stamping method, the labor time increases to more than double that of the extrusion method for a comparable number of finished blanks. Additionally, with the need to process three times the number of melts, metal employment and working margin become even larger issues than initially indicated.

Taking the material yield and labor sensitivities into consideration, it is evident that the extrusion method represented significant improvements in process effi¬ciency without sacrificing product quality.

The Last Ten Years
There have been a number of improvements to the manufacturing procedures evaluated by Johnson Matthey since 1989 / 1990. Technically advanced stamping equipment and tooling now permit the simultaneous stamping of ring and center disc in a sin¬gle operation, saving time and further insuring OD / ID concentricity. Improved tooling materials require reduced maintenance. However, excessive webbing and disc working margins, plus secondary machining operations to clean up damage material at sheared edges remain as undesirable by products of this tech¬nology.

We have also come a long way in casting Pt alloys, both in the development or flask / mold materials and casting equipment / techniques. Casting trees with reasonable gating systems can produce significant yields as demonstrated by Jurgen Maerz and Michael Epsteinfor "Friendship Rings". Unfortunately, as-cast grain and dendrite structures still prohibit expansion / sizing, an essential requirement to minimize stocking volumes for wedding band application.

Much of the manufacturing improvements for the man¬ufacture of tubing by the deep draw method has cen¬tered on the very small geometries. However, the more popular heavy wall wedding band configurations remain outside the range of PGM manufacturing, and the process remains both labor and working margin sensitive.

There have also been advancements in the ability to produce longer length center-bored wrought rod stock. There are a limited number of vendors with highly pro¬prietary processes who will quote heavy wall drilling capabilities up to 8 ft. in length. However, the toler¬ances beyond 12 inches remain undesirable. Further, recovery and refining issues with tile center-bore machine scrap by subcontract vendors remains a significant and costly security issue when dealing with platinum alloys.

There have been exciting technical improvements for the production of near net shape wedding bands for 950 Hallmark Pt alloys. High density fine-grain microstructures are attainable and the need for expensive secondary operations are no longer required for densities at 98% Plus. Unfortunately, the process continues to be relat¬ed to a reliable source of powder within defined geo¬metric parameters.

Over the last ten years Johnson Matthey has contin¬ued to improve the extrusion manufacturing methods and resulting tube quality. Extensive development of die geometry / materials, mandrel materials, and high temperature lubricants has significantly extended tool¬ing life.

The Next Ten Years
Observing how much progress has been made in the platinum jewelry industry in the past ten years leaves no doubt that there will be more technological devel¬opments to come. We are developing and learning how to work with increasingly difficult alloy composi¬tions.

What else might our industry develop over the next 10 years? Perhaps the momentum behind casting improvements will continue and techniques that promote extensive nucleation of fine grain equiaxed microstructures in aero space alloys, may be incorporated into jewelry applications. Perhaps we will overcome current tooling deficiencies for the high inching point Pt alloys.

Whether a current technology breakthrough, the development of an all new technology or a hybrid of two or more existing manufacturing methods, it will be exciting to see whether or not an emerging technology will surpass Hot Extrusion of Heavy Wall Seamless Tube and become the next standard in ring tube / wedding band manufacturing.

V11N2

Hot Extrusion of Heavy Wall Seamless Tube for Platinum Jewelry Applications
Edwin A. Crombie and Dave Arey
JOHNSON MATTHEY, NOBEL METALS N.A.

This is an abbreviated version of the original work. For full technical details, please consult the original paper.