AJ033 Cracking UP - World Gold Council

UP
MINIMIZING CRACKS DURING
JEWELRY MANUFACTURING, AND BEYOND
Cracking
By Peter Raw and Christopher W. Corti
A
s many jewelry manufacturers and goldsmiths know from hard experience,
cracking in jewelry can occur at any time during its manufacture. It can also
occur much later, after the jewelry has been sold to the consumer or during
repair. Cracking can also occur in the processing of the starting materials (the casting
grain and mill products from which the jewelry is to be made), and may not be
detected until several stages later in the manufacturing process.
Whenever it occurs, cracking is at the least an inconvenience and an undesirable
cost, and at worst may reflect adversely on the jeweler’s reputation.
In many cases, however, it can be prevented by paying careful attention
to each processing step. The challenge for jewelers (and repairers) is to
understand which of the many possible causes is responsible for a
particular incidence of cracking.
The various causes of defects, many of which manifest themselves
as cracking during manufacturing, can be attributed to the following
problems:
• Poor quality start materials, including recycled scrap, leading to contamination
and possible embrittlement;
• Poor melting practice, leading to casting defects and/or gas porosity and blisters, incorporation
of inclusions, excessive shrinkage porosity, and segregation;
• Poor ingot or material working practice, often related to changing alloy composition without
changing the working procedure;
• Incorrect annealing practice, often due to incomplete metallurgical knowledge of the karat golds;
• Stress corrosion cracking, to which some 14k and many lower karatage gold alloys are susceptible;
• Quench cracking and fire cracking in nickel white golds.
This article discusses these causes as they relate to cracking and the steps that can be taken to minimize
their occurrence. The particular focus is on karat golds, but much is also applicable to silver and
platinum jewelry.

Start MATERIALS
When making karat gold
alloys, it is essential to start with clean, oxide-free metals, whether
they are pure metals or pre-alloys (master alloys). All should be
analyzed or purchased with certificates of analysis. The purity of
gold should be at least 99.9 percent, with lead, tin, bismuth, antimony,
selenium, and tellurium impurities specified as less than 0.01
percent. These impurities can be present in mined gold and can lead
to alloy embrittlement—a tendency to crack when a load is applied,
such as when working the material.
A frequent cause of problems, however, is the use of scrap
gold, which tends to be a recurring source of contamination. This
is particularly true for scrap bought from external sources, commonly
used as a start material in some countries. But even internally-generated
scrap can be problematic, especially if it is recycled
because of prior process failures. The use of scrap to make
new product should be strictly controlled. Preferably, the gold
should be subject to melting and analysis before it is used to make
new alloy ingots, or recycled in investment casting.
Typical contaminants in scrap include refractory materials,
such as investment particles on unclean sprues, oxides
3
from dirty surfaces, silicon from casting alloy, and lead-tin
solder from repaired jewelry. Scrap jewelry containing soldered
joints may introduce indium, germanium, or tin. All
these contaminants can lead to inclusions or alloy embrittlement.
As a result, the only guaranteed safe way of utilizing
scrap is to refine it first.
Embrittlement by low melting point metals (and silicon)
tends to result from the formation of low melting point
metallic second phases. These phases are either the contaminating
metal itself, which generally has extremely low solubility
in gold, or they occur as a reaction product of the contaminant
with gold, silver, or copper. The effect is magnified if the grain size
of the alloy is large, since these second phases tend to be dispersed
as very thin films around the grain boundaries. Fine-grained alloys
will tend to have a lower concentration of embrittling phase per
grain boundary area. Often, these contaminants manifest themselves
as cracking during metal working operations. However, as will
be discussed later, there are other reasons for karat gold alloys failing
during fabrication.
Melting and CASTING A very significant
proportion of karat gold jewelry is manufactured either by investment
casting, using casting grain, or from mill products (sheet,
strip, wire, or tube) that are derived from statically-cast ingots or
continuously-cast material.
Investment (lost wax) castings are prone to embrittlement,
particularly when silicon-containing alloys and scrap are used.
Problems can also arise due to unclean gold scrap (even when it
doesn’t contain silicon), inclusions from crucibles, weak investment
molds, and shrinkage and gas porosity. Large porosity, in
particular, may act to cause cracking during subsequent processing.
Hard refractory inclusions (i.e., ceramic particles from melting
crucibles or investment molds, or “dirt” from the workshop
that falls into the melt) resist deformation during working and act
as crack initiators in the gold alloy. If present on the surface, they
can break out, leaving large surface porosity that is drawn into a
longitudinal surface crack on further working.
Continuous casting of karat gold alloys almost always uses
high density, fine grain graphite for the mold material to ensure
good quality product. It is capable of giving much higher quality
(and higher product yields) because there is no shrinkage pipe, as
occurs with statically-cast ingots. However, erosion of the mold can
lead to graphite inclusions in the melt. Surface defects are also a possibility
if mold wear or sticking
occurs to any extent. By
1
and large, though, continuously-cast
materials seldom
give rise to mechanical defects.
(An example of mech-
2
anical defects from continuously cast materials is Figure 1, which
shows cracks on ring surfaces after working. The rings were made
from continuously-cast 14k tube, and the defects are attributed to
incorrect casting conditions.)
Static casting of ingots tends to be a much simpler operation,
with melting by gas heating, oil-fired furnaces, electric resistance
heating, or induction heating. Induction heating ensures maximum
stirring of the alloy constituents, making it the preferred
method of heating, although other heating methods combined
with physical stirring of the melt with graphite or refractory rods
is commonplace. Crucibles are typically clay-graphite or graphite
(fireclay for nickel white golds, as nickel will react with graphite),
and molds are typically made from iron or water-cooled copper.
OPENING PAGE: PHOTO COURTESY OF GREG NORMANDEAU. FIGURE 1 COURTESY OF DIETER OTT, FEM. FIGURES 2 AND 3 COURTESY OF MARK GRIMWADE.
42 > AJM

FIGURE 4 COURTESY OF MARK GRIMWADE. FIGURE 5 COURTESY OF STEWART GRICE AND FRED KLOTZ. FIGURE 6 COURTESY OF DIETER OTT, FEM.
Static casting can be a source of several problems, including:
Shrinkage and pipes. When a cast ingot solidifies, it shrinks.
This becomes evident as a central pipe at the top of the ingot. (Figure
2) This pipe must be cut off before working the ingot, otherwise a
central defect will be introduced that will elongate upon working.
This central defect is likely to result in subsequent longitudinal cracking.
The pipe will be more pronounced if the casting temperature is
too high, so it is normal practice to restrict the casting temperature to
no more than 200°F/93°C above the liquidus temperature of the
alloy. High casting temperatures also encourage large grain sizes,
which decrease the ductility of the alloy and magnify the effect of any
low melting point impurities that might be present.
Blistering and porosity. Surface blistering or internal gas
porosity can show up later in fabrication operations as surface
5
4
defects or cracks. In this situation, gas from the start material (dissolved
gas or damp materials) or gas dissolved during the melting
operation (aggravated by too high a melting temperature, lack of a
protective atmosphere or a flux, and use of gas melting) evolves as
porosity during solidification. Initial working may flatten the pores
and cause small laminations and cracks, or it may close the porosity,
only to have the gas expand later during annealing operations and
reappear as blisters.
Inclusions. Inclusions can be incorporated into the melt from
several sources, including erosion of the crucible (replace crucibles
6
before significant wear occurs), from furnace insulation or lining, or
from broken stirring rods. They can also be caused by a reaction
between the atmosphere and alloying element (for example, oxygen
and copper forming copper oxide), or the use of grain refiners that
have not been dispersed correctly, particularly iridium, which is very
insoluble in gold and forms hard clusters of particles. Such inclusions
can give rise to cracks or failure during subsequent working because
they act as stress concentrators, which initiate cracks.
Surface defects. Surface defects on the ingot can also lead to
cracks. These defects can arise due to poor melting and casting practices.
They include surface inclusions, oxidation, mechanical damage,
and solidified splashes during casting that stick to the mold wall.
Many of these problems can be avoided by inspecting all ingot surfaces
and cleaning away all evidence of defects before any working
operations are undertaken. If necessary, the ingot surface might
have to be milled to ensure it is clean and flat.
Cracking During FABRICATION
Cracking can occur at any stage of fabrication, including:
Overworking. All forms of metal working—including sheet
and rod rolling, wire and tube drawing,
blanking, stamping, coining, spinning and
raising, milling, turning and machining,
and simply bending by hand—result in the
material becoming harder and less ductile.
The degree to which it hardens and loses
ductility depends on the amount of deformation
imparted. If material is overworked, the ductility reduces
to zero, and it will crack.
Annealing restores the material’s ductility, and so is normally
performed at appropriate stages in the working process. The rate
at which alloys work-harden and the extent to which they can be
worked before annealing varies from alloy to alloy. Typically, karat
gold alloys can be worked up to about 70 percent reduction in
area (strain) before they require annealing. However, there are
considerable variations; for example, nickel white golds harden
rapidly and normally require annealing after a 35 or 40 percent
reduction. On the other hand, fine gold and some of the high
karat golds can be worked well in excess of 90 percent reduction
in area before annealing.
Overworking can cause several problems. For example, edge
cracking during rolling of sheet material is normally the result of
overworking. To avoid further problems, the edges must be
trimmed, since rolling after annealing will increase the danger of
some cracks running in towards the sheet center. (Figure 3)
Problems that occur during rod rolling include the formation of
fins, which are caused when too much material is pushed into the
March 2003 < 43

44 > AJM
rolling groove, so that the rolls are forced
apart and excess metal is squeezed out sideways.
(Figure 4) These fins are then rolled
into the rod, becoming laps. Both fins and
laps can open up as cracks at later stages of
fabrication. Their formation can be prevented
by avoiding too large a reduction
and by rotating the rod through 90 degrees
between successive passes.
Localized overworking can also cause
cracking during sheet metal forming
operations, such as stamping or deep
drawing. (Figure 5) Fracturing occurs at
the weakest or thinnest point, which in
forming operations is usually where the
sheet bends around the tool. It may be
necessary to partially shape the component
in one die-set and then further form
it in another die-set. Selection of the correct
material and processing conditions
are important, and will depend on individual
circumstances.
Embrittlement by impurities. As we
discussed earlier, certain impurities, including
silicon, will embrittle gold. Any attempt
at working embrittled material will result in
cracking. (Figure 6 shows cracking in 18k
gold due to silicon embrittlement.)
Another reported source of embrittlement
is contamination from lead formers.
Manual working, such as raising, and repair
operations often involve hand-working on a
soft former, frequently made of lead, to prevent
surface damage. We know of an example
of embrittlement in which lead from the
former contaminated the surface of the
gold, with the lead diffusing into the gold
during subsequent annealing or soldering.
Because of the possibility of such contamination
that can lead to embrittlement, the
use of metallic lead in contact with gold is
risky. If the technique is considered essential,
the gold should be separated from the
lead with a tough grade of paper.
Incorrect annealing practice. Incorrect
cooling conditions after annealing
can, paradoxically, lead to hardening

ather than softening in some karat golds.
On subsequent working, the material
cracks. Golds in the red to yellow range,
8k to 18k, should be rapidly cooled after
annealing by quenching directly in water;
this maintains a soft ductile condition,
whereas slow cooling results in hardening.
Repairers should also anneal and water
quench such jewelry items before re-sizing
or repairing for this reason.
Over-annealing the metal at too high a
temperature and/or for too long a time can
also result in cracking. Over-annealing produces
a large, coarse grain size, and subsequent
deformation can lead to premature
cracking and fracture, as well as an “orange
peel” surface. This is a problem particularly
with torch annealing, where the capability to
control temperature is limited. To avoid
over-annealing, use the lowest effective temperature
and only the time needed.
The issue of cracking arising during
the fabrication of karat gold jewelry, or
later during service or repair, can be complex.
Although there are well-defined
causes for cracking, a crack’s appearance is
not uniquely associated with one particular
cause. Establishing the precise reason
for failure may require specialized equipment
and knowledge. The situation may
be further complicated by defects that
arise as a result of more than one cause.
However, there are probably two
aspects of manufacturing that contribute
most towards minimizing the production
of defective or scrap jewelry products:
First, a good understanding of the metallurgy
of the karat golds and, second, the
establishment of good manufacturing
practices for materials and products—and
the strict adherence to those practices. ◆
Peter Raw is a technology consultant in
Surrey, UK, and Christopher Corti is director,
International Technology, at the World
Gold Council, London.
AJM READERS’ RESOURCES
WORKING
WITH
GOLD
“Handbook of Casting and Other
Defects in Gold Jewelry Manufacture,”
Dieter Ott, World Gold Council, London,
1998.
“Causes and Prevention of Defects
in Wrought Alloys,” Mark Grimwade,
Gold Technology, No. 36, Winter 2002.
“Live and Let Die (Struck),” Fred
Klotz and Stewart Grice, Gold Technology,
No. 36, Winter 2002.
“But I’ve Always Done It This Way.
Technical Support—It Makes a
Difference,” Stewart Grice, Gold Technology,
No. 25, April 1999. Also in
Proceedings, Santa Fe Symposium on
Jewelry Manufacturing Technology,
1998, Met-Chem Research.
Note: Gold Technology and the
“Handbook on Defects” are published by
World Gold Council. Copies are available
from the London office. To obtain
copies, e-mail chris.corti@gold.org. Recent
issues of Gold Technology are also
available online at www.gold.org.
For additional resources, including
information on minimizing stress corrosion
cracking, quench cracking, and fire
cracking, visit AJM Online at www
.ajm-magazine.com.
March 2003 < 47