Electroforming Process and Application to Micro/Macro Manufacturing
J.A. McGeough' (I), M.C. Leu2 (2), K.P. Rajurka? (I), A.K.M. De Silva4, Q . Liu2
University of Edinburgh, School of Mechanical Engineering, Edinburgh, UK
Department of Mechanical & Aerospace Engineering and Engineering Mechanics, University of
Missouri-Rolla, USA
Industrial and Management Systems Engineering, University of Nebraska, Lincoln,USA
Department of Engineering, Glasgow Caledonian University, Glasgow, UK
'
Abstract
Electroforming is the highly specialised use of electrodeposition for the manufacture of metal parts. This
paper describes the process principles and mechanisms of electroforming, outlining its advantages and
limitations. A review of modelling and simulation of electroforming and experimental analysis work is also
presented. The metals that can be electroformed successfully are copper, nickel, iron or silver, thickness
up to 16 mm, dimensional tolerances up to 1 pm, and surface finishes of 0.05 pm %. The ability to
manufacture complex parts to close tolerances and cost effectively has meant that electroforming has
applications both in traditional/macro manufacturing and new micromanufacturing fields. These include
tooling; mould making; fabrication of microelectromechanical systems (MEMS) and the combination of
lithography, electroforming and plastic moulding in the LlGA process. Applications in micro-optics and
medicine are included.
Keywords: Electroforming, Micro/Macro part manufacture
1 INTRODUCTION
Electroforming has been known since 1837 when it was
first observed by Jacobi during the electrodeposition of
copper onto a printing plate[l]. The American
Electroplaters' and Surface Finishers' Society (AESF)
defines electroforming as "the production or reproduction
of an article by electrodeposition upon a mandrel or
mould that is subsequently separated from the deposit".
A less rigorous definition is simply, "the art of growing
parts", which excludes the above requirements for
mandrel separation, but in some cases, is more
appropriate [2]. Although it is not a universally applicable
process, its ability to produce/reproduce shapes to close
dimensional tolerances with good surface finish and
superior metallurgical properties has ensured that
electroforming is a competitive process in precision
manufacturing. Its application areas range from
consumer products to highly specialised aerospace
components. More recently electroforming has evolved
into a highly specialised micro fabrication technique. This
paper outlines the principles and practice of the
electroforming process and reviews its applications in
traditional macro- as well as the new micro- fabrication
industry.
1.1 Basic principles
Electroforming is basically a specialised form of
electroplating. In electroplating, metal is dissolved
electrolytically at an anode. The basic principles of
electroforming are shown in Figure 1. The resulting
metal ions are transported through an electrolyte
solution, usually containing a high concentration of the
same metallic ions, to be deposited at a cathode. The
difference between electroforming and electroplating lies
in the purpose of use for the deposited metal.
Electroplating is concerned with taking an existing article
and applying a metallic coating to provide a decorative
and/or protective surface. An electroform, however, is a
metallic object that has been created by utilising the
electroplating process to deposit a metal on or against a
master form or mandrel.
Its purpose is to serve
functionally or decoratively as a separate entity.
Many engineers and designers are not fully aware of the
advantages of electroforming. A major reason is that the
technology is based upon the principles of
electrochemistry and alleged "black art". However,
electroforming can be subjected to a high degree of
control and can be operated with extreme precision and
reliability, as will become evident later in this paper.
Electroforming should be thought of as a basic
manufacturing process when considering alternatives
best suited for making any particular component. Other
processes such as casting, forging, stamping, deep
drawing and machining may serve well for most
applications. However, when requirements specify high
tolerances, complexity, lightweight and miniature
geometry, electroforming is a serious contender and in
certain cases may be the only economically viable
manufacturing process.
Figure 1: The principles of electroforming.
1
Electroforming may be distinguished in three main ways
from electroplating [ I ,3]:
The deposits produced by electroforming are much
thicker. Typically electroplated deposits vary in
thickness from about 7-50 pm. Electroforms,
however, are rarely thinner than about 18 pm and
very often can be several millimetres thick.
After electrodeposition the electroformed component
is physically removed from the mandrel. Therefore,
in electroforming the deposit must have low
adhesion to the mandrel to facilitate its removal.
Conversely, in conventional electroplating the
deposit is often used to protect the base metal (for
example from corrosion) and or to provide a
cosmetically attractive appearance. For these
purposes the deposit must adhere to the cathode.
Mechanical properties and dimensional accuracy
are key features in electroforming requiring control
over deposit composition, structure and internal
stress. The physical properties of an electroform are
independent to that of the mandrel. In contrast, the
physical properties of an electroplated deposit and
base metal usually have to be complementary.
In practice, the mandrel (cathode) has, or is given, the
desired shape prior to electrodeposition, and, because of
the ionic action of the process, an exact replica of the
mandrel surface is produced. This (negative) replica may
itself then be used as a mandrel to regenerate the
original (positive) form. Therefore, a large number of
identical forms can be produced from a single master.
1.2 Features of electroforming
Advantages
The characteristics of electroforming in comparison with
traditional metal forming processes give rise to
significant advantage, which determine the type of
applications for which the technology is used. By
electroforming components can be manufactured which
would be difficult or in some cases impossible to produce
by conventional methods of fabrication. Its principal
advantages are [I]:
High dimensional precision - after a mandrel has
been made to the required dimensions, replicate
electroforms can be produced all having precisely
the same dimensions (4pm), provided the
deposited metal is without internal stresses. The
technique is therefore suitable for producing
components where such exactness of form is
required, such as moulds and dies.
Precise reproduction of surface detail - t h e process
can also be used to give extremely high accuracy in
reproduction of surface detail. This level of
reproduction of fine detail (typically 0.01 pm) is
difficult to match with other production processes.
Production of complex-shaped components provided that the correct mandrel technology is
employed electroforming can be used to produce
complex shapes in a single operation. Such shapes
might otherwise require a multiplicity of operations
such as pressing, drilling, machining, deburing and
welding in order to be manufactured by alternative
techniques.
Production of thin-walled components - particular
advantage of the electroforming process is its
capability to produce thin walled cylinders, without a
joint line. Thin walled products represent a highly
commercial use of electroforming.
2
Extensive range of size - size is usually only limited
by the capacity of plating equipment available.
Products typically may range in size from a few
millimetres to several meters. There is no direct
relationship between electroforming time and part
volume or area; the only constraint is the size of the
electrolyte bath.
Composite materials - sandwiches of various metals
can readily be built as required, with varying
properties in a controlled fashion.
Mass production - Multiple mould cavities can be
electroformed at an electrolyte bath at the same
time to reduce mould-making cycle time. One
master can be reused many times so that identical
moulds can be made easily. The electrolyte solution
can be easily reused in the next electroforming
process so there is little environmental impact.
Limitations
Like any manufacturing process, electroforming, has
certain limitations, which may hinder its use as a viable
production process. Often these drawbacks can be
avoided or reduced to make electroforming feasible. Its
main engineering limitations are [3]:
Long deposition times - these can be reduced by
operating parallel production lines or by increasing
the relative velocity between the mandrel and the
electrolyte which increases the current density, and
consequently yields a higher deposition rate.
Material restrictions - due to brittleness, oxidation
and internal stresses, usually only copper, nickel
and iron are electroformed in practice.
Electroform/Mandrel separation - this is normally
achieved by mechanical, chemical or thermal
means. For the electroform to be removed from the
mandrel undamaged, careful design procedure is
required.
Non-uniform thickness - Multifarious methods are
applied in order to obtain an electroform with a
uniform thickness. An example is the use of
supplementary anodes inserted into recessed areas
in order to increase the local deposition rate (as the
current density is lower there than at higher peaks).
Internal stress - Most electroforms are deposited
with either compressive or tensile internal stress.
This is undesirable as it can result in cracking and
peeling of the deposit. It is often controlled by the
use of stress- reducing additives, electrochemical
feed - back systems or pulse power.
1.3 Applications of Electroforming
The range of electroforming applications is wide, as
noted below [4, 5, 61:
Thin foils - A major use of electroforming in terms of
tonnage of material is the production of thin metal foils.
Copper foil is predominantly used in the production of
printed circuit boards. Nickel foil is used to produce
circuit boards, resistive heating tapes and bursting discs.
A novel application of electroformed nickel foil is the
production of solar heating panels. Iron foil can be
produced from scrap steel [7,8,9,10].
Perforated products - These represent the most
important use of nickel in electroforming. A prime
example is the screen printing cylinder, for printing
fabrics, wall papers and also carpets. Other mesh
products include filters, sieves and perforated electric
razor foils.
Record stampers - This represents one of the earliest
successful large scale uses of electroforming and still
remains the only method by which this type of article can
be produced. The degree of precision required in this
application has always been considerable. Since the
advent of compact and video discs, employing digital
recording, precision required has been even further. A
recording of this type consists of a helical track of very
shallow depressions, approximately 0.1pm deep and
about 0.6pm wide. In order to achieve this degree of
precision, extreme care is required in the production
process, the electroforming being carried out under
clean room conditions.
Moulds and dies - A popular application area is the
production of shoe moulds where the technique enables
the faithful reproduction of natural finishes such as wood
and leather.
2
SIMULATION
OF THE
ELECTRO-FORMING
PROCESS
Despite the usefulness of the electroforming process as
described above, negligible theoretical simulation has
been performed except for the work of McGeough and
Rasmussen [q.These authors threw light on the
influence of current efficiency when periodic reversal of
polarity is used to obtain a uniform thickness for the
electroformed metal [q.
The formation of a layer of nonuniform thickness is a frequent cause of fracture with
many electroformed components.
In practice, this
difficulty is often overcome by the periodic reversal of
electrode polarity. Each polarity cycle consists of a
period of deposition of metal on the mandrel followed by
a shorter time of dissolution. A series of such cycles is
considered to lead to a uniform thickness of metal layer,
provided firstly that the current efficiency for deposition is
less than that for dissolution, over the operating range of
current density. Secondly, the over-potentials at the
electrodes are thought to make a significant contribution,
although their role has not yet been clearly defined [ I I ] .
McGeough and Rasmussen developed a theoretical
analysis of this hitherto empirical practice. The resulting
equations are shown to be so complicated that numerical
methods offer the best means of solution.
These latter methods are then used to demonstrate how
the achievement of a uniform thickness for the layer is
influenced by the time intervals of the polarity cycle and
the current efficiencies for deposition and dissolution.
The latter section incorporates a basis for the practical
device whereby suitable agents are added to the
electrolyte in order to obtain sufficient contrast in the
current efficiencies for the deposition and dissolution
reactions [12, 13, 14,151.
Indeed, the electrolyte is the key to effective
electroforming. For example, in the electroforming of
copper components, cyanide copper solution is a
suitable vehicle, when a uniform thickness for the
electroformed layer is necessary, unlike, copper
sulphate-sulphuric acid mixture, the other common
electrolyte [2].
In applications for which the latter electrolyte is needed,
a constant direct current (D.C.) is applied. Similar
process conditions are usual in the electroforming of
nickel articles. A consequential non-uniform thickness for
the electroformed layer has then to be accepted.
(Nonetheless, an approximately uniform thickness of
metal is occasionally obtained, although the means by
which this result is achieved are not well understood).
Conversely, in other applications of D.C. electroforming,
the achievement of a uniform thickness is not the
problem. Instead, the lower face of the metal layer must
reproduce the surface of the mandrel whilst its upper
face must be as flat as possible [13, 141. The same
authors investigate these aspects of electroforming by
use of the model formulated in the earlier paper, for a (1)
sinusoidal - shaped mandrel electrode, and (2) an
arbitrarily shaped one. They show that a uniform
thickness is attainable in D.C. electroforming provided a
specified relationship links the constant voltage gradient
with a faraday metal-deposition parameter and current
efficiency. The meaning of this parameter and
relationship is discussed fully by McGeough and
Rasmussen [12], but for convenience their significance is
also summarised here. The deposition parameter is
equal to the electrochemical equivalent of the deposited
metal and the specific conductivity of the electrolyte,
divided by the metal density. Current efficiency is usually
defined as the ratio of the observed deposition rate to
the rate predicted from Faraday's law. From
experimental evidence for an approximately linear
decrease in current efficiency with increasing current
density, current efficiency can be expressed as the
product of slope of the current efficiency-current density
curve and the voltage gradient across the inter-electrode
gap.
The analysis demonstrates how the current density is to
be calculated if a uniform thickness of metal is desired,
and shows that higher current density is needed; e.g.
typical calculations troduce a current density of lo9
A/mm2, where as 10- Nmm2is more common. For a time
of electroforming of 3 . 6 ~ 1 0
s,~the layer thickness is 0.63
mm. Another relationship between the above three
parameters can be derived to establish conditions for
electroforming a layer within an approximately flat upper
surface.
In practice, the use of plane anodes does render difficult
the deposition of metals in recessed areas of the
mandrel, as a consequence of the locally lower current
density in those regions. This difficulty is usually
overcome by use of a shaped anode, whose form is
complementary to that of the mandrel. A uniform current
density over the latter electrode, and hence a constant
metal thickness is then obtained. When an undistributed
(insoluble) anode such as graphite is used, the
deposition action depends on its metal supply from the
electrolyte solution. In consequence, the electrolyte
requires frequent replenishment of the basic metal salts
in order to keep its concentration and other properties,
such as pH, constant. As an alternative to this tedious
procedure, soluble anodes are used. As electroforming
proceeds, these materials dissolve electrolytically, and
so maintain a constant concentration of metal salts in the
electrolyte solution.
McGeough and Rasmussen have developed analytic
treatments, which describe the effects of conforming and
soluble anodes in electroforming [14]. Their solutions
enable conclusions to be drawn concerning the physical
conditions of the process, for example, for uniform
thickness and flat upper surface. They also describe a
numerical method for investigating these effects. The
method is then applied to the practical problem involving
the electroforming of a mould for a rubber '0' ring.
These researchers show that when a soluble anode has
a flat surface at the start of D.C. electroforming, its
shape gradually changes as it dissolves, owing to the
electric field distribution between the two electrodes. If a
flat upper surface for the electroformed layer is required,
a steep slope for the current efficiency-current density
curve is required. The flatness only occurs momentarily
during electroforming. If electroforming is continued
further, the surface of the metal layer becomes wavy.
When periodic reversal of polarity is used with soluble
anodes, a uniform thickness for the electroformed layer
cannot be achieved.
In their application of their model to the practice problem
of a mould for production of '0' rings, the authors
recognise that electroforming gives the dimensional
3
accuracy and surface finish needed for the mould.
However a major problem concerns the thickness of
metal that can be electro-deposited over the regions of
the mandrel, that become the main recessed locations of
the electroformed mould. If the electroform is too thin,
the mould may distort. The theoretical model
incorporates analysis of four choices of anode, by which
the thickness that can be achieved is influenced: (i)
plane and insoluble, (ii) plane and soluble, (iii)
conforming to the shape of the mandrel and insoluble
and (iv) conforming to the mandrel shape and soluble.
The modelling reveals that little significant change in
metal thickness is obtained for this range of anode
shapes.
The main other contribution in this phase has been
simulation of the effects of a deep V-shaped scratch on
the mandrel surface. The cathodic face of the
electroformed metal is found to adopt the shape of the
scratch, whilst its upper face becomes level. The
analysis reveals that the effect of overpotentials is to
render more uniform the current density distribution,
even in such a case where the geometry of the mandrel
in the vicinity of the scratch (or notch) implies a locally
wide variation in the current density, and in metal
thickness. The combined influence of overpotentials and
current efficiency is then a reduction in the rate of metal
deposition in the high current density regions, and
simultaneous promotion of increased metal deposition in
the valley of notch. These conclusions give rise to the
gradual formation of a "cleavage plane" in the structure
of the electroform. Related experimental studies confirm
the occurrence of the cleavage plane arises from the
growth of electrodeposited metal from opposite sides of
the notch on the mandrel surface. When the two faces of
metal growth meet, the cleavage plane is formed. This
plane is a line of weakness.
Mechanical tensile testing of these electroformed
structures confirms that the cleavage plane weakens
them so much that the material fractures, even under low
loads. Fracture due to the onset of cleavage plane
formation is commonplace in electroforming practice
[ I l l . These needs for related experimental studies of
electroforming gives rise to the next section.
or olymethyl methacrylate (PMMA) for cells up to 10
dm capacity. Working with larger volumes of solution
f
was found to be problematic owing to the high operating
temperature (>85 OC): glass tanks were prone to
cracking and PMMA was found to buckle. Fuller-scale
tests were undertaken in a rubber-lined mild steel tank
(of 260 dm3 capacity). The tank was thermally insulated
on the outside. Heat and evaporation losses from the
surface of the tank were reduced by floating two layers of
polypropylene spheres. The electrolyte could be heated
electrically with quartz-sheathed elements, controlled to
f 0.5 OC. Although this test apparatus was designed for
the electroforming of iron, similar items are used in the
electroforming of other components, whether of iron,
nickel and copper, the other main metals used in the
process.
4
TRADITIONAL
ELECTROFORMING
MACROMANUFACTURING
IN
Electroforming enables the fabrication of products that
have intricate shapes including those that would
otherwise require locking dies or parting lines.
Electroforms are produced in a variety of sizes and for a
wide range of needs.
4.1
Tooling
Electroforming is widely employed in the manufacture of
metal tooling because it is suitable for the applications
that require good surface finish, tight tolerance, and
intricate detail. By using the excellent inscribing
capability of finely detailed metal pattern electroforming,
reversals of fine patterns can be accomplished. Figure 2
[I61 shows the metal patterns for urethane parts with
undercuts and rubber-type mouldings.
3
EXPERIMENTAL EQUIPMENT AND PROCEDURES
FOR ELECTROFORMING
In this section the salient features of electroforming cells
are described.
3.1 Small-scale tests
The standard Hull cell is often considered for small-scale
experiments in the examination of test solutions. A useful
aspect is the range of current densities that can be
investigated, in order to determine the most appropriate
process conditions to yield a coherent and crack-free
metal deposit. For example, Lai and McGeough [9] used
such a system to establish that a suitable electrolyte for
electroforming of iron foil consists of a mixture of 400g
dm-3aqueous ferrous chloride, 80g dm-3calcium chloride
and 2 cm3 dm-3 of a commercial anti-pitting agent. The
optimum operating conditions are a current density
range of 10 - 30 A dm-2, a pH range of 0.5 - 2.0, and a
temperature range of 85 - 108 OC [9]. These conditions
provide a useful direction towards scale-up of equipment
for practical applications of electroforming, especially in
the case of extreme operation conditions such as high
temperature.
3.2 Larger-scale tests
Since such an electrolyte is of low pH, acid proof
mandrels are required for tanks, test cells and auxiliary
equipment. These authors used mandrels such as glass
4
Figure 2: Metal patterns for urethane parts with
undercuts and rubber-type mouldings [16].
Nickel electroforming is one of the most popular
electroforming methods. It is a process by which nickel
is deposited on a contour medium to a specific thickness
[ I q . The electroformed surface is then separated from
the contour medium, which can be reused. The surface
finish of the electroformed component can range from
125 to 4 pm rms.The effective use of nickel
electroforming has proven to increase the tool life of
rotational moulded tooling components, such as those
used in the manufacture of plastic and composite
canoes and kayaks. Some of the application areas for
nickel electroformed tooling are:
polyurethane & resin moulding;
slush moulding;
injection moulding;
rotational moulding;
compression moulding.
Some researchers still focus on improvements of nickel
electroforming. Some of the results are as follows:
heat transfer properties allowing faster turnaround
time;
greater range of operating temperature;
increase in abrasion resistance hence the life of the
mould;
accurate and repeatable replacement tooling setups
from reusable mandrels.
With the development of electroforming technology,
alloys such as Ni-Co can be deposited onto the master
to form an alloy shell so that the strength and hardness
of the mould can be largely improved. Quick
turnarounds, significant cost savings, and improved part
quality are achieved by electroforming catheter tipping
dies with NiCoForm's proprietary high-strength alloy,
NiColoyTM [ I 81. This material has electromagnetic,
thermal,
strength
and
corrosion
resistance
characteristics closely matching those of stainless steel
and can be electroformed in a stress-free state on
permanent or expendable mandrels thus reducing the
per piece die cost. The high degree of polish achieved
on even the smallest tip forming surfaces assures
excellent release properties. Electroforming dies with a
low but uniform wall thickness (down to 125 pm) allows
significantly shorter heating and cooling cycles leading to
increased throughput. Unconventional configurations
including balloon ends, double lumen dies, integrated
elements such as guide pins and calibrated throughholes are possible. Figure 3 shows catheter tipping dies
made from NiColoyTM[18].
Figure 3: Catheter tipping dies from NiColoyTM[I81
1.2 Components Fabrication
As well as tooling applications, electroforming can be
used to manufacture components directly, including
those made from precious and refractory metals. The
EL-FormTMprocess, developed by Engelhard-CLAL [ I 91,
is a low cost process that can be used to fabricate
directly the components. By use of computer driven,
rapid cycle deposition techniques, high efficiencies are
achieved with substantial reduction in process time. The
EL-FormTMprocess is conducted at higher temperatures,
which reduce internal deposition stresses, permitting the
formation of relatively thick, near net-shape, freestanding products. An added feature of this innovation is
control of deposited grain size and uniformity.
Electroformed tubing made with the EL-FormTMprocess
provides the scientist and engineer with fine diameter,
seamless tubing made from materials that do not lend
themselves to traditional working procedures. Common
materials produced to date are iridium and rhenium.
Rhenium tubing has been produced which exhibits
excellent ductility permitting cold forming of gas delivery
tubes.
In addition to pure metals and selected alloys, the ELFormTM process permits high quality composite
deposition. The uses of niobium or rhenium on iridium,
on iridium/platinum, or on platinum are well-established
combinations. Almost any combination can be deposited.
This capability permits the fabrication of high strength,
near net-shape products with controlled and compatible
environmentally protective coatings as an integral part of
the structure. Depending on the location within the
electrochemical series, various elements can be
deposited as true alloys and layer-by-layer composites.
For example, alloys of 10% to 85% platinum in iridium
have been routinely produced. Experience has shown
that alloys of iridium and rhodium, tungsten, and
molybdenum can be produced in addition to the deposits
of pure metals. Other refractory metals such as niobium
and tantalum have been formed by the process.
Electroformed rocket nozzles for satellite positioning and
control have been developed using the EL-FormTM
process as shown in figure 4. Exact dimensional
conformance is assured due to the precision machining
of the mandrel upon which the material is deposited. The
process permits integral materials to be deposited so
that the iridium inner wall for oxidation resistance is
developed with the rhenium structural member. Selective
deposition of niobium on the ends of the chamber can be
provided for ease of attachment to bell cones or injector
assemblies. The present facilities permit the economic
and rapid supply of chambers from 22 to 445 N.
Figure 4: Rocket nozzles using EL-FormTMprocess by
Engelhard-CLAL [ I 91.
Electroforming can be also used to repair a mould.
Adding metal where it is needed does not necessarily
entail welding with inevitable splatter, potential for
warpage, the disadvantage of cracking and annealing of
the heat affected zone. Electroforming can help bring
mould parts back into tolerance. All plating is done from
water-based solutions of metal salts and other
ingredients necessary for the correct functioning of the
baths. Metals commonly used for mould component
repair are: nickel (soft, hard, and electroless), chromium
(hard) and nickel-cobalt alloys. Figure 5 shows mould
blocks selectively plated on two sides with 1 mm of
NiColoy" [20].
Figure 5: Plated mould blocks [20].
5
1.3 Electroforming with solid freeform fabrication
The emergence of solid freeform fabrication (SFF)
technology has brought about new opportunities for
electroforming in rapid tooling [21]. Electroforming
masters can be built by rapid prototyping (RP) machines
and then by electroforming to copy the shape of the
master. The integration of RP and electroforming can
produce complex metal cavities and inserts required in
dies and moulds. Because the outside surfaces of
electroformed shells are very rough and uneven, only the
inside working surface, which contacts the master
surface, of the electroformed shell is used as the cavity
for moulding. Unlike other metal powder sintering
processes, electroforming can produce cores, cavities
and EDM electrodes with excellent working surface
finish, dimensional and geometric accuracy, as well as
shell material properties. Integrating SFF with
electroforming meets the challenge of making products
and components otherwise difficult or impossible to
make. In this way, electroforming can make thin walled
hollow bodies of intricate shape and accurate inner
dimensions, such as waveguide tubes, wind tunnels,
venture nozzles with varying sections, etc.
Figure 6 illustrates the electroforming tooling process
recently being investigated for fabrication of dies, moulds
and EDM electrodes [22, 231. The basic process is as
follows. A CAD model is generated and transformed into
STUSLI file format. This file is input into an RP machine
to build the electroforming master. The master is
metalized and then placed in an appropriate
electroplating solution and metal is deposited upon the
master by electrolysis. When the required electroform
thickness has been deposited, the master is then
removed from the metal shell by burnout. The metal shell
is backed with low melting alloy to form a mould cavity or
an EDM electrode. If a thin walled hollow body is
required, the master is removed from the metal shell and
the shell is post-processed to form the hollow body.
1.4 EDM Tooling [21, 22, 231
EDM has been widely used to form cavities in the
production of complicated metal dies and moulds
(forging, casting and injection moulding) which are
typically difficult to machine using traditional chip
removal techniques due to complex geometry and hard
material. An EDM machined cavity is normally used to
form both metal and large plastic parts. The forming
process produces very large compression forces during
part-making, so tough and hard materials are needed for
the forming tools. There are several advantages of EDM,
such as zero contact force between the tool and the
workpiece and virtually no mechanical stresses on the
workpiece. It is therefore possible for EDM to machine
parts that are extremely fine and brittle. However, the
more complicated the geometry of the die or mould, the
more difficult it is to fabricate the corresponding EDM
tool electrode by conventional machining. With the
increase in die and mould complexity, EDM electrode
manufacture is becoming even more time consuming
and expensive. If new rapid and economical EDM tooling
methods become available, EDM could be more widely
used in the die and mould industry. Research on rapid
fabrication of EDM electrodes using RP models is
worldwide. One approach is electroplate positive
stereolithography (SL) models with copper to form EDM
electrodes. This is the most direct and convenient way to
produce the EDM electrode if the plated metal can be
uniformly distributed on the whole surface of the SL
master. However, due to the character of electroplating,
the metal layer plated on the master cannot be uniformly
distributed on the entire surface if the surface consists of
slots, cavities and sharp corners.
6
Figure 6: Electroforming of EDM electrodes.
The process of making EDM electrodes using SFF parts
and electroforming is illustrated in figure 6. Compared
with making EDM electrodes by direct electroplating of
copper on positive SL parts, this method does not
require uniform thickness of the plated copper. The
negative (complementary) geometry of the EDM
electrode is prototyped by stereolithography for use as
the RP part. The part must be rigid enough to withstand
the electroforming stress induced during the copper layer
deposition. The electroforming stress is largely
determined by the process parameters and varies widely
from compressive to tensile. Using stress reducer
additives and optimal parameters, the stress can be
controlled to less than 6 MPa. Before electroforming the
SL part, metallisation of the SL surface is needed to
make the part electrically conductive. Several techniques
for metallisation of nonconductive materials are
available. Electroless plating, a process involving an
autocatalytic or chemical reduction of aqueous metal
ions onto a base substrate, can produce a uniform
pinhole-free metal film on the entire surface regardless
of the complexity of part geometry, so the SFF parts are
metallised using electroless plating. The metallised SFF
parts are then electroplated with copper to the required
thickness. In the process of separating the SFF part from
the metal shell, prevention of the electroform
deformation is critical. The RP parts built with ceramic
and other difficult-to-melt materials are preferably
separated by mechanical extraction. Melting, burning
out, or heat softening can be applied to wax and plastic
RP
parts.
Stereolithography
(SL)
resins
are
thermosetting materials, and removal of the part by
burning out is the preferred separation method.
Complete incineration of the SL part is observed at the
temperature of about 56OoC. During the burnout process,
heat results in expansion of the part. This may crack or
deform the electroformed metal shell. The geometry and
internal structure (e.g. hollow SL parts produced with
Quickcast process) of the RP part and the thickness of
the electroformed copper shell need to be optimised.
This is in order to minimise the manufacturing cost, while
the stresses exerted on the metal shell due to the
thermal expansion of the part do not crack the copper
shell or generate unacceptable deformation. Since the
EDM electrode does not contact the workpiece during
the electrical discharge machining process, the strength
of the electrode is not critical. A low melting alloy with
good electrical and thermal conductivity is suitable for
backfilling the shell to form an EDM electrode.
Two EDM electrodes of the same geometry, as
illustrated in Figure 7, were made by this tooling process
with a copper layer thickness of 2 mm and 4 mm, which
are identified as electrode 1 and electrode 2,
respectively. The cavity of the SL part is 12 mm deep.
Figure 7: Electroformed electrode and EDM generated
workpiece
The SL cavity is polished to give a surface finish of 1.24
pm. After polishing, each dimension marked in figure
7(a) is recorded. The generated part in figure 7 (b)
shows that the corners with zero radii can be
electroformed, although the copper layer thickness is
always the smallest in these positions.
Both SL parts are electroformed at room temperature to
avoid thermal expansion of the parts. Incineration is
used to remove the SL part from the electroformed
copper shell. After the oven is heated to 56OoC, the
electroformed SL parts are put into the oven and then
kept for one hour to completely burn out the SL resin.
After burning out the SL resin, the dimensions of the
finished EDM electrodes are measured. The values are
compared with the corresponding measured data on the
SL parts. The dimensional deviation of the electrode with
4 mm thick copper shell is smaller than that of the
electrode with 2 mm thick copper shell, indicating that
the thinner the copper layer, the lower the dimensional
accuracy of the electrode (due to larger deformation
resulting from the burning process). The copper shell is
backed with a tin-lead alloy whose melting point is
103OC. The average surface roughness of the two
finished EDM electrodes is about 1.26 pm & for both
electrodes. The finished electrodes are then used to
machine a hard steel workpiece using the machining
settings typically used for EDM roughing with copper
electrodes. The cavity to a depth of 10 mm shown in
figure 7(d) is generated by electrode 2 (figure 7(c)).
Another novel hybrid method of manufacturing tools for
die casting, plastics moulding has recently be described.
It consists of electroplating a master to replicate its
surface, simultaneous spray peening to build up a
substantial body, and mounting in an appropriate
manner [24].
1.5 Mould Tooling [21,22,25,26]
The rapid mould making process is very similar to the
process for making EDM electrodes. A CAD model is
created and sliced layer by layer to produce an STUSLI
file, which is transferred to a RP machine, e.g. a
stereolithography machine to build a SL part, which is
post-cured, sanded and finished. The part is then
metalized using nickel electroless plating to a thickness
of about 0.005mm. The metalized part is then put in an
electroforming bath as a cathode to deposit a layer of
metal, which is thick enough to resist any deformation
caused by the separation and backfilling. Separation is
done by burning out the SL part. The deformation of the
metal shell generated during the separation is largely
affected by the thickness, material properties, and
geometry of the electroformed part. Backfilling of the
electroform is more critical in manufacturing a mould
cavity compared with manufacturing an electrode due to
the high strength required in the subsequent injection
moulding process. The harder the backfilled metal, the
higher the strength of the mould cavity. However, a
harder metal usually has a higher melting temperature.
Casting with a high melting temperature metal tends to
generate larger thermal stresses, which may cause
larger deformation in the electroformed metal shell. To
reduce the injection moulding cycle time, conformal
cooling lines may be put around the nickel shell before
the backfilling process.
Using the described mould generation process, the SL
part shown in Fig. 8(a) is used to produce a nickel
electroform shown in Fig. 8(b). Nickel is used in the
electroforming because its mechanical properties are
about the same as those of stainless steel and nickel is
highly wear and corrosion resistant. The electroforms are
then backed with tin-lead alloy and copper to generate
the mould cavities shown in Fig. 8(c) and Fig. 8(d),
respectively. The geometry of this SL part is difficult to
create by machining due to the sharp corners and fine
features. The parts are polished to a surface finish of
1.22 pm R,.
An RP-based process to produce sturdy prototypes and
short-run production tooling for plastic injection moulding
is being developed by CEMCOM Research Associates
Inc. [25]. The procedure has the potential to fabricate
fully functional matched die sets more inexpensively and
in one-half to one-third the time needed to machine
metal moulds/dies. Turnaround times of six weeks or
less are possible, along with lower costs. The Nickel
Ceramic Composite (NCC) tooling system uses plastic
RP models as master patterns for the fast fabrication of
NCC tooling for intermediate-volume plastic-injectionmoulding runs, usually in tens of thousands of shots.
Instead of building the mould through a powder
metallurgy process, CEMCOM builds it through nickel
electroforming. The secondary tooling method is based
on plating nickel over plastic stereolithography patterns,
then reinforcing the thin, hard nickel face with a stiff
ceramic material. The resulting rigidity lets core and
cavity shells of electroformed nickel grow simultaneously
on the model's opposite faces. The shells, each only
about 2.5 mm thick when finished, are then backed using
a thermally conductive ceramic, which fills in the gap
between shell and mould frame. Thus the two nickel
shells become two halves of a finished injection mould
as shown in figure 9. The resolution is almost on the
molecular level and the moulds have precisely the same
surface finish as the SL master. The near net-shape
process is particularly suited to larger components
(greater than 250 mm x 250 mm). This capability would
fill a niche in the RP toolmaking field, which tends to be
size-constrained because of limited RP build envelopes.
The process has the ability to produce moulds larger
than 250 cm2 economically. While the time to machine a
mould grows in proportion to mould volume, the time to
electroform does not. As a result, the cost and lead time
advantages of electroforming should grow as the size of
the mould grows. Dimensional accuracies for the moulds
can be achieved to a level about the same as those of
the original stereolithography patterns (k0.125mm).
Figures 9 and 10 show the associated injection moulds.
To date, the process has been shown to produce a
minimum of 5000 injection-moulded plastic parts. They
include parts composed of unfilled resins and reinforced
resins formed via injection moulding, gas-assisted
injection moulding, and compression moulding. A
7
benchmarking mould has been demonstrated for Kodak,
and an internal part for a mailing machine measuring 15
x 6.5 x 125mm has been built for Pitney Bowes. The
toolmaking process begins with a CAD representation of
the desired mould, which includes a 12 mm thick
separation insert at the mould parting plane. This
modified mould design is then used to create a highquality stereolithography model using an RP machine
supplied by 3D Systems. The model is then coated with
a conductive silver-based material and placed in an
electroforming bath of nickel sulfamate where a thin
nickel layer is plated over it. The typical nickel plating
thickness over the tool face varies from 0.1-5mm. The
high-resolution nickel shell reproduces fine surface
details of the mandrel and provides mechanical integrity
for the most highly stressed areas of the mould. After
electroforming, the stereolithography model continues to
serve as a fixture, holding the nickel shell in place.
Alignment and accurate dimensions are maintained
during the chemically bonded ceramic (CBC) casting
process by stabilising the shell to the stereolithography
model. Prior to casting, cooling lines are custom-fit and
fixed in place. Because the ceramic has almost zero
shrinkage, casting can take place in the mould base
pocket. The resulting precision nickel-shell and
stereolithography-model assembly is then attached to a
standard pocketed steel mould frame using a highstrength CBC called COMTEK 66. This stiff backing
material is a water-hardened, metal-filled, cement-based
composition with low-shrinkage properties. The ceramic
is then vacuum-cast through a small opening in the back
of the frame. After the CBC cures for about a day, the
opposing side is cast. Later, the two halves are
separated, the model is removed, and the CBC is postcured. Once cured, the ejector pin holes are drilled and
the pins installed. The resulting NCC mould has a hightensile-strength, abrasion-resistant surface, and the
high-compressive-strength backing provides support and
mechanical coupling to the steel mould frame, which
provides containment and alignment. The similar
thermal-expansion characteristics of the nickel mould
face, the stiff ceramic backing, the steel frame, and the
net-shape forming characteristics of the nickel and the
ceramic all help maintain an effective bond and precise
location of tooling components.
Figure 9: Plastic injection moulding through
electroforming from a SL pattern by CEMCOM.
Figure 10: The plastic-injection-moulding core and cavity
set indicates the fine feature capability rapid-toolmaking
technology.
Figure 11: Diagram of TCLD-SFF [27l.
Figure 8: Nickel electroformed mould cavity with fine
features.
8
1.6 Chemical Liquid Deposition Based Solid
Freeform Fabrication [27]
The principles of electrolysis and electroforming are
applied in many innovative ways to achieve rapid and
cost-effective methods for manufacturing of tools and
products. One of the processes that is fast gaining
ground due to its practicality is chemical liquid deposition
based freeform fabrication. The Chemical Vapour
Deposition (CVD) process has been used in solid
freeform fabrication for several years, and it has faced
some critical problems in deposition rate, product
accuracy, and facility cost. To overcome these
shortcomings and explore new chemical deposition
methods and materials, a new SFF rapid tooling
technique named Chemical Liquid Deposition based
Solid Freeform Fabrication (CLD-SFF) has been
developed.
CLD-SFF
can
be
divided
into
thermochemical (TCLD-SFF) and electrochemical
(ECLD-SFF) liquid deposition-based SFF.
The
deposition rate of CLD-SFF is much higher than that of
CVD.
TCLD-SFF is based on the following: when cold (room
temperature) liquid reactants are sprayed from a nozzle
and come in contact with a hot substrate, the reactants
can decompose or react with one another, and then the
solid products are deposited on the substrate. By
controlling the motion of the nozzle and the spray time, a
desired three-dimensional shape of deposited material
can be formed using layer-by-layer scanning. Figure 11
is a conceptual diagram of TCLD-SFF. It consists of six
sub-systems: a substrate heating system, an X-Y-Z
scanning and elevating system, a reactants providing
and controlling system, a pressure generating system, a
gaseous product treating and recovery system, and a
central computer control unit.
The main technical processes of TCLD-SFF can be
described as follows: to design a part by a computer
using a three-dimensional CAD software package and
translate the designed part files into an STL format file;
to install the substrate and heat it to a certain
temperature which is then maintained; to close the
chamber and then run the pressure which is maintained;
to run the reactant providing and controlling system and
adjust the flow and pressure of the liquid reactants; to
run the scanning systems and let the nozzle spray the
reactants towards the hot substrate at prescribed
positions and time according to the STL file to form the
part by layering; and then to remove and examine the
final product. There are many kinds of liquid reactants
that can be used to deposit materials by TCLD-SFF. The
solid deposits can be ceramics or metals. A desired
shape formed by TCLD-SFF can be obtained as long as
processing parameters, especially the temperature of the
substrate, are controlled in a defined range.
In ECLD-SFF, a special anode is designed to deposit
metallic materials among powder particles on a cathode
plate. By controlling the motion of the anode and the
voltage of the electrical field, a desired three dimensional
part made of connected powder particles can be formed
through layer-by-layer scanning. The substrate is made
of, or coated with, conductive materials (metals or
graphite), and is connected to a DC power supply, as the
negative electrode (cathode). This is followed by putting
the substrate in a plating bath that is filled with
electroplating liquid. The very thin pin electrode that is
made of deposition metal is connected to the DC power
as the positive electrode (anode). Between the substrate
and the tip of the pin there is a thin layer of metal
powder. The electric field is applied in the Z direction.
Two assistant electric fields are arranged perpendicular
to each other to form an X-Y surface electric field. A
magnetic field applied in the Z direction of the substrate
forms a tight connection of powder particles for ferrite
materials. Under the effects of the magnetic and electric
fields, metal ions from the electrode moving to chemical
liquid will deposit onto the powder particle and grow. The
metal particles will be bound by the deposited materials
to form a freeform solid. By controlling the pin movement
and electrifying time, a desired three-dimensional shape
can be formed through layer-by-layer scanning. The
formed product is further treated, such as sintering and
infiltration. The motion and control units of the ECLDSFF system shown in figure 12 are very similar to those
of TCLD-SFF system. In ECLD-SFF most conventional
electroplating liquids can be used to deposit materials
among metal powder to form desired parts by connecting
the particles of the powder. The key parameters in
ECLD-SFF are the shape and size of the anode, electric
current density and the distance between the powder
surface and the pin-tip. If these parameters are well
controlled a desired shape can be consistently obtained.
Figure 12: ECLD-SFF system [27l.
MICRO-MANUFACTURE
Over the last three decades, the microelectronics
industry has undergone unprecedented growth. A major
new
development
is
the
fabrication
of
microelectromechanical
systems
(MEMS).
The
application
of
electroforming
to
manufacturing
microstructures and MEMS will be described below.
5
5.1 LlGA
The LlGA process was developed by the German FZK
(Research Centre Karlsruhe) in the early 1980's under
the leadership of W. Ehrfeld [28]. LlGA is a German
acronym standing for the main steps of the process, i.e.,
lithography, electroforming, and plastic moulding. These
three steps make it possible to mass-produce micro
components at a low cost. Figure 13 illustrates the LlGA
process [29]. It should be noted that electroforming can
be useful in micromanufacturing only if it is combined
with the Iithography process.
Lithographic techniques use the shadow projection of an
absorber relief into a radiation-sensitive resist material
(Figure 13(a)). Depending on the resist material,
following the exposure process and the development of
the irradiated area, either there is a direct threedimensional reproduction (positive resist, see e.g. Figure
13(b)) or there is a tonal inversion through the use of
negative resist. If, however, the demand is for mass
manufacturing processes for materials such as metals,
polymers, ceramics or indeed glass, the resist structure
is filled in a further process step by electroforming
processes (Figure 13(c)). In this way the negative pattern
of the plastics structure is generated as a secondary
structure out of metals such as nickel, copper and gold,
or alloys, such as nickel-cobalt and nickel-iron. This
technique is used to produce microstructures for direct
use, but also tools made of nickel and nickel alloys for
plastics moulding (Figure 13(d) to Figure 13(f)). Plastics
moulding is the key to low-cost mass production by the
LlGA process. The metal microstructures produced by
lithography and electroforming are used as moulding
tools for the production of faithful replicas of the primary
structure in large quantities at low cost.
Products fabricated by the LlGA process may be made
of the following materials: polymers (e.g. PMMA,
polycarbonate), metals (e.g. Ni, Cu, Au), and alloys (e.g.
permalloy). They may have the following features [30]:
9
4. Mould insert
1. Irradiation
Figure 13: LlGA process.
any lateral geometry of structures;
structural height above 1 mm;
smallest lateral dimensions down to 0.2pm;
aspect ratios of free-standing individual structures
and details above 50 and 500, respectively; and
surface quality in the submicron range with
roughness, %, of 30 nm.
There are mainly three lithography techniques that can
be used as irradiation methods: X-ray, UV light, and
laser. X-ray lithography uses collimated X-rays as the
exposing energy. Being much shorter in wavelength than
ordinary light, X-ray provides increased lateral resolution.
For micromanufacturing the penetrating power of the Xrays deep into the photoresist allows microstructures
with great height to be fabricated. X-ray lithography is
expensive to perform because of the expense of
operating a synchrotron. The cost can be hundreds of
dollars per hour to recover the actual operating
expenses in addition to the initial investment of tens of
millions of dollars. Therefore, the LlGA process was
developed several decades ago to reduce the
dependency on fabrication with a synchrotron. Figure 14
shows an example of stepped structure fabricated by Xray LlGA [30].
10
UV-LIGA electroforming is a technology that allows the
fabrication of electroformed micro-components using UV
irradiation onto photo resist, and the subsequent
electroplating to form metal-plated components. This
technology is cheap and can be taken up by industries
and universities with a minimum of capital investment.
Like X-ray LIGA, it permits the fabrication of a large
range of devices of different structural height. However,
the resolution and the height-to-width aspect ratio are
lower. Study of photosensitive materials is being
performed to solve this problem. The introduction of SU8 resin patented by IBM in 1989 and commercialised by
Shell Chemicals and others under the name EPON-SU8, has created a breakthrough in the fabrication of lowcost, large-structural-height
microsystems. Largestructural-height micro devices as shown in Figure 15
can be fabricated [31]. The three components needed for
this process are: resin, an organic solvent such as GBL,
PGMEA, or MIBK, and a photo initiator like triaryl
sulphonium. The percentage in weight of resin
determines the viscosity and therefore the maximum
thickness that can be achieved under spinning.
The high cost of the X-ray LlGA process has led to
widespread research into alternative methods for
realising three-dimensional microstructures. A low-cost
'Laser-LIGA process based on electroforming of polymer
moulds formed by excimer laser ablation as shown in
Figure 16 has been developed [31]. This process can
produce nickel structures with depths of up to several
hundred microns and with surface roughness down to
100 nm R,. Current developments include the extension
of the technique to allow multi-level processing, the
formation of structures with complex surface profiles by
means of variable lithography. Figure 17 shows six-level
nickel microturbine fabricated using a combination of
UV-LIGA and Laser-LIGA.
Figure 15: Large-structural height micro-device.
The additive build up of microstructures out of metal is
best achieved by electroforming. With the introduction of
the X-ray LlGA process, bulk metals have entered the
domain of Microsystems technology. This pioneering
microfabrication technology enables the fabrication of
2.5-dimensional, micromechanical parts of metal. LlGA
has turned out to become a key technology for
manufacturing "heavy-duty" micromechanical parts. Its
drawback is the process complexity and high costs to
fabricate micromechanical systems or tool inserts. A lowcost HARMS (High Aspect Ratio MicroSystems) process
has been developed, which allows to overlap a majority
of the field of LlGA applications while enabling low costs
and short throughput times [32]. Structural feature
heights of 500pm (>1000pm with LIGA), minimal lateral
dimension of 8pm, sharp sidewalls and large aspect
ratios up to 10 (100 with LIGA) can be achieved. The
elements are produced by means of electroforming
materials such as Ni, Cu, Au, and Ag. Nickel is often
used to produce mechanical parts and elements due to
its properties. Miniature gearwheels, sets of gears,
counting wheels with virtually any tooth shape can be
produced. Typical applications include watches and
revolution counters.
Aluminium has been used in the fabrication of
microdevices [33]. One important application of
aluminium microstructures is in the fabrication of
integrated circuits. Aluminium is a predominant material
used to define electrical conductors in the integrated
circuit technology.
Figure 18: Gear wheel produced from Ni.
Microstructures such as high-aspect-ratio current
carrying traces and heat sinks fabricated from aluminium
do not have the problems associated with intermetallic
alloys at junctions between dissimilar metals. Standard
photoresist electroplating mould processes cannot be
utilised in aluminium microstructure fabrication because
they cause swelling and/or decomposition of the
polymers used in many photoresist systems. In contrast,
polyimide materials are shown to have the properties
necessary to withstand electroplating conditions for
aluminium. The basic process for fabrication of
electroplated microstructures using photosensitive
polyimide is analogous to the LlGA process, except that
the photosensitive polyimide is used as the electroplating
mask instead of PMMA, and an ultraviolet exposure
source is used instead of an X-ray synchrotron. Figure
19 shows an aluminium gear fabricated using the above
process [33]. The aluminium gear has a thickness of
45pm, an outer diameter of 300pm, an inner diameter of
50pm, and a tooth width of 40pm. The surface of the
microstructure is representative of the grain sizes
obtained using the basic aluminium electroplating
solution.
11
Figure 19: Aluminium gear fabricated using LIGA.
5.2 Micromou Iding
Micromoulding can be used as an alternative fabrication
process of devices for medical applications. One
advantage of micromoulding is that it facilitates the
fabrication of devices from a large selection of plastic
materials, which are adapted and approved, for medical
applications. Micromoulding also provides a wide range
of possibilities in terms of design.
Figure 20 shows the steps of the micromoulding
process. Usually, a master structure is created which is
copied into a metal tool by electroforming. Lasermachined microstructures can be replicated in metal or
plastic. Polymer replicas are usually formed from a
negative metal master. Metal replicas can be formed by
electroforming using polymer moulds. This solves the
problems associated with interfacing the microstructure
to the macro world. In the production process, the
component is moulded as a single part, including micro
and macrostructures. Perhaps the biggest advantage of
micromoulding is the potential to produce relatively large
devices at low cost (per unit). Cost is usually the major
driver in miniaturisation technology.
Devices may be manufactured with fluid volumes of up to
10 microliters. Standard fluidic fittings, connectors, large
sample input ports, large waste reservoirs, and handling
structures may be integrated along with the
microstructures into the plastic mount. In this manner,
complicated devices may be constructed at a cost level
acceptable for disposable application. Different types of
sensors (e.g. electrochemical) can be incorporated onto
the chip to create the self-contained ensemble.
Figure 20: Micro moulding process.
12
5.3 Micro-optics applications
Micro-optics is usually associated with one- or twodimensional waveguide structures in linear or nonlinear
media. In this technology the propagation of light is
restricted to one or two dimensions; therefore unlike
most macro-optical setups, no use is made of the natural
three-dimensional characteristics of light. Waveguide
integration is subject to topological restrictions similar to
those in integrated microelectronics. Three-dimensional
free-space
imaging
provides
extremely
large
interconnection densities, even in comparison with
microelectronics. For a 3-D integration of optical
components, different fabrication technologies are
applied to different types of components. A typical
macroscopic optical system consists of lenses, mirrors,
beam splitters, amplitude or phase masks, and suitable
mounting mechanics. Lenses and amplitude or phase
masks can be miniaturised by use of lithographic
fabrication technologies. These planar fabrication
methods, however, are restricted to components that are
arranged in a two-dimensional plane.
A new approach for fabricating micro-optical elements is
the LlGA process. With this technology, microstructures
with arbitrary cross-sectional shapes and smooth
sidewalls can be fabricated either in metals or in
polymers, characterised by little attenuation in the visible
and near-infrared regions. Recently, examples of microoptical structures such as deflection prisms, cylindrical
lenses, and beam splitters have been presented. Also,
mounting supports can be fabricated in the same
process sequence, thus offering the possibility of
assembling stable and robust complex micro-optical
systems that combine passive and active optical
components.
In Figure 21, an arrangement of microprisms fabricated
by the LlGA process is shown [MI. The structural height
of these prisms is 500 pm; the length of each
hypotenuse of the small prisms is 700pm. All prisms are
fabricated in one process step and are adjusted
precisely with respect to each other so that no further
adjustment is needed.
Figure 21: An arrangement of micro prisms fabricated by
the LlGA process [MI.
Spin-on glass (SOG) is widely used in integrated circuits
as a high-quality planarising dielectric. In integrated
circuits, SOG is typically spin coated to form very thin
layers. There are several techniques that can be used to
make nonplanar SOG microstructures [36]. The LlGA
process has been used to make nickel moulds up to
150pm in height with mechanically planarised surface.
Spin-on glass is applied to obtain glass structures in
nickel moulds.
A multiple dispensing/drying/curing process has been
developed resulting in crack-free SOG structures.
Reverse electroplating is used to remove the nickel
mould and release the glass structure. Nonplanar
dielectric structures with aspect ratios of 20:l have been
fabricated. Figure 22(a) shows the channel plate Ni
mould up to 150pm high, while Figure 22(b) shows the
SEM of a SOG channel plate structure with 96pm height.
Figure 22: Channel plate structure [36].
Electroforming can also be used for the manufacture of
compact discs (CD) [371. Figure 23 shows the steps in
the CD fabrication process.
First, a source material such as an audio recording or
computer software is pre-mastered into a digital format.
This pre-mastered source material is played into a
mastering system, causing a blue laser beam recorder to
cut away peaks and valleys into a prepared surface
coated onto a thick glass substrate. These peaks and
valleys are known as lands and pits. Once the glass
master is completed, it is inspected and sent to
electroforming. In this process, nickel is plated onto the
glass master creating a negative impression known as a
"father". For shorter production runs, the "father" can be
used to mould discs. This provides a faster turnaround
time. Using a galvanic process, four to five positive
impressions, known as "mothers," are made. If a
"mother" passes inspection, it is used to create a number
of negative impression nickel moulding matrices called
stampers. These stampers are fitted into an injection
mould cavity. Polycarbonate resin is then injected into
the centre of the mould cavity, spreading outward to
create a clear plastic disc with a positive impression of
the lands and pits on one side. These clear discs are
metallized to create a surface that a playback laser can
read in a CD player or a CD-ROM drive. The discs are
then coated with a protective acrylic, labelled, and
packaged.
Spin coating: dye polymer is coated onto cleaned
and polished glass.
Recording: DRAW lasers expose and playback
master recording .
Baking: the glass master is baked to eliminate
residual moisture.
Metallisation: a thin layer of Ni is applied to the glass
master.
The "father" is made by separating the metal from
the glass master.
The metal "mother" is made from the father.
(vii) The metal stamper is made from the mother.
Figure 23: Steps in manufacturing a CD
Fabrication of MEMS
Microelectromechanical systems (MEMS) bring together
mechanical, electrical and optical technologies to create
an integrated device that employs miniaturisation to
achieve high complexity in a small volume. This
generally involves fabricating millimetre to micron-size
structures with micron to nanometer tolerances.
Microsensors detecting parameters like pressure, flow,
force, acceleration, temperature, humidity, chemical
content, etc., have been engineered into the engine and
performance management systems of cars and aircrafi.
They also provide the key to electromechanical
microcomponents such as inkjet printer nozzles, gas
chromatographs,
gyroscopes,
galvanometers,
5.4
13
microactuators, micromotors, and micro optics. Devices
are in development that contain implantable drug
delivery systems comprising sensors, valves, and control
system, as well as a power source capable of operating
for many years. Adaptation of silicon lithography and
etch
batch-processing
as
developed
by the
semiconductor industry is currently the dominant MEMS
fabrication method. However, it is restricted to just one
material (silicon) surface and bulk-etched in three
directions (along crystallographic planes). More flexible
micromanufacturing methods including electroforming
have been employed for MEMS fabrication. In recent
MEMS applications, electroplating provides a way to
build up layers thicker than PVD or CVD can do.
Especially in the LlGA process, electroforming is used to
make the main structures of micro parts.
Figure24: Microturbine [35]
There are numerous practical applications for
micromechanical products in such varied sectors as
medicine, aeronautics, and computer design. In the
medical sector, for instance, micropumps can be made
which, when implanted under the skin, are capable of
continuously administering a precisely controlled amount
of medicine. MEMS also permit the development of
better and smaller endoscopes by which not only
diagnosis but also fairly complex surgery can be carried
out on polyps, damaged ligaments and ulcers, for
example. Figure 24 is an example showing a
microturbine [35].
There is a demand for metallic microstructures with
relatively large thickness (i.e. 10-1000pm) in the
fabrication
of
MEMS.
For
example,
metallic
microstructures with larger thickness are greatly needed
in the actuation system where higher structural rigidity
and/or higher actuation force is required. Metallic micronozzles and micro-channels, as essential fluidic parts in
inkjet heads, are tremendously needed with the growth
of printer market. Metallic microstructures with small
features and high aspect ratios are demanded as precise
hot-embossing masters for plastic micromachining.
Figure 25 shows 150pm thick SU-8 photoresist moulds,
and Figure 26 illustrates a micromotor [35].
1.5 Photo-electroforming
Photo-electroforming is a new manufacturing process for
making microelectromechanical systems (MEMS).
Photo-electroforming builds parts by an additive process
which defines geometry by depositing powder in layers
and creating regions of selective conductivity by laserenhanced electroless plating [38]. The conductive region
is then joined (fused together) by a second plating to
form an integral part. The unmetallised portion is
removed by selective etching in one step after all layers
are defined and joined. Figure 27 shows the basic
concept of photo-electroforming.
14
Figure 26: Micromotor
Single-layer and two-layer standalone parts made of
nickel/silicon carbide composites of overall size 75-100
pm and feature size 25 pm have already been created.
Writing speeds up to 24 cm/s and in-plane resolution of
15pm can be achieved. The high laser-induced plating
rates are due to elevated substrate temperatures under
the laser spot and enhanced mass transfer due to
pumping by the hydrogen bubbles resulting from the
plating reaction. A simple two-layer part made by photoelectroforming is illustrated in Figure 28 [38].
Figure 27: Basic concept of photo-electroforming [38].
[7l
[8]
[9]
McGeough, J. A. and Rasmussen, H., 1981,
"Theoretical Analysis of the Electroforming
Process" J. Mechanical Engineering Science, Vol.
23, NO. 3: 113-120.
Lai, S.H.F., McGeough, J.A. and Lau, P., 1977/78,
Electroforming of Iron Foil, J. Mechanical Working
Technology, Vol. 1: pp 231-243.
Lai, S.H. and McGeough, J.A., 1979 Electroforming
and Mechanical Properties of Iron-Nickel Alloy Foil,
J. Mechanical Engineering Science, Vol. 21, No. 2,
1979, pp 41 1-417.
[ l o ] Lai, S.H. and McGeough, J.A., 1980, Some Effects
of Heat - Treatment upon the Mechanical
Properties of Electroformed Iron-Nickel Alloy Foil,
J. Mechanical Engineering Science, Vol. 22, No. 2:
103-105.
Figure 28: Two-layer part made by photo-electroforming
[381.
6 SUMMARY
Electroforming, although is a long established technique,
has shown itself capable of being adapted to meet the
stringent demands of modern technology, as marked by
the major new uses of the process, especially in the
micromanufacturing field. As the demands in many
industries for increased accuracy and precision become
more acute the process is undergoing continuous
adaptation to meet these needs. For example, the
technology for producing conventional record stampers
has been refined to give the much higher orders of
accuracy required for digital recording. New applications
for electroforming will require an even higher order of
accuracy, for instance in the production of masters for
hologram manufacture, where the dimensional limits are
measured in Angstrom units rather than microns.
Simulation, or modelling, of the electroforming process
remains an under-researched area, mainly due to its
complexities and cross-disciplinary characteristics. Yet
modelling will become increasingly significant, if the
process is to meet the needs of micromechanical
systems, engendered by microelectronics, medical and
micro-optics industries.
7
REFERENCES
Sole, M.J., 1994, Electroforming: Methods,
Materials and Merchandise, Minerals, Metals and
Materials, Vo1.46:29-35.
Hart, A.C., 1986, Electroforming as a Production
Process, Metallurgia, Vol. 53: 534, 579-580.
Watson,
S.A.,I 989,
Modern
Electroforming,
Transactions of the Institute of Metal Finishing, Vol.
67: 89-94.
DiBari, G. A., 1996, Nickel Electroplating
Applications and Trends, Plating and Surface
Finishing, Vol. 83: 10-14.
Tang et. al., 1996, Important Parameters and
Applications for Nickel Electroforming, AESF
Annual Conference (Cleveland, Ohio, USA) 1Oh13'h June: 627-636,.
Watson,
F.A.
and
Worn,
D.K.,
1981,
Developments in Electroforming of Nickel for
Engineering
Applications,
Metallurgist
and
materials Technology, Vol. 13: 310-314.
[ I l l Crichton, I.A., McGeough, J.A. and Thomson, J.R.,
1980, Effects of Cathode-Mandrel Surface
Irregularities on the Mechanical Properties and
Structure of Electroformed Iron Foil, J. Mechanical
Engineering Science, Vol. 22, No. 2: 1980, pp 4953.
[I21 McGeough, J.A. and Rasmussen, H., 1976, A
Perturbation Analysis of the Electroforming
Process, J. Mechanical Engineering Science, Vol.
18, NO. 6: 271-278.
131 McGeough, J.A. and Rasmussen, H., 1977,
Analysis of Electroforming with Direct Current, J.
Mechanical Engineering Science, Vol. 19, No. 4:
163-166.
141 McGeough, J.A. and Rasmussen, H., 1981,
Theoretical Analysis of the Electroforming Process,
J. Mechanical Engineering Science, Vol. 23, No. 3 :
113-120.
151 Rasmussen, H. and McGeough, J. A., 1981, On the
non-linear Stability of the Electroforming Process,
Journal of Applied Mathematics, 1981, Vol. 27
:211-220.
161 Yang, B., 1998, Rapid Tooling by integration of
Solid Freeform Fabrication and Electrodeposition,
Ph.D. dissertation, New Jersey Institute of
Technology, Newark, NJ, USA.
17l Spiro, P., 1968, Electroforming: a Comprehensive
Survey of Theory, Practice and Commercial
Applications, Teddington, Draper.
[ I 81 Symposium on Electroforming: Applications, Uses,
and Properties of Electroformed Metals, 1962,
American Society for Testing and Materials, Dallas,
TX, Feb. 6-7.
[ I 91 htttx//www.enuelhardclal.com/elf. html.
[20] htttx//www.hikifune.com/enulish/woduct/denchu/nie
f.html.
[21] http://www.sate-Iite.com/Prod Electroform.htm1.
[22] http://www.nicoform.com/frame2.html.
[23] http://www.nicoform.com/ElForminqPaper.html.
[24] Chen, X. K., Hepburn, W. D. and McGeough, J. A.,
1993, Tools made by Electroforming and
Simultaneous Spray Peening, Am. SOC. Mech.
Eng., PED, Vol. 64, Manufacturing Science and
Engineering (Ed. K. E. Ehmann):645-648.
[25] Yang, B., and Leu, M.C, 1999, Integration of Rapid
Prototyping and Electroforming for Tooling
Application, Annals of the CIRP, Vol. 48, No. 1:
119-122.
[26] Ashley, S.,1997, From CAD Art to Rapid Metal
Tools, Mechanical Engineering, Vol. 119, No. 3: 8287.
15
[27l He, Z., Zhou, J., and Tseng, A., 2000, Feasibility
Study of Chemical Liquid Deposition Based Solid
Freeform Fabrication, Materials and Design, Vol.
21: 83-92.
[28] Ehrfeld, W., 1990, Micro System Technologies, (ed.
H.Reichl), Springer-Verlag, Heidelberg.
[29] http://www.imm.uniainz.de/enulish/sk a tec/basic t
e/liqa.html.
[30] Bacher, W., Menz, W., and Mohr, J., 1995, The
LlGA Technique and its Potential for MicrosystemsA Survey, IEEE Transactions on Industrial
Electronics, Vol. 42, No. 5: 431-441.
[31] http://www.cee.hw.ac.uk/microenuineerinu/microsy
stems/uv-liqal
[32] http://www.sae.orq/automaq/tech briefs 0900/13.ht
m
[33] http://www.ntb.ch/infos/lMS/Proiekte/LC/HARMS.ht
ml
[34] Frazier, A.B., and Allen, M.G., 1997, Uses of
Electroplated Aluminium for the Development of
Microstructures
and
Micromach ining
Processes,”Journal of MEMS, Vol. 6, No. 2: 91-98.
[35] Brenner, K.H., Kufner, M., and Kufner, S.,1993,
Application of Three-dimensional Micro-Optical
Components
Formed
by
Lithography,
Electroforming, and Plastic Molding, Applied
Optics, Vol. 32, No. 32: 6464-6469.
[36] Liu, R.H., Vasile, M.J., and Beebe, D.J., 1999, The
Fabrication
of
Nonplanar
Spin-On
Glass
Microstructures, Journal of MEMS, Vol. 8, No. 2:
146-151.
[37l http://www.optical-disc.com/A11%20other/CD.html
[38] Tsao, C.C., and Sachs, E., 1999, PhotoElectroforming: 3-D Geometry and Materials
Flexibility in a MEMS Fabrication Process, Journal
Of MEMS, VOI.8, No.2: 161-171.
16