In the past, chemical
processing plants have used pumps that incorporate silicon carbon
sliding bearings which have a good resistance to chemicals but poor
abrasion resistance. As a result they have a very low tolerance
when running under dry conditions, for example during start-up operations,
and often fail earlier than is desirable as a result of chipping
and cracking. Silicon nitride ceramics were known to have better
mechanical properties but this benefit was offset by the fact they
were only available in formulations with poor chemical resistance.
In 1993, a group of interested companies came together in a three-year
Brite-Euram project in a bid to develop high-performance, low-cost
silicon nitride bearings suitable for use with pumps in chemical
processing plant where the chemicals involved are corrosive or abrasive.
Raw materials and machinery
Co-ordinated by the German bearing manufacturer,
Cerobear, the project essentially had two strands of work. One part
involved research into the raw materials required for the ceramic
bearings, while the other concentrated on the machinery needed for
production. Consequently the project attracted a complete cross-section
of interested companies. Three raw material suppliers became involved
in the work: Tioxide Specialities in the UK (which was part of ICI
but has now been split between two smaller companies), the French
company Ceramiques et Composites, and the German supplier SHM Werkstofftechnologie.
On the machinery side was Cerobear, and on the material inspection
side was SKF, the Netherlands arm of the biggest bearing manufacturer
in the world.
The skills and expertise of these partners were complemented by
the Institute für Keramische Technologien under Sinterwerksroffe
(IKTS), part of the Fraunhofer Institute in Dresden, and Feodor
Burgmann Dichtungswerke, also from Germany. IKTS carried out research
into powder processing, while Feodor tested trial bearings in its
acid production works.
Improved knowledge and new technology
The project let to three specific advances.
First, knowledge of the relationship between performance of a ceramic
and its formulation, forming and sintering technology was greatly
improved. As a result a grading system for silicon nitride ceramics
with specific physical, mechanical and chemical properties for pump
applications was developed. This means that corrosion-resistant
grades of ceramics can be tailor-made for exposure to a range of
acids, bases and hydrothermal environments. For example, a ceramic
produced from a silicon nitride powder with a low amount of aluminium
and yttrium and using hot isostatic pressing as the production method
exhibited good corrosion resistance to hydrochloric acid and other
Secondly, a canned pump was designed to incorporate the newly developed
corrosion resistant bearings. The pump had magnetic couplings and
was hermetically sealed to prevent loss of the process liquids;
it was used to demonstrate the very low wear of the ceramic rolling
bearing under unlubricated operation. It has been estimated that
chemical plant operators can realise maintenance cost reductions
of up to 5% by switching to pumps with the new bearings.
A third major achievement was the development of machining and finishing
technology that enables series-scale production. This is particularly
important because it opens up high-volume markets to silicon nitride
rollers. For example, machine tool operators can replace the steel
rollers on their high-speed cutting machines with silicon nitride
rollers, allowing them to achieve higher outputs. Also, because
the technology allows the bearings and traces to be produced to
a very high tolerance and with excellent surface finishes, they
can be used in applications where high reliability is essential,
such as in the aerospace industry.
Spin-off technology and competitiveness
The lessons learned during the development
of the chemical plant pump have since been applied to a fan for
a new design of gas laser machine where similar mechanical properties,
chemical resistance and tolerance to dry running were required.
Cerobear also recognised the potential of the technology in high-tech
sports, and has now introduced it to the engineers of Formula 1
racing car teams.
In fact, thanks to this project, Cerobear is the only source of
series-produced silicon nitride rollers and raceways that are tolerant
to dry running, giving them a major advantage in the global industry.
For Cerobear this has been translated into 30-35% growth per year
and has also helped the company to obtain venture capital backing
for further development work.