Ceramics at TWI

Joining

Introduction

Design for a particular component is primarily based on requirements such as:

- functionality of component

- environment in which component will operate

- component properties, including chemical, electrical and thermal

- required resistance to loading

- ease of assembly

Designing for glasses and ceramics

The design and testing of ceramics are more critical than for any other class of materials. The reasons are that ceramics are inherently brittle and ultimately contain flaws (although production methods control their size and frequency).

The following principles should be kept in mind when designing with glasses and ceramics:

• Operating conditions and requirements on the component must be specified as closely as possible.
  It is necessary to know the external loading and temperature regime: forces (dead weight, centrifugal force, etc), thermal shock (thermal stresses in particular, can cause severe problems) and whether conditions are cyclic or static, etc.
  
  
• The design must take into account the specific properties of the glass or ceramic:
  
  
  1. High brittleness (lack of ductility) compared with other structural materials
  2. High strength in compression but lower in tension, bending and torsion
  3. Susceptibility to impact load and point load: high contact stresses, particularly at points of support or load transfer, are very dangerous in glass or ceramic because the associated stress cannot be accommodated by plastic deformation. Thin, ductile metal interlayers are sometimes inserted to distribute this stress more uniformly
  4. Most glasses and ceramics have low thermal conductivity and increased susceptibility to thermal shock: components subjected to temperature fluctuations should have simple and symmetrical form as well as thin walls. For example, a cylindrical rod will resist sudden changes of temperature better than a rod with square cross-section
  5. Sensitivity to stress concentrators: abrupt changes in shape or thickness, notches and corners, etc - which increase the stress - should be avoided or rounded to a suitable radius
  6. Ceramics and glasses contain flaws, especially at the surface: the probability of larger flaws increases with increasing body dimensions
  7. Cracks and other faults are mostly produced during manufacture: sometimes they form during use, e.g. through contact stresses or thermal shocks

It is clear from the above that design with brittle materials needs a different approach to that with conventional engineering materials such as structural steels or plastics.

Joint design for ceramic-to-metal joints

Generic rules which should be followed when producing ceramic to metal bonds include:

• the ceramic should be thick-walled in comparison with the metal
• the ceramic should be in compression
• use soft ductile metals next to ceramics

Some of the preferred joint designs are shown in Fig.2.

Fig.2. Best practice joint designs

Where joint dimensions are large, or there is a large CTE mismatch between the materials, joint design and joining process are critical.

Use of interlayers

Interlayers can be soft, ductile metals which flex during heating and cooling to absorb stress between the two materials. Their CTE value can lie between those of the two materials being joined, or they can be functionally graded materials which have a continuously graded composition and hence a continuously graded CTE, to reduce the inherent mismatch.

Fig.3. Interlayer designs

Some generic interlayer designs are given in Fig.3. For some joint configurations, the use of stress-relieving interlayers is not possible and specialised joint design is the only option for relieving stress at the interface.

Typically, the use of a compressive stress on the ceramic is desired, since this increases the potential joint strength. Such a bond takes advantage of the mechanical properties, in particular ductility, of the metal, such that the metal will yield in preference to the ceramic, thus stopping the ceramic from exceeding its fracture stress. Therefore, stress at the ceramic-metal interface should not exceed the bond strength, provided that the metal is in tension whilst the ceramic is in compression.

A final consideration is that where service performance dictates the need for inspection, parts should be capable of being inspected by simple techniques.

There are limitations to non-destructive testing of ceramics (discussed in Section 6 of this Best Practice Guide) and even a component passing inspection may not necessarily be free of flaws, although the flaws may not be significant in terms of engineering performance.

One of the most important practical features of joining ceramics is joint design. This is particularly so when joining ceramics to materials having a different coefficient of thermal expansion. In this case joint design is critical, otherwise the joint may fail either during production, or in service.

There are several ways to produce successful mixed CTE joints) including:

• compressive fitting
• flexible interlayers (to absorb residual stress)
• Finite element modelling

These techniques have all been applied at TWI to produce a range of CTE mismatched joints.

Ceramic Faced Tappets	Successful bond

See below for more information about joint design for ceramics:

• Joining ceramics - a guide to best practice. Section 2. Joining technology - the basics, plus Design issues Industrial Members and JoinIT North East users only

For more information please contact: ceramics@twi.co.uk