Variables that affect UHMWPE Quality

Ultra High Molecular Weight Polyethylene usually referred to as UHMWPE or 'poly' has been the material of choice for the bearing surfaces of most joint replacements used in the last 25 years.Its generic properties of toughness and low friction combined with exceptional abrasion resistance are well known. What is less appreciated is the variation in these properties directly attributable to the manufacturing conditions and polymer grade.

UHMWPE is not a single material but a range of materials with differing properties and performance.
For each of the polymer types, the physical properties are affected by the method and conditions used in converting from the polymer powder to the solid material. The properties of the solid are often changed by the device manufacturer by the use of post treatments varying from sterilising dose gamma radiation to high dose radiation to increase cross linking. Some of these treatments are intended to give substantial benefits, but these benefits may also be affected by the choice of polymer and processing conditions. Since these post treatments are outside the converter's control, I will restrict my comments to the properties of the bulk material as supplied to the device manufacturers.

Several different grades and types of UHMWPE have been used over the years, which has resulted in confusion in nomenclature.

Ticona (which is the business spun off by Hoechst to deal with UHMWPE etc.) currently produces four grades certified to ISO 5834 part 1 and ASTM F648. The new numbering system which has taken over from the old American and European ones gives an easily understood product definition. The first digit being one signifies medical use polymer, the second shows the addition of calcium stearate or not and the third digit is related to the molecular weight.
Therefore GUR 4150 HP (often called 415) is now GUR 1150 and the original European sheet grade (often called 412) is GUR 1120. With the general move away from calcium stearate, the most popular materials are GUR 1020 and GUR 1050 in our plant.

To complete what is believed to be the current polymer picture, there is also 1900 polymer from Montell in use.
It is often not realised that the Montell 1900 UHMWPE is also a family of grades with a similar spread to the Ticona product.

I have yet to hear a speaker refer to the Intrinsic Viscosity or IV number which relates to the molecular weight and hence the properties when discussing the wear of this material. 1900 polymer has also been available with calcium stearate but it is thought that most of the material used for devices has been without it.
It would be of great value to the converter, and aid our materials research programme, if other researches could reveal the full polymer identification as well as conversion method when reporting on their research.

Due to the extreme molecular weight of UHMWPE, it does not become fluid when heated above it's crystalline melting point, but becomes a translucent amorphous rubbery material. This restricts the processing of the pure polymer to sintering methods the most popular of which are Ram Extrusion and Compression Moulding. Direct Compression Moulding is a variant of Compression Moulding where near net shape components with a mould finished articulating surface are produced one at a time from a miniature compression press.

Ram Extrusion has been the established method of conversation in the USA for many years, and has been used to a lesser extent in Europe. The traditionally preferred polymer was a higher molecular weight, stearate containing material, which built up good back pressure in the die and contained calcium stearate to keep the die surface lubricated. More recently, the process conditions have been developed to allow processing of all available grades.
Due to the short time available, Ram extrusion has to use a large temperature gradient both on heating and cooling. This can result in property variation if the output rate is run too high. Pressure control is a result of friction between the polymer and the die wall combined with the ram speed and is always cyclic.
Ram Extrusion is the least expensive conversion method.

Compression Moulding into slab form provided the original UHMWPE material for joint surgery. The original polymer is believed to have been the equivalent of Ticona GUR 1120. Now all UHMWPE polymers can be processed by this method.
Due to the large size of compression presses, it is necessary to heat and cool slowly resulting in little temperature gradient across the material and a long processing time for the grains to sinter together. On modern presses the control of both temperature and pressure is by closed loop feedback controls resulting in high levels of accuracy and repeatability. To give an idea of scale, our medical press weighs 450 tons, stands 35 feet high and can exert over 7000 tons force. It uses up to 9000 cubic feet a minute of gas to heat the fluid circulation system to produce two slabs of about 13 feet by 6 � feet by up to four inches thick. This scale of equipment, and the longer processing time that can be beneficially used, results in a moderate cost.

Direct Compression Moulding as a variant of compression moulding provides a single part from each mould cycle. Since the material thickness is much smaller, the cycle is much shorter than bulk slab moulding. Systems have been developed to maintain the same level of process control as the large presses. The short time at temperature does change some material properties.
The overall result is a similar material to bulk compression moulded, with a mould finished surface which can be intentionally gloss or stippled, but at a higher cost.
Within each processing method there are several processing conditions that can be altered either directly or as a result of changing another condition. The effect of each of these needs to have been assessed in order to optimise the material quality.
It is not sufficient to merely pass the minimum ASTM or ISO standard - we must be aiming to maximise the properties by process condition and reduce variation by process control.
The following data sets on temperature and time were generated from mouldings made on a purpose built laboratory press using our standard production pressure cycle. Pressure was controlled to within plus or minus 0.2 bar or 3 psi and temperature was controlled to a repeatability of plus or minus 2 C by fluid circulation. Both control systems are programmed for time and ramp rate and utilise feed back correction systems.

Temperature.

It is necessary to process above 135C to get UHMWPE above it's crystalline melting point, and below the point at which oxidative degradation accelerates. This point varies according to polymer grade. Minimum temperature and time to achieve consolidation through the thickness are related by the thermal conductivity. Higher temperatures mean shorter minimum processing times.
Within this band of process temperatures, some properties such as tensile strength do not appear to be greatly affected, while the double notched impact strength is far more sensitive.
It can be seen that at temperature above normal, the polymer starts to degrade and become brittle whereas below normal there was insufficient bonding to maximise the strength.

Time

The minimum time chosen was sufficient to achieve a visually good moulding, however the morphological examination did show an increased mottling at low times. All tested mouldings were free of actual fusion defects. It can be seen from the graph that there appeared to be little change in tensile properties but significant improvement could be made to the impact strength by lengthening the cycle time.
The combined effects of temperature and time can clearly be seen in this three dimensional graph of impact strength.
The highest temperature on this graph is the normal process temperature since we have shown the degradation above that. It can be seen that the greatest impact strength is at the combined high temperature and long time point.
The maximum density and tensile properties were found at the same conditions.

The swell ratio was assessed on the same mouldings including this time, the very high temperature set. The lowest swell ratios occurred at the normal process temperature, above this the degradation caused a sudden increase.

Molecular weight

The effect of using different molecular weights of otherwise similar polymers, for example using different IV numbers of Montell 1900 or changing the Ticona grade number, can be seen in this graph.
The results are from historical data for a range of polymers - not all medical - processed under similar conditions.
The impact strength peaks at around three million nominal molecular weight which coincides approximately with the minimum in ASTM F648. As the molecular weight increases there is a gradual reduction in impact strength but this can be offset against an increase in third body abrasion resistance. Elongation is more constant over this range.

Pressure

Pressure is required to create consolidation. Very low pressure will result in physically porous mouldings. Very high pressure will result in frozen in stress. Initial very high pressure is used to expel air while the polymer is below melt point.
Medium pressure during the sintering phase gives good thermal conductivity and consolidation. Reduced pressure at the critical point in the cooling cycle reduces density variation and hence stress.
This is an example of density variation presumed caused by poor control of pressure and temperature.
Fusion defects or visible grain boundaries have been shown to increase the oxidative degradation of UHMWPE. The elimination of calcium stearate has been shown to reduce the visibility of grain boundaries, but the elimination of fusion defects requires pressure control. This is an example of a piece of polyethylene that was made into a device last year.
The International Standards Organisation has recognised the importance of this problem, and the potential effect on joint failure, with the result the 1998 issue of the standard for UHMWPE - ISO 5834 part 2 - has been sent to committee for further revision.
The requirement is to add a test method for consolidation.

Cross linking

A lot has been said about the beneficial wear properties of cross linked UHMWPE, as I said earlier, I can not comment on this as we are not device manufacturers or testers but it is interesting to note the difference in processed materials available to the device manufacturers..
The crosslink density can be calculated from the swell ratio, which in turn can be calculated from ASTM D2765 and other proposed methods.
The basic method is to place a sample of the UHMWPE in a solvent (normally xylene at 110C) at which point it is molten and adsorbs some the solvent.
The volumetric swell ratio between the original solid and the gel is calculated correcting for temperature and density.
The swell ratio is proportional to the sample's ability to change size, which is restricted by crosslinks and chain entanglement.
High numbers show low levels of crosslinking and low swell ratios show increased crosslinking.
The process-induced reduction in swell ratio was affected most by the choice of processing method or process condition and least by the molecular weight. The process with the lowest exposure to combined temperature and time is ram extrusion, and a polymer processed by this method exhibits a swell ratio nearly double that of a normal compression moulding. Short cycle low temperature compression mouldings can be similar to ram extrusions.
The molecular weight effect is much lower and the polymer range which is covered by the ASTM F648 requirements has a plus or minus 20% from average for any one processing method.

In conclusion, it can be seen that the properties of Ultra High Molecular Weight Polyethylene are not fixed. They are greatly affected by the actions of the processor and the process method chosen by the device manufacturer

Neil T Hubbard