If you deal in polymers and have not come across PEEK – it’s probably because its one of those materials which does not surface unless really needed. When it is needed – there’s little else that can be used in its place and this often confuses OEMs, because even among expensive, high-end engineering polymers PEEK sits at a price point that causes the client no small amount of shock.

It is important to talk about the price of PEEK before all its other characteristics, as this is usually the first thing the client wants to discuss. Invariably, they come knowing that they need this mystery polymer (PEEK), but knowing little else. They expect the price to be similar to Polyacetal or, at the very worst PTFE. When they find out that it is close to 10 times the price of PTFE, it comes as a huge surprise.

Why PEEK is expensive is not fully known. Perhaps it is because it has not yet reached the global scale of manufacture of more commoditized polymers, or perhaps the technology is so unique that it allows resin suppliers to charge a huge premium – knowing that alternatives are not available. As processors, we know only so much:

1. The resin is 5-8 times more expensive than PTFE
2. Processing PEEK is time consuming and expensive in comparison to PTFE
3. Machining PEEK is tricky in comparison to other polymers

Since the resin prices are not in our control, we would like to look at points 2 and 3 and discuss them in more depth. But first, let’s get a better idea of what PEEK offers.

High tensile strength

In the polymer space, it would be tough to find something tougher than PEEK. It is so strong, in fact, that machining guidelines for PEEK need to follow the same as those for metals.

This strength allows PEEK to be used in applications such as gasketing and auto components – especially where metals cannot be used, but where a metal-like durability is required

High temperature resistance

PEEK melts at about 400 Degrees Celsius and is capable of running in environments of 300-325 Degrees without deforming. While PTFE can withstand up to 250 Degrees, any pressure/ load on PTFE at this temperature will invariably cause deformation. In the case of PEEK, its hardness allows it to be in a high-load-high-temperature environment without loss of dimensional properties.

High wear resistance

Again, while both PTFE and UHMWPE can take a significant amount of wear, PEEK exhibits a high PV value and can withstand wear effects even under harsh physical and chemical conditions.

Chemical resistance

While not on the same level as PTFE for pure chemical inertness, PEEK exhibits resistance to many harsh chemicals, allowing it to be used in corrosive environments, under heavy loads
In a nutshell, PEEK’s ability to stay dimensionally stable under harsh environments makes it a highly sought after polymer. OEMs who use PEEK do so knowing well that for the properties offered, PEEK is unique and therefore expensive.

Processing PEEK

We will not delve very deep into the processing of PEEK (as this is a proprietary process unique to each processor), but we will point out the key differences between PEEK and PTFE processing (which has been looked at earlier). It should be noted that here we are referring only to compression moulding, and not injection moulding.

The main difference is that while PTFE is cold compression moulded and then loaded in batches into a sintering oven, PEEK needs to be sintered during compression itself.  Furthermore, post sintering, PEEK needs to go through an annealing process, which is time consuming. This leads to a few complications:

1. Batch processing is difficult. Since the total heating cycle for a single piece can take up to 8 hours, and since heaters are expensive, PEEK is normally moulded a few pieces at a time. So unlike PTFE, where a batch of 8-10 large pieces can be moulded in series and then put in the oven for a single cycle, PEEK will offer only a few pieces in the same amount of time.
2. Since PEEK is heated under pressure, issues of flash can arise as the resin becomes molten, but has pressure being applied on it. Furthermore, the pressure and temperature have to be balanced very carefully, since the temperature makes the PEEK molten, allowing it to reach its desired shape, but the pressure is responsible for vacating air bubbles from the material, so that there is no porosity.
3. Batch processing the PEEK parts for annealing is possible, but takes about 24 hours

So overall, the productivity in moulding PEEK is far below that of PTFE. This does answer, in part, the question of why the price of the finished material is so expensive.

Machining PEEK

As discussed above PEEK machines more like a metal than like a polymer. It is hard and has a significant impact on the tool. The same tool that might churn out 3000-4000 PTFE parts may struggle to churn out a few hundred PEEK parts. Again – this adds to the cost of the finished product significantly.

More importantly for machining though is that if PEEK is not annealed properly, the part will behave erratically during machining as different areas within the material react differently to the stress being placed by the tool. Thus, cracks can develop during machining and the dimensional stability across a batch of components can vary significantly.

As a result, PEEK machining is a difficult process and there are few who are willing to take on the risks of machining such an expensive item, knowing that the rate of rejection could be very high.

In conclusion – PEEK has remained a largely niche polymer due to its prohibitively high price. If it were cheaper – say around the price of PTFE – there are chances that it could steal a significant chunk of the PTFE market. PTFE still rates much higher than PEEK on characteristics like coefficient of friction and dielectric strength, but where it is a question of sheer strength, PEEK stands unmatched amongst polymers.

So we’re back to pricing – because until they fully stabilize, we need to be on our guard. Considering the data below, one might be allowed to assume that things are finally easing out and that the sector is slowing reaching an equilibrium of sorts, coming off the highs seen in mid-2011 to rest at about US$24/Kg. But we would rather still be wary.

Since prices spiked in July 2011, there has been a decrease of about 13% in prices – which has been gradual. There are a number of reasons one can point to for this decrease – most of which we have already touched upon in our last article on pricing. To list them out:

1. Re-entering of China and Russia into European and Indian markets at competitive rates
2. Easing out of Fluorspar supplies due to opening of new mines and reduction in China’s domestic consumption

However we remain wary for 2 very specific reasons:

1. China’s summer approaches.In our very first article on pricing, we specifically highlighted the impact that the Chinese summer was having on PTFE prices. Summer months spike domestic demand for refrigerators and air conditioning and consequently cause R22 to be diverted from PTFE and into these products. This creates the shortage in R22 and was one of the root causes for the price escalations seen last year. However, we also postulated that once summer passes, the prices would ease out – which they have. But what now? Summer is about 2 months away and there is nothing to suggest that the rest of the world’s fluorspar mines can support the industry as yet. Our own sources indicated that it would be at least 2 years before the re-opening of mines in Mexico and South America eased the supply side constraints on fluorspar.
2. The PTFE industry is far from efficient.In finance, we always assume that if an event (like China’s summer) is imminent and the effects of that event are known – then the prices of goods linked to the event should already reflect this information. In other words, if processors were aware that prices are going to spike during the Chinese summer, they would already have stockpiled raw materials to avoid against it, implying that there would be less demand during the summer and prices would not escalate again. However, this is unlikely to have happened since, (1) there are mixed opinions on whether the prices will go up or keep going down and (2) processors have already had to triple their working capital in order to keep up with the price increase in raw materials and it is unlikely that too many would have funds to stockpile materials for a full quarter. Therefore we remain nearly as exposed as we were last year.

But the news may not be all that bad. For one, China has been seriously implementing the R22 phase-out and as of August 2011 was even awarded a grant to speed up the efforts. Whether this phase-out sees any immediate impact on domestic demand remains to be seen and would possibly define the price of PTFE for the next few years. Secondly, there would be good reason to believe that PTFE resin manufacturers have hedged against such a scenario – even if the processors themselves are unable to do so. In India, our local producer has not only augmented PTFE resin capacities, but also become one of the foremost global producers of R22.

In a nutshell, the state of the future depends largely on the balancing of the Chinese summer against the precautions taken by resin manufacturers to safeguard against a further spike. I do not believe processors have any real part to play in all this – we remain, for the most part, price takers. If there is a fluctuation in prices, we would need to absorb it much the same as we did last year.

Machining PTFE, as we have touched upon before, is never a straightforward process. Most machining handbooks will suggest that PTFE should be treated much like wood when it comes to machining, as this is the material it most closely behaves like when machined. And while this is a good starting point for tool selection and CNC programme settings, as we delve deeper into the aspects of machining PTFE, we see that it behaves much like it’s own material. So learning by doing becomes the only option – since PTFE is a niche product (when compared with other known polymers).

Recently, we faced an interesting issue when creating a rather complex part. The part is approximately 200 Grams in weight and machining it involved multiple operations including CNC turning, CNC milling, drilling and finally tapping. All in all, the drawing highlighted over 28 dimensions that needed to be within a strict tolerance and it took us the better part of a week to just get 10 prototypes ready.

We were pretty happy with the result: everything measured, as it should. We almost didn’t check the tapping – which called for an M3 tap in two places. The M3 taps used were brand new and the first tap was done on the VMC as part of the programme – so there was no way it could be an issue, we thought. But we were wrong.

The no-go gauge entered in the hole all too easily and we were pretty shocked to realise that even an M3 bolt was sitting loose in the hole. At first we though we had the wrong tap – which we didn’t. We then argued that the gauge would always enter – as it was designed mainly for harder materials and PTFE would yield all too easily, since it was much softer. To check this we used the same taps on a mild steel plate and confirmed that the no-go did not enter. But this still did not answer why the bolt itself was loose.

We searched extensively for an answer online, but there was very little information on tapping and even less on the issue we in particular were facing. We then decided to start experimenting with different combinations of taps and drill holes.

On the part, we had used a 2.2mm drill with all 3 taps. The first tap was done on our VMC, while the next 2 were done by hand. We tried the following combinations:

Drill Hole	Tap 1	Tap 2	Tap 3	Result	Remark
1.5	Y	Y	Y	Reject	Bolt loose
1.5	Y			Reject	Bolt loose
1.5			Y	Reject	Bolt tight
2	Y	Y	Y	Reject	Bolt tight
2			Y	Reject	Bolt loose
2.2	Y	Y	Y	Reject	Bolt loose
2.2			Y	Reject	Bolt loose

In a couple of cases – where we used only the 3rd (finest) tap, the bolt was tight. However, none of the holes were answering to the no-go gauge, which passed equally easily in all the holes. We once again argued that this was a PTFErelated issue and that as long as the bolt was tight, it should not be a problem. But many of the consultants and experts I spoke with said that they had come across parts in PTFE that answered to the no-go gauge, and hence there must be a way to machine such a part.

The problem was finally solved when an engineer in our client’s side suggested we use a “Form Tap”. I had never heard of a form tap and when I searched it, it seemed to apply mainly to tapping soft metals (such a aluminium). There was no mention of applications to PTFE. Nonetheless, it was our last shot, so we tried it and were pleasantly surprised.

We eventually went with a 2.0mm drill and an M3 form tap to get a result that was both functionally good and which answered to the gauge.

The reason the form tap works, is because unlike a regular tap, it does not bore into the PTFE, taking out material as it does. Instead, it merely forms the tap profile within the drilled hole and as PTFE is soft, it yields quite easily. The result is that the tapped hole is much fuller than when a normal M3 tap is used – making it tighter and ensuring the pitch profile does not yield to the no-go gauge.

Surprisingly, this does again strengthen the PTFE-Wood similarity in machining. Tapping is unheard of in wood; a screw can be passed through a drilled hole and sit tight forever! In many ways, a form tap is nothing more than passing a screw/bolt into the PTFE to imprint its profile within the hole. Only that the form tap is possibly more exact and can ensure that the resulting tap is accepted when inspected with the correct gauges!