The uncertainty about uncertainty – the new ground of sufficiency in Singapore

27 Jun 2023

Incentivising research to push the frontiers of science and technology is what underpins the patent regime. It is inherently forward thinking. Which is why it is a great irony that when the validity of a patent is challenged, the exercise is predominantly retrospective. The Court needs to put on the mantle of a person skilled in the relevant art at the relevant time to determine whether the invention is new and inventive, and whether the patent can be understood. The relevant time can be a decade or more ago. This is therefore an immensely difficult task, one that requires the Court and the lawyers not only to apply the law, but also to double up as skilled persons and historians.

Navigating a patent revocation action also requires much more finesse than one assumes. It is all about striking a balance. Take the sufficiency requirement of a patent as an example. Sufficiency is the cornerstone of the patent’s bargain. To earn the right to a monopoly, the inventor must ensure that the patent clearly and completely discloses the invention such that a skilled person is able to understand the patent and its claims, and is also able to recreate the invention.

However, the Court takes a practical approach in determining the extent to which the inventor is burdened with the obligation to clearly and completely disclose his invention. Absolute clarity and completeness are not uncompromisingly required. Some latitude is afforded to the inventor to assume that the skilled person will be able to fill in some gaps by himself. Any ambiguity in the patent, while not encouraged, is also not fatal. It is well established that lack of clarity is not a recognised ground of revocation. In First Currency Choice Pte Ltd v Main-Line Corporate Holdings Ltd [2008] 1 SLR(R) 335 (“First Currency Choice”), the appellant argued that the patent was unclear and ambiguous and should be revoked because a skilled person would have difficulty understanding the terms “identifier code”, “issuer code” and “issuer identifier code”. The Court of Appeal held that this was not enough to invalidate patent, and that the “failure to define certain terms … did not mean that the Patent Specification was therefore insufficient”.  

So how unclear must a patent be before it transcends beyond merely a lack of clarity to one that causes the patent to be invalid? This question was explored to its full extent in the recent Singapore Court of Appeal’s decision in IIa Technologies Pte Ltd v Element Six Technologies Limited [2023] SGCA 5 (“IIa v E6”).

The patent in the appeal is Singapore Patent No 115872 (“SG872”), and was owned by Element Six Technologies Limited (“E6”). It had a claimed priority date of 21 November 2002 and was granted in Singapore on 30 May 2007. The patent concerns an alleged new method of growing a diamond in the lab via chemical vapour deposition (“CVD”).

Claim 1 was the main independent product claim in SG872. It claimed, amongst other things, a CVD single crystal diamond with a low optical birefringence such that when it is measured over a specified area, the birefringence of the CVD diamond remains in first order, d does not exceed p/2 and |sin d | does not exceed 0.9. To measure these parameters, the patent required the use of an equipment called “Deltascan” to measure these parameters. “Deltascan” was renamed to “Metripol” subsequent to the filing of SG872.

E6 sued IIa Technologies Pte Ltd (“IIa”) for patent infringement of SG872, and IIa countersued, amongst others, that SG872 was invalid and should be revoked because the patent was insufficient. In particular, IIa claimed that it was not possible to use the Metripol to determine whether the birefringence of the CVD diamond remains in first order, d does not exceed p/2 and |sin d | does not exceed 0.9.

Having brought this issue to the fore, as there were two inventors of the Metripol, E6 and lla each called one of the inventors as their expert witness for the trial. Since the patent had a claimed priority date of 21 November 2002, counsel and the experts had to delve into the knowledge that existed more than 20 years ago, in determining whether a person skilled in the art would know how to work the Metripol to determine whether the birefringence of the CVD diamond remains in first order, d does not exceed p/2 and |sin d | does not exceed 0.9.

First, as a matter of law, E6 argued that the flaw of the Metripol is an issue of lack of clarity, which is not a basis for revocation. The Court of Appeal accepted and reaffirmed that lack of clarity is not a ground of revocation, but went on to find that IIa was entitled to rely on a new type of insufficiency called uncertainty. “Insufficiency by uncertainty” is not to be confused with lack of clarity, although they are similar. They differ in the extent to which the lack of clarity can be said to have impugned the validity of the patent. The test for clarity is whether a person skilled in the art can understand the words used in the claim. For uncertainty, the lack of clarity must be so great that it leaves the person skilled in the art unclear as to how to determine whether a particular product or process is within the scope of the claim.

After establishing that Singapore law recognises this new type of insufficiency, the Court of Appeal went on to apply it to the facts.

It was common ground between the parties that the Metripol, on its own, does not give the person skilled in the art data to determine whether d falls within p/2. The question was whether there were any other available solutions at the relevant time to solve this problem. During the trial and on appeal, E6 claimed that there were four solutions which would enable a person skilled in the art to verify whether d falls within p/2, but the Court of Appeal found that none of these solutions were workable or, in any event, formed part of the common general knowledge. The Singapore Court of Appeal found that a person skilled in the art would not know how to determine whether a particular diamond satisfied “d falling within p/2”. Neither the Metripol nor the common general knowledge could plug this gap. Consequently, the main independent product claim was uncertain and therefore invalid. All subsequent dependent claims that relied on the main independent product claim were also similarly infected with the same uncertainty, and the whole patent was revoked.

IIa v E6 was an excellent test case for the Court of Appeal to introduce uncertainty as the new type of insufficiency into the consciousness of Singapore patent practitioners. It gave an answer to a question that was left unanswered for 15 years since First Currency Choice. But because the question of clarity operates on a spectrum, it remains to be seen what exactly the limits of insufficiency by uncertainty are i.e. how unclear the patent specification must be before it reaches the tipping point from merely being a lack of clarity to uncertainty. In IIa v E6, the uncertainty was striking. It was evident that no one, not even the inventors, could clearly identify an acceptable way to determine whether d falls within p/2. Not all cases will be so fortuitous. For example, if a patent claims a parameter of 1 to 10, but the measurement method can only confirm parameters from 1 to 5, is this patent claim still uncertain, or just ambiguous? The Singapore Courts will have to ensure that insufficiency by uncertainty will not be used as a trojan horse to disguise what is fundamentally merely a lack of clarity argument. Once more cases expound on this new type of insufficiency, it is hoped that the limits will become clearer and clearer.

More importantly, IIa v E6 must not be seen as an aversion for patentees to rely on new measurement techniques or equipment to quantify parameters in the claim. There may be circumstances where a particular characteristic can only be quantified with a particular parameter, and measured only in a particular way. Patentees should not, as a default, be penalised simply because of this. That would be the antithesis of the patent regime, which is to encourage discovery of anything novel and inventive. However, if and when a patentee chooses to rely on such an esoteric measurement technique or a typical equipment that is not in the common general knowledge, then the patentee bears the burden of fully teaching the person skilled in the art how to carry out that measurement. Afterall, a patent is only as good as it is understood.
 

• This article was first published on Managing IP website here on 13 June 2023