BREDEMA

"The science of a good nose!"
An interview with Pierre Breese 

Source: Découverte
Issue N°337
Date: April 2006

 

With a background in physics and law, Pierre Breese is a founding partner of an industrial property consulting firm based in Paris and London. He helped set up the Valorization Department at INSERM, the French National Institute of Health & Medical Research. He has published several books including “Industrial Property Strategies”, “Financial Evaluation of Industrial Property Rights”, “Olfactory Marketing” and “IP at the Service of Innovation”. He lectures in Industrial Property Law in engineering schools such as HEC, ENST and IEEPI.

After copying the packaging, bottle and brand name, perfume counterfeiters have now turned to faking the fragrance itself. Although it is possible to protect the packaging or even the formulation of a perfume, or at least to authenticate them, what can be done about protecting the fragrance?

For a recent court case between L’Oréal and the Belgian company Bellure, the industrial property consulting firm BDM took up the challenge of finding and adapting the means for identifying the unauthorized imitation of the olfactory information specific to a perfume. *

Counterfeiters have long been able to copy a perfume’s external appearance, i.e. its bottle and packaging, but more recently they have turned their attention to faking its fragrance (olfactory message) as well. The name of the perfume and the shape of the bottle can be protected under trademark law and design law respectively.

But industrial property laws have always been powerless to protect a fragrance, and so its formulation has traditionally been kept a closely guarded secret. Nowadays, any analytical laboratory is capable of recreating a perfume’s formulation. Secrecy is no longer a viable option, and so it was important to find a way of protecting a fragrance. Since a perfume is essentially the result of a creative act, just like an artistic or literary work, couldn’t a fragrance be considered to be an intellectual work and thus protected by copyright ? Even so, it would still be necessary to work out exactly how to apply copyright to olfactory creation, as the difficulty lay in objectively demonstrating the olfactory similarity of the original perfume to the allegedly fake perfume. Scientific techniques were therefore developed to compare two perfumes. This work was directed by Pierre Breese (industrial property consultant) in collaboration with ISIPCA (International Superior Institute of Perfume, Cosmetics and Food Aromas).
His background in both physics and law gave him the right expertise and experience to successfully tackle this assignment, the object of which was to provide the judge not with any clear-cut answers but with material facts and figures on which to base a decision as to whether or not there has been an infringement of copyright.

By granting us this interview, Pierre Breese guides us step by step along the path where science meets law.

 

What different types of counterfeits can be found in the field of perfumery?

Pierre Breese. First of all, there is a very basic form of counterfeit readily available on market stalls. The external appearance of the perfume is imitated, with a very similar brand name, packaging and bottle to the original perfume. However, the “jus” (term commonly used to refer to the formulation of the perfume) is very different and of poor quality and only bears a slight olfactory resemblance to the original brand. The problem is that, when it is bought from market stalls, the purchaser is told that it is an authentic product that happens to be in a different bottle and this supposedly explains its rock-bottom price. The consumer is therefore taken in.

Then, in the 1990s, new types of counterfeits began to appear where the “jus” was much closer to the original perfume but the packaging, bottle and brand name were different. With current analytical methods, a “jus” can be prepared whose fragrance (olfactory message expressed by the perfume) is identical or very similar to the fragrance of an original perfume. This fragrance counterfeiting tends to be the work of fraudsters with huge sums of money at their disposal, which means that they can procure their raw materials from the same suppliers as the major perfume houses and can purchase sophisticated analytical equipment. This type of forgery is on the increase and the highly lucrative nature of this market makes it worth while for the counterfeiters to invest in the necessary means to produce high-quality imitations.

 

Types of infringement in perfumery

Brand infringement: unauthorized copying or imitation, possession, sale or import of a sign that is used to distinguish a product or service and that has been registered with INPI (National Institute of Industrial Property). In France, a brand name is protected for an indefinitely renewable period of 10 years.

Design infringement: unauthorized copying, sale or import of an object that is distinguished by characteristics of external appearance and typography and that has been registered with INPI. Protection is for five years, renewable four times.

Copyright infringement: reproduction, import or sale of an original literary or artistic work without the author’s consent. Copyright is acquired by the mere fact of creation. Protection is for 70 years post mortem.

 

What industrial property rights are applicable to the different types of perfume counterfeiting?

PB.

The name of the perfume and the shape of the bottle are protected by trademark law and design law respectively (see section Types of infringement in perfumery). The ways to combat these types of imitations are fairly well-established, easy to apply and effective but do not extend to the heart of the creation itself.

Fragrance counterfeiting, on the other hand, is much more difficult to control as these legal remedies cannot be applied to protect the olfactory message expressed by the perfume. In practice, until recently, the only way to protect a fragrance was by secrecy. The formula was kept locked in a safe and divulged to no-one. This made it impossible to recreate a Chanel N° 5 or, for that matter, copy the exact composition of a Coca Cola.

Nowadays, any chemical analysis laboratory can work out the basic formulation in half a day or a day. Secrecy, which still afforded a real degree of protection only a few years ago, is now no longer feasible, and so it has become difficult to maintain the exclusiveness of a new, original creation. Progress in analytical techniques has led to a surfeit of new perfumes. As it is very easy to find out the formulation of a perfume, it is enough to replace one chemical molecule or group of molecules with another (similar or completely different) molecule or group of molecules. A mixture will eventually be obtained with an olfactory message that is very similar, or even identical, to the original perfume. In fact, some molecules have the same olfactory description despite being chemically different. It is possible for two compounds to be either chemically very different or to display only a minor structural modification but to release the same olfactory message. Conversely, it is possible for two chemically very similar
products to have a very different olfactory message. All it requires is for the composition of the perfume to include a molecule that produces a very strong olfactory impression, altering the nature of all the others. The olfactory message of a perfume can therefore be copied using a different chemical composition. Today, any successful, newly launched perfume soon finds very similar fragrances following in its wake. Hence the importance of finding a way to fight this type of copy.

 

How then can the fragrance of a perfume be protected?

PB

Creating a perfume consists in harmoniously combining a selection of fragrant components known as odorants. The choice of components is determined by the perfumer according to his or her inspiration and expertise in blending a large number of odorants. The creator’s personal “touch” can be recognized in a perfume. Under those conditions, a perfume can be considered to be an intellectual work and will therefore be protected by copyright in the same way as a literary or artistic work (see section Types of infringement in perfumery).

Copyright is acquired by the mere fact of creation. It protects the author of an intellectual work, which automatically becomes a legal object simply by being created. Unlike patents or trademarks, copyright arises independently of any formality or registration.

Although the law does not explicitly state that an olfactory composition is an intellectual work, the list of protected works quoted is simply given as an example and does not in any way preclude an extension of the law to other forms of creation. Copyright can therefore apply to olfactory creation in so far as it expresses its creator’s personality, on the same grounds as a literary or artistic work. Moreover, the vocabulary employed in perfumery is very similar to the terms applied to a musical work. The creator of a perfume is known as a “composer”; the composer has an organ, a piece of furniture holding a range of compounds for creating a new composition. And finally, it is usual to refer to top notes, middle notes and bottom notes to describe the changes in the olfactory perception of a perfume.

 

Why wasn’t copyright applied to olfactory creation earlier?

PB

The major problem in applying copyright to olfactory creation was finding a way of objectively comparing the fragrance of an allegedly counterfeit perfume with the fragrance of an original perfume. For a musical or artistic work, there are appropriate, accepted means of comparison (the major “mechanical” senses: touch, sight and sound). Comparisons are fairly easy to make using words as we have all been familiar with materials, colors and shapes since our earliest childhood. But for an olfactory creation, we do not have any objective means of characterizing an odor. In the 21st century, most people have little experience of describing a smell and only a limited vocabulary for doing so, and this makes it difficult to give a clear, universal description of a fragrance. However, experts - commonly known in the trade as “noses” - can describe a perfume in precise terms. Over a period of many years, the “nose” learns to memorize an increasingly large number of odors by associating them with a name or, better, an image. It would therefore be possible to entrust an expert with the work of characterizing and comparing a creation and a potential counterfeit. But in many cases, experts vary in their opinions. The very mechanism of memorizing odors is quite unusual.

First of all, how does a molecule give rise to an odor ?

This phenomenon is described in the following extract from an article by M-H. Faverger, published in Découverte N° 281, October 2000, page 30. ”To perceive an odor, molecules must first be released into the air. These are then carried to the nasal mucosa by the intake of breath. The nasal mucosa includes a structure known as the olfactory epithelium (or membrane). This part of the nasal mucosa contains millions of neurons (10 million) that transmit information on the molecules reaching them to the brain. Each of these neurons has long, filamentous extensions or cilia at one pole, and fibers (known as axons) connected to the brain structures at the other pole. Bathed in watery mucus, the cilia are in direct contact with the inhaled air and contain sensors or chemoreceptors to detect the odorant molecules. These molecules attach to the receptor proteins carried by the olfactory neurons. They then trigger physical-chemical reactions resulting in the emission of electrical signals. These signals are then transmitted by the axons to the corresponding olfactory bulb. The olfactory bulb, after processing the information, will interpret the meaning of the odorant message, associate it with a sensation, thought or emotion, or even trigger a behavior.”

We therefore start with a substance having a chemical composition that releases an olfactory form that is perceived and then interpreted by the brain. However, interpretation by the brain has a cultural and historical dimension. Two people who do not share the same history will not derive the same emotions from the same olfactory form; this is the “Proust syndrome”. Nonetheless, this does not mean that they will not have the same perception. It is precisely to this difference that perfumes owe their potency, since their memorization is always related to a life event, a culture, intense experiences or powerful emotions. To apply copyright, the word of the expert is not enough.

 

How then is it possible to objectively compare the olfactory similarity of two fragrances?

PB

To do this, the scientific techniques that have been developed for comparing an original perfume with an allegedly counterfeit product must be applied not to the respective compositions of the two products but to the olfactory messages. As we have already seen, two different formulations can give the same olfactory message: two chemically very different compounds can release the same olfactory information. Conversely, two chemically very similar products can have a very different olfactory effect. All that it takes is for one molecule to produce a very strong olfactory impression, altering the nature of all the other molecules.

A number of methods already known to perfumers and chemists were therefore envisaged and examined with a view to establishing a battery of tests for the objective comparison of two fragrances: sensory methods (triangle test with forced choice and assessor) and physical-chemical analytical methods such as the electronic nose or gas chromatography coupled with mass spectrometry (GC-MS). The combined results of these individual methods can be considered to display a reasonable degree of objectivity.

It was by referring to these scientific methods that the industrial property consulting firm BDM was able to convince the judges in the above-mentioned lawsuit.

 

What was the first method envisaged for comparing two fragrances?

PB

The electronic nose was the first physical-chemical analytical method tested. This is a well-known quality control technique in the world of perfumery. Nevertheless, as we shall see, this analytical method was ultimately rejected as a means of fragrance comparison testing for infringement control. The operating principle of the electronic nose is described in the section of the same name.

In our case, the logic behind the electronic nose test is quite simple: two identical perfumes have approximately the same profile. This technique is eminently suitable for quality control testing to see whether there is a drift in the composition of a perfume in a production line. Such a drift will immediately modify the output signal. But we cannot go much further than that. If the two signals are different or slightly different, this does not necessarily mean that the two perfumes are different. Further, the signals may be very similar but the perfumes very different, or vice-versa. The information that can be drawn from this technique is too open to criticism and the transparency of the system is poor. It was not therefore included in the battery of tests to be used for defending an olfactory creation by copyright. The next method to be examined was gas chromatography coupled with mass spectrometry (GC-MS).

 

Could you explain the advantage of using GC-MS for comparing two fragrances?

PB

GC-MS is a reproducible, precise and rigorous analytical method much used in the field of perfumery (see section Gas chromatography coupled with mass spectrometry (GC-MS)). This physical-chemical technique can be used to identify and monitor the molecular components of a perfume. The method used is as follows. Each sample is injected into the gas chromatography apparatus. For each injection a chromatogram is produced containing a number of peaks, each representing a component. The height of a peak indicates the concentration of the component in the sample of perfume. The chromatograms are then compared by juxtaposition to see whether or not any identical constituents are present in both analyzed samples. All the compounds present in one of the two samples are determined. A check is made whether that sample contains any compounds that, despite being chemically different, have the same olfactory description (as indicated by a reference work) as any compounds in the other sample. Pairs of perfumes displaying the greatest similarity are thus identified. The fact that two perfumes have a significant number of common olfactory components is a strong indication of counterfeiting or at least of resemblance. When used in isolation, this technique can be criticized, but it has its relevance. The quantitative evaluation of the similarities between perfumes forms a complement to GC-MS analysis and is done by a technique known as principal component analysis (PCA).

 

In what way is principal component analysis the essential complement to GC-MS in comparing two fragrances?

PB

The purpose of PCA is, once chromatographic analysis has been performed, to determine the molecules present in the largest numbers in the perfume and to arrange them in decreasing order of occurrence. Generally, about fifty molecules often used in perfumery to contribute to the olfactory message are examined. A comparison is then made of the arrangement of molecules in the original perfume and in the allegedly counterfeit perfume. Some molecules will be found in both perfumes, some will be present in neither and some will be so-called substitution molecules. A substitution molecule is a molecule that is different from the one in the original perfume but that is known to have the same olfactory behavior. Then, from among the main molecules, the proportion of molecules common to both perfumes is measured and a calculation made of the degree of probability that the similarity is the result of chance. In fact, a composer has a palette of about 150 odorant raw materials to create a perfume. Each perfume contains between 10 and 50 olfactory components in significant concentrations. A composer will combine about thirty molecules from among these 150 to create a perfume. It can for example be estimated that the probability that, out of these 30 molecules, 27 identical ones will be found in the allegedly counterfeit perfume is about the same as the probability of winning the lottery every day for a century.

 

Operating principle of an electronic nose

The operating principle of an electronic nose is similar to the previously described operating principle of the human nose. The electronic nose consists of three parts: an enclosed evaporation space, an array of sensors and a data-processing device.

A temperature profile is applied to the evaporation space and the molecules evaporate. This is where the difficulty in modeling the operation of the human nose begins. To fulfill this function, the device is equipped with sensors. These sensors have been developed from different technologies, each with an olfactory field of predilection: enzyme sensor, conducting polymer, etc. In order to increase the reliability and sensitivity of the apparatus, a very large number of sensors is used. They deliver an electrical signal dependent on the time, concentration and composition of the olfactory message. The signal is then digitized and processed by a data-processing device acting like the brain. Processing and analysis are carried out by comparison with known odor signatures.

Electronic noses are sophisticated detectors creating digital olfactory fingerprints. They are widely used in an increasingly large number of industries for quality control and product development.

 

What about sensory analysis methods?

PB

The last method suitable for comparison of olfactory similarity is the sensory analysis method. Sensory analysis is governed by standards defining how to put this type of analysis into practice. This method consists in placing classified advertisements to recruit a statistically representative panel of “untrained” people, i.e. people with no knowledge of the field of perfumery. They are each given a number of sets of three samples. This technique measures the average degree of confusion in assessing the level of resemblance between compared fragrances. Each assessor generally has to do 20 tests presented in the form of visually indistinguishable smelling strips. Two strips carry the allegedly counterfeit perfume B.

This method is a forced-choice method; the assessor has to indicate which of the three strips seems different. If the perfumes are very different, the assessor will always indicate the right strip. The error rate will therefore be nil. If the two perfumes are very similar, or even identical, there is a risk of confusion. The assessor will only be able to indicate the strip that is different at random. The error rate will be 66%, in other words the assessor has two out of three chances of making a mistake. By measuring the error rate for all panel members and for all sets of smelling strips, it is possible to give an objective, quantitative indication of the risk of confusion. This method is rigorous and is applied in the first instance to the two perfumes (the original and the allegedly counterfeit perfume) and separately to the original perfume and other perfumes belonging to the same olfactory family. These are “reference” perfumes classed in the same category as the original perfume. Perfumes are divided into seven fragrance families: citrus, floral, fougère, chypre, woody, amber and leather. By “fragrance family” we mean a category of perfumes having common characteristics as classified by the French Society of Perfumers. This shows that the perfume is original as compared with other perfumes belonging to same type, and not identical, and this constitutes a reference for the degree of resemblance allowed in perfumery. Typically, error rates on the order of 20% are found for authentic perfumes and on the order of 40, 50 or even 60% for counterfeit perfumes. Thus, perfumes in the same family provide a threshold of acceptability of resemblance. They can also be used to check the panel’s capacity to pick out authentic perfumes belonging to the same family.

To complete this sensory analysis, we also ask an assessor (trained in the use of sensory analysis methods), endowed with excellent sensory acuity, to comment on the two perfumes; a little like a wine expert comments on wines.

These sensory analyses give a perspective which reinforces the other analyses. All these methods finally provide information on which the judge can assess whether this olfactory resemblance is the result of two independent compositions arising by chance from the same inspiration, or whether it goes beyond what is allowed and comes under the heading of counterfeiting.

Each of these methods taken independently very quickly displays its limitations, but when all three produce similar results, this can be taken as indicating some degree of truth. I emphasize the fact that the aim is to provide the judge not with a definite answer but with material facts and figures to help him or her decide whether or not there has been a breach of copyright. The various analyses carried out in the framework of the previously mentioned lawsuit were the work of ISIPCA.

The Institute supplied us with the raw data. Our firm analyzed and interpreted them and put them into a meaningful form that could be used by the judge to assess whether or not there has been an infringement.

 

Gas chromatography coupled with mass spectrometry (GC-MS)

GC-MS is nowadays one of the most widely used techniques in analytical chemistry as it can be used to study complex mixtures present in traces. But what is GC-MS?
It is a combination of a so-called “separative” method (gas chromatography or GC) for separating out the molecules of a simple or complex mixture, and an identification method (mass spectrometry or MS) for characterizing these molecules. Minute quantities can be detected: this technique can identify concentrations of less than one picogram per microliter (pg/µ1 pg = 10-12 g = 0.000 000 000 001 g). The aim of this section is to give a brief explanation of the principles of gas chromatography and mass spectrometry and to show how important the combination of these two techniques has become in analytical chemistry. Nowadays, GC-MS has applications in a wide variety of fields, including the agri-food industry, medicine, pharmacology, the environment and perfumery, the latter being the field in which we are particularly interested here. GC-MS can “decode” an odor by identifying its main chemical constituents.

 

Gas chromatography ?

The operating principle of gas chromatography (GC) is as follows. Using a syringe, a very small volume of a liquid sample (on the order of 1µl = 0.000 001 l) is introduced into an “injection chamber”. The injection chamber is at a sufficiently high temperature (generally 200 to 300°C) to vaporize the liquid immediately. The gaseous molecules are mixed with a flow of helium, referred to as the “mobile phase”, which carries them onto an analytical column, formed from a long (10 to 100m) and very fine tube (inside diameter less than 1mm). The analytical column is the core of the system; the molecules are separated according to each one’s affinity for the internal coating of the column: the stronger a molecule’s affinity for this chemical coating, the more it is slowed down. This affinity depends on the physical-chemical properties specific to each molecule, particularly on the property known as polarity. Some molecules will pass very quickly through the column while others will take a lot longer.

They are thus separated according to the time they take to complete their journey. At the outlet to the column, a detector produces a graph called a chromatogram, in which each peak corresponds to a constituent of the injected mixture. The size of the peak indicates the concentration of the component in the sample. Of the different types of detectors available, it is the mass spectrometer which performs the best as it allows each of the peaks to be identified and enables a precise determination to be made of the concentration of the corresponding molecules in the analyzed sample.

 

Mass spectrometry

The molecules separated by chromatography enter the mass spectrometer. First, the mass spectrometer breaks down the molecules by ionizing them. Two techniques are possible but only one is described here: electron impact. To understand this phenomenon, it is necessary first to understand how the atoms in a molecule are linked to one another. In a molecule, each atom is linked to its neighbor by a chemical bond (sometimes more than one) consisting of two paired (or “married”) electrons. The aim of electron impact is to project unpaired (“bachelor”) electrons onto our molecule to split up the paired electrons: “divorce” ensues…and the chemical bond shatters. Fragmentation does not take place at random; it occurs where the molecule is most fragile. The following stage consists in sorting the pieces (known as “daughter ions” or “fragment ions”) according to size and in making them exit one by one to the detector, which counts them and produces a graph called a “mass spectrum”.

Thus, for each chromatographic peak, there is a corresponding mass spectrum consisting of a succession of strokes; each stroke represents a fragment of the broken molecule. A mass spectrum is a sort of jigsaw puzzle where the fragments must be fitted together to reconstitute the molecule. However, it is very difficult to proceed in this way on the basis of chemical criteria alone, as combinations are often multiple. Hence most spectrometers are equipped with a spectrum library containing the spectra of a large number of molecules. When a library search is possible (i.e. when the analyzed compounds are listed in the library), it has the major advantage of providing an instantaneous result.

 

Principles of copyright
Copyright does not require any initial registration.
The Berne convention governs international protection.
Moral rights (right to claim authorship, right to the integrity of the work).
Economic rights (performance right and right of reproduction).
Duration: 70 years after the death of the author.

 

Is this a legal breakthrough for perfume composers?

PB

The protection of an olfactory creation by copyright is a minor revolution (see section Principles of copyright) with a major impact on the creator’s status. The creator is no longer considered to be a chemistry technician but an author in his or her own right, with all that that implies in terms of protection.

Once a perfume is recognized as an intellectual work, the creator can exercise his or her moral right to the integrity of the work. In other words, no-one can substitute one compound for another. The creator may also invoke the right to claim authorship of the work, i.e. the right to be known as the creator of that perfume, which is not at all usual in perfumery circles. Under economic rights, the creator can also be remunerated proportionately, and a self-employed creator is entitled to receive royalties rather be paid for the provision of a technical service, which makes a considerable difference from the tax point of view. Furthermore, the perfume composer’s status in relation to the perfume houses is radically altered. The creator is no longer a marketing subcontractor but a front-line player. And finally, previously, as there were no legal tools to prevent newly launched perfumes from being too closely copied, there was a rush to produce similar fragrances. If a perfume was seen to be selling well, a company was asked to supply a formulation closely resembling the original, as that was a way of obtaining a perfume that the consumer would be sure to like. Unfortunately, we did not have any means of stopping that kind of behavior. Now, if we apply copyright, the company or perfume house that has launched an original perfume can prevent others from following too closely behind.

When a new perfume is created, there is always the possibility that it may not catch on. Take the case of Angel; when it was created, no-one imagined that it would still today be among the ten best-selling perfumes. When a perfume is successful, its competitive advantage can be maintained for a long time thanks to copyright.

Recognition of copyright rewards the boldest creators and forces the others to invest in creativity. We can finally start to leave behind the cycle where constant copying forced perfume houses to regularly launch new creations as earlier products soon lost any advantage they had had. Moreover, the application of copyright is not necessarily to the taste of all perfume houses and is thus still a topic of debate.

Perfume houses cannot say that they are not favorable to copyright. The application of copyright is automatic, whether you like it or not. Some perfume houses may find this change in the law inconvenient; others on the contrary may welcome it.

 

(*) Even though such techniques have been available since 1999, the L’Oréal vs. Bellure case is the first time that the court of appeal has acknowledged the relevance of applying science to the detection of counterfeit fragrances.

 

By Isabelle Monnerais, Scientific Mediator, Chemistry Department, Palais de la découverte.