Does a gaschromatograph = possible rebirth of a lost legend ?

Well, my knowledge in chemistry is limited, but a chromatograph doesn’t tell you about the ingredients but the molecules in the liquid.

Exactly. The gas chromatograph -- or liquid chromatograph -- can tell you what molecules are in the finished product. But, these tools do not show you how to create a finished product with that exact combination of those ratios of those molecules.

If I leave Portland on an east-bound flight, I can get to New York. If I leave Portland on a west-bound flight, I can also get to New York; it's a longer route, less efficent, but I end up equally in New York either way. So, once the ratios and molecules are known, the fragrance can be re-created. The direct approach of just measuring out those molecules may be a longer, less efficient route.

In reality, chromatographic analysis can tell you the destination you want to get to, fill in the New York of the puzzle. With that, the challenge remains of finding the best way to get to that destination.

Yes, but the hindrance to the equation is an unfortunate period on the fragrant timeline:

the IFRA years.

Maybe it would be more realistic, in case of Tabac Blond e.g., to take the current extrait and "play around" with castoreum and whatever is available and opportune to achieve the desired effect. Not a scientific approach, I know...

A good and reasoned question, but I suspect the answer is more complicated than one would expect at first blush. I'm not a fragrance chemist, but I am a chemist who has operated GCs for many years. GCs and LCs can tell you the chemical composition and relative amounts of fragrance molecules, but cannot tell you their origin. For example, eugenol is a common fragrance molecule that is present in many (if not the majority) of fragrances. It is available to the perfumer in both pure synthetic form and indirectly as a component of natural extracts. So when a GC reports the presence of eugenol, it can't tell you whether it was added as a single component or as a component in, say, clove extract. Pattern recognition might provide a clue, but then you get into the problem of deconvoluting multiple patterns, many of which may contain common ingredients. In short, GC and LC are good starting points for reverse engineering a fragrance, but It's only just that - a start.
Perhaps one of the fragrance industry pros on Basenotes can provide more insight on how fragrances are reverse engineered.

This is why I love BN.

I'll throw in my two cents as a chemist and basically agree with Snafoo's explanation. As Gollnick suggested, even knowing ratios of a group of the major, identifiable components would just get you to a rough recipe, at first just trying to reproduce the mixture at that level of detail. Then you'd have to play around with the ratios and think about what minor components may have been in there that you'd like to add, based on what plant extracts you think were actually used.

I have a doctorate in Chemistry, major in Biochemistry, and my doctoral thesis was in Mass Spectrometry and worked all my life in the chemical analysis area so I can tell you with knowledge of the subject that not only you can tell the components, but also de origins of most complex mixtures by the technique known as chemical fingerprinting. Using chemical fingerprinting, analytical chemists can tell you, for instance, if an oil tarball found in the Gulf of Mexico came from the BP blown oil well, or from other spill etc and can tell you if an attar came from India, or China, etc. The problem is that components in a perfume have different volatility, so the ratios may change with time, therefore a 40 years old perfume could be replicated identical as how smells today, not how it smelled originally.
Anyhow, it seems like a great idea to try to reproduce the old classics, I have the know how, so if anybody is interested and willing to look for investors, I would be willing to start an enterprise with the purpose of bringing back those espectacular fragrances of yesteryears.

ok - let's play .... what would your ballpark cost estimate be just to analyze and list the componenets and their origin for a fragrance ?

Exciting...

First, you have to have the instrumentation, but if I go to a laboratory for such a test, the bill would range between 500 to 1000 bucks.

The problem is that you don't have a single fingerprint, you have multiple fingerprints superimposed on each other, at least with regard to perfume ingredients derived from natural products. A more accurate example would be a tar ball from the Gulf that drifts around the world, accreting tar balls from the North Sea, the Gulf of Oman, and half a dozen other locations. The resulting tar ball will share a number of common constituents with each contributing tar ball, so deconvoluting the combined chromatogram might prove difficult. Back to fragrance, consider a perfume that is composed of, say, 15 natural products, and each natural product is itself composed of 50 individual chemical compounds - many of which are common to more than one natural product. I have to think that the resulting gas chromatogram would be formidable to reverse engineer - not in terms of identifying individual compounds (that would be straightforward), but rather in terms of identifying the sources of the compounds. You might be able to take advantage of the uniqueness of some compounds to a specific natural product, but I think it would still be a tough road. Identifying the individual components in a fragrance would be very straightforward, divining their source would be a challenge. Synthetic aromachemicals, or at least those that have no natural source, don't apply to this argument; it would be trivial to reverse engineer a fragrance made entirely of synthetic compounds.

I agree that age throws another variable into the mix that would probably be at least as challenging as determining the source of the constituents.

Good luck with that. I estimate the bill for a full reverse engineering of a fragrance would be at least 10 times that amount, if not more, and will depend largely on the analyst's expertise in the field of interest. $500-$1000 will get you a printout of the chemicals in a fragrance, but not much more.

Snafoo: Nobody said it would be easy, if it was easy to do, somebody would have done it long long time ago, but being difficult does not mean it is impossible. And I do have the experience in fingerprinting using gas chromatography/mass spectrometry. It seems that you are refering to simple gas chromatography and you are right in that respect, GC by itself will not cut the cake, but when you add mass spectrometry, and triple quadropole mass spectrometry, then things are different. By the way, tell me about your fingerprinting experience, I am profesionally very interested in hearing your experience in the field.

That is your estimate, OK, it is your estimate and I respect that. But, I have done fingerprinting and charged $750 for the full test not only giving the list of the components, but also including the interpretation, so mine is not an estimate, it is the actual price I charged for doing the test.

If you can reverse engineer Patou Pour Homme for $750, I commend you, sir.

Principal Chemist, 34 yrs experience in environmental and forensic chemistry, 5 yrs Mass Spectrometry Facility Manager, including GC/MS, LC/MS/MS, HRMS. CV on request. I'm not questioning your credentials, PC, I'm arguing the challenges in recreating a fragrance solely from chromatographic data. And yes, my earlier comments assumed MS as the detection method.

That is all that a GC -- or LC or NMRS -- can give you for any price.

The real challenge is in the interpretation -- and application -- of those data.

I did not say I can recreate the fragance for $750, I think that it was clear. I was asked by SoGent "what would your ballpark cost estimate be just to analyze and list the components and their origin for a fragrance ?" Note the words "just to analyze", not to recreate, that is a completely different ball game.
My experience:
Doctorate in Chemistry majoring in Biochemistry from the American University in Washington DC
Doctoral thesis at the National Institute of Science and Technology (NIST).
Manager of GC/MS lab at National Institutes of Health (NIH)
Research chemist at NIST
VicePresident of Research and Development at Biochemical Marketing Corporation
Technical Director of Coast to Coast Analytical
Technical Director of BC Laboratories
Laboratory Director of Halcyon Laboratory.
Laboratory Director of Central Environmental Laboratories in Kuwait working in the assesment of the environmental impact of the war.
Laboratory Director of Coachella Valley Water District Water Quality Laboratory.
Plus a few more not worth to mention

and there is where my experience counts INTERPRETATION!!!

I humbly stand corrected - a simple GC/MS analysis and printout of results should cost no more than $1K. As Gollnick stated, the real cost lies in the interpretation of those results. However, I think we can all agree that a list of chemical components would not be sufficient to recreate a fragrance, at least in a practical sense.


This is pointless. What began as a thought exercise is devolving to a CV shout down. Simple answer to the OP's question: chromatography (coupled with whatever detection method you choose) is insufficient to recreate a fragrance. How closely a recreated fragrance approximates the original will depend as much or more on the skill of the perfumer than of the analytical chemist.

Nice CV, by the way.

Nobody questions that, I was just answering the price of the cost of the analysis.

i have a question tho: how did some of the semi-fake perfume industry like onedrop perfumes and another oil-based perfume manufactorer who is able to reproduce popular scents but in oil form? did they manage to have access to formulas of Shalimar, CK etc?

Don't worry about the professional squabbling, David. Put two scientists in a room together and there's bound to be disagreement. That's the nature of science.
Despite my ramblings about how challenging it would be to recreate Patou Pour Homme, yes, it's theoretically doable - to a point. I suppose it would depend on how closely one wants to approximate the original. As I said earlier, fragrances based primarily on synthetics should be relatively easy to copy, and I suspect those in the business of making copies of popular fragrances use the methods we're discussing here. I also suspect that the devil is in the details, in other words, how to capture the nuances that make a fragrance such as PPH a masterpiece. For example, a perfumer may have a particular supply of Mysore sandalwood oil at his/her disposal which an oil from another supplier may not precisely match. Would a substitute be "good enough"? Possibly - and this is where my limited knowledge on the subject fails me. Perhaps someone from the DIY forum could provide more insight.
As for getting 1000 BNers to pony up $ for a reverse engineering project - half the time we can't seem to be able to decide what to wear from the selection we already have. Not sure how you'd manage to get 1000 guys to agree on a single fragrance.

Later: I Googled "reverse engineering fragrances" and found several sources for more information on this topic. Wikipedia has a brief section under "Perfume" devoted to reverse engineering that corroborates pretty much everything we've said here (only more eloquently). There are also numerous companies that offer the service. One such is Aroma Alternatives in Austin TX., which quotes a price of $95/hour, with the following disclaimer: "Depending on the level of difficulty of the product hours and time frame can vary widely and until the development of reverse engineering toward said product starts there will be no idea how long or how many hours it will take to develop said product."

Perhaps hiring a great nose or two and if possible sniff the original, experiment and ultimately re-create might be even better?

lol this thread is becoming a place to show off :P Keep it simple and straight to the point guys.

Fascinating chat--I assume that copying a fragrance in this way is legal ?? --after all this not an interpretation but a copy.Like taking an iphone to bits, sourcing your own parts ,making it yourself and calling it a Nexus (joke) !!

Interesting article in Fashion and Apparel Law Blog that provides an overview on the legal status of copycat fragrances:

http://www.fashionapparellawblog.com...common-scents/

GC would be a good starting point to copy a fragrance, but it would just be a start.

The IP issues of perfume formulas is interesting stuff. The 2006 Dutch ruling aside, formulas are protected by trade secrets law, the weaker cousin of patents and copyrights. These 'protections' let you sue an employee for disclosing confidential information but that's about it. If you can reverse engineer a trade secret it is yours free and clear.
Givaudan and the other big frag companies do this sometimes for older fragrances (There's some article floating around about how the did the Robert Piguet reconstructions) but it required a professional perfumer to interpret the results and they ended up replacing a number of the old ingredients. I've found a few amateur copys of formulas floating around but I've no idea how good they are.
PS: Did you ever try out that Tabac Blond recipe you posted? Find any other interesting things?

If you have a problem...and no one else can help...and if you can find them...maybe you can hire the A(lexandria Egypt Perfume store)-TEAM!

that will never work.

that's right. you could also add elimination.

but even when you would find out, its rose oil.. is it rose damascena from bulgaria or something from morocco with some ingredients added? how are you going to find the exact oil? probably not. there is a similar problem with chemicals. these molecules might not smell the same as they did decades ago. and a molecule from givaudan might smell different from the same material from iff.

however, you can get a rough sketch using gq, and some useful ideas on how it was done, if you know a good bit about perfume making. then you can get the overall shape more or less right, but not the exact behavior and details. you will need an good and experienced perfumer (and a bit of cash) to get kind of close to the original. it will likely take many trails (and errors).

interesting.

i am no expert in chemical analysis, but i heard about these fingerprints. you know, perfumery ingredients is a rather shady business. these fingerprints, they get faked. they alter a cheap rose oil, until they have the fingerprint of something much, much more expensive. it makes everything very complicated for the buyers.

i would love to hear your comments on this.

there are several suppliers that offer this service, i know http://www.thegoodscentscompany.com/index.html does it (they have a great site btw which list almost every material that is being on the market with their properties) and www.perfumersworld.com does it as well. i believe they have a fixed price of $100, but you probably just get a sheet with the constituents and their percentages.

Guys, let`s cut it out with quarrels on matters of size: David, as per answers to your question, it is clear chemists and trained noses can reconstruct classics, but... what about production costs? And what about sales: will customers will be eager to pay for reconstructed blends?

I also love the idea, but to the extent of my knowledge, the are accords in classic scents that many perceive as dated. I had the chance of testing an extensive amount of historical blends and many describe most of them as "old", "dated", "old-lady/man", etc. If I have to describe which notes are the ones to blame, I'd say iris and / or "talcum-powder-like" notes. Even scents available nowadays which are extensively liked by many BNoters are described as "passé", specially those dating from the 1970's / 1980's. To my surprise, synthetic notes common in present-day scents are among accords described as "modern". So the answer would be, "yes, but what for?"

Simple; I am in love with the idea of launching a line of non-IFRA compliant classic style scents for customers disregarding mainstream fragrances, besides the fact that a bunch of classic fragrances are described as wonderful - it happened to a 11 year old daughter of mine who had the chance of trying a chypre from the 1960s made by a local company and found it amazingly good, the best she had ever smelt so far. So I guess there is a business opportunity around.


Thanks, data for my businessplan!

I can also tell you exactly what will happen if you do duplicate one of these lost classics, if you nail it right on the head, 100% right: the critics -- including those here -- will all say, "It's just not the same," "It's just not as good."

I think it work and be successful as a niche', small-scale thing.

If it costs $50,000 to reverse-engineer a fragrance (and you start off with a fairly simple one and hope to finish up for well under 50K$), and if you can sell your product for $100/bottle (which is the ballpark for a niche' fragrance) then you'd need to sell 500 bottles to recoupe the research cost. Can you sell a few thousand bottles? I'd hope so. So, yes, I think it could work. I doubt that it's a get-rich-quick plan, but it could be sustainable.

Your next big problem will be getting authoratative, properly-preserved samples of the old fragrances to start from.

That IS a problem; some of the blends I 've got are perfectly well, others havee been altered even though they were stored in the right conditions. To this, I might add that some aromachemicals and / or EO used in the past must have been discontinued.

However, it does get my attention that local traditional manufacturers houses are launching scents for designers and / or private lables with strong classical character, so I guess there must be an "editorial line" they are following: they are not pressed for these houses do not belong to big mutinationals so they are free to do as they will. That is where my hope lies in... I think that if asked to re do classics I will get a "YES" right from the start. The question being, is gascromatography an option for them?

Will post about my endeavours...

Don't forget about taxes and all the necessary paperwork and fees for doing legal commercial business. Pretty much depends on the country you want to operate from, of course.

I'd love to see an actual production formula for a commercial fragrance!


At least you could say with conviction that it's a reformulation.


The beginning of the end of the dream...


You're right, of course, and I regret getting sidetracked earlier. It's a topic I find deeply interesting and I can get a bit carried away. Much more fun for me than debating what [insert name here] would wear.