Code of Conduct
Results 1 to 12 of 12
  1. #1

    Default Constructing a note using the terp profile of a natural sample?

    Anyone care to discuss this? I have set out to replicate natural notes using the constituent chemicals of their sources in nature. While a GC/MS of natural material is a good and logical starting point, I feel that those of you with a bit more experience might have something to say about the finer points of replicating it in the nose rather than the spectrometer.

    I have this sneaking suspicion that there may be some more refinement than simply duplicating a terpene profile. Has anyone ever published or discussed much on this? It seems like the differences in detection thresholds and partial pressures (read: volatility) may be somewhat different in nature than in the corresponding mix of aroma chemicals in liquid form. Throwing butter, flour, water, and eggs in a pan isn't just going to taste like pancakes. Oversimplified and not-quite-applicable, but you get my point.

    I am hoping to start a discussion here, as I may not even know the right sorts of questions to ask at this point.

    Cheers!

  2. #2

    Default Re: Constructing a note using the terp profile of a natural sample?

    Good thought. I explore some of this in post on my blog about what makes the scent of a flower.

    In my mind there are four key reasons why you can't just replicate a flower fragrance from a headspace analysis:
    1) each type of capture mechanism yields different results, plus time of day, age of flower etc all influence what you capture - so you have also to make a judgement about what you want to attempt to match
    2) there may be stability considerations: not everything in a flower scent is stable in a liquid perfume
    3) there are likely to be both safety and regulatory reasons why you can't replicate it anyway: not everything that is safe to smell is safe to put on your skin and even if it's safe you may not be compliant with regulations.
    4) flowers generally manufacture more scent on a constant basis, so the scent may be all or nearly all middle and top notes - to make it work in a liquid fragrance you're going to need fixatives and base notes.
    Chris Bartlett
    Perfumes from the edge . . .

    www.perfumedesigner.co.uk
    Twitter: @PellWallPerfume

    If you are looking for a perfumery consultation Iím happy to quote: if you want free advice, thatís what these forums are for
    You can also join my blog if you wish to ask questions of me.

  3. #3

    Default Re: Constructing a note using the terp profile of a natural sample?

    I have to ask why do you want to do this? Do you just want to replicate the aroma produced by a living flower,do you want to match an extracted essential oil, or do you want to create fragrance bases based on the natural material. If you want to duplicate the fragrance of a living flower using head space technology you will need the living flower as a reference. The GC analysis will not be the same as the head space. Some materials will be captured preferentially resulting in a distorted result. After making up your copy based on GC you will need to rebalance the formula to get it closer to the head space smell. If you simply want to analyse an Essential Oil and copy it, it depends how good your GC/MS set up is, how sensitive and how accurate. Some chemicals are present in Oils at such a low amount that it is impossible to detect them by GC/MS; what you need in addition to your basic equipment is a smelling or sniffer attachment. This enables you to smell each separated chemical as it passes through the GC. As you see the peaks rise and fall you can write the name of the chemical against it. What is most interesting is when there is no peak but there is a smell. This assumes you have a certain knowledge of what things smell like and a very good knowledge of retention times.

    If you are only interested in matching for the fun of it then safety regs. don't come into it. However if you want to use your matches in fragrances then everything Chris has mentioned applies. Stability of constituent, and safety of chemical must be considered. A head space based match will be very toppy (as Chris said mainly top and middle notes) and will smell quite strange. Very strong. It will smell closer to the original flower when heavily diluted.

    It is difficult to get a single chemical (synthetic) that smells quite like the same chemical in your flower mix. Synthetics often contain small amounts of contaminants which distort the smell. The same chemical smelled through a sniffer GC or in the living flower smells different. It is really difficult to use the amount of Indole found in Jasmin, for example, when matching Jasmin. If you do, the Indole overwhelms everything else. Again, a good sense of smell is required.

    Finally, not all the constituents of Essential Oils (and Head space) are Terpenes. Also, sometimes the necessary chemical is not available for use in matching, and then you need to find something that smells and behaves in the same way.

    It ain't easy! Good luck.

  4. #4

    Default Re: Constructing a note using the terp profile of a natural sample?

    Ah! Let me sort out a couple points.

    -My first method of attack is going to be to run a sample through HPLC (via an outside lab that tests these things and is calibrated for this specific material), targeting a list of identified compounds (both presumably responsible for the fragrance, and available in pure form as fragrances) and quantifying their presence by weight (mg per g).
    -This note is interestingly near-identical in the source material, concretes, and absolutes in the nose - insofar as one can identify the specific sample positively in all 3 forms and the characteristic fragrance is present in nearly identical ways. I appreciate that I may not understand the finer points and am surprised this is the case, but I think this is proof-positive that the terpenes are likely to be helpful to my cause.


    I guess I might want to be more specific here - we're trying to reproduce a cannabis note. Maybe note is the wrong word and what is in mind is actually an accord? So far as I know, the only options currently available are either something that has the general funk of that note, or the EO which doesn't replicate the scent one experiences from the real thing.

    I must head off, but looks like we're going to have a good thread!


    P.S.

    A quick list of some of the things I've got planned to work with as far as aroma chemicals (there may be more but this is off the top of my head) - myrcene, dihydro myrcenol, sabinene hydrate, linalool, ethyl linalool, d-limonene, guaiacol, geraniol, nerolidol, menthol, beta caryophyllene, caryophyllene oxide, delta-3 carene, terpineol alpha.

  5. #5

    Default Re: Constructing a note using the terp profile of a natural sample?

    I'm confused by the way you use the word "fragrance". To me, a fragrance is a mixture of chemicals that you can smell (this is the crudest definition I know), it does not mean a single chemical.

    "Note" is fine, and can mean accord, base, or oil; or indeed a single chemical. A nots is a smell found in a larger smell (e'g an Orange note found in a Fragrance).

    You are trying to duplicate the smell of Cannabis. So are you planning on analysing the head space, or the smoke, or what?

    Looking at your list of chemicals: don't see many being of use to you in this project but I've never studied what makes Cannabis smell the way it does. And DHM isn't natural, doesn't occur naturally. So if you've found it, it's been added.

  6. #6

    Default Re: Constructing a note using the terp profile of a natural sample?

    Quote Originally Posted by David Ruskin View Post
    I'm confused by the way you use the word "fragrance". To me, a fragrance is a mixture of chemicals that you can smell (this is the crudest definition I know), it does not mean a single chemical.

    "Note" is fine, and can mean accord, base, or oil; or indeed a single chemical. A nots is a smell found in a larger smell (e'g an Orange note found in a Fragrance).

    You are trying to duplicate the smell of Cannabis. So are you planning on analysing the head space, or the smoke, or what?

    Looking at your list of chemicals: don't see many being of use to you in this project but I've never studied what makes Cannabis smell the way it does. And DHM isn't natural, doesn't occur naturally. So if you've found it, it's been added.
    my use of fragrance may be incorrect, but in many words: there are subtleties that can allow experienced noses to pick up different varieties or at least families; in other words akin to a naval orange smelling different than a valencia. I do not know of these subtleties being duplicated in cannabis notes. It's my feeling that cannabis samples are all pulling from similar genomes, and so the differences must be due to the varying ratios of key components. If I am able to replicate one sample in a way that is specific, I will have achieved a smell that is more specific and more refined. I find this extremely interesting!

    as far as my list - the majority are naturally present, and a couple similar molecules are on the list in the hopes that they will be 'close' to the things they are replacing (some natural elements are not available from aroma chemical suppliers). For example, caryophyllene is what hashish detection dogs are trained on (based on my reading at least), and one of the characteristic citrus notes people pick up from some samples is correlated with a high amount of limonene.

    My initial analysis will be running both a died and 'cured' bud, and a concrete produced from the same sample batch (cannabis concretes generally provide a near-direct duplication of the smell of their source material when properly produced, but unfortunately also contain psychoactive components and thusly are not suitable for perfumery). If I find another (better?) way to go about this process, I will explore that. I will then do my best to duplicate the rtios of those components as they are present according to the HPLC sults, and then attempt to work on modifying the ratios of components as necessary to accurately duplicate the smell of the original samples.

    Just in case anyone is concerned, all the samples and lab analysis are being coordinated and handled by medicinal cannabis professionals according to their local laws and regulations, not me. I am just fascinated with what I feel is a very complex smell that has not been explored by this industry in any real meaningful way.

    - - - Updated - - -

    on a related note, if anyone has ideas on sourcing for some of the stuff i haven't been able to find already, here's a couple of ones i am still looking for:

    valencene (CAS 4630-07-3
    germacrene (CAS 28387-44-2)
    farnesene (CAS 502-61-4)
    a-bisabolol aka levomenol (CAS 23089-26-1)
    fenchol (CAS 512-13-0)
    Last edited by science; 4th October 2012 at 04:15 AM.

  7. #7

    Default Re: Constructing a note using the terp profile of a natural sample?

    OK, you have made things a bit more clear. I think you may have misunderstood me as well, when I mentioned the use of a sniffer GC. There are indeed "subtleties" between different types of the same plant (your example of Valencia and Navel Oranges) but I didn't mean that. What I meant was that simply using a GC/MS set up is not enough to provide a good match of anything. There are usually components at level too low for the GC to pick up which can be vital in producing a good match. Thus you need a sniffer, and an ability to recognise what those components are.

    Your difficulty in obtaining certain chemicals is also a problem, as I mentioned before. You may have to find out which oil contains the highest concentration of of the chemical you are interested in, and try that. Your example of Caryophyllene. I know that is obtainable but it is found at very high levels in Clove Leaf oil, and Clove Terpenes. The materials you mention will be difficult to obtain.

    If you don't mind using synthetics, have a look at Vetikol Acetate.

  8. #8

    Default Re: Constructing a note using the terp profile of a natural sample?

    Quote Originally Posted by David Ruskin View Post
    OK, you have made things a bit more clear. I think you may have misunderstood me as well, when I mentioned the use of a sniffer GC. There are indeed "subtleties" between different types of the same plant (your example of Valencia and Navel Oranges) but I didn't mean that. What I meant was that simply using a GC/MS set up is not enough to provide a good match of anything. There are usually components at level too low for the GC to pick up which can be vital in producing a good match. Thus you need a sniffer, and an ability to recognise what those components are.

    Your difficulty in obtaining certain chemicals is also a problem, as I mentioned before. You may have to find out which oil contains the highest concentration of of the chemical you are interested in, and try that. Your example of Caryophyllene. I know that is obtainable but it is found at very high levels in Clove Leaf oil, and Clove Terpenes. The materials you mention will be difficult to obtain.

    If you don't mind using synthetics, have a look at Vetikol Acetate.
    I've seen VA mentioned before as being useful, I'll have to look into it although it is second tier at the moment as it is not naturally found in cannabis as far as I know. Synthetics I'm OK with, but I would prefer to be able to accomplish this using synthetic molecules that are present in natural samples. The most major (by weight) components are for the most part already in my hands, it is really just a handful of items that I haven't found yet which I believe is likely due to them not being used much in perfumery. I can get reagents using CAS numbers but that is of course not as simple or practical as getting components from aroma chemical suppliers. I agree with finding the unobtainables via EOs of other things though if they prove to be necessary, a good example being bisabolol which I haven't found but is a primary component of chamomile EO. I just also realize I could get into a game of whack a mole if I start trying to balance out the tertiary components of EOs.

    I definitely am not expecting the HPLC / GC / MS to be a shake-and-bake (haha!) formula, but my hope is that it will be a good starting point. From there my ideal situation would be that it is only a matter of careful fiddling in order to translate things out to an end result.

  9. #9

    Default Re: Constructing a note using the terp profile of a natural sample?

    I've commented on the use of cannabis as a scent note in January here and I'll just repeat what I said then:

    There are some smelling notes in the Single Note thread.

    In addition we had a discussion about it a while back in another thread, but for convenience I’ll repeat here what I said there:

    There is an essential oil of cannabis sativa flowers that is widely available and is produced mainly for the perfumery market. Depending on the cannabis variety it can be either virtually free of THC or high in it - the latter would not be legal in the UK but the former meets the Home Office rules and can be sold freely.

    I understand however that the essential oil does contain the aromatic substance (caryophyllene oxide) used to train dogs to sniff out illegal cannabis . . .

    Hemp oil is different - that's made from the seeds and is used in cooking among other things - it has no appreciable aroma.

    Did you read my blog post (and the comments on it) about re-creating the hyacinth scent? If not I suggest you have a look because one of the key things it reveals is that majority components by weight are absolutely not the majority of the detected scent. Most of the chemicals you list would be found in a headspace analysis of almost any natural material - they won't give you the distinctive scent of cannabis that you are looking for. Though the oil I've mentioned above probably would - not as much fun but a whole lot simpler.
    Last edited by Chris Bartlett; 4th October 2012 at 10:38 AM. Reason: added quote code for clarity
    Chris Bartlett
    Perfumes from the edge . . .

    www.perfumedesigner.co.uk
    Twitter: @PellWallPerfume

    If you are looking for a perfumery consultation Iím happy to quote: if you want free advice, thatís what these forums are for
    You can also join my blog if you wish to ask questions of me.

  10. #10

    Default Re: Constructing a note using the terp profile of a natural sample?

    Quote Originally Posted by David Ruskin View Post
    OK, you have made things a bit more clear. I think you may have misunderstood me as well, when I mentioned the use of a sniffer GC. There are indeed "subtleties" between different types of the same plant (your example of Valencia and Navel Oranges) but I didn't mean that. What I meant was that simply using a GC/MS set up is not enough to provide a good match of anything. There are usually components at level too low for the GC to pick up which can be vital in producing a good match. Thus you need a sniffer, and an ability to recognise what those components are.

    Your difficulty in obtaining certain chemicals is also a problem, as I mentioned before. You may have to find out which oil contains the highest concentration of of the chemical you are interested in, and try that. Your example of Caryophyllene. I know that is obtainable but it is found at very high levels in Clove Leaf oil, and Clove Terpenes. The materials you mention will be difficult to obtain.

    If you don't mind using synthetics, have a look at Vetikol Acetate.
    David, mind that "science" is going to use HPLC/MS, not GC/MS. This means better separation of components and probably lower rilevability limits (i.e. it detects substances in lower concentration). Furthermore this detection limit depends a lot on the technical specs of the specific instrument you are using, there are highly accurate GC that can be better than a "cheap" HPLC and the MS part can vary dramatically as well. What I mean is: it's correct that it's probably impossible to detect a complete smell profile, but there are millions of different levels of accuracy.
    Sebastiano - Organic Chemist

  11. #11

    Default Re: Constructing a note using the terp profile of a natural sample?

    I think it would be hard in case you want to copy this fragrance in a perfume, in case you would use a kind of device that would release fragrances like a living flower it might work.

    The fragrance chemicals mentioned: most are available through any good assorted lab chemical supplier, like Sigma Aldrich. Most are not commonly used as aromachemicals, so it might be harder to source a perfume grade.

  12. #12

    Default Re: Constructing a note using the terp profile of a natural sample?

    My comments still stand.

Similar Threads

  1. Replies: 39
    Last Post: 31st March 2014, 05:53 AM
  2. Replies: 0
    Last Post: 21st May 2012, 11:55 AM
  3. Profile Bug
    By Grant in forum Announcements Archive
    Replies: 2
    Last Post: 28th October 2010, 11:00 PM
  4. Single Note Perfume/Aroma Sample Sets & Collections
    By warlandsboy in forum Just Starting Out
    Replies: 2
    Last Post: 3rd September 2009, 04:37 PM
  5. Base note, middle note, high note etc etc
    By lejaimlefunk in forum Just Starting Out
    Replies: 5
    Last Post: 3rd January 2007, 04:49 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •