@Ufostrahlen: thanks that book has been extremely helpful to me it has some really helpful lab techs that have improved my A/B projects and distillations anyone who is interested in chemistry should get a copy

If anyone is looking for a brand and model for UV-Vis, Jenway 6405 spectrophotometer has been used to successfully by some good people that we all know and love (and this isn't an egotrip so it wasn't Shaolin)

Ufostrahlen. you might find the following Sciencemadness thread helpful:
http://www.sciencemadnes.../viewthread.php?tid=7993

Benzyme

Wouldn't the rf value determined by hplc and the absorption data obtained from uv vis be enough to identify a compound? You stated ftir was the only affordable way to identify compounds in a mixture. I guess i dont really understand why. As in why wont gc followed by some type of detector wont work, or hplc followed by uv vis.

Another question i have is can hplc take in larger samples and separated it into fractions based on rf values. For example could relatively pure dmt crystals be dissolved into a solvent system then hplc could separate the mixture into something like 1 fraction with gramine one with dmt one with hordenine and so on?

Thank you for your time in helping with all of this i greatly appreciate it.

I really advise people to do some research about these instruments. A lot of these questions, not to discourage anyone, make me feel like if the person owned one of these instruments they would have no idea how to even use it. There's only so much spoon-feeding that can be done before someone needs to take the bib off and feed themselves.

No HPLC and UV VIS can not characterize a compound. It can tell someone who 'knows' whats inside of a mixture what fraction, might be what, but more than likely it can tell you absorption at a series of photons' wave-length(that's what a UV-VIS does...). With a UV VIS, this is just an example, it might not necessarily be true, lysergic acid would show up at the same or very similar lambda max as say LSD. Here's an example, toluene lambda max 263nm, ethylbenzene lambda max 262nm. And more than one compound absorbs at these wavelengths. Thus UV-VIS can not tell a person what exactly is in their sample even after fractionation. Hell, if you change the pH of some solutions it can shift the lambda max tremendously even with one analyte.

In the case of similar organic compounds UV VIS is not ideal. FT-IR can characterize because every organic compound has a unique IR spectra. It's that simple. Thus one can deduce that their molecule is, with certainty, __Insert Molecule Name Here__.

Rf values don't really apply to HPLC in the same way they do with TLC(http://teaching.shu.ac.uk/hwb/chemistry/tutorials/chrom/chrom1.htm). Retention factor values are used on TLC plates but differently on columns(http://www.uwplatt.edu/chemep/chem/chemscape/labdocs/catofp/chromato/tlc/tlcq.htm). Yes HPLC can do separations of materials that's what its for. It's not like you press a button on a machine and say "separate DMT gramine and hordenine for me please", it has to do with the solvent gradients which a person must set up and the chosen columns. Can it take in larger samples? Yes. However preparative HPLC columns cost lots of money. Even used ones.


It's good to dream but when you want to turn a dream into reality it takes some leg moving not more dreaming. What I'm saying is research these things, get interested. It's not only about the cash, there's a level of understanding that must be attained as well. If there's no interest then don't waste the money. If there's no money however, its' not a waste of interest.

Len1 from sciencemadness link shaolin posted


This brings up an important point. If something goes wrong with your instrument, or you purchase one that doesn't work, do you have the dedication to fix it? Keep in mind alot of these instruments are really sensitively made and if shipped improperly or turned upside down well, yea, something will need to be fixed. FT-IR's especially.

To simplify things, TLC and or column chromatography(or flash chrom. if you so desire but it has poorer resolution in general) with UV-VIS can tell you probably just as much from a hobbyist perspective as HPLC with UV-VIS would. With careful sample preparation and a sensitive instrument. I really wouldn't expect someone to ever run a HPLC without atleast running a gravity column first.

Thanks Shaolin and InMotion for the info, guess I will lurk around in some chemistry forums.

Better yet, start researching. People on forums don't want to explain to you thousands of pages of chemistry, theory, and calculations. When people research they ask better questions, and the replier thinks of replying as less of a chore.

This recent work describes how to build a Raman spectrometer for $3K. Looks interesting (sorry it is 10x the price in the OP). Paper attached.

Essentially, ilumimate a transparent sample with a green laser pointer, spread the tramsmitted light (rainbow), and take a picture of it.

The raman spectrum for DMT was measured in this paper. They go through and talk about how the molecule vibrates in our stomach (how cool are these guys?). The tables are turned in this paper since usually it is the molecule that makes US vibrate .



Would be cool of one if us posted this kind of spectrum someday (along with our other favorite molecules). Side note: seems like the signal cod be enhanced using colloidal gold/Ag for even moar fun

Here's the IR spectrum for DMT freebase. If you need the IR's for other compounds, I got them all.

Thanks. Very good match between IR absorption and Raman emmision. Raman has a couple extra peaks, but that is supposedly because it is in HCl salt form and there are extra pi-Cl vibrations (according to the paper of I understand it).

Raman seems doable. It's "only" a $15 green laser pointer plus some optics (where some money needs to be spent). However, optics are just passive parts. There are no voltages/magnets/mass injections to worry about like in mass spec. End result is an emmision "finger print". With the right database and software to search for matches we can try to identify different compounds. It may get messy when there are several compounds together, but the right algorithm may be able to break stuff out? I guess it would be like separating overlapping fingerprints.

LCMS, GCMS, and NMR are all crap compared to IR and Raman IMO.

Raman looks identical to IR but it's 5-10 times cheaper.

I looked at the Raman Spectrophotometer documents and it looks easy enough. I can get the optics for pennies, the camera electronics can be bought separately for cheap. The only thing I'm not sure about is how to process the image into a graph!? Can you explain this?

As far as the software, essentialftir.com is more than enough. It can even do mixture searches, I tested it with excedrin and it id'ed all of the components.

The libraries are available as well. If I push it I can get it made professionally for under $300, there won't be much room for profit but I'll be doing this for the masses, as charity. So basically for under $300 you'll have the device, the software, and the libraries, it will be just as good as an FTIR. Put another way, for under $300 you will be able to identify any compound (even mixtures) in the liquid or solid you are testing.

I've been wanting to give back to the masses/world for a long time. Your post has motivated me! This device will certainly change the world and take us to the next level.

MS is the gold standard of detection, and unlikely to be dethroned anytime soon, with unparalleled sensitivity and versatility.
Like with any other tech, miniaturization is the trend, and DESI and DART eliminate the need for extensive sample prep. Emerging technologies being integrated into mass detection are photodissociation ionization, and circular dichroism.
NIST and Wiley have over 600,000 entries in the MS databases. Mass detection is used extensively with cancer research and metabolomics, and surgeons will soon be using laser pens interfaced to MS to characterize and diagnose...

Mine is a 17-year old instrument, but it’s still sensitive enough to register dust particles in the air as keratin peaks. It was $260K when new, I was fortunate enough to get it complete, 3 years ago, for $5K. It does fine for small molecules, peptides, and proteins, as well as reaction monitoring.

I know what you’re thinking, Loveall.
and Raman makes more sense for applications with small molecules, particularly in forensics. definitely.

its relatively compact size lends itself well to field work. all techniques have their advantages and disadvantages. smaller instruments sacrifice range and sensitivity for portability, but they also tend to be application-specific. Raman is definitely hot, just about every Spectroscopy issue I get features it.

For sure benzyme, MS is great and the golden standard.

What I was trying to say is that for me (a kitchen worker ) MS seems out of reach, while on the other hand, Raman looks feasable thanks to the mass consumer optics market.

indeed.

I’ve actually been considering raman lately, as a secondary ID instrument. they’re not cheap, but they’re also not very involved.
hplc-ms is a lot of troubleshooting, as you might imagine.

How can you get such cheap optics? Can I get that price too?

The way I understand it we need high quality optics since the Raman scattered light is dim. The author mentions that he tried to go on the cheap with a webcam as detector and a CD to disperse light, and while he got something, the spectrums where noisy and not of good quality. However, it is interesting (and illustrates the accessibility of this) that with a $15 laser, a $50 webcam, and an old CD it is possible to get some Raman spectra lines.

To process the image into a graph the paper said they save the camera image in RAW format, convert it to 8-bit PNG using "Darktable" software, and then use "ImageJ" software to convert the PNG to a spectra vs pixel position. Then to turn the pixel position into cm-1 they calibrate with a standard home made solution of benzine, carbon tetrachloride, and naphtalene which has known Raman peaks. They also take a without a compound to do a background substraction. After that we can feed the data to essentialftir.com as you suggested.

I think the most difficult part of this project is to align all the optics and have no stray light in a well designed/built home-made plastic casing. In the paper they used a 3D printer, I'd be using drills, plastic cutters and glue . To aid in alignment when putting the parts together, an olive oil target is used in the paper since it creates a red fluorescent beam. Once everything is aligned and we get a good signal in the camera, the software and signal processing stuff seems straightforward.

I ended up getting a Thermo Nicolet IR100 for $2000. It's a mid-IR spec, fourier transform; comes with an ATR accessory.
http://www.selectscience...eter/?prodID=9508#tab-2

as far as i know, a japan made sensor alone is around $500, in bulk, 50 pieces at least, meaning a group buy. but perhaps a home made spectrometer could be assembled for around $700,
for a reasonably working device, those things made from cds that go around for $20/50/100 are for entertainment purposes only.

Good sensitivity linear CCD sensors can be bought for a few dollars. Attach the sensor to a small board (also cheap and powerful) and you got a good detector below the $50 mark. People have been working on this and have published instructions and code (example).

One does need a couple focusing mirrors and a diffraction grid to complete the spectrometer. That is where some $ need to be spent, but in the lens surplus market you can get these pretty cheap. At the end of the day, I think you can get a decent spectrometer for $300.

After looking at this a lot, I think the best thing to do is to go to the used spectrometer surplus market, where a 1nm resolution capable spectrometer can be found for $211 (including shipping). This is from a link Pitubo graciously posted in the chat.

Lasers are very cheap. To complete Raman setup a couple filters are needed to clean up the light/signal.

I think this is very accessible for the amateur.

Currently I have a spectrometer, a laser, and a rapid edge filter in the mail. Dreaming of one day getting a spectrum of our favorite molecules.

as noted in another thread, I am following this project, and purchasing the same components (laser and spectrometer).

after a week of experimenting with the FTIR-ATR, I can see the appeal of the method. Whenever I get new instrumentation, I always try to determine the limitations. That being said, there is no "best" instrument for characterization. In science/method dev, there are only advantages/disadvantages, with respect to application.

The clear advantage to the methods being considered here are lower cost and less need for sample prep. Also, analysis time is rather short. The disadvantage is the lack of sensitivity/signal strength (this is where mass spec reigns supreme, albeit at a much higher cost and complexity). Nevertheless, the compact form factor and mobility of Raman lends itself well to field work, where compact gas chromatographs ($$$) have traditionally been employed. The tuneability for application-specific purposes is also attractive for this method. With proper tuning/alignment, it appears one could gather some useful structural information with decent resolution. I'll be following/contributing with my findings.

great stuff, btw. This is a far cry from the camera phone/dvd hacks from a few years ago, which is why I am spending money on this. This is an actual useful DIY solution, approaching comparable to high-end name brand products from oligopolies (Thermo, Agilent, Perkin-Elmer, Biorad, etc. ) that charge a kidney for instrumentation.

:
this thread.