Sciencemadness Discussion Board

desktop NMR?

spirocycle - 23-2-2011 at 19:41

http://www.picospin.com/products/
Is this NMR machine good?

Im sorry for such a simple question, but I have no idea how the specs listed fare in terms of quality.


Specifications
Larmor frequency 45 MHz
Nucleus proton
Sample form liquid
Resolution better than 100 ppb
Signal-to-noise ratio 300 for water, single shot
Magnet type permanent
Capillary 300 micron ID
Weight 7 lbs.
Dimensions 8"W x 5"H x 11"D
Power 115/230 VAC, 50/60Hz, 150 W

[Edited on 24-2-2011 by spirocycle]

bfesser - 23-2-2011 at 20:53

Wish I had $20,000...

Dotty - 13-3-2011 at 02:46

What do you want to use it for?

I don't know anything about the software, it would be nice to know what pulse sequences are available or if you could write your own. The frequency spread of peaks wouldn't be great, so if you want to use it for qualitative analysis, molecules with a many unique sites might give you a bloody mess of overlapping peaks.

arsphenamine - 13-3-2011 at 08:49

Picospin lists possible applications and gives spectra. Their essential oils bit caught my eye:
http://www.picospin.com/applications/process-control/essenti...

It is adequate for some quality assurance in an industrial setting and comparisons against costly 200 or 300MHz 1H NMR are unfair.

45MHz 1h-nmr of wintergreen oil

300MHz 1H-NMR of wintergreen oil


[Edited on 13-3-2011 by arsphenamine]

Nicodem - 13-3-2011 at 11:36

The peaks are absolutely terribly wide and couplings horrible!

But speaking relatively, the peaks are absolutely fine and coupling gorgeous - after all, it is a freaking 45 MHz machine we are talking about!
I would love to have one such machine at home, but at the job I would hate it.

A major drawback I noticed about these weak field machines is that they appear to require very concentrated samples. The methyl salicylate spectra above was taken neat and with 72 scans! A regular 300 MHz machine currently costs bellow 200 k$ (though the maintenance costs can not be ignored with the current price of liquid helium), yet a sample of 10 mg per 0.7 mL solvent will give an excellent spectra with less than 16 scans. A 800 MHz machine can give you an 1H and 13C spectra from about 1 mg sample under optimum conditions. Though this 45 MHz machine seems not to use normal NMR tubes for sampling. I saw they mention some capillary tube sampling system, so the use of the neat sample might be due to this peculiar sampling system.

arsphenamine - 13-3-2011 at 11:56

Quote: Originally posted by Nicodem  
A regular 300 MHz machine currently costs bellow 200 k$ (though the maintenance costs can not be ignored with the current price of liquid helium)...
My hope is that the high temp superconductor NMR devices arrive on the market Real Soon Now.

A 200MHz 1H-NMR will still cost <$200k but the cooling requirements are less onerous. Still higher temp supercons can be cryogen-free.

http://www.hts110.co.nz/wp-content/uploads/2010/05/hts-110-2...

spirocycle - 13-3-2011 at 13:20

I was looking at this device for a basic organic classroom setting.
My professor is looking to get a grant for one.
even if it is not accurate to a high enough degree to identify complex unknowns, it would still probably be good enough to confirm the identity of a sample if you think you know what it is.

redox - 13-3-2011 at 16:05

The resolution on this NMR is too low. 45 MHz is far below the minimum (60 MHz is minimal, 100 MHz is good).

Dotty - 16-3-2011 at 12:41

Quote: Originally posted by arsphenamine  

A 200MHz 1H-NMR will still cost <$200k but the cooling requirements are less onerous. Still higher temp supercons can be cryogen-free.

http://www.hts110.co.nz/wp-content/uploads/2010/05/hts-110-2...


What are they using for the solenoid in the high temps? YBCO wire?

To spirocycle: The 45 MHz unit would probably be fine for that. Since it's a classroom setting, you can just choose unknown compounds appropriately such that the spectra are easily analyzed. My supervisor has a Terranova Earth's Field NMR and we use it as a teaching tool, though more for demonstrations of things like T1 and T2, etc. Like I said before, it would be useful to know what the software/hardware limitations were... for example, the resolution of the spectra on your 45 MHz could be improved given a longer acquisition time, etc.

[Edited on 16-3-2011 by Dotty]

[Edited on 16-3-2011 by Dotty]

Ozone - 16-3-2011 at 17:56

"I would love to have one such machine at home, but at the job I would hate it."

This.

But, Anasazi makes permanent magnet NMR (they are large, not bench-top) in the 90 MHz range. They wanted ~$70k, installed (software, etc.) with a HETCOR probe.

Cheers,

O3

loveoforganic - 16-3-2011 at 23:02

I would imagine with the low field strength you'd have trouble discerning anything with pretty proximal peaks, e.g. many aromatics. Confirming synthesis of an expected compound is about what I'd expect it to be limited to. Probably not much use in determining relative purity or identifying unknown organics with complexity beyond hexane, chloroform, etc.

arsphenamine - 17-3-2011 at 03:00

The Anasazi's are refurbished Varian chassis, definitely old school using permanent magnets, back when "real NMR's weighed a half ton". Their power and space requirements are similar to that of a refrigerator.

The picospin45's applications are explained well enough on the web site so any discussion of utility is moot.

One does _not_ develop such a narrowly focused product without assurance of a ready market, particularly in the monied ones like pharmaceuticals or food&beverage.

You can bet that Picospin's first business prospectus for venture capital was preceded by some cogent market research.

watson.fawkes - 17-3-2011 at 05:17

Quote: Originally posted by loveoforganic  
I would imagine with the low field strength you'd have trouble discerning anything with pretty proximal peaks, e.g. many aromatics. Confirming synthesis of an expected compound is about what I'd expect it to be limited to. Probably not much use in determining relative purity or identifying unknown organics with complexity beyond hexane, chloroform, etc.
Yes, but if you have an application where you don't need to distinguish between them, such as determining contamination levels in a water sample, it will work just fine.

There's far more in the wider chemistry industry than synthesis.

unionised - 17-3-2011 at 11:44

Quote: Originally posted by redox  
The resolution on this NMR is too low. 45 MHz is far below the minimum (60 MHz is minimal, 100 MHz is good).


I once used one with a 30MHz magnet. Since it was just for a student practical it didn't matter that it was a bit crap.

spirocycle - 18-3-2011 at 04:54

^this is what seems to be the general idea that I agree with.

arsphenamine - 18-3-2011 at 07:06

Quote: Originally posted by watson.fawkes  
There's far more in the wider chemistry industry than synthesis.
+1

Having a quick, simple test procedure that is undemanding in sample preparation is a Good Thing (tm).

Until spectrophotometric and HPLC tools were ubiquitous, thin layer chromatography was widely used because you got results in 10 minutes.

TLC did not generally effect or display the widest possible separation of compounds on a plate.

Does that mean it is useless?

It was more than adequate to show presence or absence of compounds of interest.

A knowledgable technician can cobble up a TLC test from coffee filters, alcohol, lighter fluid, and a mayo jar.

Low-rez NMR is often all you need for a fast, rough, determination that requires little of no sample preparation, e.g., beverages and fuels.

Of course, if some asshole demands a 300 MHz superconducting NMR for monitoring their reaction kinetics, that's between them and their grant foundation.

loveoforganic - 19-3-2011 at 23:38

Fair enough :) My perspective is admittedly colored - research was all in organic synthesis and generally had access to 350 - 600 MHz

arsphenamine - 20-3-2011 at 06:11

Quote: Originally posted by loveoforganic  
Fair enough :) My perspective is admittedly colored - research was all in organic synthesis and generally had access to 350 - 600 MHz
"Fair?"

Sir, you misunderstand me. I did not seek to be fair.

I intended an excoriating rebuke.

(cough, chortle)

Okay, maybe it wasn't.

The existential fact is that research is a small part of the entirety of chemical endeavors.
A rough estimate is the ratio of money spent on research vs. money toward industrial production. Research was as high as 7% in the early 70's.

The industry doesn't generally need research-grade tools, much the same as we don't need analytical reagent grade salt when we cook.

[Edited on 20-3-2011 by arsphenamine]

loveoforganic - 22-3-2011 at 12:59

Well in that case...

*&#@ you! :P

arsphenamine - 22-3-2011 at 13:24

Quote: Originally posted by loveoforganic  
Well in that case...

*&#@ you! :P
Thank you, sir.

That is the proper order of things.

Lest that be misunderstood ...:D

about the picoSpin-45

johncprice - 24-3-2011 at 09:13

Greetings, this is John Price, president and chief scientist at picoSpin. Let me try to answer some of these questions.

Software: The unit has an embedded web server and an Ethernet connector on the rear panel. You can connect it to your LAN and control it from any web browser. After an experiment is finished you download the data and analyze it with an NMR analysis package, such as MNova from Mestrelab Research. We provide a free one-year license to this software with every unit. There are also other packages available that can read our data format (JCAMP-DX).

Pulse sequences: The first units will ship with version 0.8 of the embedded software, which only supports 1-d spectroscopy. Over the first year we will be introducing several other experiment scripts as free upgrades, including T1 and T2 spin-echo sequences. When version 1.0 is released users will be able to write their own experiment scripts. The scripts specify the pulse sequence, the user interface created by the embedded web server, what data is displayed as the experiment is run, what data files are written and so forth.

HTS NMR systems: These may eventually be available, but they will be far more expensive than our unit.

About the capillary: The capillary is a tube with inside diameter of about 0.3 mm and length of about 6 inches. There are inlet and outlet fittings on the front panel. You can fill the capillary however you want, for example with a syringe and a syringe port. The NMR coil is only about 0.5 mm long so the volume of fluid that generates the NMR signal is very small; less than 1 microliter.

About sample size: The small sample size is an intrinsic feature of this technology. It is what allows us to use a small magnet and make the whole system shoe-box sized and much less expensive than any other NMR spectrometer.

About applications: Our product is excellent for classroom use. However, we are also seeing strong interest from industry. Anyone who has small organic molecules at high concentration may have an application. NMR provides unique information, and no other product can provide resolved NMR spectra at any where near this cost.

Regards,
John



[Edited on 24-3-2011 by johncprice]

NurdRage - 24-3-2011 at 15:22

Hmm... i wonder if i could get an evaluation model for a few weeks. ;)

spirocycle - 24-3-2011 at 19:12

^yea, me too

[Edited on 25-3-2011 by spirocycle]

arsphenamine - 24-3-2011 at 21:30

Quote: Originally posted by johncprice  
Software: The unit has an embedded web server and an Ethernet connector on the rear panel. You can connect it to your LAN and control it from any web browser. After an experiment is finished you download the data and analyze it with an NMR analysis package, such as MNova from Mestrelab Research. We provide a free one-year license to this software with every unit. There are also other packages available that can read our data format (JCAMP-DX).
[Edited on 24-3-2011 by johncprice]
Some comments for emphasis:

It's an NMR web appliance. How insanely cool is that?!

The bundled MNova software is on a 1-year ticker.
The perpetual license is another ~$430, not cheap but reasonable for a research tool.

The JCAMP-DX data format is a mature industry standard for spectroscopic data exchange. Everybody uses it these days.

picoSpin started in a home lab

Polverone - 14-11-2011 at 17:41

I found this interesting comment about the origins of picoSpin:
Quote:
Greetings, this is John Price, president and chief scientist at picoSpin. I’d be happy to answer any questions you or your readers might have about our spectrometer. By the way, the picoSpin project got its start a few years ago in a home lab. We found that it was not possible to use conventional chemical supply companies. Fortunately, very little chemistry was needed in the beginning and we were able to get by using common solvents available at our local hardware store. I did find a really good book on this subject: “Illustrated Guide to Home Chemistry Experiments,” by Robert Bruce Thompson.


Even a talented scientist/entrepreneur can't get cooperation from chemical vendors if he doesn't have the imprimatur of an institution.

Both John Price and Robert Bruce Thompson actually have accounts here on SM, though RBT hasn't posted in a while and John Price's one post is earlier in this thread. It is interesting what a small world home chemistry is.

aliced25 - 16-11-2011 at 02:02

Be a bloody sight smaller soon with luck:cool:

Anyone interested in a prototype modified halbach array that will be being tested with a simplified spectrometer? The design is going to be being optimized for use at around 50-51MHz and will use opensource software (at the moment)... It is potentially a one-off chance to get involved in something that involves some absolutely "mad science" at home, is fairly cutting edge and a boon to amateur/beginner scientists.

The array is Patent Pending (otherwise the Company involved would potentially/probably have to pay a fucking license fee to build their own design). I've spoken with the parties involved and their aim is to get a 50-51MHz machine, using thick walled 4-5mm tubes (of lower than standard length - but fairly standard all the same), that has similar performance characteristics to the PicoSpin (in terms of size, ease of use & output), to the market at ~$1,000US within 12 months (obviously they are aiming for maximum market disruption). Similar advances in ease of use, size and accessibility in conjunction with low-cost compared to high-end markets, have been successful before (http://mitpress.mit.edu/books/NORVH/chapter2.html):P

[Edited on 16-11-2011 by aliced25]

A wealth of desktop NMR machines: comparisons?

Polverone - 20-9-2013 at 00:04

A lot has changed since the picoSpin 45 was launched in 2010. The company was acquired by Thermo Scientific. They now offer a higher resolution model, the picoSpin 80, at 19 kg mass instead of 4.8 kg for the original. It still uses small capillary tubes.

There's a <a href="https://en.wikipedia.org/wiki/Benchtop_NMR_spectrometer">list on Wikipedia</a> of benchtop NMR spectrometers, all based on rare earth permanent magnets as far as I can tell. Some use small capillaries to hold samples while others use standard NMR tubes. For some you'll need a sturdier bench: 170 kg or more.

The Wikipedia page is somewhat out of date. It lists:
The <a href="http://www.magritek.com/products-spinsolve">Magritek Spinsolve</a>, 42.5 MHz.
The <a href="http://www.picospin.com/products/picospin-45/">Thermo Scientific picoSpin 45</a>, 45 MHz.
The <a href="http://www.nanalysis.com/nmready60p.html">nanalysis NMReady</a>, 60 MHz.
The <a href="http://www.oxford-instruments.com/businesses/industrial-products/industrial-analysis/campaigns/something-is-missing/pulsar-delivering-nmr-to-you r-benchtop">Oxford Instruments Pulsar</a>, 60 MHz.

It is missing:
The <a href="http://www.picospin.com/products/picospin-80/">Thermo Scientific picoSpin 80</a>, 80 MHz.
The <a href="http://www.bruker.com/products/mr/compact-mr/fourier-60/overview.html">Bruker Fourier 60</a>, 60 MHz.

Apart from price, what are strengths and weaknesses of these various systems? Anyone well versed in NMR care to offer an opinion? For example, there are three nominally 60 MHz instruments available -- what else is important in choosing between them?

Edit: some <a href="http://process-nmr.com/WordPress/?p=273">external commentary</a>.

[Edited on 9-20-2013 by Polverone]

bfesser - 20-9-2013 at 03:45

Quote: Originally posted by Polverone  
The Wikipedia page is somewhat out of date.
I'll see what I can do.

Interesting find, <strong>Polverone</strong>. Thanks for this update. :)

[edit] Sloppily added the new spectrometers to the Wikipedia page. Are you considering purchasing one of these, or are you just curious?

[Edited on 20.9.13 by bfesser]

Dr.Bob - 20-9-2013 at 07:45

There is an older discussion of tabletop NMRs on the technology section of the forum, a bit old now.

http://www.sciencemadness.org/talk/forumdisplay.php?fid=9

and go back a few pages. or just search NMR there.

But I am very amazed at what can be done now, I remember when T60's were considered great machines, and then new 90 Mhz machines came along and even had floppy disk drives. Thanks to Picospin for starting to make it better and smaller.

Once someone makes a 45 MHz machine, then people will figure out how to improve it, that is the great part. Eventually we might have much better small systems, but incremental change is sometimes part of the systems, so every little bit of improvement is great.

Polverone - 20-9-2013 at 08:18

I am just interested in the capabilities of the machines. The picoSpin 45 was around $20,000 when announced, which made it a cost breakthrough but still out of my price range. I am guessing that the newer machines are at least as expensive. Some day I'd like to have something from this class of machines, but I will have to wait until the second hand market is more developed or until there is more price competition among new models.

bfesser - 20-9-2013 at 09:09

I'm waiting until they incorporate a miniscule GC-MS/NMR/HPLC/UV-Vis./FT-IR/STM/XRD/bag of chips on a chip into a cellular telephone. :P

Oh, and better throw in some lasers, as well&mdash;just in case. Who knows, maybe some day in the distant future you'll even be able to use a cellular telephone to make an actual phone call&mdash;one where you can understand eachother!

pdm55 - 9-4-2016 at 07:13

Found this page comparing different benchtop NMRs
http://www.cbrnetechindex.com/Chemical-Detection/Technology-...

Prices range from $13 for SpinnSolve (Magritek), $30k for the picoSpin (ThermoFisher) to $65k for the Pulsar (Oxford Instruments).

[Edited on 9-4-2016 by pdm55]