Volume 93 Issue 37 | pp. 19-21
Issue Date: September 21, 2015

A Renaissance For NMRs, Big And Small

Some nuclear magnetic resonance instruments shrink, even as the more complex ones get larger
Department: Business
News Channels: Analytical SCENE
Keywords: nuclear magnetic resonance, NMR, electronics, magnets, permanent magnets, protein analysis, teaching, quality control

In the field of nuclear magnetic resonance (NMR) spectroscopy, bigger and more expensive have always meant better. Although that maxim is still true, a new generation of companies contends that smaller benchtop or even handheld NMRs can do much of the yeoman’s work of identifying molecular structures that room-sized NMRs do.

A scientist analyzes a sample in a benchtop NMR.
Credit: Thermo Fisher
A person’s arms work on a laptop next to a white box about the size of a shoebox.
A scientist analyzes a sample in a benchtop NMR.
Credit: Thermo Fisher

Makers of the smaller tools—firms such as Magritek; Thermo Fisher Scientific, which purchased start-up picoSpin in 2012; Nanalysis; T2 Biosystems; and WaveGuide—don’t expect their modestly priced machines will replace expensive high-field instruments tricked out with cryo-cooled magnets and separate operator consoles. But aided by modern electronics, improvements in permanent magnet design, and advances in software, compact-NMR makers say their small tools will become more commonplace over the next five years.

What’s more, makers of smaller NMRs say their tools will be accessible to more than just scientists skilled at using the traditional machines. Expect to see the new instruments, they say, in quality-control and teaching applications, under lab hoods monitoring reactions and verifying samples, and in hospitals diagnosing disease.

As for the two surviving makers of the larger, more complex NMRs, Bruker and JEOL, they say modern electronics and magnet design are driving development of their instruments as well. Agilent Technologies shut down its money-losing NMR operation in 2014.

Magnets are at the heart of NMR. Placed in a magnetic field, carbon- and hydrogen-containing samples give up their structural secrets when they are interrogated with pulsed radio frequencies using Fourier transform methods. Pulses of a single radio frequency are used to interrogate molecular structure in time-domain techniques.

Rare-earth-based permanent magnets have made possible compact benchtop NMRs, whereas their larger brethren use powerful helium-cooled magnets to provide more detailed information at higher magnetic fields.

Industry observers see a developing NMR landscape in which large and small machines will coexist. “An application like protein structure determination requires large and expensive high-field NMRs, and therefore they will persist,” says Stan Sýkora, an Italy-based physicist and NMR expert. “But if an application can be handled with a lower-field less costly machine, then why not?

“Simply put, the application is what matters; the instrument is secondary,” Sýkora adds. “If I were to start a company today to build an NMR, I would ask myself ‘What for?’ and not ‘What kind?’ ”

A 1.02-GHz NMR magnet developed at NIMS with JEOL and other partners.
Credit: National Institute for Materials Science
A man in a suit stands next to a giant cylinder with scaffolding around it.
A 1.02-GHz NMR magnet developed at NIMS with JEOL and other partners.
Credit: National Institute for Materials Science


Room-sized and benchtop NMRs each have their own advantages


Large NMR
◾ Cost can exceed $1 million for the highest-resolution instruments
◾ More sensitive, with higher-resolution spectra
◾ Requires helium to cool superconducting magnet
◾ Operated from a large console by an experienced specialist


Magritek’s Spinsolve benchtop NMR.
Credit: Magritek
A woman places a sample in a red box on a lab bench.
Magritek’s Spinsolve benchtop NMR.
Credit: Magritek

Benchtop NMR
◾ Cost is generally below $100,000
◾ Less sensitive, with lower-resolution spectra
◾ Does not require helium, and uses a permanent magnet at room temperature
◾ Can be operated by a technician using automating software
 SOURCE: NMR manufacturers

Many benchtop-NMR makers have already determined that their machines are good for the education market, points out Mark Dixon, a Thermo Fisher product manager. Small NMRs such as the firm’s picoSpin line “are plug-and-play systems for an educator” to help undergraduates get hands-on experience in analyzing samples.

But Dixon sees many other applications that suit small NMRs. A new accessory for the picoSpin line allows academic researchers and small laboratory operators to monitor chemical reactions in real time, he says.

Technology advances have made the new machines possible. “Electronics in cell phones were the inspiration behind benchtop NMRs,” Dixon explains. Whereas the big NMRs are often built with bulky printed circuit boards, the smaller NMRs depend on miniature components such as microchips and flash memory modules, he points out.

Improvements in permanent rare-earth magnets also allow makers of portable NMRs to build compact instruments that don’t have the complex magnet cooling requirements of their larger brethren. But permanent magnets currently remain limited to magnetic fields of less than 100 MHz, Dixon points out. By comparison, big helium-cooled NMRs with fields exceeding 1 GHz are now under development.

Those limitations explain the difference between the benchtop NMRs and the larger instruments. “High-field NMRs are environmentally sensitive, sophisticated to operate, and very expensive,” says Andrew Coy, chief executive officer of New Zealand-based Magritek, which makes the Spinsolve line of benchtop NMRs.

The high-field instruments excel at elucidating the structure of large molecules such as proteins typically exceeding 1,000 daltons, Coy explains. The lower magnetic fields of the smaller instruments make them more suitable to the analysis of molecules of less than 1,000 Da, he says.

Although the big high-field instruments can be set up for many different experiments, the benchtop machines are “more relevant for quick routine measurements,” Coy notes. Simpler design and intuitive software make the benchtop instruments easy to use.

“People will purchase benchtop NMRs to solve two or three specific problems,” Coy predicts.

“The market for NMR is evolving in part because of the benchtop instruments,” says Garett Leskowitz, chief technology officer of the Canadian firm Nanalysis. The education market is the “low-hanging fruit” for simple-to-use NMRs such as Nanalysis’s NMReady line of benchtops, he says.

But quality analysis and control are also developing markets, Leskowitz says. Makers of portable Raman and near-infrared spectrometers “should see us as the new competition,” he says.

NMR users have long had to line up to use high-field instruments at a specialized university center or send samples out to a commercial lab. Such delays now can be avoided, Leskowitz says, by using no-wait benchtop machines for routine analysis. As a bonus, costs to use the benchtops are much lower, he points out.

The benchtops from Thermo Fisher, Magritek, and Nanalysis all use Fourier transform techniques to characterize samples, as do the larger high-field NMRs.

Other benchtops use time-domain techniques, often for specific diagnostic tests. Among them are machines from nine-year-old T2 Biosystems, based in Lexington, Mass., which has developed a Food & Drug Administration-approved NMR diagnostic unit and test for candida fungal infections in the blood. The firm’s technology, developed by scientists at Massachusetts General Hospital and Massachusetts Institute of Technology, including Priestly Medal winner Robert Langer, is based on the activity of water molecules in a magnetic field.

According to Tom Lowery, the firm’s chief scientific officer, magnetic nanoparticles coated with an analyte-specific binding agent are mixed with a blood sample and placed in the firm’s benchtop T2Dx. The particles bind to and cluster around the pathogen, changing the microenvironment of water and altering the magnetic resonance signal. After a three-hour concentration and amplification step, a reading reveals the presence of candida, he says.

An alternative to blood cultures, which take up to five days for results, T2’s test identifies positive results 91% of the time, Lowery says. Blood cultures only do so 50 to 60% of the time, he says. The firm is now selling the system to hospitals.

Enabling the NMR reading, he adds, is a proprietary permanent magnet “about the size of a Rubik’s Cube” producing a field of 23 MHz. In addition, the firm is developing tests for pathogens associated with sepsis and Lyme disease.

Also using time-domain techniques, but for a handheld diagnostic device, is WaveGuide. Like T2 Biosystems, WaveGuide will rely on binder-coated magnetic particles to test for diseases such as tuberculosis in sputum or whole blood, says Nelson Stacks, CEO of the Cambridge, Mass.-based firm. Specifics on how the test will work are not available.

But the technology behind WaveGuide’s device is a tiny chip-based relaxometer, which measures relaxation of nuclear spin, from the lab of Donhee Ham, a Harvard University physics professor.

WaveGuide has so far raised $27 million from angel investors and has created a number of prototypes. The firm plans to hire staff and rent lab space soon to develop them.

Whereas WaveGuide is relatively new to time-domain-enabled NMR, Bruker, more often thought of as a maker of large high-field NMRs, has been active with time-domain-based benchtop NMRs for about 40 years. Many of them are targeted at specific quality-control and analysis applications, says Stefan Jehle, an applications scientist with the firm.

For instance, the firm’s minispec mq Toothpaste Analyzer measures fluorine content in toothpaste. The minispec mq7.5 Large Seed Analyzer characterizes seed oil and moisture content. “A dedicated system is cost-effective and reduces subjectivity while also speeding analysis,” Jehle says.

Bruker says it is developing a line of Fourier-transform-based benchtop NMRs, but they are not yet ready for the market. A spokesman says that “we are interested in new markets and applications.”

And Bruker is certainly interested in expanding its large-NMR offerings. It continues to develop superconducting magnets capable of operating at 1.2 GHz and plans to deliver them to customers starting next year. The largest magnet Bruker has installed to date operates at 1.0 GHz.

Although the government funding to pay for larger systems is in short supply, Clemens Anklin, vice president for NMR applications at Bruker, says they are needed to push advances in materials research, structural biology, and intrinsically disordered proteins.

So the firm has vowed to develop the new high-temperature superconducting wire that will enable higher field systems. Anklin notes that the firm owns a superconducting materials business, allowing it to stay on top of the effort to push ultra-high-field NMR.

Trying a bit of one-upmanship on Bruker, large-NMR competitor JEOL says it has plans for a spectrometer capable of operating at 1.3 GHz. The firm is collaborating on the development of a 1.3-GHz magnet with the Japan Science & Technology Agency, the National Institute for Materials Science of Japan (NIMS), and other partners. In July, the group said it successfully built a 1.02-GHz NMR magnet.

Michael Frey, a product manager for JEOL, says the firm has redesigned the electronics built into its NMRs to include “state-of-the-art digital and radio-frequency designs to greatly improve NMR experiment flexibility and data collection.” The improvements include high-speed signal and pulse generators and receivers, he says.

These advances not only have improved the instruments’ reliability but also have helped shrink the footprint of operator consoles by more than 20%, Frey says. The firm is also offering the upgraded consoles to owners of Agilent and Bruker magnets.

JEOL’s traditional strengths are in small-molecule characterization, Frey says. But the firm is now increasing its efforts “in all areas of NMR, including bio-NMR, real-time NMR, and cold probe development,” he says.

Although JEOL plans to be aggressive in big NMRs, it is leaving the development of benchtop systems to others. But whether for big or small systems, consultant Sýkora says, improvements in NMR electronics, magnets, and software have been profound.

Those improvements are not just enabling bigger and better tools for the high priests of academic science. They are also making possible simple and useful NMR gear for everyday students, bench chemists, and quality-control technicians.  


Advances in electronics and nuclear magnetic resonance (NMR) science mean new opportunities for entrepreneurs.

Marcus Semones, one of the people behind NMR start-up WaveGuide, had worked in academia and as a pharmaceutical researcher. After leaving GlaxoSmithKline in 2010, he began visiting university technology transfer offices looking for commercial opportunities.

T2 Biosystems’ benchtop diagnostic NMR is based on research done at MIT and Massachusetts General Hospital.
Credit: T2 Biosystems
An open sample port in a small analytical instrument.
T2 Biosystems’ benchtop diagnostic NMR is based on research done at MIT and Massachusetts General Hospital.
Credit: T2 Biosystems

Semones found what he was looking for at Harvard University’s Office of Technology Development: a 2-mm2-microchip-based relaxometer for a miniature NMR that its inventor, Nan Sun, now at the University of Texas, Austin, developed in the lab of Harvard physicist Donhee Ham. The relaxometer measures relaxation of nuclear spin.

Semones licensed the technology in 2012 after four months of talks with the technology office and formed WaveGuide. The chip, he says, should allow his firm to do for NMR what others have done for infrared spectrometry: “turn clunky benchtop scientific instruments into handheld devices.” His goal is to make a handheld NMR that works with nanoparticles to diagnose infectious diseases and cancer.

At first, Semones and partner Nelson Stacks, WaveGuide’s chief executive officer and an entrepreneur and venture capitalist, financed development of the handheld NMR on their own. Earlier this year, they raised $27 million from investors, including Peter Farrell, founder of respiratory disease treatment firm ResMed.

The partners are now on their way to raise an additional $30 million. They plan to hire researchers and set up lab space in Cambridge, Mass., to further develop their handheld device.

Harvard is eager to license out additional NMR technology, says Sam Liss, business development manager at the university’s technology office. Now ready and waiting is a second-generation microchip from Ham’s group that incorporates an NMR transmitter, arbitrary pulse sequencer, and receiver all on a 4-mm2 piece of silicon. The more advanced chip is another step toward compact, low cost NMRs.

“Our goal is to help advance science and ensure new technology is developed and realized as products with impact,” Liss says.

Large, established companies have “deep capabilities and broad distribution channels” to get technology to market, Liss says. “But start-ups move faster and are willing to take greater risk.” At this point, he adds, the jury is still out on the type of business partner best suited to bring the Ham group’s latest microchip design to market.

Like WaveGuide, T2 Biosystems also converted university technology into an NMR-based tool. T2’s benchtop analyzer tests for candida fungal infection in the blood in a matter of hours instead of days. The business got started in 2006 with technology developed at Massachusetts Institute of Technology and Massachusetts General Hospital, says Tom Lowery, T2’s chief scientific officer.

The technology itself arose from a research collaboration among six scientists associated with the two institutions. Among the scientists was Priestley Medalist and serial entrepreneur Robert Langer.

To get its technology off the ground, T2 raised about $100 million over several investment rounds from venture capital firms including Flagship Ventures, Aisling Capital, and Flybridge Capital Partners. In August, the firm went public, raising about $57 million. It recently filed a preliminary prospectus with the Securities & Exchange Commission to sell another $100 million in stock at some point in the future.

But now comes the hard part for T2 and any start-up that has made it as far as T2 has: making money. The firm reported a $22 million loss in the first half of the year.

But that may change, as T2’s sales force is now actively selling its NMR-based T2Dx System to hospitals in the U.S. and Europe, Lowery says. Through the end of June, T2 had secured 10 hospital contracts, he said.

Chemical & Engineering News
ISSN 0009-2347
Copyright © American Chemical Society
Richard Shoemaker (Tue Sep 22 14:07:35 EDT 2015)
The appearance of quality benchtop NMR systems is definitely important and promises to be a welcome, growing part of the NMR market. I applaud those pioneering companies that are pushing that market forward.

However, this article leaves the impression that high-field, superconducting magnet NMRs will only be needed for studying large molecules like proteins, and benchtop NMR system can handle structure elucidation/verification of molecules "below 1000 Da." Most complex molecules under 1000 Da molecular weight rely upon 300-500MHz for unambiguous structure characterization, and it is very difficult to imagine solving a new structure like paclitaxel (854 Da) at concentrations of 1mg/ml or less on a 45-90MHz benchtop system.

Modern mid-field NMR spectrometers (300-600MHz)represent the lion's share of the NMR market and also the NMR research demand. Not only for structure determination, but for studying conformational dynamics, molecular diffusion, and properties of materials (including solid-state and other Magic Angle Spinning methodologies).

Publishing an article that represents such a critical research field like NMR as being primarily divided into simply Ultra High Field (i.e. biomolecular NMR) vs. low-field Benchtop systems seems to be highly misleading, if not downright irresponsible.
Josh J. (Wed Sep 23 14:47:02 EDT 2015)
Thanks, Richard! Spot on!
Stan Sykora (Sat Sep 26 09:25:59 EDT 2015)
The article, in my opinion, reacts to a current shift in paradigm: so far, NMR instruments dictated the applications, and they were carefully refined for just the single application you mention (determining the molecular structure of a steroid-size compound). That highly optimized application with 50 years of history behind it certainly calls for a 400+-100 MHz field (always use the tools most appropriate for a given task!). I do not see the "new" and "small" instruments as something that should replace an existing and highly successful application; that would be foolish.

But there is a whole world of thousands of potential NMR applications (not necessarily all spectroscopic): industrial process monitoring/control, education, mobile and/or portable NMR deployments, NMR in poor Countries, materials research, agriculture, food technology, even things like early sepsis detection. For example, you just can not dissolve a lubrificant in chloroform to study its lubricating properties, right?

Regarding the historically important but single application you are mentioning: First of all, it will probably not remain the economically dominant one; it will become just one of many. Second: it will also greatly benefit from the unavoidable advent of compact, table-top 200, 300, and 400 cryogen-free superconducting magnets and novel types of microsensors to go with them. They are not yet on the market, but just wait 5 years, not more. Unfortunately, I may not say more :-)
Åsmund Hattemaker (Wed Sep 23 14:53:57 EDT 2015)
I very much agree with you Richard Shoemaker.

A litte rant on the side: At my workplace, we were demonstrated a 60 MHz benchtop model and the failure of the sales representative to invite the chemists in the room to use the instrument to acquire real world spectra of compounds in solution, as well a reluctance to admit that the seeminngly decent/usable NMR spectra presented were acquired with a neat, low-vicosity liquid compound, left me surprised that they had even spent the time and money to launch such an instrument. On top of this, the representative of the company presenting the instrument in question, was not the least bit interested in hearing about alternatives to Fourier processing, such as harmonic inversion noise reduction (HINR), that likely would allow for the aquisition of usable data from an in-solution sample...

Mind you, I am not discrediting the efforts of the companies/researchers trying to make an NMR instrument smaller/more affordable/more accessible, but as you, I hardly see a market for these instruments in the academic sector, if not a market for developing them into more generally usable instruments.

I'd say I have more faith in higher temperature superconductors, new lower noise electronics and alternative signal processing in the development of NMR instruments that are affordable for budget users.

Stanislav Sykora (Sat Sep 26 09:01:27 EDT 2015)
Marc, thank you for the article which is excellent, despite the fact that you mention me :-) As you might guess, I take an issue with one of your sentences ("pulses of a single radio frequency are used to interrogate molecular structure in time-domain techniques") which is technically inaccurate, but that is really off-topic with respect to the scope and target of the article. Actually, I should thank you for that, too, because it will give me an opportunity to expand and explain on my own website (Stan's Hub).

The article illustrates nicely the current spasms that are agitating NMR technology and application areas. It will take some time to sort all this out. Bruker is presently hoarding orders for their huge 1.2 GHz system, selling for about $25 million a piece! In my opinion, there will always be a need even for such systems, but limited to very few specific applications and operated in huge governmental labs. While on the low-cost side (down to $10 kilo) we will see an explosion of applications and application-defined instrument types, many barely recognizable as NMR. Which does not mean, however, that all the new makers will get it right ...

But that you have made quite clear.
Leverett Smith (Thu Oct 01 21:26:45 EDT 2015)
I'm a bit puzzled that C&EN's article about an "NMR Renaissance", while describing a welcome growth of the smaller instrument market, doesn't mention Anasazi Instruments among the various manufacturers discussed. We acquired one of their machines at Contra Costa College some fifteen years ago, and for our sophomore organic students and our community college budget it was pretty much the perfect instrument...
Salih HACINI (Sun Oct 04 16:02:46 EDT 2015)
Very good article information and comparison of NMR analyzes of very different devices in their designs, uses ... and in their prices!
I just bought the small device for use in organic synthesis, but I have not the right procedures for efficient use.
Robert Buntrock (Mon Oct 05 10:39:03 EDT 2015)
Although NMR spectrometers are rated in MHz, that's the strength of the Rf field that's scanned while the corresponding magnetic field remains constant. I'm more concerned with the strength, in Teslas or whatever, of the magnet. Those of us with pacemakers could not operate any of these machines.
Leave A Comment