A benchtop NMR spectrometer as a tool for monitoring mesoscale continuous-flow organic synthesis: equipment interface and assessment in four organic transformations

Abstract

An approach is reported for monitoring continuous-flow reactions by means of a low-field benchtop NMR spectrometer. The spectrometer is interfaced with a mesofluidic reactor and used as a tool for optimising four organic transformations, namely an acid-catalysed esterification, a Knoevenagel condensation, a Diels–Alder reaction, and an alkylation. Reactions need to be performed either solvent-free or at relatively high concentration in order to monitor them effectively using the NMR spectrometer, but this allows for the leveraging of one of the key advantages of flow processing, namely process intensification.

---

Introduction

By using continuous-flow processing, reactions can often be streamlined, multiple steps can be linked together, and reactions that were challenging in batch are now available for chemists to use.^1,2 There is also better control of heating and mixing, allowing for reactions to be performed under very precise and reproducible conditions. For these reasons, flow processing is seeing increasing interest both in academic and industrial settings. When evaluating the outcome of reactions performed using flow chemistry and optimizing reaction conditions, the best option is to employ in-line analytical tools. This opens the avenue for fast, reliable assays in comparison with the traditional approach in which performance is evaluated based on off-line product analysis. To this end, in-line analysis of flow processes has taken significant strides in recent years, particularly when interfaced with microfluidic devices.^3,4 Spectroscopic tools such as infrared,⁵ UV-visible,⁶ Raman,^7,8 and mass spectroscopy⁹ have all been successfully employed. Other reaction assay tools are also attracting attention, such as the use of digital cameras,¹⁰ or analytical balances.¹¹ By combining the available tools, either individually or together, with flow chemistry, it is possible for these processes to be self-optimizing.^4,12,13 In essence, this involves performing the reaction and then varying parameters such as stoichiometry, temperature and residence time, until optimal conditions are found. These iterations can be performed automatically by means of feedback loops between in situ reaction monitoring tools and the flow apparatus.

Of all the analytical techniques available to organic chemists on a day-to-day basis, NMR spectroscopy is probably the most widely used. A key advantage of NMR spectroscopy is that the signals observed are proportional to the concentration of the analytes. It is therefore possible to calculate conversion and yields without needing first to derive calibration curves or apply other post-analysis corrections. Traditionally, the compounds are dissolved in a deuterated solvent, placed into a sample tube and the spectrum recorded. The combination of chromatography and NMR has also made its way into the analytical laboratory. A liquid-chromatography (LC) system can be coupled directly to an NMR unit, the sample being transferred into the spectrometer “as is”.¹⁴ That is, it arrives at the concentration and in the (non-deuterated) solvent supplied by the LC. This in essence is the genesis of NMR analysis of flow chemistry.¹⁵ A range of NMR spectrometers have been interfaced with microfluidic equipment for real-time analysis of chemical reactions and biological processes.¹⁶ The majority, however, require complex engineering either on the part of the spectrometer or the flow cell used. Traditional high-field NMR spectrometers have been used for real-time analysis of batch reactions by means of a flow system removing aliquots over time for analysis.¹⁷ More recently, a benchtop NMR spectrometer comprising of a permanent magnet and operating at 43 MHz has been used in a similar way to monitor a transfer hydrogenation reaction.¹⁸ The NMR spectrometer was installed directly next to a Schlenk tube containing the reaction mixture and a peristaltic pump used to pass the solution through the magnet and back to the tube.

The application of NMR spectroscopy for real-time monitoring reactions performed using mesofluidic (flow rates of 0.5–10 mL min⁻¹) equipment is less explored. In our laboratory we have had success interfacing a simple benchtop Raman spectrometer with a mesoflow unit.^7,19 This has allowed us to monitor reactions from both a qualitative and quantitative perspective. Cronin and co-workers recently described the use of the aforementioned benchtop NMR spectrometer¹⁸ for monitoring continuous-flow reactions.²⁰ They performed reactions using a flow cell inside the spectrometer as the reactor. The approach works well and can be extended to self-optimization by means of feedback loops. However, using this configuration, reactions need to be performed at room temperature. A Grignard reaction has also been monitored using this NMR unit.²¹ These reports sparked our interest in interfacing a basic NMR spectrometer that we use primarily in teaching settings with one of our continuous-flow units and employing it for in-line reaction monitoring of a number of organic transformations both at ambient and elevated temperatures. Our results are presented here.

Results and discussion

Apparatus

As our NMR spectrometer, we selected a benchtop unit containing a 2 Tesla magnet and operating at 80 MHz (picoSpin™ 80 Series II).²² It operates in a lock-free manner and therefore can be employed with non-deuterated organic and aqueous solvents. This instrument and its 45 MHz analog have been used with success previously as a teaching tool as well as for rapid analysis of reaction mixtures in a research and development environment.²³ In these settings, samples are manually injected into the inlet of the spectrometer using a syringe (100 μL glass or 1 mL plastic). The fluid capillary system is contained within a cartridge and has a working volume of 40 μL. Material is injected through the cartridge and then out of a drain tube. Solvent can be passed through to clean the system. What drew our attention was that both the inlet and drain assemblies on the instrument are compatible with the standard PEEK nuts used on mesoscale flow equipment. We therefore envisaged we could connect our flow unit directly to the spectrometer with no modification required to either apparatus (Fig. 1). We decided initially to operate in a bypass mode whereby we would briefly divert the stream exiting the flow unit through the spectrometer. We could then record the NMR spectrum of the sample while the material was static. When we were ready to take another sample, we planned to use that to push out the prior one and refill the spectrometer with new analyte. Also, we decided to place the spectroscopic interface just after the back-pressure regulator assembly. This meant that we did not need to engineer a flow cell capable of holding significant pressure and also we were able to record the NMR spectra at room temperature, the product mixture having cooled by the time it reached the spectrometer.

Fig. 1 A continuous-flow unit interfaced with a benchtop NMR spectrometer.

Test reactions

With an experimental design established, we decided on four test reactions to probe its application as a reaction monitoring tool (Scheme 1).

Scheme 1 Test reactions selected.

(i) Acid-catalysed esterification reaction. We selected the acid-catalysed esterification of acetic acid with methanol as our first test reaction. We envisaged monitoring the disappearance of the signal due to the protons on the methyl group of methanol (3.48 ppm) and growth of that due to the methyl group attached to the oxygen of the methyl acetate product (3.72 ppm). Product conversion could then be obtained by integrating the two signals selected for monitoring. We prepared two stock solutions, one of acetic acid (solution A) and one of methanol and sulfuric acid (solution B). Pumping from solution A at a flow rate of 0.34 mL min⁻¹ and from solution B at a flow rate of 0.66 mL min⁻¹ and combining them by means of a T-piece, we passed the reaction mixture through a 10 mL capacity coil (1 mL min⁻¹ combined flow rate; 10 min residence time) at room temperature. At these flow rates, the reagent stoichiometry was such that there was a 1 : 2 molar ratio of acetic acid to methanol and a sulfuric acid catalyst loading of 2.5 mol%. We recorded the NMR spectrum of the product mixture, making sure that at least one coil volume had passed before we sent a sample in to our bypass loop and the spectrometer. Without stopping the flow of reagents, we raised the reactor coil temperature in 20 °C increments, recording the NMR spectrum of the product mixture at each point. Again we ensured that at least one coil volume of reagents had passed through the reactor at the desired temperature before we recorded the NMR spectrum. A plot of product conversion vs. temperature shown in Fig. 2. The product conversion begins to plateau at around 85 °C, this being as expected since the reaction reaches equilibrium in the sealed system. We repeated the reaction at a catalyst loading of 5 mol%, the data also being shown in Fig. 2. Again, product conversion is seen to plateau, equilibrium being reached slightly faster in this case. These results showed that we were able to monitor the reaction effectively using our apparatus.

Fig. 2 Monitoring an acid-catalysed esterification.

(ii) Base-catalysed Knoevenagel reaction. With the results of the esterification reaction in hand, we turned next to a transformation performed using a solvent. We selected the Knoevenagel condensation reaction between benzaldehyde and ethyl acetoacetate catalyzed by piperidine. This is a reaction we had monitored previously using Raman spectroscopy.^7,24 When performing a reaction in a non-deuterated solvent and recording spectra on a low-field benchtop NMR spectrometer, reagent concentration and the number of scans taken are key factors to take in to account. If the reaction mixture is too dilute, signals from the reagents and products will be swamped by those of the solvent and make quantification of reaction progress significantly challenging. This can be overcome by increasing the reagent concentration, increasing the number of scans taken, or a combination of the two. In the case of the esterification reaction, we had set the instrument to record 10 scans, this taking around 90 s. We did not want to increase the acquisition time when moving to the Knoevenagel reaction. With an eye to obtaining spectra with good signal-to-noise ratio in a timely manner, we set about determining what the optimal reagent concentration at which to operate. We prepared a series of stock solutions of benzaldehyde and ethyl acetoacetate in ethyl acetate at analyte concentrations up to 2 M and recorded their NMR spectra. We focused attention on the aldehyde signal from benzaldehyde at 10.11 ppm and the methylene group in ethyl acetoacetate at 3.55 ppm. Being able clearly to see these signals would be key to reaction monitoring. As shown in Fig. 3, a reagent concentration at least 1 M would be needed to be able to monitor the reaction reliably.

Fig. 3 NMR spectrum of benzaldehyde and ethyl acetoacetate in ethyl acetate as solvent, recorded across a range of analyte concentrations.

Deciding to proceed at reagent concentration of 1 M, we took an approach similar to that for the esterification reaction, performing the Knoevenagel condensation at a series of temperatures. In one reservoir we placed an ethyl acetate solution of benzaldehyde and ethyl acetoacetate (2 M in each) and in another we placed a 0.2 M ethyl acetate solution of piperidine (10 mol%). Each solution was flowed at 0.5 mL min⁻¹ giving a combined flow rate of 1 mL min⁻¹ through the reactor coil. To quantify the outcome of the reaction, we focused attention on the signal from unreacted benzaldehyde (10.11 ppm) and the signal for the CH₃ moiety of the acetyl group in the product. Using this signal, we were able to differentiate between the E- and Z-isomers of the product (2.48 ppm vs. 2.39 ppm respectively). A superimposition of the spectra recorded is shown in Fig. 4 and product conversion data shown in Table 1. Increasing the temperature from 40 °C to 85 °C led to a concomitant increase in product conversion. However, moving to 105 °C resulted in a slight decrease, perhaps due to the competitive decomposition of the product. There is a correlation between these results and those obtained using Raman monitoring for the same reaction.⁷

Fig. 4 NMR spectra of the base-catalysed Knoevenagel condensation, recorded across a range of reactor coil temperatures: Top – full spectrum, bottom – aldehyde signal from benzaldehyde and methyl signal from alkene product.

Table 1 Product conversion as a function of temperature in the base-catalysed Knoevenagel reaction of benzaldehyde and ethyl acetoacetate

Entry	Temperature (°C)	Product conversion (%)
1	40	26
2	50	43
3	65	53
4	85	63
5	105	55

---

(iii) Diels–Alder reaction. We selected the Diels–Alder reaction as a further test reaction for our study. We again would need to operate at high reagent concentration. We selected isoprene as the diene and maleic anhydride as the dienophile. A test reaction, even at room temperature, showed the significantly exothermic nature of the process when performed at high concentration in common organic solvents. This is not unexpected, given the previous reports on the significantly exothermic nature of solvent-free Diels–Alder reactions. An advantage of continuous-flow processing is that it is often possible to mitigate the problems associated with performing reactions that are potentially exothermic in nature.²⁵ The same Diels–Alder reaction has been successfully performed in a microcapillary flow disc reactor. In that case, the reagents are dissolved in acetonitrile and passed through the disc reactor at 60 °C with a residence time of 28–113 min. Operating with a slight excess of dienophile (concentrations of 5.9 M in isoprene and 6.3 M in maleic anhydride), and using acetonitrile as the solvent, we performed the reaction across a series of reactor temperatures, keeping the residence time constant at 10 min. To assay the reaction, we focused attention on a signal from unreacted maleic anhydride (7.26 ppm) and one that is unique to the product (3.57 ppm). As shown in Fig. 5, at 75 °C the conversion to product plateaus at 75%.

Fig. 5 Monitoring a Diels–Alder reaction.

We then performed the reaction, keeping the temperature constant at 75 °C and varying the flow rate. By decreasing the flow rate to 0.5 mL min⁻¹, hence increasing the residence time from 10 min to 20 min, the product conversion rose to 98%.

(iv) Alkylation reaction. Given the wide use of alkylation reactions in synthetic chemistry, we selected this class of transformation as our final example to probe in this study, our inspiration again came from a reaction previously performed at high reagent concentration using a microcapillary flow disc reactor, namely the alkylation of salicylaldehyde using allyl bromide. In that case, a twofold excess of allyl bromide was used and the process performed at room temperature. We decided to focus our attention on the less reactive 1-bromobutane so that we could probe the reaction across a range of temperatures. We operated at a salicylaldehyde to 1-bromobutane stoichiometric ratio of 1 : 2 and employed DBU (1.5 equiv.) as a base to remove the HBr byproduct. We decided to operate at a concentration of 2 M in salicylaldehyde and 3 M in 1-bromobutane. As in the case of the Diels–Alder reaction, this alkylation is exothermic when performed at such high concentration in batch on scale. Focusing on the aldehyde signal from the salicylaldehyde starting material (10.42 ppm) and the triplet from the methylene group of the alkyl chain next to the oxygen atom in the product (4.18 ppm), we monitored the reaction over a range of reactor coil temperatures. The results are shown in Fig. 6. While at room temperature there was no product formation observed, by the time we reached 100 °C, full conversion was obtained. To probe the effects of varying the reagent stoichiometry, we changed the salicylaldehyde to 1-bromobutane ratio from 1 : 2 to 1 : 1.5 and recorded spectra over the same temperature range. There was little difference in product conversion, showing that the reaction can be effectively performed at the lower stoichiometric ratio.

Fig. 6 Monitoring an alkylation reaction.

Conclusions

In summary, we have interfaced a low-field benchtop NMR spectrometer with a mesofluidic reactor, allowing us to monitor reactions and optimize reaction conditions in a streamlined manner. We show the application of the apparatus in four synthetic organic transformations. Reactions need to be performed at relatively high concentration in order to be able to monitor them effectively using the NMR spectrometer. This however allows us to leverage one of the key benefits of flow chemistry, namely process intensification. By positioning the NMR spectrometer directly after the back pressure regulator, we are not only able to monitor reactions without the need for a specially engineered flow cell but also can probe the effect of temperature on the outcome of the reaction without issue. In future work we will explore further the application of this benchtop NMR spectrometer for process control and optimization of a broader range of medicinally and industrial relevant transformations performed using flow approaches.

Experimental section

Apparatus configuration

The benchtop NMR spectrometer used was a Thermo Scientific picoSpin 80 Series II tuned at 82.699017 MHz for ¹H with the permanent magnet at 36 °C. Spectra were recorded using 10 scans, 3072 data points, 8 s relaxation delay, and a −8.6° pulse of 45 μs. The continuous-flow unit used was a Vapourtec E-series equipped with three pumps and, in the configuration used here, interfaced with one PFA reactor coil of total internal volume of 10 mL. The spectrometer was interfaced with the flow unit after the back-pressure regulator by attaching to the “waste” port of the waste/collect valve using 1 mm i.d. PFA tubing. Material was loaded in to the spectrometer by diverting the flow from “collect” to “waste”. During a reaction, spectral data was recorded at pre-determined time intervals using software provided with the instrument. The data was then processed using Mnova NMR (Mestrelab Research, S.L.).

Acid-catalysed esterification reaction

Two stock solutions were prepared – solution A: acetic acid (43 mL, 750 mmol, 1 equiv.); solution B: concentrated sulfuric acid (2 mL, 37.5 mmol, 0.05 equiv.) in methanol (61 mL, 1506 mmol, 2 equiv.). Two pumps were employed (pump A and pump B), the outlets of which were joined to a T-piece and the outlet of the T-piece attached to the “reagent in” port of the 10 mL capacity reactor coil. The “reagent out” port of the second reactor coil was directly interfaced with a variable back pressure regulator set to 4 bar, after which there was a short length of tubing leading to the waste/collect switch. The flow system was primed using the equipment manufacturer's suggested start-up sequence. The entire system was flushed with methanol for 5 min at a flow rate of 1 mL min⁻¹ on each pump. The flow rate of pump A was then decreased to 0.34 mL min⁻¹ and that of pump B decreased to 0.66 mL min⁻¹ (giving a combined flow rate of 1.0 mL min⁻¹) and the reactor coil heated to 20 °C. Once at the target temperature, the flow was then changed from solvent to reagents by means of a switch on the flow unit. Solution A was pumped by pump A and solution B by pump B. The reaction mixture was then flowed through the reactor coil. After 15 min, an aliquot of the product stream was diverted to the spectrometer and the NMR spectrum recorded. The temperature was then increased to 40 °C and, once at temperature and after at least one reactor volume of reagents had passed through the coil, the NMR spectrum was again recorded. The process was repeated across a range of temperature values. Once the trial was complete, the flow was changed back to solvent and the system run until the remainder of the reaction mixture has passed through the entire configuration. Once all the product had exited the unit, heating was ceased and, once the reactor coil had returned to room temperature, the flow of methanol solvent was stopped.

Base-catalysed Knoevenagel reaction

Two stock solutions were prepared – solution A: benzaldehyde (21.224 g, 200 mmol, 2 M) and ethyl acetoacetate (26.028 g, 200 mmol, 2 M) in ethyl acetate (total volume 100 mL); solution B: piperidine (1.703 g, 20 mmol, 0.2 M) in ethyl acetate (total volume 100 mL). An identical approach to that for the esterification was taken with the exceptions that ethyl acetate was used in the place of methanol for priming and flushing the system, and that the flow rate of pumps A and B was set at 0.5 mL min⁻¹ (combined flow rate of 1.0 mL min⁻¹) for the trials across a range of temperatures.

Diels–Alder reaction

Two stock solutions were prepared – solution A: maleic anhydride (47.718 g, 487 mmol, 6.3 M) in acetonitrile (total volume 77 mL); solution B: isoprene (65 mL, 650 mmol, 5.9 M) in acetonitrile (total volume 110 mL). Given the low boiling point of isoprene, solution B was kept in an ice bath. An identical approach to that for the esterification was taken with the exception that acetonitrile was used in the place of methanol for priming and flushing the system, and that the flow rate of pumps A and B was set at 0.5 mL min⁻¹ (combined flow rate of 1.0 mL min⁻¹). When probing the effect of flow rate on the reaction, the temperature of the reactor coil was set constant at 75 °C.

Alkylation reaction

Two stock solutions were prepared – solution A: salicylaldehyde (21 mL, 197 mmol, 2.0 M) and DBU (45 mL, 301 mmol, 3.0 M) in acetonitrile (total volume 100 mL); solution B: 1-bromobutane (32 mL, 298 mmol, 3.0 M) in acetonitrile (total volume 100 mL). An identical approach to that for the esterification was taken with the exception that acetonitrile was used in the place of methanol for priming and flushing the system, and that the flow rate of pumps A and B was set at 0.5 mL min⁻¹ (combined flow rate of 1.0 mL min⁻¹).

Acknowledgements

We gratefully acknowledge financial support from the National Science Foundation Research Experience for Undergraduates (REU) program (CHE-1359081) and the University of Connecticut. The University of Connecticut Department of Chemistry teaching laboratory services is thanked for the loan of a picoSpin NMR unit. Input from Shelli Miller is gratefully acknowledged.

References

1. For books on flow chemistry see: (a) Flow Chemistry, ed. F. Darvas, G. Dormán and V. Hessel, DeGruyter, Berlin, 2014; (b) Microreactors in Preparative Chemistry: Practical Aspects in Bioprocessing, Nanotechnology, Catalysis and more, ed. W. Reschetilowski, Wiley-VCH, Weinheim, 2013; (c) C. Wiles and P. Watts, Micro Reaction Technology in Organic Synthesis, CRC Press, Boca Raton FL, 2011; (d) Chemical Reactions and Processes under Flow Conditions, ed. S. V. Luis and E. Garcia-Verdugo, Royal Society of Chemistry, Cambridge UK, 2010.
2. For selected recent reviews, see: (a) B. Gutmann, D. Cantillo and C. O. Kappe, Angew. Chem., Int. Ed., 2015, 54, 6688–6728; (b) K. F. Jensen, B. J. Reizman and S. G. Newman, Lab Chip, 2014, 14, 3206–3212; (c) I. R. Baxendale, J. Chem. Technol. Biotechnol., 2013, 88, 519–552; (d) L. N. Protasova, M. Bulut, D. Ormerod, A. Buekenhoudt, J. Berton and C. V. Stevens, Org. Process Res. Dev., 2013, 17, 760–791; (e) D. T. McQuade and P. H. Seeberger, J. Org. Chem., 2013, 78, 6384–6389.
3. For a perspective, see: (a) N. Holmes and R. A. Bourne, in Chemical Processes for a Sustainable Future, ed. T. Letcher, J. Scott and D. A. Patterson, Royal Society of Chemistry, Cambridge UK, 2014; (b) J. Yue, J. C. Schouten and T. A. Nijhuis, Ind. Eng. Chem. Res., 2012, 51, 14583–14609; (c) W. Ferstl, T. Klahn, W. Schweikert, G. Billeb, M. Schwarzer and S. Loebbecke, Chem. Eng. Technol., 2007, 30, 370–378.
4. V. Sans and L. Cronin, Chem. Soc. Rev., 2016, 45, 2032–2043.
5. For selected examples, see: (a) S. T. R. Müller, A. Murat, D. Maillos, P. Lesimple, P. Hellier and T. Wirth, Chem.–Eur. J., 2015, 21, 7016–7020; (b) J. W. Rydzak, D. E. White, C. Y. Airiau, J. T. Sterbenz, B. D. York, D. J. Clancy and Q. Dai, Org. Process Res. Dev., 2015, 19, 203–214; (c) J. S. Moore and K. F. Jensen, Angew. Chem., Int. Ed., 2014, 53, 470–473; (d) J. S. Moore and K. F. Jensen, Org. Process Res. Dev., 2012, 16, 1409–1415; (e) T. Brodmann, P. Koos, A. Metzger, P. Knochel and S. V. Ley, Org. Process Res. Dev., 2012, 16, 1102–1113; (f) C. F. Carter, H. Lange, S. V. Ley, I. R. Baxendale, B. Wittkamp, J. G. Goode and N. L. Gaunt, Org. Process Res. Dev., 2010, 14, 393–404.
6. (a) F. Benito-Lopez, W. Verboom, M. Kakuta, J. G. E. Gardeniers, R. J. M. Egberink, E. R. Oosterbroek, A. van den Berg and D. N. Reinhoudt, Chem. Commun., 2005, 2857–2859; (b) H. Lu, M. A. Schmidt and K. F. Jensen, Lab Chip, 2001, 1, 22–28.
7. T. A. Hamlin and N. E. Leadbeater, Beilstein J. Org. Chem., 2013, 9, 1843–1852.
8. (a) G. Chaplain, S. J. Haswell, P. D. I. Fletcher, S. M. Kelly and A. Mansfield, Aust. J. Chem., 2013, 66, 208–212; (b) S. Mozharov, A. Nordon, D. Littlejohn, C. Wiles, P. Watts, P. Dallin and J. M. Girkin, J. Am. Chem. Soc., 2011, 133, 3601–3608; (c) G. Rinke, A. Ewinger, S. Kerschbaum and M. Rinke, Microfluid. Nanofluid., 2011, 10, 145–153.
9. (a) T. W. T. Bristow, A. D. Ray, A. O'Kearney-McMullan, L. Lim, B. McCullough and A. Zammataro, J. Am. Soc. Mass Spectrom., 2014, 25, 1794–1802; (b) S. E. Hamilton, F. Mattrey, X. Bu, D. Murray, B. McCullough and C. J. Welch, Org. Process Res. Dev., 2014, 18, 103–108; (c) D. L. Browne, S. Wright, B. J. Deadman, S. Dunnage, I. R. Baxendale, R. M. Turner and S. V. Ley, Rapid Commun. Mass Spectrom., 2012, 26, 1999–2010; (d) S. Koster and E. Verpoorte, Lab Chip, 2007, 7, 1394–1412.
10. R. J. Ingham, C. Battilocchio, D. E. Fitzpatrick, S. Sliwinski, J. M. Hawkins and S. V. Ley, Angew. Chem., Int. Ed., 2015, 54, 144–148.
11. L. Guetzoyan, R. J. Ingham, N. Nikbin, J. Rossignol, M. Wolling, M. Baumert, N. A. Burgess-Brown, C. M. Strain-Damerell, L. Shrestha, P. E. Brennan, O. Fedorov, S. Knapp and S. V. Ley, MedChemComm, 2014, 5, 540–546.
12. For reviews, see: (a) S. V. Ley, D. E. Fitzpatrick, R. J. Ingham and R. M. Myers, Angew. Chem., Int. Ed., 2015, 54, 3449–3464; (b) D. C. Fabry, E. Sugiono and M. Rueping, Isr. J. Chem., 2014, 54, 341–350.
13. For commentaries, see: (a) R. A. Skilton, R. A. Bourne, Z. Amara, R. Horvath, J. Jin, M. J. Scully, E. Streng, S. L. Y. Tang, P. A. Summers, J. Wang, E. Perez, N. Asfaw, G. L. P. Aydos, J. Dupont, G. Comak, M. W. George and M. Poliakoff, Nat. Chem., 2015, 7, 1–5; (b) M. Peplow, Nature, 2014, 512, 20–22.
14. For background, see: F. Dalitz, M. Cudaj, M. Maiwald and G. Guthausen, Prog. Nucl. Magn. Reson. Spectrosc., 2012, 60, 52–70.
15. For a historical perspective, see: A. Nordon, C. A. McGill and D. Littlejohn, Analyst, 2001, 126, 260–272.
16. For examples, see: (a) M. V. Gomez, H. H. J. Verputten, A. Díaz-Ortíz, A. Moreno, A. de la Hoz and A. H. Velders, Chem. Commun., 2010, 46, 4514; (b) J. Bart, A. J. Kolkman, A. J. Oosthoek-de Vries, K. Koch, P. J. Nieuwland, J. W. G. Janssen, P. J. M. van Bentum, K. A. M. Ampt, F. P. J. T. Rutjes, S. S. Wijmenga, J. G. E. Gardeniers and A. P. M. Kentgens, J. Am. Chem. Soc., 2009, 131, 5014–5015.
17. For examples, see: (a) J. Y. Buser and A. D. McFarland, Chem. Commun., 2014, 50, 4234–4237; (b) A. C. Barrios Sosa, R. T. Williamson, R. Conway, A. Shankar, R. Sumpter and T. Cleary, Org. Process Res. Dev., 2011, 15, 449–454; (c) M. A. Vargas, M. Cudaj, K. Hailu, K. Sachsenheimer and G. Guthausen, Macromolecules, 2010, 43, 5561–5568.
18. E. Danieli, J. Perlo, A. L. L. Duchateau, G. K. M. Verzijl, V. M. Litvinov, B. Blümich and F. Casanova, ChemPhysChem, 2014, 15, 3060–3066.
19. T. A. Hamlin and N. E. Leadbeater, J. Visualized Exp., 2015, e52393.
20. V. Sans, L. Porwol, V. Dragone and L. Cronin, Chem. Sci., 2015, 6, 1258–1264.
21. M. Goldbach, E. Danieli, J. Perlo, B. Kaptein, V. M. Litvinov, B. Blümich, F. Casanova and A. L. L. Duchateau, Tetrahedron Lett., 2016, 57, 122–125.
22. Accessed July 24, 2016 – https://tools.thermofisher.com/content/sfs/brochures/picoSpin%2080%20Datasheet_EN.pdf.
23. M. V. Silva Elipe and R. R. Milburn, Magn. Reson. Chem., 2016, 54, 437–443.
24. J. R. Schmink, J. L. Holcomb and N. E. Leadbeater, Chem.–Eur. J., 2008, 14, 9943–9950.
25. See, for context: (a) A. M. Thayer, Chem. Eng. News, 2005, 83, 43–52; (b) J. Antes, D. Boskovic, H. Krause, S. Loebbecke, N. Lutz, T. Tuercke and W. Schweikert, Chem. Eng. Res. Des., 2003, 81, 760–765.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19662d

---