Letter pubs.acs.org/ac
Monitoring Isomerization of Molecules in Solution Using Ion Mobility Mass Spectrometry James N. Bull,† Michael S. Scholz,† Neville J. A. Coughlan,† Akio Kawai,‡ and Evan J. Bieske*,† †
School of Chemistry, University of Melbourne, Melbourne, Victoria 3010, Australia Department of Chemistry, Graduate School of Science and Engineering, Tokyo Institute of Technology, Ohokayama, Meguro-ku, Tokyo 152-8551, Japan
‡
S Supporting Information *
ABSTRACT: An ion mobility spectrometer (IMS) with an electrospray ion source is used to investigate photo and thermal isomerization of photoactive molecules in the electrospray syringe. A light emitting diode adjacent to the syringe establishes a photostationary state that relaxes thermally toward the more stable isomer once illumination ceases. The arrangement is demonstrated by measuring Z−E thermal isomerization rates for several azoheteroarene compounds. The IMS technique has a distinct advantage over UV−vis spectrophotometry for measuring isomer populations in situations where there are multiple isomers with overlapping absorption profiles. In another development, an LED array adjacent to the silica capillary connecting the syringe to the electrospray ion source, is used to activate photochromic molecules, and investigate sequential photoswitching events.
M
olecular isomerization plays a key role in organic synthesis, the properties and function of biological molecules, including retinal and lipids, and in the operation of molecular photoswitches that are proposed elements of lightactivated logic gates and molecular motors.1−3 Generally, photoisomerization is extremely rapid, occurring on a subnanosecond time scale, whereas thermal isomerization can be much slower, often happening on time scales exceeding 1 ms. Slow Z−E thermal isomerization is conventionally characterized using UV−vis spectrophotometry, fluorescence spectroscopy, or time-resolved NMR.4 However, measuring isomer populations and determining reaction rates can be difficult when reactants, products, intermediates and degradation products have overlapping or indistinct spectral signals. Consequently, there is a need for alternative techniques for measuring isomer populations and isomerization rates. One strategy involves combining online electrospray ionization and mass spectrometry, an approach employed to monitor reactants and products in kinetic studies of bioorganic systems, enzymology and protein folding.5−7 This approach features sub-100 ms time resolution, reflecting the throughput and duty cycles of conventional mass spectrometers.8 Alternatively, information on reaction rates can be obtained by varying capillary lengths or flow rates to modify reaction times.9 Although single-stage mass spectrometers are suited to quantifying reactants and products having different masses, they are normally unable to determine abundances of isobaric ions, hampering characterization of isomerization reactions. In this letter we demonstrate that Z−E thermal isomerization rates for 2-phenylazo-methylimidazolium cations (Figure © XXXX American Chemical Society
Figure 1. 2-Phenylazo-methylimidazolium cations considered in this study. Key: 2PA-Mmim, 2-phenylazo-1,3-dimethylimidazolium; 2PABmim, 2-phenylazo-1-butyl-3-methylimidazolium; 2PA-Hmim, 2phenylazo-1-hexyl-3-methylimidazolium; and 2PA-MOEmim, 2-phenylazo-1-methoxyethyl-3-methylimidazolium.
1)10−13 can be measured by analyzing the ion population using a drift tube ion mobility spectrometer. Ion mobility spectrometry (IMS) is a gas-phase analog of electrophoresis, whereby isomers are separated based on their collision cross sections with an inert buffer gas,14,15 and has found widespread use for distinguishing isobaric ions, including Z and Eisomers.14,16−21 The target azoheteroarene cations (Figure 1) satisfy the requirements for molecular photoswitches, including high photoisomerization quantum yield, reversible switching, and resilience to photoinduced damage.3 The cations are also ideal Received: October 11, 2016 Accepted: December 6, 2016 Published: December 6, 2016 A
DOI: 10.1021/acs.analchem.6b04000 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
Figure 2. Illustration of the ion source and ion mobility mass spectrometer. The LED arrays adjacent to syringe and the capillary include 627, 520, 450, and 385 nm LEDs (see photographs in the Supporting Information). The drift region typically has N2 buffer gas at 6 Torr and a drift field of 44 V cm−1.
Figure 3. Example [2PA-MOEmim]+ ATDs obtained with dark and 385 nm irradiation conditions. The mobility resolution for the peaks is 70−80, consistent with them being associated with singly charged ions.28
subjects for exploring the utility of IMS for monitoring the kinetics of molecules in solution because their E and Z isomers are easily resolved by IMS, they efficiently photoisomerize from the more stable E-isomer to the less stable Z-isomer following exposure to near-UV light, and the Z-isomer thermally isomerizes back to the E-isomer over minutes to hours at room temperature. The current work builds on earlier IMS investigations of E-Z photoisomerization in solution.22−27 For example, E-Z photoisomerization of 4-[4-{2-(pyridin-4-yl)ethenyl]phenyl}2,2′:6′,2″-terpyridine in methanol was monitored using a traveling wave IMS.22 In this case, isomerization in the electrospray ion source prevented direct determination of isomer abundances, a drawback overcome by coordinating the molecules to Fe2+ to “freeze out” the isomer populations, following which the metal complexes were analyzed with IMS to determine isomer abundances. In other examples, IMS was deployed to monitor the photoconversion of protonated merocyanine to protonated sypropyran23,24 and bolaamphiphile−amphiphile photoisomerization, initiated in the tip of a nanoelectrospray source.27 The IMS instrument is illustrated in Figure 2. Briefly, a 10− 100 μmol L−1 solution of azoheteroarene dissolved in methanol was loaded into a syringe that was gently stirred with a magnetic flea. The syringe was connected to an electrospray ion source by a short section (≈300 mm) of transparent silica capillary. Flow rates were typically 10−40 μL min−1. Light from an LED array (LUMILEDS Rebel Series, 627, 520, 450, and 385 nm) was used to photoactivate the solution in the syringe
Figure 4. (a) ATDs for [2PA-Bmim]+ at 3 selected times following 385 nm irradiation of the solution in the syringe; (b) Z and E-isomer abundances from ATDs fit with first-order kinetic functions; (c) reversion kinetics probed at 385 nm in a UV−vis spectrophotometer. Inset in part c is UV−vis absorption spectra of the E-isomer (dark) and E + Z-isomers. The arrow indicates the 385 nm probe wavelength.
and establish a photostationary state. Alternatively, if desired, the solution could be illuminated in a short section of the capillary by a second LED array, which together with the capillary was encased in a reflective metal tube. The LED arrays were switched through a computer controlled relay board and, B
DOI: 10.1021/acs.analchem.6b04000 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
are consistent with calculated collision cross sections for the E and Z-isomers (see the Supporting Information) and with the UV−vis absorption spectra of the dark and irradiated solutions (inset in Figure 4c). The three other 2-phenylazo-methylimidazolium compounds produce similar ATDs and have essentially identical UV−vis absorption spectra with a weak n → π* transition in the 550− 420 nm range and a much stronger π → π* transition in the 420−300 nm range.10−13 The E-isomer absorption at 385 nm is 4−5-fold more intense than the Z-isomer absorption, consistent with net E−Z photoisomerization at 385 nm. Ceasing illumination of the syringe by 385 nm light was followed by a progressive decline of the Z-isomer peak and corresponding growth of the E-isomer peak due to Z−E thermal reversion. To monitor the reversion kinetics, we accumulated sequential ATDs at short intervals (ranging from 3 s for [2PA-Mmim]+ to 30 s for [MOEmim]+) for time periods that extended for 5000 s. Example ATDs for [2PA-Bmim]+ are shown in Figure 4a. All ATDs were normalized to have the same total area to account for slow drift in the electrosprayed ion current. The Z-isomer and E-isomer ATD peaks were integrated and fit with first-order kinetic curves, as shown in Figure 4b, yielding a [2PA-Bmim]+ thermal isomerization rate constant of 6.8 ± 0.2 × 10−4 s−1 at 294 K. Rate constants for Z−E thermal reversion for the four 2phenylazo-methylazoimidazolium cations are summarized in Table 1. For each azoheteroarene cation, there is good agreement between the IMS rates and rates determined using UV−vis spectrophotometry by monitoring the Z-isomer absorption at 385 nm (see Figure 4), with any variation presumably reflecting a slight temperature difference for the two sets of measurements. Examination of the isomerization rate constants (Table 1) reveals that a longer alkyl substituent is associated with more rapid Z−E thermal isomerization, agreeing with previous measurements using UV−vis spectrophotometry.12,29 The relatively slow isomerization of the [2PA-MOEmim]+ cation is presumably due to hydrogen bonds between the ether group and methanol solvent molecules hindering rearrangement (see the Supporting Information). Indeed, studies of similar azobenzene systems revealed that Z−E thermal isomerization is slowed considerably in solvents that hydrogen bond with the substituent chain.30−32 In the present experimental setup, the sampling time (≈1 s) required for each ATD makes the technique best suited to monitoring thermal isomerization processes occurring over time scales of minutes to hours. The IMS methodology has several advantages over conventional UV−vis spectrophotometry for monitoring isomer populations, chiefly that isomerization kinetics can be followed simultaneously for several species undergoing photo- or thermally induced interconversion. For example, all four 2phenylazo-methylimidazolium cations could be characterized from a single mixed solution with the quadrupole mass filter tuned alternately to the mass of each cation. Moreover, the technique’s versatility would be enhanced if the quadrupole mass filter was replaced by a time-of-flight mass spectrometer, allowing simultaneous detection of ions with different masses. Another advantage of the technique is the ability to monitor isomer populations of charged species that do not absorb UV or visible light. Finally, we demonstrate that IMS can be used to monitor sequential photoswitching in the capillary connecting the syringe to the electrospray ion source. Two [2PA-MOEmim]+
Table 1. Thermal Isomerization Rate Constants for 2Phenylazo-Methylimidazolium Cations in Methanol at 294 K k ( × 10−4 s−1)
k ( × 10−4 s−1)
species
ion mobility
UV−vis
[2PA-Mmim]+ [2PA-Bmim]+ [2PA-Hmim]+ [2PA-MOEmim]+
4.2 ± 0.1 6.8 ± 0.2 17.6 ± 0.4 1.18 ± 0.02
4.6 ± 0.1 7.1 ± 0.2 18.2 ± 0.5 1.21 ± 0.01
Figure 5. (a) ATDs for [2PA-MOEmim]+ with illumination of the solution in the syringe at 385 nm and with irradiation in the syringe at 385 nm and in the capillary at 450 nm and (b) depletion of the Zisomer (formed by 385 nm irradiation of the solution in the syringe) with 450 nm irradiation of the capillary for 3 different capillary flow rates.
as explained below, can be operated to study sequential twocolor photoswitching processes. The electrospray ionization source and IMS sections of the apparatus have been discussed elsewhere.28 Electrosprayed ions were transferred via a heated capillary into an ion funnel, the RF drive voltage for which was minimized to limit collisional heating and isomerization of the azoheteroarene ions. Pulses of ions were injected at 20 Hz from the ion funnel through an ion gate into a 0.9 m drift region where they were propelled by a ≈44 V cm−1 electric field through N2 buffer gas (pressure ≈ 7 Torr). At the end of the drift region, a second ion funnel was used to collect the ions and introduce them into a differentially pumped octopole ion guide and quadrupole mass filter that mass-selected the ions before they reached the ion detector. The detector was connected to a multichannel scaler that generates a histogram of ion counts against arrival time, corresponding to an arrival time distribution (ATD). As shown in Figure 3, the ATD of [2PA-MOEmim]+ electrosprayed from a solution shielded from light displays a single peak, evidence for the presence of just one isomer. In contrast, following irradiation of the solution with 385 nm light, a second, faster isomer also appears. The slower peak can be readily assigned to the more stable E-isomer, whereas the faster peak corresponds to the more compact Z-isomer which, as shown in previous studies, is formed when the solution is exposed to 385 nm light.10−13 These ATD peak assignments C
DOI: 10.1021/acs.analchem.6b04000 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry ATDs are shown in Figure 5a, the first of which was recorded with 385 nm irradiation of the solution in the syringe to produce the Z-isomer. The second ATD was recorded with 450 nm irradiation of the capillary to convert the Z-isomer back to E-isomer, a process that occurs because the Z-isomer absorbs more strongly at 450 nm than the E-isomer (see solution spectra in Figure 4c). Figure 5b shows the Z-isomer abundance as a function of flow rate, where in each measurement the 450 nm LED illuminating the capillary was switched on at 30 s and off at 150 s. As expected, the Z-isomer signal responds with a time lag inversely proportional to the flow rate. This 2-color back-and-forth photoswitching between Z and E isomers can obviously be generalized to multistage reactions and isomerizations. There are distinct advantages in combining capillaries or microflow reactors with IMS detection to study photochemical processes, namely, small sample volumes, short photoinitiation times, and ability to probe multistage switching processes.
■
(11) Asaka, T.; Akai, N.; Kawai, A.; Shibuya, K. J. Photochem. Photobiol., A 2010, 209, 12−18. (12) Kawai, A.; Kawamori, D.; Monji, T.; Asaka, T.; Akai, N.; Shibuya, K. Chem. Lett. 2010, 39, 230−231. (13) Yoshida, T.; Monji, T.; Kawamori, D.; Akai, N.; Shibuya, K.; Kawai, A. Chem. Lett. 2013, 42, 1490−1492. (14) Eiceman, G. A.; Karpas, Z.; Hill, H. H., Jr. Ion Mobility Spectrometry, 3rd ed.; CRC Press: Boca Raton, FL, 2014. (15) May, J. C.; McLean, J. A. Anal. Chem. 2015, 87, 1422−1436. (16) Zhong, Y.; Hyung, S.-J.; Ruotolo, B. T. Expert Rev. Proteomics 2012, 9, 47−58. (17) Bohrer, B. C.; Merenbloom, S. I.; Koeniger, S. L.; Hilderbrand, A. E.; Clemmer, D. E. Annu. Rev. Anal. Chem. 2008, 1, 293−327. (18) Wyttenbach, T.; Pierson, N. A.; Clemmer, D. E.; Bowers, M. T. Annu. Rev. Phys. Chem. 2014, 65, 175−196. (19) May, J. C.; McLean, J. A. Annu. Rev. Anal. Chem. 2016, 9, 387− 409. (20) Ewing, M. A.; Glover, M. S.; Clemmer, D. E. J. Chromat. A 2016, 1439, 3−25. (21) Lanucara, F.; Holman, S. W.; Gray, C. J.; Eyers, C. E. Nat. Chem. 2014, 6, 281−294. (22) Santos, J. J.; Toma, S. H.; Lalli, P. M.; Riccio, M. F.; Eberlin, M. N.; Toma, H. E.; Araki, K. Analyst 2012, 137, 4045−4051. (23) Rogers, R. A.; Rodier, A. R.; Stanley, J. A.; Douglas, N. A.; Li, X.; Brittain, W. J. Chem. Commun. 2014, 50, 3424−3426. (24) Markworth, P. B.; Coughlan, N. J. A.; Adamson, B. D.; Goerigk, L.; Bieske, E. J. Phys. Chem. Chem. Phys. 2015, 17, 25676−25688. (25) Coughlan, N. J. A.; Adamson, B. D.; Gamon, L.; Catani, K.; Bieske, E. J. Phys. Chem. Chem. Phys. 2015, 17, 22623−22631. (26) Coughlan, N. J. A.; Wallace, C. M.; Adamson, B. D.; Bieske, E. J. J. Phys. Chem. A 2016, 120, 6557−6562. (27) Urner, L. H.; Thota, B. N. S.; Nachtigall, O.; Warnke, S.; von Helden, G.; Haag, R.; Pagel, K. Chem. Commun. 2015, 51, 8801−8804. (28) Adamson, B. D.; Coughlan, N. J. A.; Markworth, P. B.; Continetti, R. E.; Bieske, E. J. Rev. Sci. Instrum. 2014, 85, 123109. (29) Otsuki, J.; Suwa, K.; Sarker, K. K.; Sinha, C. J. Phys. Chem. A 2007, 111, 1403−1409. (30) Schanze, K. S.; Mattox, T. F.; Whitten, D. G. J. Am. Chem. Soc. 1982, 104, 1733−1735. (31) Schanze, K. S.; Mattox, T. F.; Whitten, D. G. J. Org. Chem. 1983, 48, 2808−2813. (32) Gille, K.; Knoll, H.; Quitzsch, K. Int. J. Chem. Kinet. 1999, 31, 337−350.
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b04000. Details of calculated isomer geometries, energies, and collision cross sections, and [2PA-MOEmim]+−methanol hydrogen bonded structures (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
James N. Bull: 0000-0003-0953-1716 Evan J. Bieske: 0000-0003-1848-507X Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported under the Australian Research Council’s Discovery Project funding scheme (Project Numbers DP150101427 and DP160100474).
■
REFERENCES
(1) Dugave, C.; Demange, L. Chem. Rev. 2003, 103, 2475−2532. (2) Beharry, A. A.; Woolley, G. A. Chem. Soc. Rev. 2011, 40, 4422− 4437. (3) Bandara, H. M. D.; Burdette, S. C. Chem. Soc. Rev. 2012, 41, 1809−1825. (4) Connors, K. A. Chemical Kinetics: The Study of Reaction Rates in Solution; Wiley-VCH: New York, 1990. (5) Norris, A. J.; Whitelegge, J. P.; Faull, K. F.; Toyokuni, T. Biochemistry 2001, 40, 3774−3779. (6) Ben-Nissan, G.; Sharon, M. Chem. Soc. Rev. 2011, 40, 3627− 3637. (7) Lee, J. K.; Kim, S.; Nam, H. G.; Zare, R. N. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 3898−3903. (8) Urban, P. L.; Chen, Y.-C.; Wang, Y.-S. Time-Resolved Mass Spectrometry: From Concept to Applications; John Wiley & Sons, Ltd.: Chichester, U.K., 2016. (9) Wilson, D. J.; Konermann, L. Anal. Chem. 2003, 75, 6408−6414. (10) Ida, A.; Cohen, B.; Asaka, T.; Kawai, A.; Organero, J. A.; Shibuya, K.; Douhal, A. Phys. Chem. Chem. Phys. 2011, 13, 20318− 20325. D
DOI: 10.1021/acs.analchem.6b04000 Anal. Chem. XXXX, XXX, XXX−XXX