Chester Clinic

Recent developments in liquid chromatography–mass spectrometry and related techniques

Contents lists available at Journal of Chromatography A Recent developments in liquid chromatography–mass spectrometry and related Michal Holˇcapek , Robert Jirásko, Miroslav Lísa Department of Analytical Chemistry, Faculty of Chemical Technology, University of Pardubice, Studentská 573, 53210 Pardubice, Czech Republic This review summarizes the state-of-art in liquid chromatography–mass spectrometry (LC–MS) and Available online 31 August 2012 related techniques with the main focus on recent developments in the last decade. LC–MS records an enormous growth in recent years due to the application potential in analytical chemistry, biochemistry, pharmaceutical analysis, clinical analysis and many other fields, where the qualitative and quantitative Mass spectrometry characterization of complex organic, bioorganic and organometallic mixtures is needed. Beginners and moderately experienced LC–MS users may be confused by the number of different LC–MS systems on the market, therefore an actual overview of mass spectrometers designed for the LC–MS configuration and matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) from main manufacturers is compiled here together with an independent assessment of their advantages and limitations. Current trends in terms of mass analyzers, ionization techniques, fast LC–MS, LC–MALDI-MS, ion mobility spec- trometry used in LC–MS, quantitation issues specific to MS and emerging mass spectrometric approaches complementary to LC–MS are discussed as well.
2012 Elsevier B.V. All rights reserved.
Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
constant improvement of operating parameters of mass spectro- meters, the situation is reviewed here again with two main goals: Several reviews devoted to new trends in instrumental develop- (1) to highlight recent innovations in LC–MS especially during the ments in LC–MS and related techniques were published in previous last decade, (2) to prepare a list of mass spectrometers currently special issues of "Mass spectrometry: Innovation and application" offered by main manufacturers for LC–MS and MALDI-MS config- published in Journal of Chromatography A in other places urations together with their technical specifications ( as well to numerous new LC–MS developments and the and the comparison of their application potential. This task could not be performed without the close cooperation with representa- tives of individual manufacturers and their websites we ∗ Corresponding author. Tel.: +420 466 037 087; fax: +420 46 603 7068.
do our best to prepare fair and balanced scientific overview with- E-mail address: (M. Holˇcapek).
out any advertisement of individual technical solutions presented 0021-9673/$ – see front matter 2012 Elsevier B.V. All rights reserved.
M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 Overview of commercial mass spectrometers designed for LC–MS with their technical specifications provided by individual manufacturers.
Instrument name, manufacturer Mass accuracy (ppm) Acquisition speed (Hz) (FWHM defined at m/z) Flexar SQ 300 MS, PerkinElmer LCMS-2020, Shimadzu LC/MS Purification System, Gilson MSQ plus, Thermo Scientific SQ Detector 2, Waters Amazon Speed ETD, Bruker Daltonics LCQ Fleet, Thermo Scientific LTQ Velos Pro, Thermo Scientific LCMS-8030, Shimadzu TQ Detector, Hitachi Triple Quad 5500, AB SCIEX TSQ Vantage, Thermo Scientific XEVO TQ-S, Waters QTRAP 5500, AB SCIEX 6230 TOF, Agilent 24,000 (m/z 1522) AxION 2 TOF MS, PerkinElmer 100,000 (m/z 609) micrOTOF II focus, Bruker Daltonics XEVO G2 TOF, Waters LCMS-IT-TOF, Shimadzu 10,000 (m/z 1000) maXis 4G, Bruker Daltonics 60,000 (m/z 1222) 30 (MS), 10 (MS/MS) micrOTOF-Q II, Bruker Daltonics TripleTOF 5600, AB SCIEX 25 (MS), 100 (MS/MS) XEVO G2 QTof, Waters 6550 QTOF, Agilent Synapt G2-S HDMS, Waters Exactive, Thermo Scientific 100,000 (m/z 200) 10 (at RP = 10,000) Q Exactive, Thermo Scientific 140,000 (m/z 200) 12 (at RP = 17,500) Orbitrap Elite, Thermo Scientific 240,000 (m/z 400) 8 (at RP = 15,000) SolariX 15T, Bruker Daltonics 2,500,000 (m/z 400) LTQ FT Ultra 7T, Thermo Scientific 750,000 (m/z 400) 2 (at RP = 50,000) a If manufacturers have more instruments in particular series, then only the instrument with the best performance is listed here. Individual manufacturers take the full responsibility for the correctness of technical specifications. Instruments in individual classes are sorted alphabetically according to the instrument name. This list contains only main manufacturers and may not be comprehensive.
b Acquisition speed for low RP mass analyzers is typically specified by manufacturers in Da/s, but we have recalculated these values into Hz units for the mass range of in this review. Authors take no responsibility for the correctness incorrect terms is Mass Spectrometry Desk Reference compiled by of technical specifications provided by manufacturers. MS terms Sparkman some recommendations are not consistent with and definitions used in this paper are in agreement with the IUPAC previously mentioned glossary.
sponsored project the Standard Terms and Definitions for Mass Spectrometry This database is now being maintained and 2. Overview of up-to-date mass spectrometers
updated by Murray to build a reference tool and a glossary of MS terms with 780 entries as of July 12, 2012 Recommended The market of MS and LC–MS is extremely dynamic and MS terms related to the separation sciences have been published individual manufacturers invest into the development of new in the previous special MS issue in Journal of Chromatography A technologies. This competition has a positive effect on frequent valuable literature source on the use of correct and launches of new products and technical solutions. For better Common parameters of mass spectrometers used in LC–MS.
m/z range (upper limit) Acquisition speed a TOF, Orbitrap and ICR also include common hybrid configurations with Q or LIT as the first mass analyzer.
b Qs with hyperbolic rods provide mass accuracies better than 5 ppm.
M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 Overview of commercial mass spectrometers designed for MALDI-MS with their technical specifications provided by individual Instrument, manufacturer Mass accuracy (ppm) Laser (wavelength, frequency) FWHM (defined at LTQ XL, Thermo Scientific N2 (337 nm, 60 Hz) (depending on scan Autoflex Speed, Bruker Daltonics 26,000 (m/z 3147) Nd:YAG (355 nm, 1000 Hz) Axima Confidence, Shimadzu 15,000 (m/z 3660) N2 (337 nm, 50 Hz) Axima Resonance, Shimadzu N2 (337 nm, 10 Hz) MALDI Synapt G2-S HDMS, Waters 32,000 (m/z 3495) Nd:YLF (349 nm, 1000 Hz) Axima Performance, Shimadzu 20,000 (m/z 3660) N2 (337 nm, 50 Hz) JMS-S3000 SpiralTOF, Jeol 60,000 (m/z 2093) Nd:YLF (349 nm, 250 Hz) TOF/TOF 5800 System, AB SCIEX 33,000 (for m/z range Nd:YLF (345 nm, 1000 Hz) UltrafleXtreme, Bruker Daltonics 40,000 (m/z 3147) Nd:YAG (355 nm, 1000 Hz) MALDI LTQ Orbitrap XL, Thermo 100,000 (m/z 400) N2 (337 nm, 60 Hz) SolariX 15T, Bruker Daltonics 2,500,000 (m/z 400) Nd:YAG (355 nm, 1000 Hz) a If manufacturers have more instruments in particular series, then only the instrument with the best performance is listed here. Individual manufacturers take the full responsibility for the correctness of technical specifications. Instruments in individual classes are sorted alphabetically according to the instrument name. This list contains only main manufacturers and may not be comprehensive.
overview of current LC–MS systems, we have compiled a list of second sprayer or high-confidence ions from the sample, which commercial mass spectrometers designed for LC–MS is used for the correction of previously done external calibration by for MALDI-MS (with their technical specifications, such as locking particular m/z value example of lock-mass resolving power, resolution, mass accuracy specified for internal internal calibration is the use a subset of high confidence pep- and external calibrations, mass-to-charge (m/z) range and acqui- tide identifications from a first pass database search sition speed. These tables will become obsolete quite rapidly, but improvements of MA can be achieved by software tools to eliminate we still believe that such current snapshot of MS technology as of a lower dynamic range of time-to-digital converters in some mass spring 2012 is useful. typical operating parameters analyzers way of calibrant introduction is the use for five basic types of mass analyzers used in LC–MS.
of dual sprayer in ESI-MS using a rapid modulation (switch in less This review is primarily intended for low to moderately experi- than 70 ms) between ESI emitters external calibration is enced LC–MS users, therefore basic MS terms used in the procedure, where the sample and the calibrant are not present explained here. More detailed information can be found in several in the ion source at the same time. If the mass spectrometer is stable excellent textbooks devoted to MS basic parameter enough without any mass drift, the external calibration may pro- for the characterization of mass analyzer ability to resolve peaks vide almost comparable results, but the time difference between in mass spectra is a resolving power (RP), which is defined as the sample and the calibrant introduction should be as low as the m/z value of particular peak divided by the peak full width at possible, e.g., the calibrant can be introduced between the LC injec- half maximum (FWHM): RP = (m/z)/m/z. The RP must be always tion time and the void time or immediately after the elution of defined for the particular m/z value (e.g., RP is 20,000 at m/z 922), last peak in the chromatogram. The m/z ranges of individual mass because the RP grows with increasing m/z value on condition of analyzers in the maximum possible measure- identical peak width. The older definition of RP (established for ment span of mass detectors. The acquisition speed is typically magnetic sector analyzers) based on two neighboring peaks of expressed in technical specifications in Da/s for low-resolution and identical heights and 10% valley is not used in the current LC–MS in Hz for high-resolution mass analyzers. We have re-calculated practice. The resolution is the inverse of RP expressed as m/z for a specifications for low-resolution analyzers from Da/s to Hz for the given m/z value the above-mentioned example, the reso- typical measurement range of 1000 m/z to allow the comparison lution is calculated as m/z = (m/z)/RP = 922/20,000 = 0.046. Better among all instruments By the way, numerical values of quality of mass analyzer is associated with lower values of resolu- acquisition speed expressed in kDa/s and Hz for 1000 m/z range are tion and higher values of RP (see Mass accuracy (MA) is defined as the relative difference between the experimen- tal m/z value and theoretical m/z value related to this theoretical 2.1. Mass spectrometers in LC–MS coupling value including the sign (plus or minus) and expressed in ppm: MA = 106 × ((m/z)exp − (m/z) The best values of MA The standard resolution of Q analyzer is a unit resolution, while theor)/(m/z)theor. are achieved with the internal calibration, i.e., the sample and somewhat better resolution can be obtained at cost of a lower the internal calibrant are introduced into the ion source at the ion transmission and therefore lower sensitivity, as understood same time. The introduction of internal calibrant during LC–MS from the stability diagram resolution of spherical ITs and may be impractical in some instances (e.g., interference of cal- especially linear ion traps (LITs) is slightly higher compared to the ibrant with the chromatographic separation or identical masses Q analyzer, but the inverse relation between the resolution and of calibrant and analyte). The "lock-mass" calibration is based the sensitivity based on the three-dimensional stability diagram on well-defined ion with known elemental composition coming is again valid. The typical MA of low-resolution mass analyzers from the background (known impurities occurring from previous is below 100 ppm (calculated for m/z error 0.1 at m/z 1000), samples, mobile phase, air components, etc.), introduced by the but such low MAs are not sufficient for the elemental formula

M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 determination. Q rods with an ideal hyperbolic profiles provide a higher resolution than for regular round Q rods with the MA <5 ppm comparable to high-resolution mass analyzers (see QqQ TSQ Vantage in Another advantage of QqQ with hyperbolic rods is a narrow isolation width of precursor ions used for the selected reaction monitoring (SRM), while the drawback is lower acquisition speed. The m/z range is typically up to m/z 2000–3000 for Qs and m/z 4000–6000 for ITs The special Q with significantly higher transmission for large ions (up to m/z 32,000) can be fabricated in the transmission non-resolving mode only, which means that such Qs can be used only for the transmission of all ions but not for their mass resolution.
The comparison of high-resolution mass analyzers listed in needs additional comments. The ICR has superior val- ues of RP and MA among all analyzers followed by Orbitrap and then TOF based analyzers. It is also important to realize the rela- tion between the RP and the acquisition speed, because the highest acquisition speed of Fourier transform (FT) mass analyzers ICR and Orbitrap does not correspond to their highest RP. The best parame- ters reported for FT analyzers require longer acquisition times due to the fact that higher number of image currents has to be recorded.
The incorporation of FT mass analyzer into the fast LC–MS concept is possible, but at cost of significantly reduced RP in comparison to the best values reported for slow scan speeds in TOF Fig. 1. Overview of installed LC–MS systems in the Czech Republic according to the
mass analyzers have the highest scanning speed among all mass type of mass analyzer (in total 233 systems, update September 2011).
analyzers and also their m/z range is theoretically unlimited (e.g., measurements in hundred thousands Da in MALDI-TOF linear con- dominant in the quantitative analysis, because SRM scan typical figuration), but the m/z range of TOF based analyzers in LC–MS for this type of analyzer is a golden standard for any LC/MS or systems is limited to several tens of thousands. The linear dynamic shotgun MS based quantitation. Hybrid Q-TOF instrument is range (depends on the particular instrument and applica- the most common in the structural characterization due to the tion, but in general FT mass analyzers exhibit slightly lower linear possibility of measurements of high-MA in both full-scan and dynamic range.
MS/MS modes. In recent years, the shift from low-resolution The price parameter in intended only as a rough guide toward high-resolution systems including TOF based mass ana- for typical configurations, but the real price strongly depends on lyzers and ultrahigh-resolution (RP > 100,000) FT mass analyzers the particular configuration and individual offers from the man- (Orbitrap and ICR) occurs, because ultrahigh-RP and ultrahigh-MA ufacturer. In general, Q analyzer is simplest and cheapest device values open new possibilities in both qualitative and quantitative followed by spherical and linear ITs. TOF analyzer is the cheapest analyses, e.g., SRM or even SIM approaches based on ultrahigh-RP high-resolution mass analyzer with some impressive characteris- and therefore the possibility of very narrow widths for precursor tics in terms of acquisition speed, m/z range and relatively good ion isolation, which leads to increased selectivity and sensitivity.
RP and MA. FT mass analyzers, Orbitrap and especially ICR, are high-end MS technologies with the best operational parameters, but the instrumental complexity is obviously reflected in increased investment costs. be understood as an overview of common operating parameters but excluding extreme values obtained at specific conditions, e.g., slow scan speed, reduced m/z shows an overview of installed LC–MS systems in the Czech Republic. The world statistics would be more representa- tive, but reliable data does not exist unlike to our statistics for this local central European market, where we monitor and annu- ally update the situation from the first LC–MS system installed at the University of Pardubice in 1995. The local statis- tics could be affected by regional differences, but a reasonable agreement with trends in the world. Some differences between be explained – at least in part – by the different way of the preparation of these graphs. the world statistics based on the Web of Science search, while is prepared from exact numbers of installed LC–MS systems in the Czech Republic. In our opinion, certain overestimation of top-class expensive instruments occur in because new technologies are purchased primarily for research purposes yielding a higher number of papers compared to low-cost and low-resolution mass analyzers used mainly for routine analyses and quantitation in industrial and clinical laboratories with a lower publication activity.
Prevailing LC–MS configurations (see are based Fig. 2. Relative use of individual types of mass analyzers in LC–MS papers based on
on IT, TOF and Q mass analyzers. QqQ tandem mass analyzer is the Web of Science search from March 1, 2012.
M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 Some papers compared a real performance of different types of modern tandem mass analyzers for particular applications, which Suggested definitions of low, high and ultrahigh resolving power and mass accuracy of mass analyzers.
provides valuable complementary information to LC–MS with QqQ and IT have been compared for the determi- Resolving power (RP, FWHM) Mass accuracy (MA, ppm) nation of 6 pesticides in fruits QqQ provides better linear dynamic range, higher precision, less matrix interferences and bet- ter robustness, while IT provides an excellent sensitivity for product ion measurements. Four LC–MS systems equipped with Q, QqQ, IT and Q-TOF have been compared the quantitative anal- ysis (sensitivity, precision and accuracy) of carbosulfan and its (Orbitrap and ICR). MALDI-QqQ-MS configuration is designed for main transformation products. QqQ provides at least 20-fold higher the sensitive quantitation similarly as for LC-QqQ-MS.
sensitivity compared to other mass analyzers and better linear The instrument characteristics with their advantages and dynamic range. The repeatability (within-day) is slightly better for limitations are more or less identical as for the LC–MS coupling. The Q (5–10%) and QqQ (5–9%) compared to IT (12–16%) and Q-TOF m/z range of MALDI-TOF analyzers in hundred thousands is valid (9–16%). Although the QqQ is more sensitive and precise, mean only for the linear mode, while it is limited to ca. 100,000 or less values obtained by all instruments are comparable. QqQ, TOF and for the reflectron mode. Another issue is the sensitivity, which can Q-TOF were compared for the qualitative and quantitative analy- dramatically decrease for very large m/z values in the range of hun- ses of 10 anabolic steroids in human urine allowed the dred kDa. Measurements of large proteins in the MDa range have detection of all analytes at the minimum required performance been reported in the linear mode interesting configu- limit established by the World Anti-Doping Agency (between 2 and ration of TOF analyzer has been reported recently by Jeol, where the 10 ng/mL in urine). TOF and Q-TOF approaches were not sensitive traveling path of ions is increased approximately up to 17 meters enough to detect some analytes. Most compounds were detected due to the multiple reflections the declared RP = 60,000.
by all techniques, however QqQ was necessary for the detection of MALDI sources can be equipped with 2 basic types of lasers in some metabolites in a few samples. TOF-based analyzers showed a UV/vis region (gas-phase laser (nitrogen laser is used benefit to detect non-target steroids and their metabolites in some in all commercial applications) or solid-state lasers (neodymium- samples. Human liver microsomal incubations with amitriptyline doped yttrium aluminum garnet, Nd:YAG, and neodymium-doped and verapamil were used as test samples, and early-phase "one yttrium lithium fluoride, Nd:YLF). Nitrogen lasers are used as a lab visit only" approaches were used with different instruments golden standard in MALDI-MS especially due to the lower price was the only approach detecting all metabolites, shown and good performance with a wide range of matrices, but they to be the most suitable instrument for elucidating as comprehen- have certain limitations, such as the maximum repetition rate only sive metabolite profile as possible leading also to lowest overall up to 60 Hz and the average life span about 107 shots. The advan- time consumption together with the LIT-Orbitrap approach. The tage of solid-state lasers is the higher repetition rate (>1000 Hz, latter however suffered from lower detection sensitivity and false see and longer life time (109 shots). The combination of negatives, and due to slow data acquisition rate required slower advantages of nitrogen and solid-state lasers is a Nd:YAG laser chromatography. Approaches with QqQ and Q-LIT provided the with a modulated beam profile the superior performance highest amount of fragment ion data for the structural elucidation, in MALDI imaging and LC-MALDI-MS coupling. Infrared lasers in but they were unable to provide high-MA data, suffered from many MALDI were proposed as a valuable alternative to UV/vis lasers due false negatives, and especially with QqQ, from very high overall to increased life time and absorbance of virtually all (bio)organic time consumption.
compounds in the infrared region infrared lasers are still The 2002/657/EC European Commission Decision established not yet available in commercial MALDI setups.
the need to obtain at least three identification points in order to con- firm organic contaminants in animal products, which was applied 3. Current trends in LC–MS
for pesticide analyses in environmental matrices using LC–MS with QqQ, TOF and Q-TOF QqQ instrument allowed the confir- Basic characteristics of the quality of mass analyzer are RP mation of detected pesticides even at very low concentrations and MA. At present time, definitions of high-RP and also high- (ng/L) achieving between four and five identification points when MA are not sufficient to differentiate between high-resolution and adding confirmatory transitions. The direct confirmation with a ultrahigh-resolution mass analyzers, therefore we suggest updated TOF instrument was only feasible for those compounds showing definitions for low, high and ultrahigh RP and MA We sug- sufficient sensitivity, isotopic pattern, or easy in-source fragmen- gest to distinguish three basic categories of RP: low-RP (<10,000), tation. Q-TOF provided up to 20 identification points in a single high-RP (10,000–100,000), and ultrahigh-RP (>100,000). In fact, it run at relatively high concentrations (sub-mg/L). Moreover, TOF- means that most Q and IT mass analyzers belong to the low-RP based mass analyzers allow to finding additional non-target organic category, TOF based analyzers to the high-RP, and the ultrahigh- RP contains two FT mass analyzers—Orbitrap and ICR. The Orbitrap has started to approach closer to parameters typical for ICR mass analyzers after the launch of new type of Orbitrap with 240,000 RP 2.2. Matrix-assisted laser desorption/ionization mass recently Nikolaev et al. published a new ICR cell design spectrometry (MALDI-MS) they demonstrated 24 millions RP at m/z 609 recorded over 3 min for only 7 T magnetic field. This new development in the The widely accepted standard in the MALDI technology is the ICR cell technology will again widen the gap between the ICR and coupling with TOF or TOF/TOF mass analyzers, because both devices Orbitrap. It should be kept in mind that such values of RP cannot are working in a pulse regime and such connection is straightfor- be achieved in the LC–MS time scale.
ward. TOF mass analyzer can be preceded by Q or IT analyzer, where The conventional definition of high-MA is 5 ppm and better.
the limitation of such configurations is the ion transmission and Technical specifications of most recently launched high-RP mass the scanning speed of first Q/IT analyzer. Other alternatives are the analyzers report MA 3 ppm and better even for the external cal- coupling of MALDI source with different types of mass analyzers ibration, for FT mass spectrometers with the internal calibration than TOF, either low-resolution (LIT and QqQ) or high-resolution or the lock-mass approach below 1 ppm (ultrahigh-MA). The best M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 specification for 15 T ICR instrument is reported better than niques can be coupled to MS as well, as illustrated in recent works 0.25 ppm for the internal calibration FT-ICR mass ana- on thin-layer chromatography (TLC) coupled to MALDI-MS lyzers are rather expensive, so their application is typical in or other atmospheric pressure surface sampling/ionization tech- the most demanding analytical tasks, such as proteomics niques as well is simple and cheap technique still used for petroleomics metabolomics The reasonable the routine analysis, for example in lipidomics definition of potential ranges for individual expected elements and the inclusion of isotopic ratios in the searching algorithm may 3.1. Fast LC–MS further improve the reliability of elemental composition deter- mination In fact, the level of MA required for the reliable The group of fast LC–MS techniques comprises various elemental composition determination strongly depends on the approaches with the common goal to achieve the highest sam- actual m/z value, because the number of possible combinations of ple throughput and the good separation efficiency. The most given elements exponentially grows with the m/z value widespread and well established approach is UHPLC small m/z values (ca. below m/z 200), the MA <5 ppm is satisfactory, is based on the use of small particle size (sub-two ␮m particles) but such MA bring too many possible combination in the m/z range at ultrahigh-pressures (up to 1300 bars) yielding fast analyses and of 500–1000. For biomolecules with MW > 1000 Da, MA better than narrow chromatographic peaks. On the other hand, it requires 1 ppm may not be sufficient for the unequivocal determination of higher acquisition speed of mass spectrometer to obtain enough elemental formula.
sampling points for the reliable peak integration. Typical peak The miniaturization is an important issue considered in all widths in routine UHPLC–MS bioanalyses are 3–10 s fields of analytical instrumentation including both parts of LC–MS peak widths in the fast/ultrafast LC–MS are generally in the range coupling. The trend in LC is a reduction of the column diameter 1–3 s, but they can be narrower than 1 s under well optimized con- from standard bore (3–4.6 mm ID) or narrow bore (1–3 mm ID) to ditions For good reproducibility and precision in LC–MS capillary columns (<1 mm ID) or even separations on chips quantitation, at least 12–15 points per chromatographic peak are Two commercial solutions for chip-based separations are offered recommended, but the current practice often rely on the lower by Agilent Technologies and Waters The reading of number of data points per peak (8–10) in the qualitative and semi- specialized reviews is recommended for more details quantitative analysis, but it may compromise the peak shape. The terms of LC–MS, capillary columns and chips work with flow rates minimum acquisition speed to acquire 10 sampling points per peak in the range of nL/min, which is ideally suited for the coupling with is 3–10 Hz for the average peak width 1–3 s and 10–20 Hz for the nanoelectrospray ionization average peak width 0.5–1 s, but of course higher scanning speeds The two-dimensional (2D) LC either in off-line on-line are useful to generate more sampling points per peak for a better mode the separation of highly complex mixtures, quantitation. Such acquisition speeds are achieved by modern TOF such as in proteomics the analysis of nat- based mass analyzers and also some ion traps (see the ural compounds The multidimensional option is considered acquisition speed of individual analyzers).
not only on the LC side of the LC–MS system, but parallel (multidi- Other approaches used in the fast LC–MS are the use of core-shell mensional) use of different types of mass spectrometers have been particles LC (HTLC) mono- described in LC–MS coupling with the goal to obtain complemen- lithic columns Several specialized reviews and book tary information from various MS configurations most chapters on these novel approaches have been published recently comprehensive LC–MS system reported so far has been developed our opinion, the use of core-shell particles with ID <3 ␮m by Byrdwell for the parallel use of three different mass spectrome- is highly promising area because comparable results as for ters (Q-LIT, QqQ and IT) plus three additional non-MS detectors UHPLC–MS can be obtained on conventional LC–MS systems with- (UV detector, evaporative light-scattering detector and corona out the need of additional investments. shows the direct charged aerosol detector) goal of this "dilute-and-shoot" comparison of individual approaches in terms of maximum peak approach is to obtain the maximum amount of analytical informa- capacity vs. throughput. Detailed comparison and discussion of tion in a single run, as illustrated on examples of vitamin D3 and related aspects is available in the original paper triacylglycerols. Each mass spectrometer is used for obtaining com- The fastest mass analyzer is obviously the TOF analyzer plementary information from different scan types, Q-LIT is used for with common acquisition speeds 10–50 Hz, which fits well with recording SIM, SRM and enhanced MS scans in APCI mode, QqQ fast/ultrafast LC requirements. The technical specification of fastest operates in the full-scan APCI mode, and IT provides information Q-TOF instrument on the market declares 100 Hz for tandem mass spectrometry (MS/MS) measurements which allows Nowadays, important tasks in LC–MS and the analytical numerous parallel SRM scans even in fast/ultrafast LC–MS. The chemistry in general are the sample throughput, automation and fastest TOF instrument for LC–MS reports 200 Hz acquisition speed non-supervised system operation, because the clinical studies of The additional parameter important for the quantitation ten have an enormous number of samples to be analyzed. The is the linear dynamic range, which is at least 5 orders of magnitude robotic NanoMate TriVersa system can be used for this purpose in or better for modern Q, LIT, TOF instruments and their combinations various operation modes including the automatic liquid extraction from the tissue surface followed by ESI-MS analysis auto- mated fraction collection followed by ESI-MS analysis in supervised 3.2. Ionization techniques in LC–MS or non-supervised mode .
For many decades, the role of chemical derivatization was fully The status quo in ionization techniques is that nearly all recognized in gas chromatography–mass spectrometry (GC–MS), LC–MS systems are equipped with ESI, sometimes accompanied where the derivatization enabled the analysis of analytes with by atmospheric pressure chemical ionization (APCI) for less polar insufficient volatility Now the potential of derivatiza- compounds and the normal-phase LC operation tion is also realized in LC–MS to increased sensitivity shows the relative percentage of individual atmospheric pressure improved bioanalytical quantitation improved reten- ionization techniques used in published LC–MS papers accord- tion behavior of problematic analytes possible integration ing to the Web of Science search, where the dominant role of of derivatization of polar analytes and their extraction followed ESI (82% of papers) is evident. Main application areas of ESI by LC–MS determination present, planar separation tech- are in the characterization of biomolecules, ionic and very labile

M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 Fig. 4. Relative use of individual atmospheric pressure ionization techniques in
LC–MS papers based on the Web of Science search from March 1, 2012.
and synthetic polymers (16% of papers, The atmospheric pressure photoionization (APPI) is less widespread (2% of papers, compared to two above mentioned ionization techniques, which can be probably explained by a comparable application range as for APCI. The application potential of ESI, APCI and ESI was com- pared for 5 polar pharmaceuticals ESI showed the best performance in terms of sensitivity and selectivity.
Ionization mechanisms in APPI rather complex con- sisting of two basic ionization modes: without dopant and with the assistance of dopant. The ionization process in APPI is initiated by photons emitted by a discharge lamp (typically krypton, 10 eV and minor 10.6 eV). These photons ionize compounds with ionization energies lower than their energy (10 eV), which includes analyte molecules, but not typical gases and solvents used in LC–MS separation and nebulization processes. Analyte molecules are ionized rather selectively without background interferences. The ionization of analytes is dependent on their ionization energies rather than their proton affinities unlike to ESI and APCI. Toluene and acetone are the most common dopants in APPI often providing significantly better sensitivity for low polar analytes compared to ESI/APCI techniques. Dopants are first ionized by photoionization and then they ionize target analytes by ion-molecules reactions, e.g., by the proton transfer in the positive-ion mode. The presence of radical molecular ions M+. in positive-ion APPI mode is not rare unlike to ESI/APCI and they can be formed by direct photoioniza- tion (conditions without dopant) or charge-exchange mechanisms, which depend mainly on the solvent polarity, flow rate and the Fig. 3. Performance comparison of LC strategies in terms of throughput (t
presence of additives Recently, the experimental and P = 100) and maximum peak capacity (Pmax for tgrad = 3 h) in the gradient elution for quantum mechanical studies were used to revisit the mechanism model compounds: (A) butylparaben (MW = 200 Da), (B) rutin (MW = 600 Da), and of [M+H]+ formation in APPI show that both electron (C) peptide triptorelin (MW = 1300 Da).
transfer and hydrogen transfer can occur as a concerted reaction Adapted with a permission from through the ion-molecular complex precursor state.
The present LC–MS practice moves toward fast/ultrafast LC organic and organometallic compounds Fundamentals, analyses for high-throughput, which puts demands on the speed of instrumentation and biological applications of ESI and MALDI have mass spectra recording. For numerous applications, mass spectra been described in a monograph edited by Cole recently recorded in both polarity modes provide valuable complemen- the current knowledge on the mechanism of ion formation in ESI tary information both for qualitative and quantitative analyses, has been reviewed application potential of APCI is mainly because certain compound classes can be ionized only in one polar- in the area of medium polar to non-polar organic compounds ity mode, e.g., (poly)sulfates and (poly)sulfonates do not provide

M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 a signal in the positive-ion mode unless the presence of other functional groups with easy ionization in the positive-ion mode For this reason, the mass spectrometer with a fast polarity switching is desirable (<50 ms for low-resolution and <1 s for high- resolution mass analyzers), because the parallel measurement of both polarity modes can be performed within one run. The fast polarity switching in case of high-resolution mass analyzers is more demanding technical task, because the electronics usually need cer- tain time for the stabilization of high voltages in the range of kV, so only few high-resolution systems are capable of relatively fast polarity switching, but the equilibration time of few minutes is Combined ion sources can be considered as an option merg- ing the advantages and application ranges of atmospheric pressure ionization techniques, but on the other hand their sensitivity may be a compromise between both modes. The advantage of combined ESI/APCI ESI/APPI ion sources possi- ble detection of both polar and non-polar analytes in one run, which can increase the number of identified components for highly complex matrices, such as traditional Chinese medicine Fig. 5. Scheme of off-line LC–MALDI-MS coupling with the supplementary liquid
simultaneous detection of cyclodextrins, pharmaceuticals and their addition and deposition mechanisms. Junctions used for coupling of microcol- binding interactions combination of APCI, ESI and APPI can umn techniques include the following: (A) coaxial sheath flow, (B) T-junction, (C) be useful for combinational chemistry and high-throughput bio- sheathless interface, (D) porous junction, (E) liquid junction, and (F) droplet elec- trocoupling. A, B and C are typical for capillary LC–MS, while D, E and F are common logical screening. ESI can normally analyze around 80% of samples, in capillary zone electrophoresis–MS.
which can be complemented by combined source operating with Reprinted with a permission from the polarity switching within a single run combined ion source with a computer-controlled switch between MALDI and ESI modes in the ICR configuration has been developed with a possible some authors report comparable results for stable-isotope-labeled exchange between ESI and MALDI in less than 1 min proteomic quantitation Due to the complex character of studied proteomic samples, one-dimensional chromatography is 3.3. LC–MALDI-MS coupling often not sufficient and multidimensional separation approaches are needed for the adequate fractionation of studied samples LC–MALDI-MS coupling has some specific advantages over e.g., the combination of orthogonal separation LC–ESI-MS, mainly the possibility of decoupling of separation and principles of ion-exchange and reversed phase LC mass analysis steps, which allows re-analysis of peaks of inter- For on-line LC–MALDI-MS coupling effluent from est later on, lower suppression effects (possibility to use more LC is delivered directly to the mass spectrometer. Contrary to harsh LC conditions compared to LC–ESI-MS) and high m/z range off-line coupling devices, on-line devices are not yet commer- of TOF mass analyzer. On the other hand, the critical step in cially available. Methods for the liquid sample introduction can be LC–MALDI-MS coupling is the transfer of effluent from LC exit to performed by continuous-flow MALDI using frits, aerosol MALDI, the MALDI plate and matrix introduction, which may be responsi- moving wheel or moving (rotating) ball methods, but these appli- ble for certain band broadening. Off-line and on-line approaches in cations are more common for the MALDI coupling with capillary LC–MALDI-MS coupling have been described for capillary LC separations on microfluidic chips used off-line methods are based on the deposition of LC effluent 3.4. Ion mobility spectrometry on the MALDI plate using a continuous trace or discrete spots.
The continuous sample deposition is better for preserving the Ion mobility spectrometry (IMS) was developed over the past chromatographic resolution. The MALDI targets with pre-coated few decades as a method for the separation and subsequent detec- matrix are easier for the sample preparation than mixing the LC tion of volatile and semi-volatile organic compounds. IMS enables effluent with matrix. Liquid samples can be also applied on spe- the differentiation of ions by size, shape, charge as well as mass, cial nanostructured surfaces used in matrix-free approaches which can provide important supplementary information to the The deposition of continuous streak (called in-line coupling) or chromatographic separation of molecules and mass spectrometric discrete fractions can be accomplished using laboratory-built or separation of ions. Detailed description of ion mobility principles commercial robotic spotters. The spotting on the MALDI target have been discussed previously principle, the sepa- plate is achieved in several ways, i.e., most frequently contact ration of gas-phase ions at atmospheric pressure is based on their deposition (using T-junction), spray deposition (electrospray, neb- different mobilities in the low or high electric fields. Four methods ulizer), electric pulse deposition, impulse driven deposition, heated of ion mobility separation can be combined with MS, i.e., drift- droplet interface or piezoelectric microdispensor time ion mobility spectrometry (DTIMS), aspiration ion mobility line approach is often used in proteomics (e.g., analysis of protein spectrometry (AIMS), differential-mobility spectrometry (DMS) digests of post-translational modifications also called field-asymmetric waveform ion mobility spectrometry etc.), but applications in small molecule syn- (FAIMS) and traveling-wave ion mobility spectrometry (TWIMS).
thetic polymer analysis be also found, typically with Only TWIMS and DMS/FAIMS are commercially available in the microparticular monolithic columns MALDI is LC–MS coupling so far construction of TWIMS originates less prone to the ion suppression effects than ESI has higher from the traditional IMS analogous to the TOF separation, where throughput for a large number of deposited samples and higher tol- formed ions are moved to the drift region via a shutter grid. These erance toward salts and buffers. In general, MALDI is known as less ions are then separated based on different ion mobilities in a convenient for the quantitative analysis unlike to LC–ESI-MS, but weak electric field with the opposite direction of the inert gas

M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 Fig. 7. 3D plot of retention times in the LC separation, m/z values in the MS separa-
tion and drift times in the ion mobility separation of proteins obtained by LC–IMS-MS Fig. 6. Principles of ion mobility separation of ions in: (A) traveling-wave ion
Reprinted with a permission from mobility spectrometry (TWIMS) including the scheme of time-aligned parallel (TAP) fragmentation and (B) field-asymmetric waveform ion mobility spectrometry ground compounds and cluster ions from doubly charged ions with Adapted with a permission from the aim to simplify the spectra). The reduction of interferences fol- lowed by increased selectivity can be further enhanced by the use of flow. Unlike the traditional IMS, where the low electric field is suitable volatile chemical modifier (called dopant) applied continuously on the cell, a sequence of symmetric potential advantage results in the reduced chemical noise, increased dynamic waves (high-field) is continuously applied through the series of range and enhanced peak separation, and it is closely associated segmented electrodes of cell in the same direction with the ion with chemical properties of used dopant and its concentration.
migration Ions are introduced from Q at reduced MALDI and ESI are commonly used in IMS-MS. Due to the easy cou- pressure, and their motion in the electric field of IMS cell depends pling with LC and possible ionization of less volatile compounds, on particular ion mobilities (On the contrary, the ion the ESI is the best choice for LC–IMS-MS applications mobility device for DMS/FAIMS is placed in the ion source region In principle, all types of mass analyzers can be used after the ion and ions are separated under ambient conditions. DMS/FAIMS mobility separation. The commercial solution of TWIMS is followed works as a scan filter and sorts ions by the difference between by TOF-MS, while DMS/FAIMS forms a part of the ion source and ion mobilities at high and low electric field with the opposite can be combined with any type of mass analyzer.
polarity, induced by a periodic asymmetric field (application of The ion mobility separation is different compared to chromato- so called separation or dispersion voltage) orthogonal to the ion graphic and mass spectrometric separations, and the combination path. The different mobility of ions during the application of high of these separation modes provides a better resolution for com- and low voltages causes the ion drift toward one of two electrodes plex samples. Three-dimensional data set consists retention times, trajectory of particular ions along the radial axis drift times and m/z values, as illustrated in the 3D separa- can be corrected using the application of compensation voltage tion of peptide digest of human plasma proteome use to avoid ion discharge (This approach is similar to the of MS/MS provides an additional level of structural information.
Q filtering. DMS and FAIMS instruments are based on the same In addition to traditional collision-induced dissociation (CID), new principle of ion separation, but they differ in the instrumental fragmentation approaches have been introduced by Waters, such design. Electrodes are not segmented and the alternating electric as energy dependent fragmentation (MSE) and time-aligned par- field is placed between two electrodes (plate electrodes for DMS allel (TAP) fragmentation. MSE can be applied in all tandem mass vs. cylindrical electrodes for FAIMS).
spectrometers, while application of TAP is restricted to IMS instru- Instruments equipped with the ion mobility spectrometry (IMS) ments. The TWIMS ion mobility separation plays an important role were commercially introduced by Waters in 2006 (TWIMS) and in TAP approach (The precursor ion is fragmented in the now they are also provided by Thermo Scientific (FAIMS) and AB trap with subsequent separation of the first generation of product SCIEX (DMS). IMS can be applied for the separation of isobaric com- ions by IMS. The second generation of fragment ions is formed in the pounds (on condition that their cross-sections differ at least by transfer, and they are associated to the first generation parent based about 3%), the reduction of high background noise, the separation of on individual drift-times combination of IMS and MS endogenous matrix interferences from target analytes to increase with high-RP and advanced fragmentation experiments (MSn, elec- the selectivity and enables the charge state screening used mainly tron transfer/capture dissociation, MSE or TAP) is a powerful tool in proteomics (the separation of unwanted singly charged back- for the structural determination.
M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 The ion mobility MS was mainly associated with the analysis spectra (SWATH) mode in which sequential precursor ions win- of volatile compounds for homeland security and environmental dows (typically 20 m/z) are used to collect the same spectrum applications, such as explosives, chemical warfare agents, chemical precursor and fragment ions using a collision energy range. High- pollutants or drugs detection. The LC sample introduction followed resolution SRM (HR-SRM) on a rapid acquisition (<50 ms) Q-TOF by ESI has extended the range of applications to the field of biolog- instrument with the resolving power above 20,000 leads to new ical, biomedical and pharmaceutical research wide range possibilities in the integrated quantitative and qualitative bioanal- of applications for LC–IMS-MS can be found in the literature, e.g., profiling of plasma proteome of isomeric transforma- MALDI couple to TOF analyzer is not a typical method of choice tion of phenolic compounds present in foodstuff for the quantitative analyses due to worse scan to scan repro- of indole alkaloids in yohimbe bark high-throughput pro- ducibility, but the coupling of MALDI with QqQ mass analyzers teomic studies of rat urinary metabonome combines important advantages of both approaches. MALDI is the of separation of drug-related materials ultrafast technique without the need of chromatographic separa- drug metabolism study The 2D-LC–IMS-MS coupling has tion, while QqQ with SRM is the best technique for the sensitive a great potential in the biomarker discovery due to the possible quantitation usefulness of MALDI-QqQ configuration in orthogonal character of these separation techniques in liquid and the quantitative analysis has been demonstrated on the ultrafast gas phase, as illustrated on the example of 8 pharmaceutical com- quantitation (6 s for one sample) of selected drugs pro- pounds with the identical nominal mass m/z 316 can teomic analysis et al. reported MALDI SRM separate these molecules in a liquid-phase, ultrahigh-RP MS can quantitation on QqQ mass analyzer may be a serious alternative to separate their protonated molecules in a gas-phase according to established LC–ESI-MS methods in terms of linearity, limit of quan- their accurate m/z values, while IMS separates them according to titation, precision and accuracy. However, MALDI assay was at least their size-to-charge ratio.
50 times faster than LC–ESI-MS.
Another important issue is the sample preparation preceding 3.5. Mass spectrometric quantitation LC–MS analysis Obviously, the internal standard must be added before any sample pre-concentration step. The LC–MS The QqQ mass analyzer with SRM scans is a golden standard quantitation approach has a clear advantage in terms of reduced in any mass spectrometric quantitation either in the LC–MS con- ion suppression effects, trace analysis and retention times bring figuration or MS stand-along systems All MS quantitation an additional dimension in the selectivity. On the other hand, approaches obviously require the use of internal standards to elim- advantages of shot-gun approach (typically used for example in inate any possible variations during the ionization process and lipidomics) are mainly the analysis speed and simplicity. The the mass analysis, such as the ion suppression/enhancement, the fastest QqQ mass analyzers enable the determination of numerous contamination of the ion source or the mobile phase, etc., extrac- species within few minutes without any chromatographic sepa- tion losses or any other unpredictable reasons. The best internal ration, but the information on isobaric species with the identical standard in MS and LC–MS is the addition of isotopically labeled fragmentation pattern is lost. In the quantitation of complex pro- analogs, where all physico-chemical properties including the reten- tein mixtures, the isotope-coded affinity tags (ICAT) approach tion behavior, fragmentation behavior and extraction efficiency are and related tag techniques are often used alter- almost identical except for characteristic mass shifts caused by the native approach for the relative quantitation is the use of response number of labeled isotopes. For higher number of deuterium atoms factors determined from the calibration curves of pure standards (at least about 5), small shifts in retention times can occur, but it and then applied for real samples internal standard does not constitute any problem in the LC–MS quantitation. The ion addition and response factors approach can be combined in one suppression/enhancement effects play an important role in LC–MS platform together with well-optimized chromatographic separa- quantitation and the extend of these effects needs to be quantita- tion, as illustrated on the lipidomic class quantitation The tively assessed, as suggested in few recent works stable isotope labeling by amino acids in cell culture (SILAC) The novel approach for LC–MS quantitation uses is a simple approach for the incorporation of the isotopic label into ultrahigh-RP in the full-scan single stage mode using reconstructed proteins for MS-based quantitative proteomics. Two cell popula- ion currents for very narrow extraction windows (e.g., 5 ppm tions are grown in the culture media that are identical except for around the theoretical m/z value). This approach provided com- light (non-labeled) and heavy (labeled with deuterium, 13C or 15N) parable detection specificity, assay precision, accuracy, linearity form of a particular amino acid, which is incorporated into the and sensitivity for 17 therapeutic drugs as for the conventional SRM acquisition on QqQ without the need of the optimization of MS/MS parameters for SRM transitions. The full-scan mass spectra information is still retained unlike SRM measurements, which can 4. Mass spectrometric approaches complementary to
be beneficial for the detection of co-eluting species, unexpected adducts of analytes, etc.
The recent trend in MS based quantitation is an integrated In the last decade, several new approaches designed for the quantitative and qualitative bioanalysis which essentially direct mass spectrometric analysis at ambient conditions without requires the use of high-resolution tandem mass analyzers cou- the chromatographic separation have been introduced. The main pled to LC (preferably in fast LC mode). Hybrid FT tandem mass advantage of such approaches is the fast analysis without any (or analyzers are convenient for this purpose due to the ability to col- minimum) sample preparation, which significantly increases the lect full-scan high-RP mass spectra at scan speeds required for sample throughput. On the other hand, some drawbacks must be UHPLC together with routine measurements of MA <5 ppm also mentioned in terms of reduced amount of analytical informa- Another possibility is the use of the following acquisition schemes tion due to the absence of separation and sample preparation steps.
on Q-TOF mass analyzer (1) information-dependent acqui- The ion suppression and matrix effects can cause severe problems sition with TOF survey scan and product-ion scan as dependent with the quantitation and the trace analysis.
scan, (2) MSALL by collecting TOF mass spectra with and without The term ambient ionization technique has been first intro- fragmentation by alternating low and high collision energy, and duced by Takáts et al. now there is an explosion of (3) sequential window acquisition of all theoretical fragment-ion new ambient ionization techniques and associated acronyms M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 Fig. 9. Surgical mass spectrometry: scheme of ion transfer from the tissue to the
atmospheric interface to mass spectrometer.
Reprinted with a permission from tissues another interesting application of MS has been published and referred as rapid evaporation ionization mass spectrometry (REIMS) (the electrosurgical dis- section, the tissue is locally exposed to high-frequency electric current resulting in the ionization of molecules contained in this Fig. 8. Typical applications of MALDI mass spectrometry imaging in proteomics,
lipidomics and drug metabolites.
tissue, preferably lipids. The lipidomic composition of dissected Reprinted with a permission from tissue can be obtained within less than second and used for the verification of tissue type (cancer vs. healthy tissue) The main group is a family of ambient desorption 5. Current state and future trends
ionization techniques, such as desorption ESI (DESI) desorption APCI (DAPCI) desorption APPI (DAPPI) Some trends in the area of LC–MS and related techniques The direct analysis in real time (DART) is the name of ionization are already recognized now: (A) the shift from low-resolution technique introduced by JEOL the soft ionization of analyte to (ultra)high-resolution tandem mass analyzers providing high- molecules on surfaces, in liquid-phase or gases without any sample MA below 1 ppm, (B) the shift from conventional HPLC–MS to preparation. The Penning ionization mechanism describes based UHPLC–MS or other fast LC–MS techniques (core–shell particles, on interactions between excited helium atoms and target analyte high-temperature LC and monolithic columns) requiring fast MS at ambient conditions. Another ionization technique applicable for analyzers (typically TOF based systems), (C) the use of 2D-LC–MS different types of gaseous, liquid and solid samples is termed the for complex samples, and (D) other dimension also in MS, such as atmospheric solids probe analysis (ASAP) ASAP and DESI IMS-MS, parallel use of more mass spectrometers, ionization tech- can be combined in one ionization source which extends niques and polarity modes. Present LC–MS systems generate huge the range of analyte compounds in terms of their polarities and amounts of analytical data, which is often impossible to interpret molecular weights. Recently, the paper spray ionization has been manually, so dedicated softwares can help with the automation of described, where the analyte (e.g., the whole blood) is spotted on data processing and interpretation significant impact on the paper (so called dry blood spot analysis then the selected the whole mass spectrometric community had the invention solvent is automatically added, and the solution is electrosprayed and commercialization of the sixth type of mass analyzer—Orbitrap.
from the paper into the mass spectrometer technique is The notable improvement has been recently reported in the field of intended for high-throughput clinical analyses. The new ionization ICR cell construction, where Nikolaev et al. the technique solvent based direct inlet MS suitable for both RP exceeding 24 millions, which shifts the limits of MS. Ground- small and large molecules in solids or liquid solvents including in breaking news (such as the discovery of new type of mass analyzer) LC–MS configuration. The principle of this ionization technique are not probable in the near future, but such discoveries cannot be is based on the heated inlet used for reversed-phase system with anticipated. On the other hand, improvements in the area of ion- some similarity to former thermospray ionization, but author ization techniques, ion optics, fast electronics, dedicated scans and report better sensitivity and applicability to peptides.
consequently the sensitivity and selectivity will certainly continue.
Promising MS techniques designed for the determination of spatial distribution of analyte molecules on the surface is mass spectrometry imaging (MSI) is typically used for the spatial imaging of biomolecules in biological tissues. At present, the spatial resolution of MALDI-MSI is routinely in the range of Authors would like to express gratitude to numerous colleagues tens micrometers reported values are below 5 ␮m and company representatives for their help with the compilation secondary ion mass spectrometry (SIMS) can provide of tables and critical reading of the manuscript as well as to anony- the spatial resolution even below 1 ␮m additional mous reviewers for their insightful comments.
level of information can be obtained by 3D spatial imaging of bio- logical tissues using SIMS Most typical application areas of MSI (are lipidomics the distribution of [1] W.M.A. Niessen, J. Chromatogr. A 794 (1998) 407.
drugs and other small molecules in tissues while the [2] W.M.A. Niessen, J. Chromatogr. A 856 (1999) 179.
number of MSI papers in proteomics is slightly lower due to the [3] M. Himmelsbach, J. Chromatogr. B 883 (2012) 3.
sensitivity limitations and essential removal of abundant lipids [4] Y. Liang, G.J. Wang, L. Xie, L.S. Sheng, Curr. Drug Metab. 12 (2011) 329.
[5] W.M.A. Niessen, J. Chromatogr. A 1000 (2003) 413.
The MSI with high spatial resolution has been sug- [6] , downloaded on 11.01.12.
gested as an alternative technique to the histological staining of [7] downloaded on 11.01.12.
M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 [8] , downloaded on 11.01.12.
[74] D. Guillarme, J. Schappler, S. Rudaz, J.L. Veuthey, Trends Anal. Chem. 29 (2010) [9] , downloaded on 11.01.12.
[10] downloaded on 11.01.12.
[75] Y.F. Cheng, Z. Lu, U. Neue, Rapid Commun. Mass Spectrom. 15 (2001) 141.
[11] downloaded on 11.01.12.
[76] E. Uliyanchenko, P.J. Schoenmakers, S. van der Wal, J. Chromatogr. A 1218 [12] downloaded on 11.01.12.
(2011) 1509.
[13] downloaded on 11.01.12.
[77] K.K. Unger, R. Skudas, M.M. Schulte, J. Chromatogr. A 1184 (2008) 393.
[14] downloaded on 11.01.12.
[78] E. Olah, S. Fekete, J. Fekete, K. Ganzler, J. Chromatogr. A 1217 (2010) 3642.
[15] downloaded on 11.01.12.
[79] D.T.T. Nguyen, D. Guillarme, S. Heinisch, M.P. Barrioulet, J.L. Rocca, S. Rudaz, [16] , downloaded on 11.01.12.
J.L. Veuthey, J. Chromatogr. A 1167 (2007) 76.
[17] downloaded on 12.07.12.
[80] S. Heinisch, J.L. Rocca, J. Chromatogr. A 1216 (2009) 642.
[18] K.K. Murray, J. Chromatogr. A 1217 (2010) 3922.
[81] R. Wu, L.G. Hu, F.J. Wang, M.L. Ye, H. Zou, J. Chromatogr. A 1184 (2008) 369.
[19] O.D. Sparkman, Mass Spectrometry Desk Reference, Global View Publishing, [82] S. El Deeb, Chromatographia 74 (2011) 681.
Pittsburgh, 2000.
[83] D. Guillarme, J. Ruta, S. Rudaz, J.L. Veuthey, Anal. Bioanal. Chem. 397 (2010) [20] J. Barker, Mass Spectrometry, Second ed., J. Wiley & Sons, Chichester, 1999.
[21] R.B. Cole, Electrospray and MALDI Mass Spectrometry, Second ed., J. Wiley & [84] M. Lísa, H. Velínská, M. Holˇcapek, Anal. Chem. 81 (2009) 3903.
Sons, Chichester, 2010.
[85] M. Lísa, M. Holˇcapek, J. Chromatogr. A 1198 (2008) 115.
[22] E. de Hoffmann, V. Stroobant, Mass Spectrometry: Principles and Applications, [86] M. Holˇcapek, R. Jirásko, M. Lísa, J. Chromatogr. A 1217 (2010) 3908.
Third ed., John Wiley & Sons Ltd., Chichester, 2007.
[87] M. Holˇcapek, H. Dvoˇráková, M. Lísa, A.J. Giron, P. Sandra, J. Cvaˇcka, J. Chro- [23] W. Henderson, J.S. McIndoe, Mass Spectrometry of Inorganic and matogr. A 1217 (2010) 8186.
Organometallic Compounds, J. Wiley & Sons, Chichester, 2005.
[88] R. Jirásko, M. Holˇcapek, Mass Spectrom. Rev. 30 (2011) 1013.
[24] J.V. Olsen, L.M.F. de Godoy, G.Q. Li, B. Macek, P. Mortensen, R. Pesch, A.
[89] P. Kebarle, U.H. Verkerk, Mass Spectrom. Rev. 28 (2009) 898.
Makarov, O. Lange, S. Horning, M. Mann, Mol. Cell. Proteomics 4 (2005) 2010.
[90] A. Garcia-Ac, P.A. Segura, L. Viglino, C. Gagnon, S. Sauve, J. Mass Spectrom. 46 [25] J. Cox, A. Michalski, M. Mann, J. Am. Soc. Mass Spectrom. 22 (2011) 1373.
[26] V.V. Mihaleva, O. Vorst, C. Maliepaard, H.A. Verhoeven, R.C.H. de Vos, R.D. Hall, [91] J.A. Syage, J. Am. Soc. Mass Spectrom. 15 (2004) 1521.
R. van Ham, Metabolomics 4 (2008) 171.
[92] L.C. Short, S.S. Cai, J.A. Syage, J. Am. Soc. Mass Spectrom. 18 (2007) 589.
[27] A.G. Chambers, J.M. Ramsey, Anal. Chem. 84 (2012) 1446.
[93] A. Ahmed, C.H. Choi, M.C. Choi, S. Kim, Anal. Chem. 84 (2012) 1146.
[28] , downloaded on 11.01.12.
[94] Y. Zhou, Y. Wang, R.F. Wang, F. Guo, C. Yan, J. Sep. Sci. 31 (2008) 2388.
[29] Y. Pico, C. Blasco, M. Farre, D. Barcelo, J. AOAC Int. 92 (2009) 734.
[95] L.C. Short, J.A. Syage, Rapid Commun. Mass Spectrom. 22 (2008) 541.
[30] C. Soler, B. Hamilton, A. Furey, K.J. James, J. Manes, Y. Pico, Rapid Commun.
[96] R.T. Gallagher, M.P. Balogh, P. Davey, M.R. Jackson, I. Sinclair, L.J. Southern, Mass Spectrom. 20 (2006) 2151.
Anal. Chem. 75 (2003) 973.
[31] O.J. Pozo, P. Van Eenoo, K. Deventer, H. Elbardissy, S. Grimalt, J.V. Sancho, F.
[97] G. Baykut, J. Fuchser, M. Witt, G. Weiss, C. Gosteli, Rapid Commun. Mass Hernandez, R. Ventura, F.T. Delbeke, Anal. Chim. Acta 684 (2011) 107.
Spectrom. 16 (2002) 1631.
[32] T. Rousu, J. Herttuainen, A. Tolonen, Rapid Commun. Mass Spectrom. 24 [98] M. Lasaosa, N. Delmotte, C.G. Huber, K. Melchior, E. Heinzle, A. Tholey, Anal.
Bioanal. Chem. 393 (2009) 1245.
[33] F. Hernandez, M. Ibanez, J.V. Sancho, O.J. Pozo, Anal. Chem. 76 (2004) 4349.
[99] O. Peˇs, J. Preisler, J. Chromatogr. A 1217 (2010) 3966.
[34] R.J. Wenzel, U. Matter, L. Schultheis, R. Zenobi, Anal. Chem. 77 (2005) 4329.
[100] H. Wei, K. Nolkrantz, M.C. Parkin, C.N. Chisolm, J.P. O‘Callaghan, R.T. Kennedy, [35] A. Holle, A. Haase, M. Kayser, J. Hohndorf, J. Mass Spectrom. 41 (2006) 705.
Anal. Chem. 78 (2006) 4342.
[36] S. Berkenkamp, F. Kirpekar, F. Hillenkamp, Science 281 (1998) 260.
[101] D.L. Devoe, C.S. Lee, Electrophoresis 27 (2006) 3559.
[37] E.N. Nikolaev, I.A. Boldin, R. Jertz, G. Baykut, J. Am. Soc. Mass Spectrom. 22 [102] J. Lee, S.A. Soper, K.K. Murray, J. Mass Spectrom. 44 (2009) 579.
(2011) 1125.
[103] C.W. Tsao, S. Tao, C.F. Chen, J.K. Liu, D.L. DeVoe, Microfluid. Nanofluid. 8 (2010) [38] B. Bogdanov, R.D. Smith, Mass Spectrom. Rev. 24 (2005) 168.
[39] R.P. Rodgers, T.M. Schaub, A.G. Marshall, Anal. Chem. 77 (2005) 20a.
[104] K. Marcus, H. Schafer, S. Klaus, C. Bunse, R. Swart, H.E. Meyer, J. Proteome Res.
[40] A.G. Marshall, R.P. Rodgers, Accounts Chem. Res. 37 (2004) 53.
6 (2007) 636.
[41] D. Ohta, S. Kanaya, H. Suzuki, Curr. Opin. Biotechnol. 21 (2010) 35.
[105] E. Mirgorodskaya, C. Braeuer, P. Fucini, H. Lehrach, J. Gobom, Proteomics 5 [42] H.Y. Wang, X. Chu, Z.X. Zhao, X.S. He, Y.L. Guo, J. Chromatogr. B 879 (2011) [106] J. Wohlgemuth, M. Karas, T. Eichhorn, R. Hendriks, S. Andrecht, Anal. Biochem.
[43] T. Pluskal, T. Uehara, M. Yanagida, Anal. Chem. 84 (2012) 4396.
395 (2009) 178.
[44] J.P. Kutter, J. Chromatogr. A 1221 (2012) 72.
[107] E.S. Kuznetsova, A.K. Buryak, G.A. Petukhova, Chromatographia 73 (2011) S55.
[45] S. Koster, E. Verpoorte, Lab Chip 7 (2007) 1394.
[108] G. Montaudo, F. Samperi, M.S. Montaudo, Prog. Polym. Sci. 31 (2006) 277.
[46] N.V. Lavrik, L.T. Taylor, M.J. Sepaniak, Anal. Chim. Acta 694 (2011) 6.
[109] S.M. Weidner, J. Falkenhagen, Anal. Chem. 83 (2011) 9153.
[47] S.L. Lin, H.Y. Bai, T.Y. Lin, M.R. Fuh, Electrophoresis 33 (2012) 635.
[110] M.X. Gao, J. Zhang, C.H. Deng, P.Y. Yang, X.M. Zhang, J. Proteome Res. 5 (2006) [48] M. Holˇcapek, H. Velínská, M. Lísa, P. ˇ Cesla, J. Sep. Sci. 32 (2009) 3672.
[49] M. Lísa, E. Cífková, M. Holˇcapek, J. Chromatogr. A 1218 (2011) 5146.
[111] F. Gosetti, E. Mazzucco, D. Zampieri, M.C. Gennaro, J. Chromatogr. A 1217 [50] P. Dugo, F. Cacciola, T. Kumm, G. Dugo, L. Mondello, J. Chromatogr. A 1184 (2010) 3929.
[112] M.A. Kuzyk, L.B. Ohlund, M.H. Elliott, D. Smith, H. Qian, A. Delaney, C.L. Hunter, [51] F. Cacciola, P. Jandera, E. Bláhová, L. Mondello, J. Sep. Sci. 29 (2006) 2500.
C.H. Borchers, Proteomics 9 (2009) 3328.
[52] T. Hájek, V. ˇSkeˇríková, P. ˇ nuchalová, P. Jandera, J. Sep. Sci. 31 (2008) [113] X. Gu, C.H. Deng, G.Q. Yan, X.M. Zhang, J. Proteome Res. 5 (2006) 3186.
[114] C.L. Liu, X.M. Zhang, J. Chromatogr. A 1139 (2007) 191.
[53] T. Hyotylainen, T. Andersson, M. Jussila, S.K. Wiedmer, M. Rautiainen, M.L.
[115] H. Orsnes, R. Zenobi, Chem. Soc. Rev. 30 (2001) 104.
Riekkola, J. Sep. Sci. 24 (2001) 544.
[116] K.K. Murray, Mass Spectrom. Rev. 16 (1997) 283.
[54] P. Donato, F. Cacciola, L. Mondello, P. Dugo, J. Chromatogr. A 1218 (2011) 8777.
[117] C.W. Huck, R. Bakry, L.A. Huber, G.K. Bonn, Electrophoresis 27 (2006) 2063.
[55] X.H. Guo, E. Lankmayr, Bioanalysis 2 (2010) 1109.
[118] H. Borsdorf, T. Mayer, M. Zarejousheghani, G.A. Eiceman, Appl. Spectrosc. Rev.
[56] X.L. Han, K. Yang, R.W. Gross, Mass Spectrom. Rev. 31 (2012) 134.
46 (2011) 472.
[57] W.C. Byrdwell, W.E. Neff, Rapid Commun. Mass Spectrom. 16 (2002) 300.
[119] A.B. Kanu, P. Dwivedi, M. Tam, L. Matz, H.H. Hill, J. Mass Spectrom. 43 (2008) [58] W.C. Byrdwell, Anal. Bioanal. Chem. 401 (2011).
[59] V. Kertesz, G.J. Van Berkel, J. Mass Spectrom. 45 (2010) 252.
[120] B.M. Kolakowski, Z. Mester, Analyst 132 (2007) 842.
[60] K. Schuhmann, R. Almeida, M. Baumert, R. Herzog, S.R. Bornstein, A.
[121] I. Marquez-Sillero, E. Aguilera-Herrador, S. Cardenas, M. Valcarcel, Trends Shevchenko, J. Mass Spectrom. 47 (2012) 96.
Anal. Chem. 30 (2011) 677.
[61] P. Huˇsek, P. ˇSimek, P. Hartvich, H. Zahradníˇcková, J. Chromatogr. A 1186 (2008) [122] R.M. O‘Donnell, X.B. Sun, P.D. Harrington, Trends Anal. Chem. 27 (2008) 44.
[123] A.A. Shvartsburg, A.J. Creese, R.D. Smith, H.J. Cooper, Anal. Chem. 82 (2010) [62] P. ˇSimek, P. Huˇsek, H. Zahradníˇcková, Anal. Chem. 80 (2008) 5776.
[63] F.G. Xu, L. Zou, Y. Liu, Z.J. Zhang, C.N. Ong, Mass Spectrom. Rev. 30 (2011) 1143.
[124] J.H. Sun, A. Baker, P. Chen, Rapid Commun. Mass Spectrom. 25 (2011) [64] M. Niwa, Bioanalysis 4 (2012) 213.
[65] P. Deng, Y. Zhan, X.Y. Chen, D.F. Zhong, Bioanalysis 4 (2012) 49.
[125] S.J. Valentine, M.D. Plasencia, X.Y. Liu, M. Krishnan, S. Naylor, H.R. Udseth, R.D.
[66] M. Holˇcapek, H. Virelizier, J. Chamot-Rooke, P. Jandera, C. Moulin, Anal. Chem.
Smith, D.E. Clemmer, J. Proteome Res. 5 (2006) 2977.
71 (1999) 2288.
[126] X.Y. Liu, S.J. Valentine, M.D. Plasencia, S. Trimpin, S. Naylor, D.E. Clemmer, J.
[67] J. Cimlová, P. Kruˇzberská, Z. ˇSvagera, P. Huˇsek, P. ˇSimek, J. Mass Spectrom. 47 Am. Soc. Mass Spectrom. 18 (2007) 1249.
[127] C. Xie, K. Yu, D.F. Zhong, T. Yuan, F. Ye, J.A. Jarrell, A. Millar, X.Y. Chen, J. Agric.
[68] T. Tuzimski, J. Chromatogr. A 1218 (2011) 8799.
Food Chem. 59 (2011) 11078.
[69] S.P. Pasilis, G.J. Van Berkel, J. Chromatogr. A 1217 (2010) 3955.
[128] E.S. Baker, E.A. Livesay, D.J. Orton, R.J. Moore, W.F. Danielson, D.C. Prior, Y.M.
[70] B. Fuchs, R. Suss, K. Teuber, M. Eibisch, J. Schiller, J. Chromatogr. A 1218 (2011) Ibrahim, B.L. LaMarche, A.M. Mayampurath, A.A. Schepmoes, D.F. Hopkins, K.Q. Tang, R.D. Smith, M.E. Belov, J. Proteome Res. 9 (2010) 997.
[71] D. Guillarme, J.L. Veuthey (Eds.), UHPLC in Life Sciencies, The Royal Society of [129] E.L. Harry, D.J. Weston, A.W.T. Bristow, I.D. Wilson, C.S. Creaser, J. Chromatogr.
Chemistry, Cambridge, 2012.
B 871 (2008) 357.
[72] L. Nováková, H. Vlˇcková, Anal. Chim. Acta 656 (2009) 8.
[130] C. Eckers, A.M.F. Laures, K. Giles, H. Major, S. Pringle, Rapid Commun. Mass [73] L. Nováková, L. Matysová, P. Solich, Talanta 68 (2006) 908.
Spectrom. 21 (2007) 1255.
M. Holˇcapek et al. / J. Chromatogr. A 1259 (2012) 3–15 [131] G.J. Dear, J. Munoz-Muriedas, C. Beaumont, A. Roberts, J. Kirk, J.P. Williams, I.
[158] M. Haapala, J. Pól, V. Saarela, V. Arvola, T. Kotiaho, R.A. Ketola, S. Franssila, T.J.
Campuzano, Rapid Commun. Mass Spectrom. 24 (2010) 3157.
Kauppila, R. Kostiainen, Anal. Chem. 79 (2007) 7867.
[132] E. Varesio, J.C.Y. Le Blanc, G. Hopfgartner, Anal. Bioanal. Chem. 402 (2012) [159] R.B. Cody, J.A. Laramee, H.D. Durst, Anal. Chem. 77 (2005) 2297.
[160] C.N. McEwen, R.G. McKay, B.S. Larsen, Anal. Chem. 77 (2005) 7826.
[133] G. Hopfgartner, E. Varesio, Trends Anal. Chem. 24 (2005) 583.
[161] J.A. Lloyd, A.F. Harron, C.N. McEwen, Anal. Chem. 81 (2009) 9158.
[134] B.K. Matuszewski, M.L. Constanzer, C.M. Chavez-Eng, Anal. Chem. 75 (2003) [162] W.K. Li, F.L.S. Tse, Biomed. Chromatogr. 24 (2010) 49.
[163] H. Wang, J.J. Liu, R.G. Cooks, Z. Ouyang, Angew. Chem. Int. Ed. 49 (2010) 877.
[135] I. Marchi, V. Viette, F. Badoud, M. Fathi, M. Saugy, S. Rudaz, J.L. Veuthey, J.
[164] V.S. Pagnotti, E.D. Inutan, D.D. Marshall, C.N. McEwen, S. Trimpin, Anal. Chem.
Chromatogr. A 1217 (2010) 4071.
83 (2011) 7591.
[136] D. Remane, M.R. Meyer, D.K. Wissenbach, H.H. Maurer, Rapid Commun. Mass [165] M.P. Greving, G.J. Patti, G. Siuzdak, Anal. Chem. 83 (2011) 2.
Spectrom. 24 (2010) 3103.
[166] R.M.A. Heeren, D.F. Smith, J. Stauber, B. Kukrer-Kaletas, L. MacAleese, J. Am.
[137] H. Henry, H.R. Sobhi, O. Scheibner, M. Bromirski, S.B. Nimkar, B. Rochat, Rapid Soc. Mass Spectrom. 20 (2009) 1006.
Commun. Mass Spectrom. 26 (2012) 499.
[167] J. Pól, M. Strohalm, V. Havlíˇcek, M. Voln ´y, Histochem. Cell Biol. 134 (2010) [138] R. Pinhancos, S. Maass, D.M. Ramanathan, J. Mass Spectrom. 46 (2011) 1175.
[139] G. Hopfgartner, D. Tonoli, E. Varesio, Anal. Bioanal. Chem. 402 (2012) 2587.
[168] M. Setou, N. Kurabe, J. Electron Microsc. 60 (2011) 47.
[140] J.J. Corr, P. Kovarik, B.B. Schneider, J. Hendrikse, A. Loboda, T.R. Covey, J. Am.
[169] E.R.A. van Hove, D.F. Smith, R.M.A. Heeren, J. Chromatogr. A 1217 (2010) 3946.
Soc. Mass Spectrom. 17 (2006) 1129.
[170] K. Schwamborn, R.M. Caprioli, Mol. Oncol. 4 (2010) 529.
[141] R.J.W. Meestersy, E. den Boer, R.A.A. Mathot, R. de Jonge, R.J. van Klaveren, J.
[171] D. Touboul, O. Laprevote, A. Brunelle, Curr. Opin. Chem. Biol. 15 (2011) Lindemans, T.M. Luider, Bioanalysis 3 (2011) 1369.
[142] A. Lesur, E. Varesio, G. Hopfgartner, Anal. Chem. 82 (2010) 5227.
[172] S. Guenther, M. Koestler, O. Schulz, B. Spengler, Int. J. Mass Spectrom. 294 [143] D.A. Volmer, L. Sleno, K. Bateman, C. Sturino, R. Oballa, T. Mauriala, J. Corr, Anal. Chem. 79 (2007) 9000.
[173] J.S. Fletcher, N.P. Lockyer, J.C. Vickerman, Mass Spectrom. Rev. 30 (2011) 142.
[144] D. Vuckovic, X. Zhang, E. Cudjoe, J. Pawliszyn, J. Chromatogr. A 1217 (2010) [174] J.S. Fletcher, J.C. Vickerman, N. Winograd, Curr. Opin. Chem. Biol. 15 (2011) [145] S.P. Gygi, B. Rist, S.A. Gerber, F. Turecek, M.H. Gelb, R. Aebersold, Nat. Bio- [175] K.A.Z. Berry, J.A. Hankin, R.M. Barkley, J.M. Spraggins, R.M. Caprioli, R.C. Mur- technol. 17 (1999) 994.
phy, Chem. Rev. 111 (2011) 6491.
[146] K. Kristjansdottir, S.J. Kron, Curr. Proteomics 7 (2010) 144.
[176] D. Touboul, A. Brunelle, O. Laprevote, Biochimie 93 (2011) 113.
[147] R.L. Gant-Branum, T.J. Kerr, J.A. McLean, Analyst 134 (2009) 1525.
[177] V. Vidová, P. Novák, M. Strohalm, J. Pól, V. Havlíˇcek, M. Voln ´y, Anal. Chem. 82 [148] M. Lísa, F. Lynen, M. Holˇcapek, P. Sandra, J. Chromatogr. A 1176 (2007) 135.
(2010) 4994.
[149] M. Holˇcapek, M. Lísa, P. Jandera, N. Kabátová, J. Sep. Sci. 28 (2005) 1315.
[178] A. Svatoˇs, Trends Biotechnol. 28 (2010) 425.
[150] E. Cífková, M. Holˇcapek, M. Lísa, M. Ovˇcaˇcíková, A. Lyˇcka, F. Lynen, P. Sandra, [179] M.L. Reyzer, R.M. Caprioli, Curr. Opin. Chem. Biol. 11 (2007) 29.
Anal. Chem., submitted for publication.
[180] A. Römpp, S. Guenther, Y. Schober, O. Schulz, Z. Takáts, W. Kummer, B. Spen- [151] M. Mann, Nat. Rev. Mol. Cell Biol. 7 (2006) 952.
gler, Angew. Chem. Int. Ed. 49 (2010) 3834.
[152] Z. Takáts, J.M. Wiseman, B. Gologan, R.G. Cooks, Science 306 (2004) 471.
[181] J. Balog, T. Szaniszlo, K.C. Schaefer, J. Denes, A. Lopata, L. Godorhazy, D. Szalay, [153] G.A. Harris, A.S. Galhena, F.M. Fernandez, Anal. Chem. 83 (2011) 4508.
L. Balogh, L. Sasi-Szabo, M. Toth, Z. Takáts, Anal. Chem. 82 (2010) 7343.
[154] D.J. Weston, Analyst 135 (2010) 661.
[182] F. Rasche, K. Scheubert, F. Hufsky, T. Zichner, M. Kai, A. Svatoˇs, S. Böcker, Anal.
[155] A. Venter, M. Nefliu, R.G. Cooks, Trends Anal. Chem. 27 (2008) 284.
Chem. 84 (2012) 3417.
[156] Z. Takáts, J.M. Wiseman, R.G. Cooks, J. Mass Spectrom. 40 (2005) 1261.
[183] A. Makarov, Anal. Chem. 72 (2000) 1156.
[157] Z. Takáts, I. Cotte-Rodriguez, N. Talaty, H.W. Chen, R.G. Cooks, Chem. Commun.
(2005) 1950.

Source: http://ftp.ch.bme.hu/oktatas/konyvek/anal/MSc-Elvalasztastechnika/MSc-2012-2013/MS%20stateof%20the%20art.pdf