A Differentially Pumped Dual Linear Quadrupole Ion Trap (DLQIT

Oct 31, 2013 - Mass Selective Ion Transfer and Accumulation in Ion Trap Arrays. Yuzhuo Wang , Xiaohua Zhang , Yanbing Zhai , You Jiang , Xiang Fang , ...
5 downloads 9 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

A Differentially Pumped Dual Linear Quadrupole Ion Trap (DLQIT) Mass Spectrometer: A Mass Spectrometer Capable of MSn Experiments Free From Interfering Reactions Benjamin C. Owen,†,‡ Tiffany M. Jarrell,†,‡ Jae C. Schwartz,§ Rob Oglesbee,‡ Mark Carlsen,‡ Enada F. Archibold,‡ and Hilkka I. Kenttam ̈ aa*,†,‡ †

Center for Direct Catalytic Conversion of Biomass to Biofuels, Bindley Bioscience Center, Purdue University, 1203 W. State Street, West Lafayette, Indiana 47907, United States ‡ Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana 47907, United States § Thermo Fisher Scientific, 355 River Oak Parkway, San Jose, California 95134, United States ABSTRACT: A novel differentially pumped dual linear quadrupole ion trap (DLQIT) mass spectrometer was designed and built to facilitate tandem MS experiments free from interfering reactions. The instrument consists of two differentially pumped Thermo Scientific linear quadrupole ion trap (LQIT) systems that have been connected via an ion transfer octupole encased in a machined manifold. Tandem MS experiments can be performed in the front trap and then the resulting product ions can be transferred via axial ejection into the back trap for further, independent tandem MS experiments in a differentially pumped area. This approach allows the examination of consecutive collision-activated dissociation (CAD) and ion−molecule reactions without unwanted side reactions that often occur when CAD and ion−molecule reactions are examined in the same space. Hence, it greatly facilitates investigations of ion structures. In addition, the overall lower pressure of the DLQIT, as compared to commercial LQIT instruments, results in a reduction of unwanted side reactions with atmospheric contaminants, such as water and oxygen, in CAD and ion−molecule experiments.

I

energetic collisions with gaseous atoms or molecules, causing the ions to fragment [i.e., collision-activated dissociation (CAD)].6,7 The ability to do this many times in a consecutive manner (i.e., to examine structures of fragment ions and their fragment ions) provides a substantial advantage over scanningtype mass spectrometers. Another way to provide structural information is using diagnostic ion−molecule reactions. The ability to store isolated ions in a confined space for a variable time period is ideal for facilitating these type of studies. Ion− molecule reactions have been used to identify functional groups,8−13 to provide detailed structural information for ions,14−16 and even to differentiate isomers17−20 when CAD does not provide useful information. MSn utilizing ion− molecule reactions, especially when combined with information gained from CAD on isolated reaction products, has proven to be a powerful tool for the structural elucidation of unknown analytes.21−24 Previously, the combined use of ion−molecule and CAD reactions in dual-cell FT-ICR mass spectrometers has allowed the differentiation of protonated ketones, esters,

on trap mass spectrometers have had a substantial impact on the field of mass spectrometry.1 In these mass spectrometers, ions are accumulated in a confined space where they are later manipulated and/or detected. This approach provides numerous advantages over mass filter type mass spectrometers, such as quadrupole and magnetic sector mass spectrometers, which separate ions by using electric and/or magnetic fields that allow only ions of a selected m/z window to have a stable trajectory through the analyzer at a given time. For example, modification of a tandem mass spectrometry (MS/MS) quadrupole or magnetic sector instrument to enable MS/MS/MS experiments requires the physical addition of another mass-analyzing device, while for an ion trap, no hardware changes are necessary since unwanted ions can be ejected from the trap many times during an experiment. As a result, ion traps have become a standard choice for multiple-stage tandem mass spectrometry (MSn) experiments.1 Furthermore, ion trap mass spectrometers are highly sensitive, as ions can be accumulated for a long period of time prior to detection.2−5 This provides for a significant advantage in complex mixture analysis when the ions of interest may have low abundance. In addition, the structures of the accumulated ions can be examined by subjecting them to © XXXX American Chemical Society

Received: June 28, 2013 Accepted: October 31, 2013

A

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 1. Schematic of the differentially pumped dual LQIT (DLQIT), where MP00, MP0, and MP1 are the first, second, and third multipoles of the instruments’ ion optics, respectively, LQIT1 and LQIT2 are the first and second linear quadrupole ion traps (LQIT), respectively, and API is the atmospheric pressure ionization source. The arrows indicate points of differential pumping.

two stage vacuum baffle to lower the pressure of additional downstream vacuum stages of the instrument from 10−3 to 10−4 Torr via the first stage of the triple-stage turbo pump and the two Edwards EM30 roughing pumps. In the new configuration, this is also not necessary. A vacuum manifold was constructed to connect LQIT1 and LQIT2 and house an 11.8 in. (300 mm) long octupole provided by Thermo Scientific. The front and back faces of the new bridging vacuum manifold were designed to mimic that of the back vacuum manifold flange of LQIT1 and the front vacuum manifold flange of LQIT2, respectively, for facile integration. The vacuum manifold was left open at the top to allow for easy introduction of the octupole and to create a cohesive ion optics system that allows ions to travel from LQIT1 into LQIT2. Thus, a top flange was also constructed for this connecting manifold. Inside the above “boat-shaped” manifold, a support was provided for the octupole to minimize any sag. A second support was created for the end of the octupole placed into LQIT2 that mimicked the vacuum baffle housing for lens 0. Both supports are circular sections of PEEK plastic material that were shaped to fit the middle and end sections of the transfer octupole. Specifically, the support created for the end of the octupole placed into LTQ2 contained long screws to replace the old contact points used to supply the MP00 voltages. This enabled the use of the same power supplies to drive this new octupole that would have supplied MP00. In addition, the vacuum flange normally used for the evacuation of the API stack region was plugged to allow for efficient forepumping of the turbo pump that pumps the main instrument vacuum manifold and for monitoring of the forepumping pressure. Scan Function Considerations. Once constructed, the manifold was placed under vacuum and confirmed to be vacuum tight. In order to transfer ions axially out of the back of LQIT1, modified ion trap control language (ITCL) code was created (by Thermo Fisher Scientific). The modified ITCL code provides the ability to increase and decrease the DC voltages applied to various ion trap sections and lenses to facilitate efficient transfer of ions out of LQIT1 and into LQIT2. A schematic of the sections pertinent to transferring ions is given in Figure 2. New values were given to several ion trap sections when in axial ejection mode to allow for easy control and tuning of the process of ejection of ions via the use of the ion optic tuning parameters in the user interface (UI). A schematic of the axial ejection process is given in Figure 2 and shows the DC potential wells created by the DC offsets applied to the different trap sections. Initially, an oscilloscope was connected to the existing test points on the analog and RF boards of the LQITs to monitor that the desired changes were

amides, aldehydes, polyols, and many bifunctional oxygencontaining compounds.25−29 Ideally, to avoid the occurrence of unwanted side reactions, CAD and ion−molecule reactions need to be performed in separate (differentially pumped) environments, such as in the different cells of the dual-cell FT-ICRs discussed above. However, dual-cell FT-ICR mass spectrometers have become obsolete and are no longer manufactured due to their high purchase price, need for highly trained operators, and expensive maintenance. As a result, no commercial instruments are available that can be used to perform consecutive ion−molecule reactions and CAD, each for mass-selected ions, in separate, differentially pumped regions, as in the experiments described above. The aim of this research was to design and construct a novel instrument capable of those experiments. This was accomplished by combining two commercial LQIT mass spectrometers with differentially pumped vacuum chambers. This novel dual linear quadrupole ion trap (DLQIT) mass spectrometer was then evaluated based on its ability to perform experiments involving consecutive CAD and ion−molecule reactions of mass-selected ions without interference from each other.

■

EXPERIMENTAL SECTION Two Thermo Scientific LTQ linear quadrupole ion trap (LQIT) mass spectrometers were used to construct the new instrument. The front instrument was equipped with ESI and APCI sources and was coupled to a Thermo Scientific Surveyor Plus high-performance liquid chromatography (HPLC) stack consisting of a quaternary pump, autosampler, heated column compartment, and photodiode array detector. This instrument will be henceforth referred to as LQIT1. The second, or back LQIT, will be referred to as LQIT2 and was utilized without an ionization source or HPLC but otherwise was just like LQIT1. It was coupled with LQIT1, as described below. Instrument Design. Figure 1 gives a schematic of the differentially pumped dual LQIT instrument. To construct the instrument, the back vacuum manifold cover from LQIT1 and the front vacuum manifold cover of LQIT2 were removed. When removing the front cover of LQIT2, various ion optics necessary for traditional ion transfer from the atmospheric pressure ionization (API) source into the LQIT were removed. The most notable is the API stack, which includes the sweep cone, ion transfer capillary, tube lens, and skimmer lenses. Additionally, the housing for the API stack was removed. The housing not only separates the main vacuum manifold chamber from the atmosphere but also contains the electrical connections for the API stack as well as the first multipole (MP00) and a subsequent lens (lens 0), all of which were no longer necessary. The API stack housing also normally acts as a B

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

into a tighter packet in the center of the trap. This step is typical of all mass analysis scans performed in the LQIT. (2) All DC voltages applied to the trap sections and lenses are raised (trapoffset) simultaneously to create a start point for the DC gradient. (3) Next, the DC voltage on the back section of the trap is lowered while the DC voltage on the front and center sections are simultaneously set to be higher than the back section. This action takes the previous tightly concentrated packet of ions and transfers it into the back section of the trap. The trap voltages are held at these values for a predetermined amount of time (transfertime) to ensure efficient concentration of the ion packet in the back section of the ion trap. (4) While the applied DC voltage on the back section of the trap is held constant, the DC voltage applied to the back lens is dropped below the voltage applied to the back section. Simultaneously, the DC voltage applied to the center section is ramped higher above the DC voltage applied to the back section of the trap. This causes the concentrated ion packet located in the back section of the ion trap to begin exiting the trap through the aperture of the back lens. The trap voltages are held at these values for a predetermined amount of time (axialejecttime) to ensure efficient ejection of the ion packet. (5) Once the ion packet is ejected, the voltage on the back lens is raised up to close the ion gate and (6) all applied voltages are returned to preinjection values. The semiautomatic tuning feature of the LTQ Tune Plus interface was utilized to efficiently tune the transfer of ions from LQIT1 into LQIT2. The ITCL code of LQIT2 was not significantly modified; however, to ensure efficient overlap between ejection of the ion packet from LQIT1 and its injection into LQIT2, the integrated triggering function of the LTQs was used. The source trigger from LQIT1 was connected to the external trigger analog input of LQIT2, and conversely, the source trigger from LQIT2 was connected to the external trigger analog input of LQIT1. In this way, the position of both triggers (0 or +3 V) allows for the integration of qualifiers into the ITCL of the instrument to wait or perform certain scan functions based on the trigger position of each LQIT. This means that LQIT2 receives a signal from LQIT1 once ions are ejected from LQIT1 and, thus, injected into LQIT2. Upon scanning out ions from LQIT2 for detection, LQIT2 sends a signal to LQIT1 and LQIT1 injects a new packet of ions into LQIT2 and the process repeats. Integrating the triggering of the two instruments ensured synchronization of the two instruments and that the most efficient duty-cycle was obtained. Due to the required synchronization process, automatic gain control (AGC) was temporarily disabled for all experiments, which requires more complicated ITCL code and scan functions. Much effort was given in all experiments to ensure that a stable API spray was accomplished and that an acceptable injection time was chosen to reach the AGC target value for ions in LQIT1 in order to avoid space charge effects. The triggering scheme allowed for MSn experiments to be performed in each LQIT independently and the timing for transfer and scanning ions out of the LQITs to be coordinated. Sample Analysis. Analytes at a concentration of 1 mg/mL in 50:50 methanol:water (v/v) were introduced into LQIT1 via an integrated syringe drive (5 μL/min) and combined with HPLC eluent [50:50 methanol:water (v/v) at a flow rate of 100 μL/min] via a tee connector to facilitate a stable spray. The sample stream was introduced into either the ESI or APCI source. The ESI conditions were 3−4 kV spray voltage, 20 and 10 (arbitrary units) N2 sheath and auxiliary gas flow,

Figure 2. Definition of the sections of the ion trap and 1−6: Schematic of the axial-ejection sequence performed in LQIT1. The center section corresponds to the lowest point of the DC potential well in traditional operation.

occurring appropriately, but the use of the oscilloscope is not necessary to use the instrument. Axial ejection of ions from LQIT1 is achieved by the following steps (Figure 2): (1) First, the DC voltages of the back and front sections of LQIT1 are raised, thus heightening the walls of the pseudopotential well and concentrating ions C

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 3. Positive-ion mode ESI mass spectra of Thermo Scientific calibration solution measured in LQIT1 (left) and in LQIT2 after ion transfer from LQIT1 into LQIT2 (right).

Figure 4. Schematics of the optimal DC voltages applied to the trapping lenses and ion trap sections (defined in Figure 2) and timing for the ejection of ions from LQIT1 into LQIT2.

respectively, and an inlet temperature of 275 °C. APCI conditions were 3−5 μA corona discharge current, 300 °C vaporizer temperature, 30 and 10 (arbitrary units) N2 sheath and auxiliary gas flow, respectively, and an inlet temperature of 275 °C. In both LQIT1 and LQIT2, experiments were performed using the LTQ Tune Plus interface. For CAD experiments, the advanced scan features of the LQIT were used to isolate ions at an m/z window of 2 units. After isolation, the ions were subjected to CAD at a normalized collision energy of 20−40% using helium as a collision gas for 30 ms. CAD experiments were also performed nonspecifically or without the isolation of ions via in-source CAD. In in-source CAD, the kinetic energies of all ions entering the mass spectrometer are increased via the application of a larger DC offset to all ion optics from MP00 back to the ion trap of LQIT1 (set in the LTQ Tune Plus interface). This causes the ions to undergo more energetic collisions than normal with atmospheric gas molecules (mostly N2) natively present in the higher-pressure region of the LQIT (MP00; 50−100 mTorr), inducing fragmentation of the ions before entering the mass analyzer. Ion−molecule reactions were examined in the ion trap by isolating the desired ions by using an m/z window of 2 units and allowing them to react with neutral molecules in the trap for a preselected reaction time (from 0.03 ms to 10 s). The temperature of the ions was assumed to be room temperature based on studies by Gronert and Donald.29−31 The reagents were introduced into the helium line of the instrument via a previously described mixing manifold.29

■

Ion Transfer. To evaluate the efficiency of transfer of ions from LQIT1 into LQIT2, Thermo Scientific calibration solution was analyzed by positive-ion mode ESI. After recording a mass spectrum in LQIT1, axial ejection of the ions was performed and the mass spectrum was recorded in LQIT2. All LQIT1 ejection voltages and their timing and LQIT2 injection voltages and their timing were tuned for maximum total ion current (TIC) after ion transfer. Figure 3 gives the results of this experiment. The optimal voltages and timing are given in Figure 4. On the basis of the above results, the transfer efficiency of ions with a wide mass range is calculated to be about 30%, meaning that about 30% of the original ions in LQIT1 (TIC) were transferred into LQIT2 (TIC). In a different experiment, an ion of a single m/z value (m/z 524 from the calibration solution) was isolated before transfer by ejecting all other ions out, and the ion was transferred into LQIT2 by optimizing the voltages and their timing for the selected ion. The transfer efficiency was 40−50% (note that the gains of both LQITs’ electron multipliers were carefully calibrated for this measurement). Differentially Pumped, Independently Operating Ion Traps. Differential pumping was accomplished in DLQIT through the use of two ion traps in different vacuum manifolds, each of which was evacuated through the use of a different turbo pump and separated by lens elements serving as conductance limits. LQIT1 used the final stage of a tripleported Oerlikon Leybold turbo pump to reach the final pressure in the ion trap vacuum manifold. This turbo was forepumped by two Edwards EM30 rough pumps (foreline pressure of ∼1 Torr). The trap section of the vacuum manifold of LQIT2 was also evacuated using the final stage of a second triple-ported Oerlikon Leybold turbo pump. The remainder of the vacuum manifold of LQIT2 was evacuated using both of the first two stages of the triple-ported Oerlikon Leybold turbo pump for LQIT2. Also, this turbo pump was forepumped by two Edwards EM30 rough pumps (foreline pressure lower than

RESULTS AND DISCUSSION

To assess the performance of the new instrument, different samples were analyzed by using experiments involving consecutive CAD and ion−molecule reactions of mass-selected ions. The performance of the instrument in these experiments is discussed below with regard to ion transfer and benefits of differentially pumped, independently operating ion traps. D

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Scheme 1. Proposed Mechanisms for CAD (MS2) of Protonated 9-Fluorenone-4-Carboxylic Acid (m/z 225) via Loss of Water (to Form Ion of m/z 207) Followed by Isolation of the Product Ion and Another CAD (MS3) via Loss of CO (to Form Ion of m/z 179)a

a

Addition of water occurs to the CO loss product ion (to form ion of m/z 197).32,33.

100 mTorr). The vacuum manifold connecting the two linear ion traps was not monitored for pressure. The background pressures of LQIT1 and LQIT2 were measured by individual ion gauges (calibrated by the vendor with nitrogen) when the He line was closed and the API inlet of LQIT1 was open (unplugged). Under such conditions, LQIT1 was maintained at a nominal pressure of 0.21 × 10−5 Torr, while LQIT2 was maintained at 0.11 × 10−5 Torr. The somewhat lower pressure observed in LQIT2 suggests that the ambient gases from the source do not contribute substantially to the pressure in LQIT2 (which is supported by the results discussed later in this paper). The overall background pressure of both LQITs is somewhat lower when compared to the background pressure of an unaltered LQIT without helium (0.34 × 10−5 Torr). However, helium buffer gas is always present in the traps and the amount of helium cannot be regulated accurately with the commercial set up. Therefore, the pressures that we measured using the ion gauges are not very accurate. Experiments were carried out to investigate the utility of the new instrument in situations wherein background gases, such as water and oxygen, interfere with CAD studies of ions. For example, 9-fluorenone-4-carboxylic acid (MW 224 Da) was protonated by using positive-ion mode APCI, isolated and subjected to CAD in LQIT and DLQIT. Upon CAD (MS2 experiment), protonated 9-fluorenone-4-carboxylic acid rapidly loses water (to yield an ion of m/z 207). When this product ion (m/z 207) was isolated and subjected to CAD (MS 3 experiment), it yielded two product ions: an ion of m/z 179 from the loss of CO and an ion of m/z 197 from addition of adventitious water to the ion resulting from the loss of CO (m/ z 179). The latter ion is undesirable since it is not a CAD product of the ion of interest (m/z 207). Scheme 1 shows the reaction sequence.32,33 This MS3 experiment was performed in three ways: (1) MS3 in LQIT (Figure 5a), (2) MS3 in the front trap of DLQIT (Figure 5b), and (3) MS2 in the front trap of DLQIT followed by the transfer of ions of m/z 207 into the back trap where MS3 was performed (Figure 5c). The reaction time (30 ms) was the same for all of these MS3 experiments. The branching ratios of the product ions for these experiments are given in Table 1. As can be seen in Figure 5 and Table 1, the amount of the undesirable water addition product (m/z

Figure 5. MS3 CAD mass spectra measured for a product ion (m/z 207) formed upon collision-activated loss of water from protonated 9fluorenone-4-carboxylic acid (m/z 225) in an MS2 experiment. Loss of CO followed by addition of water (see Scheme 1) was observed in (a) LQIT, (b) the front trap of DLQIT, and (c) the back trap of DLQIT. Much less water addition was observed in the back trap of DLQIT.

Table 1. Branching Ratios of Product Ions of m/z 179 (Loss of CO) and m/z 197 (Loss of CO Followed by Addition of Water) Generated upon CAD of Protonated 9-Fluorenone4-Carboxylic Acid (m/z 225) in LQIT and Front and Back Traps of DLQIT LQIT m/z 179 m/z 197

44% 56%

front trap of DLQIT

back trap of DLQIT

m/z 179 m/z 197

m/z 179 m/z 197

45% 55%

74% 26%

197) is the same for LQIT and LQIT1. However, the experiment performed in the back trap of DLQIT produced E

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

less than half of the amount of the ion of m/z 197 compared to LQIT and LQIT1. In order to test the usefulness of the new instrument in MS3 experiments that involve consecutive ion−molecule reactions and CAD, CAD of the isolated product ion of an ion−molecule reaction between trimethyl borate (TMB) and protonated furfural (a molecule relevant for the pyrolysis of cellulose34) was examined. In this experiment, the reagent (TMB) was introduced through a reagent manifold31 connected to the helium line of the front trap of DLQIT. After generation of protonated furfural (m/z 97) via positive-ion mode ESI, the ion was transferred into the front trap, isolated, and allowed to react with TMB for 30 ms to give a product ion formed by the addition of TMB (+104 Da) followed by the loss of methanol (−32 Da) to produce a product at (m/z 169). The formation of this ion (an ion with a m/z value of 72 units greater than the m/z value of the protonated molecule) is a diagnostic reaction that reveals the presence of an oxygen-containing functionality. In order to examine whether further structural information could be obtained, the product ion (m/z 169) was isolated and subjected to CAD (MS3 experiment) in the front trap of DLQIT, where the TMB reagent was still present, as it would be in LQIT (Figure 6a). In addition to the real CAD fragment

in the front trap were observed. Furthermore, this allowed more information to be gained on the product ion of the reaction of TMB with protonated furfural. Specifically, a new product ion (dimethoxyborenium cation of m/z 73) was observed. When TMB is present, this ion of m/z 73 reacts away very quickly to form an adduct with TMB (m/z 177), as can be seen in Figure 6a. Finally, the influence of reactive background gases that may interfere with CAD and ion−molecule reactions was studied by examining the reactivity of the 5-dehydroisoquinolinium cation toward cyclohexane in LQIT (Figure 7a) and DLQIT (Figure

Figure 7. Mass spectra measured after 500 ms reaction of the 5dehydroisoquinolinium cation with cyclohexane in (a) LQIT and (b) the back trap of DLQIT. Interfering ions produced upon reactions of the 5-dehydroisoquinolinium cation with O2, H2O, and other background gases are marked by red stars. Ion of m/z 161 in top spectrum corresponds to the O2 adduct of the radical and ion of m/z 145 corresponds to the O adduct of the radical. The identities of the other impurity ions are unknown.

Figure 6. CAD mass spectra of the product ion (m/z 169) formed from protonated furfural upon reaction with TMB in (a) the front trap of DLQIT in the presence of TMB and (b) the back trap of DLQIT without the presence of TMB. Interfering ions formed upon ion− molecule reactions of the CAD products with TMB are marked by red stars in the top spectrum. Ion of m/z 209 in the top spectrum corresponds to a proton-bound dimer of TMB, the ion of m/z 177 corresponds to the (CH3O)2B+ adduct of TMB and the ion of m/z 163 corresponds to the CH3OBOH+ adduct of TMB.

7b). To accomplish this, 5-iodoisoquinoline was protonated by a positive-ion mode APCI to yield an ion of m/z 256. When subjected to ion source CAD upon injection into LQIT or DLQIT, this ion generated a distonic radical cation (m/z 129) by a homolytic cleavage of the iodine−carbon bond. This distonic ion (m/z 129) was allowed to react for 500 ms with cyclohexane introduced via the reagent manifold into LQIT. In addition to the only real product ion of m/z 130 arising from hydrogen atom abstraction, several misleading product ions were formed in reactions of the distonic ion with background gases (O2, H2O, etc.). When this same reaction was performed in the back trap of DLQIT, these unwanted product ions were entirely eliminated (Figure 7b).

ions, several ions were observed that were products of unwanted ion−molecule reactions of the CAD product ions with TMB. In another experiment, the ion of m/z 169 was isolated in the front trap and immediately transferred into the back trap of DLQIT where CAD was performed (Figure 6b). None of the undesired ion−molecule reaction products formed

CONCLUSIONS Traditionally, tandem mass spectrometry experiments using either collision-activated dissociation (CAD) or ion−molecule reactions of isolated ions have been a vital tool for the structural characterization of unknown compounds directly in mixtures. When these two tandem mass spectrometry methods are used together, they are much more powerful than when used

■

F

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

(21) Brodbelt, J. S. Mass Spectrom. Rev. 1997, 16, 91. (22) Eberlin, M. J. Mass Spectrom. 2006, 41, 141. (23) Green, M.; Lebrilla, C. Mass Spectrom. Rev. 1997, 16, 53. (24) Osburn, S.; Ryzhov, V. Anal. Chem. 2013, 85, 769. (25) Campbell, K. M.; Watkins, M. A.; Li, S.; Fiddler, M. N.; Winger, B.; Kenttamaa, H. I. J. Org. Chem. 2007, 72, 3159. (26) Guler, L. P.; Yu, Y. Q.; Kenttamaa, H. I. J. Phys. Chem. A 2002, 106, 6754. (27) Petucci, C.; Guler, L.; Kenttamaa, H. I. J. Am. Soc. Mass Spectrom. 2002, 13, 362. (28) Somuramasami, J.; Duan, P.; Watkins, M. A.; Winger, B. E.; Kenttämaa, H. I. Int. J. Mass Spectrom. 2007, 265, 359. (29) Thompson, R. S.; Guler, L. P.; Nelson, E. D.; Yu, Y. Q.; Kenttämaa, H. I. J. Org. Chem. 2002, 67, 5076. (30) Donald, W. A.; Khairallah, G. N.; O’Hair, R. A. J. J. Am. Soc. Mass Spectrom. 2013, 24, 811. (31) Gronert, S. J. Am. Soc. Mass. Spectrom. 1998, 9, 845. (32) Amundson, L. M.; Gallardo, V. A.; Vinueza, N. R.; Owen, B. C.; Reece, J. N.; Habicht, S. C.; Fu, M.; Shea, R. C.; Mossman, A. B.; Kenttaemaa, H. I. Energy Fuels 2012, 26, 2975. (33) Amundson, L. M.; Owen, B. C.; Gallardo, V. A.; Habicht, S. C.; Fu, M. K.; Shea, R. C.; Mossman, A. B.; Kenttämaa, H. I. J. Am. Soc. Mass Spectrom. 2011, 22, 670. (34) Shen, D. K.; Xiao, R.; Gu, S.; Luo, K. H. RSC Adv. 2011, 1, 1641.

separately. However, they cannot be used together in an optimal manner with current commercially available instrumentation due to their lack of two distinct differentially pumped reaction chambers wherein ions can be independently manipulated. As a result, a novel tandem mass spectrometer, a dual linear quadrupole ion trap mass spectrometer (DLQIT), was designed, constructed, and characterized for its ability to facilitate investigations of ions’ structures via ion−molecule reactions followed by CAD without the experiments interfering with each other. The back trap of the DLQIT mass spectrometer contains smaller amounts of reactive background gases that complicate CAD and ion−molecule reaction studies in LQIT, thus resulting in cleaner tandem mass spectrometry experiments. Also, being able to study ion−molecule reactions and CAD in differentially pumped areas wherein ions can be manipulated independently affords for less complicated product ion mass spectra and a greater degree of certainty of the true product ions formed in these reactions.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (765) 494-0882. Fax: (765) 494-0239. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported as part of the Center for Direct Catalytic Conversion of Biomass to Biofuels (C3Bio), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0000997.

■

REFERENCES

(1) McLuckey, S. A.; Wells, J. M. Chem. Rev. 2001, 101, 571. (2) Douglas, D. J.; Frank, A. J.; Mao, D. M. Mass Spectrom. Rev. 2005, 24, 1. (3) Hager, J. W. Rapid Commun. Mass Spectrom. 2002, 16, 512. (4) March, R. E.; Todd, J. F. J. Chem. Anal. Series Mon. 2005, 165, 1. (5) Schwartz, J. C.; Senko, M. W.; Syka, J. E. P. J. Am. Soc. Mass Spectrom. 2002, 13, 659. (6) Busch, K. L. G.; G., L.; Mcluckey, S. A. Biol. Mass Spectrom. 1988, 18, 942. (7) Sleno, L.; Volmer, D. A. J. Mass Spectrom. 2004, 39, 1091. (8) Kenttamaa, H. I.; Cooks, R. G. J. Am. Chem. Soc. 1989, 111, 1654. (9) Meurer, E.; Chen, H.; Riter, L.; Cooks, R.; Eberlin, M. J. Am. Soc. Mass Spectrom. 2004, 15, 398. (10) Meurer, E.; Sparrapan, R.; Eberlin, M. J. Mass Spectrom. 2003, 38, 1075. (11) Petzold, C.; Leavell, M.; Leary, J. Anal. Chem. 2004, 76, 203. (12) Watkins, M.; WeWora, D.; Li, S.; Winger, B.; Kenttämaa, H. Anal. Chem. 2005, 77, 5311. (13) Zhang, J.; Ji, H.; Sun, S.; Mao, D.; Liu, H.; Guo, Y. J. Am. Soc. Mass Spectrom. 2007, 18, 1774. (14) Cole, M. J.; Enke, C. G. J. Am. Soc. Mass Spectrom. 1991, 2, 470. (15) Lennon, J.; Cole, S.; Glish, G. Anal. Chem. 2006, 78, 8472. (16) Reid, G. E.; O’Hair, R. A. J.; Styles, M. L.; McFadyen, W. D.; Simpson, R. J. Rapid Commun. Mass Spectrom. 1998, 12, 1701. (17) Gao, H.; Petzold, C. J.; Leavell, M. D.; Leary, J. A. J. Am. Soc. Mass Spectrom. 2003, 14, 916. (18) Mosi, A. A.; Cullen, W. R.; Eigendorf, G. K. Int. J. Mass Spectrom. 1999, 190−191, 195. (19) Niemeyer, E.; Brodbelt, J. J. Am. Soc. Mass Spectrom. 2007, 18, 1749. (20) Zhang, J.; Brodbelt, J. J. Am. Chem. Soc. 2004, 126, 5906. G

dx.doi.org/10.1021/ac401956f | Anal. Chem. XXXX, XXX, XXX−XXX