您所在位置： &
&nbsp&&nbsp&nbsp&&nbsp
超高效液相色谱_高分辨质谱法快速筛查土豆中的多种农药残留_陈达炜.pdf全文-资格考试-在线文档 6页
本文档一共被下载：
次 ,您可全文免费在线阅读后下载本文档。
下载提示
1.本站不保证该用户上传的文档完整性，不预览、不比对内容而直接下载产生的反悔问题本站不予受理。
2.该文档所得收入(下载+内容+预览三)归上传者、原创者。
3.登录后可充值，立即自动返金币，充值渠道很便利
需要金币：120 &&
超高效液相色谱_高分辨质谱法快速筛查土豆中的多种农药残留_陈达炜.pdf
你可能关注的文档：
··········
··········
网络出版时间： 11:44
分析化学 (FENXI HUAXUE)摇 研究报告
网络出版地址：/kcms/detail/22..046.html
摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇
摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇
2014 年4 月
Chinese Journal of Analytical Chemistry
DOI:10.3724/ SP.J.989
超高效液相色谱鄄高分辨质谱法快速筛查土豆中的多种农药残留
陈达炜摇 高 洁摇 吕冰摇 朱 盼摇 杨 欣 摇 赵云峰摇 苗 虹
(卫生部食品安全风险评估重点实验室,国家食品安全风险评估中心,北京 100021)
摘摇 要摇 采用超高效液相色谱鄄四级杆鄄静电场轨道阱高分辨质谱联用技术(UPLC/ Q Orbitrap),建立土豆中农
药多残留的快速筛查方法。 样品采用乙腈为提取溶剂,PSA分散固相萃取净化。 以BEH C 色谱柱进行色谱
分离,通过静电场轨道阱质谱全扫描获得农药的精确质量数,以Full Scan/ ddMS 进行定性筛查和定量检测。
对欧盟考核样品土豆中175种农药残留进行分析,共从考核样品中定性筛查出17种农药残留。 17种农药的
高分辨质谱分析方法定量限为1~5 滋g/ kg,在1~250 滋g/ L 的浓度范围内均呈良好的线性关系(R &0.99);
平均加标回收率为83.1% ~115.5%,相对标准偏差在1.5% ~11.8%之间。 定量测定 17种农药的含量范围
在0.002~1.714 mg/ kg之间,z评分在-1.00~1.24之间。 本方法简单精确,灵敏度高,样品处理快捷简便,适
用于农产品中农药多残留的快速筛查。
关键词摇 四极杆鄄静电场轨道阱高分辨质谱仪;农药残留;土豆
1摇 引摇 言
农药残留问题一直受到全球关注,各国在严格限制高毒农药使用的同时,加强了农药残留限量标准
的制定。 我国最新颁布的GB《食品中农药最大残留限量》在食品品种及限量方面均提出了
更高要求。 尽管如此,我国出口农产品常因农药残留超标而影响国际贸易。 而且,因农药残留引起的食
物中毒事件时有发生
。 因此,建立农产品中农药多残留的快速筛查方法十分必要。
目前,常采用多残留分析技术开展食品及农产品中农药残留检测 ,应用的主要方法有气相色谱/
及液相色谱/ 质谱法
,且为保证检测方法的特异性,多采用串联质谱技术。 但是,由于食
物样品基质的复杂性,在多组分残留分析中,低分辨质谱常出现假阳性,造成结果误判。 四级杆飞行时
间质谱(QTOF)和静电场轨道阱高分辨质谱(Orbitrap)的发展在很大程度上弥补了这个缺陷,被广泛应
用于农药残留的快速筛查[10~13]。 普通高分辨质谱虽能降低假阳性结果,但其定量灵敏度不及三重四级
杆质谱。 Q Exac
正在加载中，请稍后...您的访问出错了(404错误)
很抱歉，您要访问的页面不存在。
1、请检查您输入的地址是否正确。
2、感谢您使用本站，通过
进行查找。
3、使用搜索框直接进行搜索Orbitrap mass spectrometry: Instrumentation, ion motion and applications - Perry - 2008 - Mass Spectrometry Reviews - Wiley Online Library
Advertisement
Since its introduction, the orbitrap has proven to be a robust mass analyzer that can routinely deliver high resolving power and mass accuracy. Unlike conventional ion traps such as the Paul and Penning traps, the orbitrap uses only electrostatic fields to confine and to analyze injected ion populations. In addition, its relatively low cost, simple design and high space-charge capacity make it suitable for tackling complex scientific problems in which high performance is required. This review begins with a brief account of the set of inventions that led to the orbitrap, followed by a qualitative description of ion capture, ion motion in the trap and modes of detection. Various orbitrap instruments, including the commercially available linear ion trap&orbitrap hybrid mass spectrometers, are also discussed with emphasis on the different methods used to inject ions into the trap. Figures of merit such as resolving power, mass accuracy, dynamic range and sensitivity of each type of instrument are compared. In addition, experimental techniques that allow mass-selective manipulation of the motion of confined ions and their potential application in tandem mass spectrometry in the orbitrap are described. Finally, some specific applications are reviewed to illustrate the performance and versatility of the orbitrap mass spectrometers. & 2008 Wiley Periodicals, Inc., Mass Spec Rev 27: 661&699, 2008INTRODUCTIONThe increasing complexity of biological samples encountered in proteomic and metabolomic studies continues to push the technological limits of analytical instrumentation. Mass spectrometry (MS) has increasingly become an analytical tool of choice in these areas owing to its speed, wide dynamic signal range, quantitative capability and the facility to interface with chromatographic separation methods. Reliable identification of metabolites and the sequences and post-translational modifications (PTMs) of proteins in complex mixtures necessarily requires robust mass spectrometers with high resolving power, mass accuracy, sensitivity and dynamic range. In addition, tandem mass spectrometry (MS/MS) (Busch, Glish, & McLuckey, ; Cooks et al., ; de Hoffmann, ; McLafferty, , , ) serves as a requisite and invaluable tool for structure elucidation and peptide/protein sequencing (Biemann, ; Biemann & Scoble, ; Hunt et al., ). Although the three-dimensional Paul trap (Paul & Steinwedel, ) and Fourier transform ion cyclotron (FT-ICR) (Comisarow & Marshall, ) are widely used mass analyzers, the low mass resolution and accuracy of the Paul trap (unit resolving power and 20&40 ppm accuracy for the quadrupole ion trap (Cooks, Cox, & Williams, )), and the high complexity and cost of the FT-ICR, as well as the relatively low space-charge capacities of both analyzers, suggests why new approaches to ion trapping are welcome in tackling the increasingly complex problems in biological mass spectrometry (Makarov, ).These bioanalytical demands were addressed by such hybrid instruments as the quadrupole/time-of-flight (QqTOF, where Q refers to a mass-resolving quadrupole, q to a radio frequency (RF)-only quadrupole or hexapole collision cell and TOF to a time-of-flight mass spectrometer) (Morris et al., ) and the linear quadrupole ion trap/Fourier transform ion cyclotron resonance (LIT/FT-ICR) (Syka et al., ) that perform in the high resolution (&10,000) and high mass accuracy (&5 ppm) regimes (Marshall, Hendrickson, & Jackson, ; McLuckey & Wells, ; Williams et al., ). The QqTOF, which can be regarded as a triple quadrupole (QqQ) instrument where the third quadrupole is replaced by an orthogonal TOF, was originally used for rapid de novo peptide sequencing (Shevchenko et al., ) but has found application in many areas of research including metabolite (Levsen et al., ), nucleic acid (Oberacher, Niederstatter, & Parson, ) and glycoprotein (Morris et al., ) analysis. However, for scan-types in which a single ion is monitored such as precursor ion and neutral loss scans (important experiments for structural characterization), the sensitivity of the QqTOF (5&30% duty cycle) is lower than QqQ instruments because more ions are lost in TOF compared with a third quadrupole. These losses occur during transfer into the orthogonal TOF, on grids and at the detector (Chernushevich, ; Chernushevich, Loboda, & Thomson, ).Many of these limitations were overcome with the LIT/FT-ICR, which combines the speed, large trapping capacity, MSn capability and versatility of a LIT (Hager, ) with the unsurpassed high mass accuracy, resolving power, sensitivity and dynamic range of the FT-ICR (Marshall, Hendrickson, & Jackson, ). However, the large size, complexity and cost of the FT-ICR restrict the laboratory settings where it can be utilized. Therefore, a more compact, less costly, easier-to-maintain analyzer with comparable performance (for relatively short acquisition times &1.8 sec) was desired to supplement the FT-ICR. This technological gap was filled by the LTQ-Orbitrap (LTQ stands for &linear trap quadrupole&) hybrid mass spectrometers (Thermo Electron Corporation, San Jose, CA) (Makarov et al., ,; Olsen et al., ).Currently, there are two commercial LTQ-Orbitrap instruments, the Discovery and XL models. One of the primary differences is that the XL has a linear octopole collision cell (absent in the Discovery model), in which collisional activation and fragmentation can be performed. Although this feature provides additional versatility to MS/MS experiments, the analytical performance and fundamental principles of operation of the orbitrap analyzers in both instruments are identical. In the LTQ-Orbitrap, precursor ions can be dissociated in (i) the LIT (collision induced dissociation (CID), both models), (ii) the C-trap, a curved RF-only quadrupole ion trap between the LIT and orbitrap that injects ions into the orbitrap (both models) or (iii) the octopole collision cell (XL model only). Another difference between the two models is that the Discovery model has a limited resolving power of 30,000 (nominal) in MS/MS mode owing to restrictions placed on the transient acquisition time. For simplicity, the term &LTQ-Orbitrap& will be used hereafter to refer to both models and MS/MS spectra refer to CID fragments produced in the LIT unless specified otherwise. Orbitrap instruments that use electrostatic acceleration lenses (Makarov, ) and modified LITs (&storage quadrupole& (StQ)) (Hardman & Makarov, ; Hu et al., ) from which ions are ejected axially and thence injected into the orbitrap have also been described. A major section of this review will focus on the development and performance of these various injection methods. Much of the work in our laboratory has been done using a prototype orbitrap of smaller radius coupled to a StQ in a system without a C-trap (Hu et al., ).Despite its relatively recent commercial introduction, the LTQ-Orbitrap has already proven to be an important analytical tool with a wide range of applications. The high resolving power (&150,000) and excellent mass accuracy (specified as &2&5 ppm, but demonstrated to be as low as 0.2 ppm under favorable conditions) (Makarov et al., ,), significantly reduce false positive peptide identifications in bottom-up protein analyses (Adachi et al., ; Charneau et al., ; Dave et al., ; de Souza, Godoy, & Mann, ; Forner, Arriaga, & Mann, ; Graumann et al., ; Hanke et al., ; Hoopmann, Finney, & MacCoss, ; Kamphorst et al., ; Li et al., ; Lu et al., , ; Manes et al., ; Mayampurath et al., ; Park et al., ; Scherl et al., ; Scigelova & Makarov, ; Shi et al., ; Swatkoski et al., , ; Thevis et al., ; Usaite et al., ; Venable et al., ; Wilson-Grady, &n, & Gygi, 2008; Wisniewski et al., ; Zougman et al., ), and improve the accuracy of de novo interpretations of MS/MS spectra (DiMaggio et al., ; Ford et al., ; Frank et al., ; Scigelova & Makarov, ; Yates et al., ). The LTQ-Orbitrap has also been used to analyze intact small (3&10 kDa) (Bredeh&ft, Sch&nzer, & Thevis, ; Thevis et al., ; Thevis, Thomas, & Sch&nzer, ), medium (10&25 kDa) (Macek et al., ; Waanders, Hanke, & Mann, ) and large (&150 kDa) (Zhang & Shah, ) proteins where the MSn capability of the LIT coupled with the high performance features of the orbitrap can facilitate unambiguous determination of the charge states of fragment ions, as well as identification of PTMs by database searching. The high performance attributes of this hybrid mass spectrometer have also found applications in environmental chemistry (Barcel& & Petrovic, ; Krauss & Hollender, ), drug and metabolite analysis (Breitling, Pitt, & Barrett, ; Cuyckens et al., ; Ding et al., ; Karu et al., ; Li et al., ; Lim et al., , ; Madalinski et al., ; Nielen et al., ; Peterman et al., ; Ruan et al., ; Thevis et al., ; Thevis, Kamber, & Sch&nzer, ; Thevis, Krug, & Sch&nzer, ; Thevis et al., ,, , ; Zhu et al., ), lipidomics (Davis et al., ; Ejsing et al., ; Schwudke et al., ; Silipo et al., ), small-molecule (Chen et al., ) and reaction product (Jamin et al., ) identification, as well as molecular structure characterization using hydrogen/deuterium exchange measurements (Chen et al., ; Hu et al., ).In addition to these applications, experimental methods have been devised that allow confined ion populations to be mass-selectively manipulated (Section V) (Hu et al., ). By applying radio frequency (RF) signals to the outer electrode of the orbitrap analyzer, Hu, Cooks, and Noll () demonstrated that ions can be mass-selectively ejected (i.e., collided with the outer electrode) from the trap with a resolution of 28,400 (up to 100,000 predicted), much higher than the current isolation capabilities of commercial Paul traps. In addition, ions can be de-excited to the equatorial plane of the orbitrap to extend trapping time or reduce specific ion signals without physically ejecting them from the trap. De-excited ions can then be subsequently re-excited for mass analysis (Hu et al., ). Although these experiments were carried out in a prototype StQ-Orbitrap instrument (Hardman & Makarov, ; Hu et al., ), the ability to manipulate the motion of confined ions is independent of the method used to inject them into the orbitrap. Thus, it should be possible to apply these techniques and principles in all orbitrap instruments. Development of these ion motion control (IMC) techniques is a significant advance because they provide the potential for implementing MS/MS inside the orbitrap analyzer and interrogating the dynamics of confined ion packets.This review offers a brief introduction to the fundamental principles (Sections II and III) and applications (Section VI) of the recently introduced orbitrap mass analyzer (Makarov, ). Various orbitrap instruments are also discussed with emphasis on the different methods used to inject ions into the trap (Section IV). The review is directed to readers with no prior experience with orbitrap MS and should provide a broad but thorough understanding of the theory, development, utility and future potential of this novel mass analyzer.ORBITAL TRAPPING AND KINGDON TRAPSThe orbitrap mass analyzer is based on an earlier ion storage device, the Kingdon trap (Kingdon, ), which employs orbital trapping in a purely electrostatic field. In this device a thin-wire cathode (central electrode) is run coaxially through an outer cylindrical anode (outer electrode) with flat end-cap electrodes to enclose the trapping volume (Fig. ). A direct current (DC) voltage is applied between the wire and the cylinder producing a radial logarithmic potential (&P) between the two electrodes given by (1) where r is the radial coordinate, and A and B are constants at a particular value of applied voltage (Hu et al., ). The axial field component due to the end-cap electrodes and non-ideal fields are neglected in Equation . When a positive ion of mass m and charge q is created inside the trap or introduced into it with a sufficient initial velocity component (v) perpendicular to the wire electrode, the ion will adopt a stable orbit around the wire (orbital trapping) provided the potential difference (V) between the central electrode and outer electrode is greater than the value given by (2) where R and r represent the radius of the anode and cathode, respectively (Hull, ; Kingdon, ). Application of a repulsive potential to the end-caps is essential to achieve simultaneous ion trapping in the axial direction (Sekioka, Terasawa, & Awaya, ). Evidently, stable ion motion in the Kingdon trap involves both rotation around and axial motion along the central wire (Lewis, ).Figure&1. Schematic showing a Kingdon trap coupled to a time-of-flight mass spectrometer. (Reproduced, with permission, from Sekioka, Terasawa, and Awaya (). Copyright & 1991 Gordon and Breach, S. A.)The Kingdon trap has been coupled to a variety of detectors such as quadrupole (Vane, Prior, & Marrus, ), time-of-flight (Fig. ) (Sekioka, Terasawa, & Awaya, ) and FT-ICR (Gillig, Bluhm, & Russell, ) mass analyzers, in which it is primarily used for external ion accumulation. It has also been coupled with electron multipliers, Faraday cups, micro-channel plates and photomultiplier tube detectors for spectroscopic interrogation of trapped ions (Church et al., ; Prior & Wang, ; Yang & Church, ; Yang et al., ). In typical experiments, lifetimes of the metastable electronic states of multiply charged ions are measured (Church, Moehs, & Bhatti, ; Moehs & Church, , ; Moehs, Church, & Phaneuf, ; Moehs et al., ; Moehs, Bhatti, & Church, ; Smith, Chutjian, & Greenwood, ; Smith et al., ; Smith, Chutjian, & Lozana, ; Yang & Church, ; Yang et al., ), providing diagnostic empirical information about the electron density and temperature in astrophysical and laboratory plasmas (Church, ; Church et al., ). Submillimeter glass and copper particles (approximately 50&60 &m in diameter) have been confined in the Kingdon trap to study orbital mechanics, with possible implications for understanding the dynamics of asteroids, galaxies, and planetary rings (Biewer et al., ; Robertson, ; Robertson & Alexander, ). Kingdon traps have also been used to study electron capture from neutral atoms by confined ions (Prior, Marrus, & Vane, ; Vane, Prior, & Marrus, ), measure rates of collisional quenching (Calamai & Johnson, , , ; Church, Yang, & Tu, ; Prior, ) and to develop the orbitron ion pump (Douglas, Zabritski, & Herb, ).Variations on the Kingdon trap include (i) using two parallel wires for the central electrode (McIlraith, ), (ii) superimposing an alternating current (AC) voltage on the DC potential (the &dynamic Kingdon trap&) for improved trapping efficiency (Bl&mel, ), and (iii) modifying the outer electrode to produce a harmonic axial potential in addition to the radial logarithmic term of Equation
(Knight, ). The latter configuration (Fig. ), approximating the &ideal Kingdon trap& (Gillig, Bluhm, & Russell, ; Makarov, ), was a step towards development of the orbitrap because it uses the frequency of oscillations of the confined ions along the central electrode for determination of their m/z ratio. The cylindrically symmetric electrostatic potential in the ideal Kingdon trap can be regarded as the combination of a quadrupolar potential (3) and the radial logarithmic potential of Equation
to give the potential distribution (4) where r and z are cylindrical coordinates (the plane of symmetry of the potential is z&=&0), while A and B are constants related to the electrode geometry and applied voltages. The logarithmic potential between the central and outer electrodes provides orbital ion trapping in the radial direction, as in the conventional Kingdon trap, while the quadrupolar potential confines ions axially, allowing them to undergo harmonic oscillation in the z-direction (Knight, ).Figure&2. Schematic showing the Knight-style Kingdon trap (&ideal Kingdon trap&(Gillig, Bluhm, & Russell, ; Makarov, )). The modified shape of the outer electrode produces a quadrupolar potential superimposed upon the logarithmic radial potential of a cylindrical capacitor. Ions are injected through the gap in the outer electrode (at z&=&0), application of RF to the outer electrode excites the ions to begin harmonic oscillations along z. No mass analysis was reported. Trapped ions are monitored by (i) measuring the time-dependent ion current due to continuous axial loss or by (ii) pulsing the central wire electrode positive and collecting the radially ejected ions on the collector plate. (Reproduced, with permission, from Knight (). Copyright & 1981 American Institute of Physics.)Knight used his version of the trap to monitor ions produced by pulsed laser ablation of solid targets. The outer electrode was split at the middle (z&=&0), allowing ions to be injected into the trap from an external source. The device could be used to monitor trapped ions by (i) measuring the time-dependent ion current due to continuous axial loss, or (ii) pulsing the central wire electrode positive and collecting the radially ejected ions on a collector plate located at the equator, z&=&0. Applying an AC frequency between the split outer electrodes allowed the observation of resonances in both the axial and radial ion signals. In both cases however, observed resonances were considerably weaker, broadened, and shifted in frequency from values expected for a quadrupolar field, which led Knight to propose that the central wire might distort the quadrupolar nature of the axial potential (Knight, ). This was confirmed by SIMION calculations for the ideal Kingdon trap (Fig. ), which showed that the central electrode (and trapping volume) should have a spindle-like geometry to achieve a purely harmonic potential in the z-direction (Gillig, Bluhm, & Russell, ). Thus, the ideal Kingdon trap can be created by either (i) shaping the electrodes to match the equipotential lines or (ii) setting voltages on a system of electrodes to form the potential described by Equation . The latter method was demonstrated by Gillig et al. using an FT-ICR wire ion guide cell (Gillig, Bluhm, & Russell, ; Solouki, Gillig, & Russell, ), where the m/z ratio was derived from the cyclotron resonance frequency of the confined ions as in a typical FT-ICR experiment. The former approach was employed by Makarov to create the orbitrap (Fig. ) (Hardman & Makarov, ; Makarov, ; Makarov et al., ), a novel mass analyzer in which the m/z ratio is derived from the frequency of harmonic ion oscillations along the z-axis of the trap.Figure&3. SIMION plot of the equipotential lines for ideal Kingdon trap parameters, end plates at 14 V and wire at &1 V. Potentials indicated for selected contours. Notice the typical quadro-logarithmic shape of the trapping volume (region of negative potential). (Reproduced, with permission, from Gillig, Bluhm, & Russell (). Copyright & 1996 Elsevier Science B. V.)Figure&4. Cutaway view of the orbitrap mass analyzer. Ions are injected into the orbitrap at a point (arrow) offset from its equator (z&=&0) and perpendicular to the z-axis, where they begin coherent axial oscillations without the need for any further excitation. (Adapted, with permission, from Hu et al. (). Copyright & 2005 John Wiley & Sons, Ltd.) [Color figure can be viewed in the online issue, which is available at .]THE ORBITRAP MASS ANALYZEROverviewThe orbitrap mass analyzer (which can also be considered a refined Knight-style Kingdon trap) is composed of a spindle-like central electrode and a barrel-like outer electrode (Fig. ). A DC voltage is applied between the two axially symmetric electrodes, resulting in the following electrostatic potential distribution (Gall et al., ; Gillig, Bluhm, & Russell, ; Makarov, , ): (5) where r and z are cylindrical coordinates, Rm is the characteristic radius, k is the axial restoring force, and C is a constant. The parameter k is determined by the exact shape of the electrodes and the applied potential. The field can be viewed as the sum of a quadrupole field of the ion trap and a logarithmic field of a cylindrical capacitor, that is, a quadro-logarithmic field (Gall et al., ; Korsunskii & Bazakutsa, ). The geometrical shape of the electrodes can be deduced from Equation , as (6) where z&=&0 is the equatorial plane of symmetry, the subscripts 1 and 2 represent the central electrode and outer electrode, respectively, and R1 and R2 represent the maximum radii of the central and outer electrodes, respectively.Stable ion trajectories involve both orbiting motion around the central electrode (r, &-motion, where & is the angular coordinate) and simultaneous oscillations in the z-direction. Equations of motion for the confined ions have been previously described in detail by Makarov (). Only ions with orbital radii less than Rm will be trapped. In addition, Equation
shows that the specially shaped electrodes produce an electrostatic potential containing no cross terms in r and z, meaning that motion along z is independent of r, &-motion.Previously, Oksman () proposed that the Kingdon trap could be used as a self-contained mass spectrometer by deriving the m/z ratio from the frequency of radial oscillation. However, this approach leads to poor mass resolution because of the strong dependence of the rotation frequency on the initial ion velocity and initial radius (Makarov, ). By contrast, in the orbitrap, the axial frequency is used to derive the m/z ratio, since it is independent of the initial properties of the ions. It is this independence that is responsible for the high resolution and mass accuracy of the orbitrap (Makarov, ). The motion along z describes a simple harmonic oscillator and its exact solution is (7) where z0 is the initial axial amplitude, Ez the initial ion kinetic energy along the z-axis and (8) is the frequency of axial oscillation (in rad/sec), where m and q are the mass and charge of the ion, respectively. The frequency of ion oscillations along the z-axis depends solely on the m/z ratio of the ion and the potential (which is held constant) between the electrodes.The image current in the outer electrodes (split at z&=&0), induced by ion axial motion, is acquired as a time-domain transient and fast Fourier-transformed to produce a frequency spectrum (Senko et al., ). Frequencies are converted to m/z by Equation . The magnitude of the image current produced by N ions with frequency &, axial amplitude &Dz and average radius r is given by (9) where &(r) depends on the geometry of the orbitrap and is a monotonically decreasing function of r (Makarov, ). Fast FT-based image current detection of ion axial motion has been previously demonstrated in 3D quadrupole (Badman et al., ; Goeringer, Crutcher, & McLuckey, ; Parks, Pollack, & Hill, ; Soni et al., ; Syka & Fies, ) and cylindrical ion traps (Badman et al., ), as well as the ICR cell (Schweikhard et al., ).The commercial orbitrap mass spectrometer has the following performance characteristics: (i) mass resolution up to 150,000, (ii) mass accuracy of 2&5 ppm (internal and external calibration, respectively), (iii) an ion abundance range of 1:5,000 over which accurate mass measurements can be made (&extent of mass accuracy&), (iv) as good as 0.2 ppm mass accuracy for peaks with signal-to-noise (S/N) ratio &10,000, (v) published upper mass-to-charge (m/z) limit of at least 6,000, (vi) increased space-charge capacity at higher masses due to independence of the trapping potential on m/z ratio, (vii) in-spectrum linear dynamic range up to four orders of magnitude and (viii) larger trapping capacity compared to FT-ICR and the 3-D Paul trap (Hardman & Makarov, ; Hu et al., ; Makarov, ; Makarov et al., ,).Ion CaptureWhen ions are injected into the orbitrap at a z-position offset from z&=&0, the ion packet will begin coherent axial oscillation without the need for any additional excitation (&excitation by injection&) (Makarov, ). Ions are injected into the orbitrap after the voltage on the central electrode is turned on (typically 50&90 &sec) but before the voltage has reached its final value (typically &3,400 V for ions with an initial kinetic energy &1,330 eV in the instrument described by Hu et al., ). Consequently, as the ions enter the orbitrap they experience a monotonic increase in electric field strength, a process termed &electrodynamic squeezing& (Makarov, , ). This has the effect of contracting the radius of the ion cloud, as well as pulling the ion packet closer to the z-axis (i.e., reducing the rotational radius), thereby preventing collisions with the outer electrode as the packets begin their axial oscillations (Fig. ). The rise-time of the field strength (typically 20&100 &sec) determines the trapped m/z squeezing allows trapping of a wide mass range, Mmax/Mmin&&&50 (where M is the mass/charge ratio of the ion) (Hu et al., ; Makarov, ). Since ions of different m/z values are injected at different times, with larger m/z ions arriving later, electrodynamic squeezing results in larger final amplitude of axial oscillation as well as larger mean orbital radius for ions of larger m/z ratio. Both effects will tend to increase the induced ion image current for larger m/z, although this effect may be partially or completely offset by the dependence on the axial frequency (and therefore m&1/2) in Equation . Squeezing is stopped when there is no more possibility of losing ions to collisions with the outer electrode, which is maintained at virtual ground.Figure&5. Principle of electrodynamic squeezing (Hu et al., ). The inset shows the voltage (V) on the central electrode as a function of time (t). Typically, ions are injected into the orbitrap &50 &sec after the central electrode is turned on and detection is started &120 msec later when the voltage on the central electrode (and deflector) stabilizes. (Adapted, with permission, from Hu et al. (). Copyright & 2005 John Wiley & Sons, Ltd.)After ions of all m/z values have entered the orbitrap, the voltage on the central electrode and deflector is held constant to prevent mass shifts during detection. The deflector is switched to a voltage level (&500 V) that compensates for fringing fields caused by the injection slot (Hu et al., ). This is necessary to ensure that ions experience the harmonic axial potential throughout the volume of the orbitrap, thereby minimizing differences in frequency for ions of a given m/z value, which in turn could result in mass errors, peak splitting and lower resolution. After both the central electrode and deflector voltages are stabilized, image current detection may take place.Rotational MotionAfter stabilizing the central electrode voltage, stable ion trajectories within the orbitrap involve axial oscillations along and rotation (composed of radial (r) and angular (&) motion) around the central electrode. The radial (motion towards and away from the central electrode) and angular frequencies vary for ions with slightly different initial positions and kinetic energies so that in these directions ions dephase in approximately 50&100 oscillations (Hu et al., ), orders of magnitude faster than in the axial direction.The r, &-motion, although not used for mass analysis, is still important because the ions must be trapped in the radial plane. When we imagine the orbitrap as a &360& electrostatic analyzer& (Fig. ), then (10) where r is the radius of the electrostatic analyzer as well as the radius of the ion trajectory through the analyzer, qV is the ion kinetic energy before injection, and qE is the force due to the electric field (directed radially inward towards r&=&0) experienced by the ion (de Hoffmann & Stroobant, ; Hu et al., ). This relationship may be derived by noting that stable rotational motion requires a balance between centrifugal and centripetal forces acting on the ion.Figure&6. SIMION simulation showing trajectories of ion packets (at z&=&0) with different initial kinetic energies directed perpendicularly to central electrode. (a) Initial ion kinetic energy poorly matched to the radial component of the electric field, resulting in highly eccentric, non-circular orbit. (b) Same incoming ion kinetic energy as in (a) but with hundreds of rotations shown. (c) Initial ion kinetic energy (1,620 eV) well matched to radial component of the electric field. This orbit is nearly circular, resulting in a locus of orbits that appears as a thin ring. (d) Locus of orbits of two ion kinetic energies, 1,570 and 1,670 eV. Nearly circular orbits demonstrate the orbitrap's kinetic energy acceptance range. (Reproduced, with permission, from Hu et al. (). Copyright & 2005 John Wiley & Sons, Ltd.) [Color figure can be viewed in the online issue, which is available at .]So, at a given electric field strength, ion trajectories are determined by their initial kinetic energy. SIMION simulations of the orbitrap depicted as a 360& electrostatic analyzer show that unstable radial motion is highly elliptical with the perigee precessing very rapidly (Fig. a and b). However, when the initial kinetic energy is appropriately matched to the radial component of the electric field (Fig. c and d) the ions have nearly circular orbits (i.e., relatively constant orbital radius) (Hu et al., ). In addition, the simulations show that the orbitrap can accept a range of initial perpendicular kinetic energies (Fig. d), which, in conjunction with nearly-circular orbits around the central electrode (vs. elliptical), makes for a larger trapping capacity and therefore increases space-charge capacity.Axial MotionUnlike rotational and radial frequencies, axial frequencies are completely independent of initial ion parameters. Therefore, ions of the same m/z ratio continue to oscillate along the z-axis together, remaining in-phase for hundreds of thousands of oscillations, significantly longer than radial oscillations. As the outer electrode is split only at z&=&0, angular frequencies cannot be observed in the frequency spectrum. To minimize side bands and mixed harmonics associated with radial oscillations in the frequency spectrum, detection is started after the ions lose coherence (i.e., dephase) in the radial direction (20&100 &sec). After angular and radial dephasing, the ion packet of a given m/z assumes the shape of a thin ring, with ions uniformly distributed along its circumference and the axial thickness of the ion ring remaining small compared with the axial amplitude. Moving towards one end of the orbitrap and then towards the other, this ring will induce opposite currents in the outer electrode halves, thus creating a signal that is detected using differential amplification. Eventually, the axial width of the ion packets will increase due to collisions with background gas molecules present even at the ultrahigh vacuum in the orbitrap (&10&10 mbar), space-charge effects both between and within ion packets, as well as traversal of non-ideal regions of the orbitrap electric field capable of interconverting radial and axial motion. When axial width becomes comparable to the axial amplitude, then the induced image currents due to different parts of the ion packet will completely cancel each other out, thus reducing the magnitude of the signal until it is completely lost in the noise (Hu et al., ). Collisions with background gas can also result in ion loss from the trap, which also reduces the signal.Modes of DetectionImage CurrentThe image current of the oscillating ions is differentially amplified from each half of the outer electrode and then undergoes analog-to-digital conversion to produce a time-domain transient. An example of a typical transient used to record a positive ion spectrum of reserpine (theoretical m/z of the protonated molecule is 609.281) appears in Figure , which is Fourier-transformed to create the mass spectrum shown in Figure . Negative ion spectra can also be acquired using the orbitrap. For example, Ejsing et al. () elucidated the fragmentation pathways of three inositol-containing sphingolipid classes from Saccharomyces cerevisiae by acquiring complementary information using QqTOF, LCQ (Thermo Electron Corporation) and LTQ-Orbitrap mass spectrometers in the negative ion mode. Negative ion nanoelectrospray ionization (nESI) LTQ-Orbitrap MS2 and MS3 spectra (MS/MS performed inside the LIT with the fragments being injected into the orbitrap for analysis) of an inositolphosphoceramide are shown in Figure , where the high mass accuracy and resolving power of the orbitrap facilitated unequivocal structural assignment (Ejsing et al., ).Figure&7. Typical transient acquired from the prototype StQ-Orbitrap for reserpine using a scan rate of 5,000 kHz and 8 million data points.Figure&8. Positive ion full-scan ESI StQ-Orbitrap mass spectrum of reserpine.Figure&9. (a) Structure of inositolphosphoceramides (IPC). The molecular composition is denoted using the following convention: number of carbon atoms in the long-chain base (LCB): number of double bonds in the LCB; number of hydroxyl groups in the LCB/number of carbon atoms in the fatty acid (FA): number of double bonds in the FA; number of hydroxyl groups in the FA. (b) LTQ-Orbitrap CID MS2 spectrum of the de-protonated IPC 18:0;3/26:0;1 (shown in (a); sum formula 44:0;4) ion at m/z 952.6859 (structure shown in (a)). (c) LTQ-Orbitrap MS3 spectrum produced from dissociation of the dehydrated ceramide phosphate (CerP) fragment ([CerP&H2O]&) (&3 ppm mass accuracy) in (b). Fragmentation of the [CerP&H2O]& yields [FA26:0;1&+&O]& (0.7 ppm) and [FA26:0;1&CH2O]& (0.6 ppm). LCB fragments are: [LCBP18:0;3&H2O]& (0.3 ppm), [LCBP18:0;3&H3NO]& (0.1 ppm), [LCBP18:0;3&2H2O]& (1.4 ppm), and [LCBP18:0;3&H2O&+&CO]& (1.5 ppm). (Adapted, with permission, from Ejsing et al. (). Copyright & 2006 John Wiley & Sons, Ltd.)Mass-Selective InstabilityMakarov suggested that adding an RF to the static voltage on the central electrode would result in modification of the axial and radial amplitudes of the confined ion populations. Under these conditions, the radial equation of motion is more complex and nonlinear (although radial excitation is self-quenching), while motion along z is governed by the Mathieu equation (Makarov, ). The stability diagram in the axial direction shows that ions with a specific m/z become unstable as au&&&1 (Fig. ), that is, when the angular frequency of the RF is 2&. Applying an RF signal with this frequency to the central electrode results in a parametric resonance (Landau & Lifshitz, ) that increases the axial amplitude of the ions at the appropriate m/z until they are eventually ejected from the trap to a detector (such as an electron multiplier) located along the axis of the orbitrap (Makarov, ). This method of mass analysis and detection remains hypothetical and has not been demonstrated in the literature.Figure&10. Stability diagram of the Mathieu equation in the axial direction and scan line for parametric oscillation (au&=&4(&2/&O2); qu&=&2&(&2/&O2)), where &O is the RF angular frequency in rad/sec and & is the ratio of the RF amplitude to the average voltage between central and outer electrodes of the trap. The scan line passes through two regions (au&=&1 and 4) where the ion motion becomes unstable. Excitation to these points will result in axial ejection of the ions. (Reproduced, with permission, from Makarov (). Copyright & 2000 American Chemical Society.)Mass Resolving Power, Mass Resolution, and Mass AccuracyTypically, in MS the frequency and mass resolving power (or resolution) are defined as &/&D&50% or m/&Dm50%, respectively, where subscript 50% indicates the full width of a spectral peak at half-maximum peak height (FWHM). In orbitrap MS, the frequency resolving power is twice the mass resolving power owing to the square root in Equation : (11) and from Equation , the experimental mass resolution is (12)Equation
also shows that the frequency of the axial oscillations is inversely proportional to (m/q)1/2 in contrast to the cyclotron frequency (&c) in FT-ICR MS, which is inversely proportional to m/q: (13) where B is the strength of the magnetic field (de Hoffmann & Stroobant, ). As a result, for fixed acquisition times (Tacq), the resolving power of the orbitrap mass analyzer diminishes as the square root of m/z, that is, more slowly than in FT-ICR, where the mass resolving power is equal to the frequency resolving power (Makarov et al., ). For example, when the initial resolving powers of the orbitrap and a 7 T FT-ICR are 60,000 and 100,000 at m/z 400, respectively, then the resolving power of the orbitrap will exceed that of FT-ICR at m/z&&&1,100. However, FT-ICR analyzers can have longer acquisition times than the orbitrap (the latter currently limited to &1.8 sec) resulting in significantly higher resolving power (Macek et al., ). The transient decay time in the orbitrap is determined by factors that result in ion loss and dephasing such as collisions with background gas, field imperfections, electric field instability, and space-charge effects.Resolution is affected by the transient length and by field imperfections that produce cross-terms and anharmonicities in the orbitrap potential, the latter causing the frequency of a given m/z to depend on its initial kinetic energy in the axial direction. According to Makarov (Hardman & Makarov, ), field perturbations are the result of (i) intentional imperfections in the electrode geometry, such as the injection slot of the orbitrap and the center gap between the two halves of the outer electrode, (ii) unintentional imperfections, viz., accuracy of machining, and (iii) stability of the high-voltage power supplies. The effects of these imperfections are significantly offset in the commercial instruments by employing highly accurate manufacturing and very stable power supplies, so that resolving power is primarily determined by Tacq (Makarov et al., ).Resolving power also decreases with increasing mass due to increased collision cross-section with background gas molecules. Collisions can lead to fragmentation of ions, loss of ion packet phase coherence, or ejection of ions from the trap. These processes are more pronounced in the orbitrap than in FT-ICR instruments because the ion energy is independent of m/z in the orbitrap, while in FT-ICR it decreases as (m/z)&1 (Makarov et al., ). Thus, ultrahigh vacuum is critical in the orbitrap for high resolution measurements of high-mass compounds such as proteins (Makarov et al., ).Mass accuracy is dependent on an instrument's ability to resolve adjacent m/z peaks, that is, on its resolving power. As a result, factors that limit resolution in the orbitrap (described above) will concomitantly worsen mass accuracy relative to an external calibrant. Additional factors also affect mass accuracy. Internal calibration is affected by factors that change the frequency to m/z relationship in a non-proportional way, such as (i) variations in initial injection positions as a function of m/z, (ii) space-charge effects which differ for the various m/z ion packets as they are injected into the orbitrap, and (iii) very weak r,z-cross-terms in the potential that would slightly change the apparent harmonic restoring force as a function of radial position, such that the axial frequency of an ion is dependent on its radial position.The orbitrap mass analyzer is capable of producing spectra with mass accuracies in the range of 2&5 ppm. At low S/N ratios, noise is the main contributor to mass error so that there is no difference between measurements obtained using internal or external calibration. As the S/N ratio increases, the precision of mass measurement improves to better than 1 ppm for internal calibration (Makarov et al., ). In addition, variability in the central electrode potential due to shot noise, thermal sensitivity and/or small variations of the output of the high-voltage power supply over time causes mass errors (when using external calibration only) that are greater than those produced by space-charge effects (Makarov et al., ). To optimize resolving power with external calibration in the commercial instrument, the orbitrap and its associated power supplies are thermally regulated (Makarov et al., ), which provides enough stability to keep mass errors below &5 ppm for more than 20 hr (Fig. ). Generally, the LTQ-Orbitrap gives accurate mass measurements (&5 ppm root-mean-square) over an ion abundance ratio range of at least 1:5,000 (Makarov et al., ).Figure&11. Stability of mass accuracy as a function of time for two ions with a large difference in ion abundance and using external calibration (black trace: nominal m/z 1,422 at 100%; gray trace: nominal m/z 524 at &0.02%). (Adapted, with permission, from Makarov et al. (). Copyright & 2006 American Society for Mass Spectrometry. Published by Elsevier Inc.)High performance mass spectrometers continue to play significant roles in many areas of scientific investigation. For example, high performance liquid chromatography (HPLC) retention times and MS/MS information in conjunction with accurate mass measurements can greatly limit peptide candidates to just a few sequences (Zubarev, Hakansson, & Sundqvist, ) and also facilitate elucidation of elemental composition. These capabilities can also provide valuable structural information about the location and identification of PTMs. So, the high mass accuracy (better than 1 ppm with internal calibration), resolving power (up to 150,000) and MSn capabilities of the LTQ-Orbitrap make it a valuable instrument for chemical analysis.ION INJECTION INTO THE ORBITRAPThe main requirement for injection into the orbitrap is that entering ion packets must have narrow spatial (& few mm) and temporal distributions (&100&200 nsec), to ensure the stability and coherency of the trapped ion packet during image current detection. To obtain these short pulses, three injection methods have been employed at different stages in the development of the orbitrap. These are (i) electrostatic acceleration lenses (Makarov, ), (ii) axial ejection from a linear quadrupole ion trap (Hardman & Makarov, ; Hu et al., ), and (iii) radial ejection from a &C&-shaped linear quadrupole ion trap (Makarov et al., ,; Olsen et al., ). These systems are reviewed below.Electrostatic Acceleration LensesThe first (prototype) orbitrap instrument (HD Technologies Ltd, Manchester, U.K.), described by Makarov in 2000 (Makarov, ), employed a set of electrostatic lenses to extract ions produced by laser desorption (LD) and inject them into the orbitrap analyzer (Fig. ). The electrodes of the trap were machined according to the geometry defined by Equation
with R1&=&7 mm and R2&=&20 mm. Ions produced by LD were accelerated by a voltage drop of 1.1&1.2 kV across a 10 mm acceleration gap to a set of accelerating electrostatic lenses and a conductivity restrictor and directed into the orbitrap analyzer. The distance between the sample plate and the orbitrap was minimized (&38 mm) to ensure that entering ion packets were temporally and spatially compacted as much as possible. This is essential because larger ion packets have lower motion stability and coherence in the orbitrap, which results in diminished performance. A deflector electrode was used to bend the ion beam into the orbitrap, where ion capture via electrodynamic squeezing and image current detection occur as previously described in Section III (Makarov, ).Figure&12. Schematic of an early prototype instrument that uses electrostatic acceleration lenses to inject ions produced by LD into the orbitrap. (Adapted, with permission, from Makarov (). Copyright & 2000 American Chemical Society.)The 56Fe+ ion was found to have a peak width of 2.39 Hz at a frequency of 711 kHz, corresponding to a mass resolving power at FWHM of 150,000 (frequency resolving power is 300,000 from Eq. ). To determine the mass calibration accuracy of the orbitrap, Makarov used the frequency of 133Cs+ to calculate k from Equation , and then calculated the experimental mass of 23Na+ from Equation
using its measured frequency. Results for 13 individual laser shots acquired over a 20-hr period showed deviations up to &20 ppm from the theoretical mass of 23Na+ with root-mean-square (RMS) deviation (for individual shots) of 5 ppm (Makarov, ).These results represent a remarkable achievement at this early stage of orbitrap development because the orbitrap's resolution surpassed that of commercial TOF mass spectrometers (approximately 10,000 (Chernushevich, Loboda, & Thomson, )). In addition, comparison with theoretical estimates of FT-ICR mass resolving power in the low-pressure limit (i.e., no ion-neutral collisions during the detection period) (Marshall, Hendrickson, & Jackson, ), shows that the orbitrap resolving power for 23Na+ is approximately midway between the resolution obtained with acquisition times of 1 msec and 1 sec with FT-ICR. Figure
shows theoretical FT-ICR resolution plotted versus m/z for these acquisition times and at magnetic field strengths of 1.0, 3.0, 4.7, 9.4, and 14.5 T. At m/z&=&1,000, the orbitrap's experimental resolving power is approximately 80,000 (estimated from Fig.
in Hardman & Makarov, ), which is higher than the theoretical FT-ICR resolution for field strengths &4.7 T and for 1 sec acquisition times. In addition, the slopes of these plots indicate that the orbitrap resolution will eventually even exceed that of 7, 9.4, and 14.5 T FT-ICR at 1 sec acquisition time at higher m/z values, because of its weaker mass dependence (see Section ). Although the measured mass accuracy was worse (compared to &5 ppm for 7.0 T LTQ-FT-ICR and TOF instruments), changes to the injection optics in later instrument designs achieved comparable performance of 2&5 ppm (Hardman & Makarov, ; Hu et al., ; Makarov et al., ; Nielen et al., ). Mass accuracy remains worse than that of magnetic sectors which can routinely achieve &1 ppm accuracies (Bristow, ).Figure&13. The experimental orbitrap resolving power (gray diamonds and solid line) is &150,000 at m/z 23 (Tacq&=&&1&1.5 sec), which is equal to or better than the theoretical resolving power for FT-ICR instruments of 1&14.5 T field strength (solid black lines) using 1 msec acquisition time (Tacq) and field-swept instruments such as magnetic sectors (dashed lines) using 1 sec Tacq. Calculated FT-ICR plots assume no ion-neutral collisions during the detection period and a fixed ion-neutral collision frequency. The orbitrap resolving power is &80,000 at m/z &1,000 (estimated from Fig.
in Hardman & Makarov ()), which exceeds that of ICR field strengths of 1.0, 3.0, and 4.7 T FT-ICRs (Tacq&=&1 sec). Extrapolating the slope of the orbitrap plot indicates that its resolving power will eventually exceed that of 9.4 and 14.5 T FT-ICRs (Tacq&=&1) at higher m/z values as well. (Adapted, with permission, from Marshall, Hendrickson, & Jackson (). Copyright & 1998 John Wiley & Sons, Inc.)Matrix-assisted laser desorption ionization (MALDI) experiments using polyethylene glycol (PEG)-1000 produced spectra with a broad distribution of PEG oligomer ions, spanning an m/z range of &700 (Makarov, ), demonstrating that the orbitrap was capable of accepting and trapping a wide range of masses. No mass discrimination was observed over this m/z range (relative to typical MALDI spectra of PEG-1000 standard) but the transient decay time for these polyatomic ions was found to be much shorter at 20&30 msec compared to 500 msec observed for atomic ions. Since the pressure was the same in both experiments, Makarov () suggested that this difference was caused by metastable dissociation of the polyatomic ions, as well as faster scattering and dephasing rates due to the increase in ion mass and hence collision cross section with background neutrals.These results represent the first demonstration of the analytical capabilities of the orbitrap mass analyzer. Its high mass resolution, surpassed only by the FT-ICR using Tacq&&&1 sec as shown in Figure
and by double focusing sector instruments, potential for low cost and simple design suggested that it would be suitable to carry out high performance mass spectrometric analyses. However, to make the instrument more versatile, it was desirable to couple continuous atmospheric pressure ionization sources such as ESI to the orbitrap and to provide capabilities for MS/MS experiments.Linear Quadrupole Ion TrapAtmospheric pressure ionization (API) (Bruins, ; Carroll et al., ; Horning et al., ) sources such as ESI (Dole, Mack, & Hines, ; Fenn et al., ; Whitehouse et al., ; Yamashita & Fenn, ,) and atmospheric pressure chemical ionization (APCI) (Carroll et al., ; Horning et al., ,) coupled with mass spectrometers, especially high performance instruments such as FT-ICR, have become important tools for analyzing complex biological samples. However, the increasing speed of chromatographic methods sets an upper limit on the transient acquisition time in both FT-ICR and the orbitrap, and, therefore, on mass resolution.Because the orbitrap operates in a pulsed fashion, external ion accumulation is desirable for interfacing with a continuous API source. This technique is already widely used for other pulsed mass analyzers such as TOF (Chernushevich, ) and FT-ICR (Senko et al., ), where gating of the voltage on the exit aperture of the external trap provides for controlled injection into the subsequent mass analyzer. The temporal width of these injection pulses is typically tens to hundreds of microseconds, which is too long to ensure axial coherency of the ion packets in the orbitrap (Hardman & Makarov, ). To obtain pulses &100&200 nsec long (measured using a secondary detector located behind the orbitrap), a modified linear quadrupole trap (storage quadrupole, StQ) was developed to accumulate and then inject ions from an ESI source into the orbitrap (Hardman & Makarov, ; Hu et al., ).Ions generated by an ESI source at atmospheric pressure are transported through an RF-only guide/collision quadrupole (2.5 MHz, 0.1&1 kVp-p, 10&2 mbar) and then a transport quadrupole (920 kHz, Vp-p&=&&600 V) to the StQ (up to 7 kVp-p at 3.45 MHz), which has an internal pressure greater than 10&4 mbar (Fig. ). The StQ has a set of rods (RF and DC) with a ring electrode (DC-only) over the end closest to the exit lens (lens 1). An axial potential well is created in the StQ by biasing the ring electrode. The depth of this well is approximately 1% of the potential difference between the ring electrode and the DC offset of the StQ. Due to collisions with the N2 bath gas, ions entering the trap lose sufficient kinetic energy that they accumulate in the well with small axial extent (i.e., a few mm). The potential energy of the accumulated ions is increased to the acceleration voltage by synchronously increasing the DC offset on the StQ and intermediate lens, which ensures that ions injected into the orbitrap will have sufficient kinetic energy to acquire stable trajectories (see Section ). As the DC offset is increased, the potential well deepens until it eventually becomes the lowest point on the potential energy surface (dotted line in Fig. ). Implementation of this &energy lift& in the StQ (Hu et al., ) was an improvement over an earlier orbitrap instrument (Hardman & Makarov, ) in which the entire ion path from the ESI source down to lens 1 was floated at a potential close to the acceleration voltage. The &energy lift& allows both the ion source and orbitrap to be maintained at near-ground potential while simultaneously accelerating ions to &1 keV in between (Hu et al., ).Figure&14. Schematic of an orbitrap instrument that couples a continuous source (such as ESI) to the orbitrap mass analyzer using a modified linear quadrupole ion trap (storage quadrupole (StQ)). The StQ is composed of four rods with a ring electrode positioned over the end closest to the exit aperture. The potential difference between the ring electrode and DC offset on the rods creates an axial potential well in the StQ. Ions are accumulated, bunched, and then ejected axially from the StQ into the orbitrap. In the StQ-Orbitrap instrument described by Hardman and Makarov (), the transport quadrupole is absent and the StQ is composed of a set of long and short rods, the latter closest to the exit aperture. In that configuration, the axial well is determined by the potential difference between the long and short rods. The pressure of each region is the same in both instruments. (Reproduced, with permission, from Hu et al. (). Copyright & 2005 John Wiley & Sons, Ltd.)Figure&15. Schematic showing the axial potential distribution in the storage quadrupole during accumulation (solid line) and &energy lift& (dotted line) for the StQ-Orbitrap instrument described by Hu et al., . For ion extraction, the voltage on lens 1 Figure
is opened by pulsing its potential large and negative for positive ions, forming the potential distribution indicated by the dashed line. (Adapted, with permission, from Hu et al. (). Copyright & 2005 John Wiley & Sons, Ltd.)The exit lens is pulsed open (negative voltage pulse for positive ions) and the ion packets are rapidly extracted and accelerated to the entrance of the orbitrap. The low pressure (10&5 mbar) between lenses 1 and 2 minimizes collisions during acceleration. Ions are accelerated through an &S-shaped& lens system composed of two symmetrical deflector electrodes (lenses 3 and 4, Fig. ) to minimize the possibility of direct gas transport into the orbitrap. Finally, a voltage applied to the deflector electrode bends the ion beam into the orbitrap through an injection slot present in the outer electrode.A secondary detector located just behind the orbitrap allows characterization of the temporal/spatial properties of ion packets at a point similar in space to where they would have entered the orbitrap. Although these experiments were only performed using the earlier instrument described by Hardman and Makarov (), the characteristics of the extracted ion packets should be similar between the two instruments. Temporal profiles (Fig. ) show peak widths (FWHM) that are approximately 150&400 nsec, which is (at most) 5&8% of the period of axial oscillations in the orbitrap over the whole mass range. Such temporal profiles indicate that the axial width of the ion packets are small relative to their axial amplitude within the orbitrap, thus ensuring that a transient can be recorded for a suitably long time before losing packet coherence.Figure&16. TOF spectrum of a mixture of ions ejected from the StQ, acquired at a multiplier sitting just behind orbitrap entrance. Each division is 400 nsec. Temporal widths are 150&400 nsec, consistent with the small temporal and spatial extent needed for ion packets injected into the orbitrap. (Reproduced, with permission, from Hardman & Makarov (). Copyright & 2003 American Chemical Society.)A charge-sensitive amplifier connected to the dynode of the multiplier was used to evaluate the ejection efficiency of the StQ by comparing the signal (proportional to total ion charge in the packet) due to ions ejected as short pulses with that for all ions extracted from the StQ. The charge available for off-axis injection into the orbitrap is up to 2&&&105 elementary charges, which is &30% of the maximum charge measured on the amplifier. Hardman and Makarov () showed that the final number of charges trapped by the orbitrap was &1&&&105 elementary charges, below the space-charge limit of the orbitrap, which indicates that the combined efficiency of the injection and trapping processes is only about &15%.Transients acquired on these instruments had decay times of 1.0&1.5 sec (the transient for reserpine extends to &1.6 sec in Fig. ) (Hardman & Makarov, ; Hu et al., ), which was longer than that measured on the LD instrument (&20&60 msec for PEG-100 oligomers in the m/z range 525&1,230) owing to differences in the injection methods. In the LD instrument, the ion packet was not compacted nor controlled in any way before injection. Additionally, ion packet arrival times at the orbitrap entrance were &4-times longer for the LD instrument compared to the StQ-Orbitrap instrument, so initial ion packet widths were larger, resulting in shorter times before the transients faded into background noise. The ability to control and reduce packet width in the StQ-Orbitrap instruments resulted in better performance in terms of reproducibility, resolving power and mass accuracy at higher masses.For the StQ instruments, the resolving power was found to be 150,000 at m/z 280, which decreased to &80,000 above m/z 1,200; the root-mean-square mass accuracy was 1.6 ppm with most measurements under 5 ppm (Hardman & Makarov, ). However, for the LD instrument, the estimated resolving power of a PEG-1000 oligomer peak at m/z 965 was &2,200 (estimated from Fig.
in (Makarov, )) and the average mass accuracy was only 11 ppm at m/z 23, with individual measurements varying by as much as 20 ppm. The resolving power at higher masses, as well as the reproducibility and accuracy of the mass measurement, were significantly improved in the StQ instruments by compacting the ion packet before injection into the orbitrap. Both the LD and ESI instruments suffer some performance loss due to temperature drift and cycling of the power supplies. To minimize these effects in the commercial instruments, a lock mass is used and the orbitrap and power supplies are thermally regulated (Olsen et al., ).Hu et al. (, ) applied the StQ-Orbitrap prototype instrument to a number of chemical systems to showcase its high performance capabilities. ESI mass spectra of a mixture of bovine insulin and Ultramark 1621 showed a wide mass range from m/z 1,000 to 2,200 which contained the 3+, 4+, and 5+ charge states of insulin. The resolving power of the instrument was shown to be &100,000 by comparing the experimental and theoretical (calculated using IsoPro 3.0, MS/MS software (Senko, )) isotopic distributions. Using the Ultramark oligomers as internal mass calibrants and a linear calibration over the range m/z 1,121&2,021, RMS errors in the insulin masses (all isotopomers) of 2.9, 1.5, and 4.9 ppm for the 5+, 4+, and 3+ charge states, respectively, were observed. In this example, no simple relationship between charge state and mass
ion frequencies are sufficiently spaced such that space-charge effects are not observed. Space charge effects on mass accuracy have been observed in the orbitrap for two doubly charge peptides with &D(m/z)&=&0.018 Th (Hu, Cooks, & Noll, ), which is one order of magnitude closer in m/z than the peaks of the 5+ charge state of insulin (&D(m/z)&=&0.2 Th).In addition, the nanospray mass spectrum of alcohol dehydrogenase (ADH) from S. cerevisiae (Fig. ) shows the upper m/z limit of the orbitrap to be at least &6,000 (Hu et al., ). Although the peaks for the 28+ (m/z 5,272) to 25+ (m/z 5,903) charge states had low S/N ratio and resolution, the spectrum compares favorably with those recorded using triple quadrupole mass spectrometers (Rogniaux et al., ). Possible reasons for reduced performance at higher m/z include (i) m/z range limitations of the StQ; (ii) poor ion transmission of high m/z ions through the z-shaped pa (iii) more dephasing of the ion packet brought on by increased collision cross section in the orbitrap compared to ions with lower molecular weights (these effects become less severe as the mass of the ion becomes significantly greater than the mass of the neutral collision partner); (iv) increased coalescence of isotopomer signals at high molecular weight&documented in FT-ICR for degrading S/N (Easterling et al., ); and finally, (v) multiply charged ions may interact more strongly by space-charge at high m/z and sufficiently close (1/n) spacing.Figure&17. Mass spectrum of alcohol dehydrogenase (ADH) from Saccharomyces cerevisiae showing the upper m/z limit of the prototype StQ-Orbitrap to be &6,000 Th. (Reproduced, with permission, from Hu et al. (). Copyright & 2005 John Wiley & Sons, Ltd.)The dynamic range of the instrument was also measured as a function of analyte (reserpine) concentration and found to be 103&104, which compares favorably with the dynamic ranges of quadrupole ion traps (QITs) (102&103) (McLuckey & Wells, ) and LIT/FT-ICR (103&104) (Syka et al., ) instruments. Hu et al. () suggested that the dynamic range may be limited by the StQ rather than by the orbitrap.C-TrapThe need to extract ion packets axially from the StQ with temporal lengths shorter than a few hundred nanoseconds necessarily limits the instrument's space-charge capacity because there is a physical limit to the number of ions that can be compacted in the axial plane of the orbitrap. As a result, injection of large numbers of ions leads to broad angular, spatial and kinetic energy distributions that limit the performance of the orbitrap (Makarov et al., ). These problems have been minimized in the commercial LTQ-Orbitrap hybrid instruments (Thermo Electron, Bremen, Germany) by using radial, rather than axial, ion ejection from a curved RF-only quadrupole ion trap (&C-trap&) (Makarov et al., ,). Figure
shows schematics of the commercial LTQ-Orbitrap instruments. This method provides fast and uniform injection for large ion populations. A linear ion trap upstream from the C-trap provides increased trapping efficiency, automatic gain control (AGC, a short prescan that enables storage of a defined number of ions N) (Schwartz, Zhou, & Bier, ) and high-quality accurate mass MSn data from mass analysis detection of fragment ions injected into the orbitrap.Figure&18. Schematics of the commercial LTQ-Orbitrap instruments. (1) Discovery model LTQ-Orbitrap mass spectrometer (Makarov et al., ): (a) (b) curved RF-only quadrupole (C-trap); (c) (d) (e) (f) inner orbitrap electrode (central electrode); (g) outer orbitrap electrode. (2) LTQ-Orbitrap XL model, with an additional octopole at the rear of the C-trap, used to dissociate precursor ions injected from the C-trap (Olsen et al., ). (Reproduced, with permission, from Makarov et al. () and Olsen et al. (). Copyright & 2006 American Chemical Society and & 2007 Nature Publishing Group, respectively). [Color figure can be viewed in the online issue, which is available at .]The instrumentation and its operation have been described in detail (Makarov et al., ). Briefly, ions from an electrospray ion source are transferred by RF-only multipoles to the linear trap of the LTQ. Typical AGC target values N are 5,000&30,000 for the linear trap detector and 20,000&2,000,000 for the orbitrap detector (N is within a factor of 2 of the actual number of ions in the LIT}