Magnetic Sector and Ion Trap Analyzers

Double-focusing magnetic sectors remain the reference method wherever the absolute highest mass accuracy and isotope-ratio precision matter and where sample throughput is secondary to precision: uranium and plutonium isotope-ratio measurements in nuclear safeguards and forensics, $^{87}\text{Sr}/^{86}\text{Sr}$ and $^{143}\text{Nd}/^{144}\text{Nd}$ geochronology by thermal ionization mass spectrometry, and noble-gas isotope work in cosmochemistry (measuring, for instance, $^3\text{He}/^4\text{He}$ in mantle-derived basalts to trace deep-Earth outgassing). Multicollector sector instruments — several Faraday cups or ion-counting detectors positioned simultaneously along the focal plane — remain the gold standard against which newer multicollector ICP-MS instruments are calibrated, precisely because simultaneous collection sidesteps the source-drift problems discussed in the worked isotope-ratio example above. Quadrupole ion traps, by contrast, dominate wherever compactness, cost, and native tandem-MS capability matter more than ultimate mass accuracy: benchtop GC-MS instruments for environmental and forensic screening, LC-MS/MS platforms for targeted metabolomics and pharmacokinetic drug-level quantitation, and — in hybrid form, as the linear ion trap (a two-dimensional variant with improved ion capacity and faster scan speed) — as the workhorse front-end or standalone analyzer in many clinical proteomics and peptide-sequencing instruments, often coupled to an Orbitrap or time-of-flight stage for the high-mass-accuracy readout described in high-resolution-and-accurate-mass. The trap's core operating principle, isolate–activate–fragment–scan within one stored population, is the physical basis of the multistage MS$^n$ experiments central to structural elucidation and is a direct instantiation of the general strategy generalized in tandem-mass-spectrometry. Linear ion traps are also central to real-time surgical tissue analysis (the "intelligent knife," iKnife) and to planetary science instruments (miniaturized ion traps flown on Mars rovers), where the small size and low power draw of the trap, impossible to replicate with a magnetic sector's heavy iron core, are the decisive engineering advantages.

Ask the tutor: How does resonance ejection quantitatively change the effective stability boundary and resolving power compared to the bare q_z = 0.908 threshold?

An ion sees two very different worlds depending on which analyzer it enters. In a magnetic sector, the ion has already been accelerated through a large, fixed potential $V$ (typically 3–10 kV) and travels at high, nearly constant speed along a beam line. It then enters a region of uniform magnetic field $B$, oriented perpendicular to its velocity. The Lorentz force $F = zevB$ supplies exactly the centripetal force needed for circular motion, $mv^2/r$, so the ion follows an arc of radius $r$ set by its momentum-to-charge ratio. Ions of different $m/z$ describe different radii and are spatially separated at the exit slit; scanning $B$ (or $V$) brings each mass in turn onto the fixed detector slit. Because all ions nominally share the same kinetic energy $zeV$, the sector is fundamentally a momentum analyzer that becomes a mass analyzer only because the accelerating step fixes $zeV$ for every ion — but real ion sources produce ions with a small spread in kinetic energy ($\Delta V$ of a few volts out of several kilovolts), and this spread limits resolution unless corrected. The correction is the electrostatic analyzer (ESA): a pair of curved plates with a radial electric field that bends ions according to $zeE = mv^2/r_e$, i.e., by kinetic energy $\tfrac12 mv^2$ alone, independent of mass. Passing the beam through an ESA before or after the magnetic sector, in a Nier–Johnson or Mattauch–Herzog geometry, makes the overall dispersion insensitive to first order in the energy spread — this is double focusing: simultaneous direction focusing (many ions leaving the source at slightly different angles reconverge) and energy focusing (many ions leaving with slightly different energies reconverge). The result is a narrow image at the collector slit and resolving powers of $10^4$–$10^6$, the highest routinely available in mass spectrometry without Fourier-transform methods. An ion trap operates in an entirely different regime. Ions are not moving in a fast, fixed-energy beam; they are nearly at rest (or in small-amplitude oscillation), confined near the center of a three-electrode assembly: a ring electrode and two end-cap electrodes shaped as hyperboloids of revolution. An RF voltage $V_0\cos(\Omega t)$ applied between ring and end caps creates a quadrupolar potential that alternately pulls ions toward the center along one axis and pushes them away, then reverses. Averaged over many cycles, this time-varying push-pull produces an effective restoring force toward the trap center — a "pseudopotential" well — but whether a given ion's motion actually stays bounded (stable) or grows without limit (unstable) depends on its $m/z$, the RF amplitude $V_0$, the RF frequency $\Omega$, and the trap's physical size. The equation governing this is the Mathieu equation, borrowed from the mathematics of parametric oscillators; solving it defines regions of $(a_z, q_z)$ parameter space — functions of $m/z$, $V_0$, and $\Omega$ — inside which ion motion is stable. Scanning $V_0$ moves ions of successively lower $m/z$ out of the stable region and into resonant ejection through a hole in an end cap, where they strike a detector. This "mass-selective instability" scan converts a stored population into a mass spectrum without ever needing a moving beam.

Magnetic sector mass equation. An ion of mass $m$ and charge $z e$ accelerated through potential $V$ acquires kinetic energy

where: $z$ = number of elementary charges, $e$ = elementary charge, $V$ = accelerating voltage, $m$ = ion mass, $v$ = ion speed after acceleration. In the magnetic field the Lorentz force provides centripetal acceleration:

where: $B$ = magnetic flux density, $r$ = radius of curvature of the ion path in the sector. Combining these (eliminate $v$) gives the fundamental mass-to-charge equation of a magnetic sector:

where: $m/z$ is expressed with $e$ folded in so that $m/z$ has units of mass per elementary charge (as conventionally reported, using $z$ as an integer and $e$ absorbed into a constant), $B$ in tesla, $r$ in meters (fixed by instrument geometry — the radius of the flight tube), $V$ in volts. Reading: at fixed geometry ($r$ fixed by the analyzer) and fixed accelerating voltage $V$, mass-to-charge ratio scales as the square of the magnetic field. A linear scan in $B$ therefore sweeps through $m/z$ nonlinearly (quadratically); real instruments scan $B$ in a way (often $B \propto \sqrt{t}$ or with digital compensation) chosen to give a convenient, ideally linear-in-mass, spectrum on readout electronics. Equivalently, at fixed $B$, scanning $V$ downward brings progressively higher $m/z$ into focus since $m/z \propto 1/V$.

Mathieu equation for the quadrupole ion trap. Ion motion along the trap's axial ($z$) and radial ($r$) coordinates in the applied RF/DC quadrupolar field obeys, in dimensionless form,

where: $u$ stands for either $z$ or $r$ coordinate, $\xi = \Omega t/2$ is a dimensionless time, $\Omega$ = angular frequency of the applied RF, and the dimensionless trapping parameters are

where: $U$ = DC offset voltage (often zero in simple mass-selective instability scans), $V_0$ = zero-to-peak RF amplitude, $r_0$ = ring-electrode radius, $z_0$ = end-cap half-separation, $m/z$ enters through the ion's mass $m$ and charge number $z$ appearing in the denominator/numerator respectively. Reading: for fixed trap geometry ($r_0, z_0$), frequency $\Omega$, and RF amplitude $V_0$, the parameter $q_z$ is inversely proportional to $m$ (at $U=0$, $a_z = 0$). Ions of different mass sit at different points along the $a$–$q$ stability diagram; stable trapped motion exists only for $|q_z| \lesssim 0.908$ on the $a=0$ axis (the boundary of the first stability region). As $V_0$ is ramped upward, $q_z$ for every trapped ion increases; the lowest-mass ions reach $q_z \approx 0.908$ first and are ejected, then progressively heavier ions follow — this is exactly how mass-selective instability scanning builds a spectrum in order of increasing $m/z$.

Example 1 (foundational) — Radius and resolving-power requirement for a single-focusing sector. A magnetic sector instrument accelerates singly charged ions through $V = 4.00\times 10^3$ V and uses a magnet with flux density $B = 0.3800$ T. Find the radius of curvature needed to bring an ion of $m/z = 127.0$ (protonated iodide-related fragment, nominal mass units, $z=1$) to focus at the fixed detector slit, and determine whether resolving power $R = 500$ (10% valley) would separate this ion from one at $m/z = 127.2$. Rearrange the mass equation for $r$:

Here $m/z$ in SI mass units requires converting: $1$ u $= 1.6605\times10^{-27}$ kg, so $m/z = 127.0 \times 1.6605\times10^{-27}\text{ kg} = 2.1088\times10^{-25}$ kg (per unit charge, $z=1$, so $e$ appears once in the denominator to keep $m/z$ in kg per elementary charge equivalent). Using $e = 1.602\times10^{-19}$ C:

Compute the numerator: $2 \times 4.00\times10^3 \times 2.1088\times10^{-25} = 1.6870\times10^{-21}$. Compute the denominator: $1.602\times10^{-19} \times 0.1444 = 2.3133\times10^{-20}$.

So the sector radius must be set (by geometry, i.e., the physical curvature of the flight tube, adjustable in practice only by scanning $B$ or $V$ around a fixed mechanical $r$) to $r = 27.0$ cm for this ion to strike the exit slit — a physically reasonable sector radius for a compact instrument. Now check separation of $m/z = 127.0$ from $127.2$. Required resolving power (10% valley definition):

Since the instrument is specified to deliver only $R = 500$, it cannot resolve these two peaks to the standard 10% valley criterion — $635 > 500$. This illustrates why nominal-mass work (integer resolving power of a few hundred) is adequate for many organic EI spectra but insufficient for closely spaced isobars; achieving $R > 635$ would require either a double-focusing geometry or narrower entrance/exit slits (at the cost of transmitted ion current, i.e., sensitivity).

Example 2 (applied) — Ejection order and scan rate in a quadrupole ion trap MS/MS experiment. A quadrupole ion trap operates at RF drive frequency $\Omega/2\pi = 1.00$ MHz (so $\Omega = 6.283\times10^6$ rad/s), with $r_0 = 1.00\times10^{-2}$ m and $z_0 = 0.707\times10^{-2}$ m (a common ratio, $z_0 = r_0/\sqrt2$, chosen to minimize higher-order field imperfections). Using the simple boundary condition $q_z = 0.908$ (ignoring the finer resonance-ejection offset for this estimate), find the RF amplitude $V_0$ needed to eject a precursor ion at $m/z = 524.6$ (a doubly protonated peptide, $z=2$) and compare to the amplitude needed to eject a fragment ion at $m/z = 175.1$ ($y_1$ immonium-related ion, $z=1$) produced by collision-induced dissociation of that precursor. First compute the geometric factor $r_0^2 + 2z_0^2$:

From Step 2 of the derivation,

Compute $\Omega^2 = (6.283\times10^6)^2 = 3.948\times10^{13}\ \text{s}^{-2}$. Compute the mass-independent prefactor:

(units work out because $m/z$ will be supplied in kg per unit charge, giving volts). For the precursor, $m/z = 524.6$ (mass units per charge, $z=2$ meaning the ion carries two charges but nominal mass 1049.2 u — here $m/z$ as conventionally reported is 524.6, so we convert this ratio directly): convert $524.6$ u to kg: $524.6 \times 1.6605\times10^{-27} = 8.712\times10^{-25}$ kg.

For the fragment ion, $m/z = 175.1$ ($z=1$): $175.1 \times 1.6605\times10^{-27} = 2.9075\times10^{-25}$ kg.

Both values (4874 V and 1627 V) fall comfortably within a typical trap's RF range (0–7500 V zero-to-peak), confirming this is a realistic MS/MS scenario: in a full MS$^2$ experiment, the trap first isolates the precursor at $m/z=524.6$ by ejecting all other stored ions (ramping $V_0$ past their individual ejection thresholds while keeping $V_0$ below 4874 V, then applying a notched broadband waveform to eject everything except a narrow window around 524.6), then applies a resonant excitation ("tickle") at the precursor's secular frequency to drive collision-induced dissociation, then finally ramps $V_0$ from low values upward to scan out all product ions — the fragment at $m/z=175.1$ would be ejected at $V_0 \approx 1627$ V, well before the (surviving, undissociated) precursor itself is ejected near 4874 V. The proportionality $V_0^{\text{eject}} \propto m/z$ derived earlier directly explains why lower-mass fragments always appear earlier in the ramp.

Example 3 (exam/competition level) — Designing a double-focusing sector for isotope-ratio precision. A geochemist needs to resolve $^{87}\text{Sr}^+$ ($m = 86.9089$ u) from an isobaric interference at $m = 86.9074$ u (a hypothetical polyatomic interference) to a 10% valley, and additionally wants the peak centroid reproducible enough to measure the $^{87}\text{Sr}/^{86}\text{Sr}$ ratio to a relative precision of $2\times10^{-5}$ (20 ppm), typical of high-precision thermal ionization mass spectrometry (TIMS). (a) Compute the required resolving power to separate the isobars. (b) Given a double-focusing sector with $r = 0.5000$ m (magnet radius) operated at $V = 1.000\times10^4$ V, find $B$ needed to center $^{87}\text{Sr}^+$ on the collector slit. (c) Discuss, using the double-focusing derivation, why energy focusing rather than raw resolving power is actually the binding constraint on the 20 ppm precision goal.

(a) Resolving power for isobar separation.

This is a substantial resolving power — well beyond single-focusing capability ($R\sim10^2$–$10^3$) and squarely in the regime that requires a double-focusing geometry (typical maximum $R \sim 10^5$–$10^6$ for Nier–Johnson sector instruments), consistent with the problem's premise.

(b) Magnetic field for centering $^{87}\text{Sr}^+$. Convert mass: $m = 86.9089\ \text{u} \times 1.6605\times10^{-27}\text{ kg/u} = 1.44295\times10^{-25}$ kg. Rearranging the mass equation for $B$ (singly charged ion, $z=1$):

Numerator: $2 \times 1.000\times10^4 \times 1.44295\times10^{-25} = 2.8859\times10^{-21}$. Denominator: $1.602\times10^{-19} \times (0.5000)^2 = 1.602\times10^{-19}\times0.2500 = 4.005\times10^{-20}$.

So the magnet must be set to $B = 0.2685$ T (268.5 mT) to focus $^{87}\text{Sr}^+$ onto the collector at this geometry and voltage — a field readily achievable with an electromagnet of this size, consistent with real TIMS instrument specifications.

(c) Why energy focusing, not raw R, actually limits isotope-ratio precision. Resolving power of $5.79\times10^4$ tells you the two isobars can be baseline- (or 10%-valley-) separated in principle; it says nothing about how precisely you can measure the area (or peak height, used as a proxy for ion abundance) of an isolated, resolved peak over repeated scans. Isotope-ratio precision at the 20 ppm level is limited instead by (i) ion counting statistics (shot noise, $\propto 1/\sqrt{N}$ for $N$ ions collected — requiring $N \gtrsim (1/2\times10^{-5})^2 = 2.5\times10^9$ ions per ratio measurement, achievable with Faraday-cup collection of stable thermal-ionization beams over integration times of tens of seconds) and (ii) the residual second-order energy-dispersion term left uncancelled by imperfect double focusing (recall from the derivation that the first-order term vanishes, $x_{\text{ESA}}(\delta)+x_{\text{mag}}(\delta) = O(\delta^2)$, but any drift in accelerating voltage $V$ or magnet hysteresis during a scan reintroduces small, correlated peak-position shifts that bias the ratio if the two isotope beams are not measured perfectly simultaneously). This is precisely why the highest-precision Sr, Nd, and Pb isotope-ratio work is done with multicollector magnetic-sector instruments (several fixed Faraday cups positioned at the focal plane, one per isotope, all ions collected simultaneously) rather than by scanning a single collector across peaks sequentially — simultaneous multicollection cancels time-dependent (source drift) noise that a single-collector scan cannot, even at the same nominal resolving power.

Alfred O. C. Nier is the figure most associated with the practical, high-precision magnetic sector, though the mass spectrograph itself traces to J. J. Thomson and Francis Aston in the 1910s–20s. Nier, working at the University of Minnesota from the 1930s onward, redesigned earlier mass spectrographs into robust, quantitatively reliable sector instruments — the "Nier-type" 60° or 90° single-focusing sector with electron-impact ionization became the workhorse geometry for isotope-ratio and residual-gas analysis for decades, and his collaboration with Ernest Johnson on double-focusing geometry (the Nier–Johnson design, 1953) established the ESA-plus-magnet arrangement described here as an industry standard still manufactured today. Nier's meticulous attention to measurement precision — he pioneered high-precision isotope-ratio measurement of uranium, carbon, and noble gases, work that directly underpinned the discovery of uranium-235's fissile abundance relevant to the Manhattan Project — set the metrological culture that modern isotope geochemistry (TIMS, MC-ICP-MS) still inherits. The quadrupole ion trap has a different intellectual lineage, rooted in Wolfgang Paul's development of RF quadrupole devices (the linear quadrupole mass filter and the three-dimensional "Paul trap") at the University of Bonn in the 1950s, for which Paul shared the 1989 Nobel Prize in Physics with Hans Dehmelt and Norman Ramsey (for ion-trapping and precision spectroscopy techniques generally). The instrument as a practical mass spectrometer, however, owes much to George Stafford and colleagues at Finnigan (later Thermo Fisher Scientific), who in the early 1980s developed mass-selective instability scanning with resonance ejection — turning Paul's trap, originally conceived as a tool for fundamental ion-storage physics and spectroscopy, into the compact, robust benchtop GC-MS and LC-MS/MS instruments now ubiquitous in clinical, forensic, and proteomics laboratories.