The Mass Spectrometer: Components and Workflow

A mass spectrometer weighs molecules by turning them into ions. It times or steers their flight through a vacuum. That single sentence hides an enormous amount of engineering. A caffeine molecule leaves an espresso shot at room temperature. It must be vaporized, then stripped of an electron or given a proton. It is accelerated to a precise kinetic energy. It is sorted by its mass-to-charge ratio. The resolution must be fine enough to separate 194.080 Da from 194.117 Da. Finally, it is converted into an electrical current. That current is small enough to require single-ion counting. Every step happens in roughly a millisecond. The whole instrument fits on a benchtop. The mass spectrometer matters because it is the most information-dense detector in the analytical chemist's toolkit. A UV-Vis spectrum gives one number (absorbance) at each wavelength. A mass spectrum gives an entire fingerprint instead. It shows the molecular weight, the isotope pattern, and a catalog of fragment ions. Together these reconstruct the molecule's connectivity. Coupled to a chromatograph, it becomes a powerful machine. It separates a mixture in space or time. Then it identifies and quantifies every component. This is why LC-MS and GC-MS underpin drug metabolism studies, doping control, environmental forensics, proteomics, and semiconductor trace-metal analysis. The big picture is architectural, not chemical. Every mass spectrometer, regardless of what it is built to analyze, is a chain of five modules. An inlet introduces the sample. An ion source converts neutral molecules into gas-phase ions. A mass analyzer separates ions by mass-to-charge ratio ( $m/z$ ). A detector converts ion arrivals into an electrical signal. A vacuum system keeps ions from colliding with air molecules along the way. The chemistry and physics inside each module vary wildly. Examples include electron ionization versus electrospray, a quadrupole versus a time-of-flight tube, and an electron multiplier versus a Faraday cup. But the architecture itself is invariant. Understanding it is the necessary scaffold. It must come before any single ionization method or mass analyzer can be studied in depth. This article builds that scaffold. It walks through each module in the order a molecule experiences it. It works out the vacuum and ion-optics numbers that make the instrument function. These include mean free path, ion transit time, and resolving power. It also shows how a raw stream of ion arrival times becomes the plotted mass spectrum every chemist recognizes. It deliberately stops short of the mechanistic detail of any one ionization technique or analyzer type. Those live in their own dedicated treatments. But by the end, the reader will know exactly where those techniques slot into the machine. They will also understand why the vacuum system is not a minor plumbing detail. It is the reason the whole method works at all.

Key takeaways

A mass spectrometer is five modules in series: inlet → ion source → mass analyzer → detector → vacuum system.
The ion source converts neutral analyte into gas-phase ions; only ions can be steered and mass-sorted by electric and magnetic fields.
The mass analyzer separates ions by mass-to-charge ratio $m/z$ , not mass alone — charge state is essential information.
High vacuum (down to $10^{-6}$ – $10^{-10}$ torr) is required so ions travel from source to detector without colliding with residual gas.
The detector converts a stream of individual ion arrivals into a current or count, which software renders as intensity versus $m/z$ .
The mass spectrum is fundamentally a histogram of ion arrivals sorted by mass-to-charge ratio, plotted as relative abundance.
Key idea: a mass spectrometer produces a spectrum by making ions, sorting them by $m/z$ in vacuum, and counting how many arrive at each value.

Where this lives today

Modern instrumentation is almost always described in terms of exactly the five-module architecture developed here, even as each module has diversified enormously. Clinical laboratories run LC-MS/MS panels for newborn screening (amino acid and acylcarnitine profiling from dried blood spots) and for therapeutic drug monitoring, where a triple-quadrupole mass analyzer performs multiple-reaction monitoring for exquisite selectivity in complex biological matrices. Doping-control laboratories use GC-MS and LC-MS/MS to detect anabolic steroids and their metabolites at parts-per-trillion levels in urine, exploiting the same electron-multiplier detection physics worked out above. Proteomics laboratories run nanoflow LC coupled to Orbitrap or quadrupole–time-of-flight hybrids, resolving thousands of peptide ions per chromatographic run and relying on the resolving-power calculation shown here to distinguish near-isobaric peptide fragments. Semiconductor fabs use inductively coupled plasma mass spectrometry (ICP-MS) — a completely different ion source (an argon plasma at ~7000 K) feeding the same downstream architecture — to certify ultratrace metal contamination in ultrapure water and wafer-cleaning chemicals at parts-per-quadrillion levels. Space missions carry miniaturized quadrupole and time-of-flight mass spectrometers (as on the Curiosity and Perseverance rovers, and the Rosetta comet mission) to analyze planetary atmospheres and volatile organics, where the same vacuum and ion-optics principles must be engineered to survive launch and operate with minimal power. Across all of these, the conceptual chain — ionization-methods-overview to generate the ions, mass-analyzers-overview to sort them, and the-mass-spectrum-and-m-z to interpret what comes out — is the invariant intellectual skeleton beneath a bewildering variety of hardware.

Ask the tutor: How does an electrospray source actually turn a liquid droplet into bare gas-phase ions?

The physical picture

Picture the analyte molecule as it moves through the instrument. It begins as a neutral species — dissolved in a mobile phase eluting from a liquid chromatograph, or as a vapor swept along in a stream of helium from a gas chromatograph, or simply as a solid placed on a probe tip. The inlet is whatever hardware delivers this material into the region where ionization occurs, and it must reconcile two conflicting demands: chromatographic inlets operate near atmospheric pressure, while the rest of the instrument lives in high vacuum, so the inlet is also the pressure-reduction interface. Once inside the source region, the analyte must become charged. This is the job of the ion source, and it is here that chemistry (not just physics) enters: electron ionization (EI) knocks an electron out of a gas-phase molecule using a 70 eV electron beam; electrospray ionization (ESI) sprays a charged droplet that shrinks by solvent evaporation until bare ions are ejected; matrix-assisted laser desorption/ionization (MALDI) uses a laser pulse absorbed by a crystalline matrix to eject analyte ions from a solid deposit. All of these techniques share one output: a beam or packet of gas-phase ions, each carrying an integer number of elementary charges, $z$ , and each having a definite mass $m$ . These ions are then accelerated and focused by a set of electrostatic lenses into the mass analyzer, the module that exploits the ion's motion in electric or magnetic fields to sort ions by $m/z$ . A magnetic sector bends ion trajectories with radius proportional to $\sqrt{m/z}$ ; a quadrupole filters ions by the stability of their oscillation in an oscillating radiofrequency field; a time-of-flight (TOF) tube sorts ions purely by their flight time over a fixed distance, since lighter ions of the same charge and kinetic energy travel faster. Whichever physical principle is used, the output is the same: ions of a selected (or continuously scanned, or simultaneously dispersed) $m/z$ emerge and strike the detector. The detector's job is deceptively simple — count or measure the current produced by ion impacts — but the currents involved are minuscule. A beam of $10^6$ ions per second carrying single positive charge corresponds to a current of only

I = (10^6 \ \text{s}^{-1})(1.602\times10^{-19}\ \text{C}) = 1.6\times10^{-13}\ \text{A}

which is far below what a simple ammeter can resolve, so essentially all mass spectrometer detectors incorporate gain — an electron multiplier that amplifies a single ion impact into a cascade of $10^5$ – $10^8$ electrons, or a Faraday cup used with a very sensitive electrometer for high-abundance ion beams. Underlying every module downstream of the ion source is the vacuum system: turbomolecular and rotary/scroll pumps (sometimes with additional cryopumping) that hold the analyzer and detector regions at pressures low enough that an ion's mean free path between collisions with residual gas exceeds the physical path length of the instrument. Without this, ions would scatter, lose energy, undergo unwanted collisions, and the mass separation would blur into noise.

Figure

Overall architecture of a mass spectrometer, module by module

There is also a directional logic to the pressure drop across the instrument. Atmospheric-pressure ion sources like ESI operate at roughly 760 torr right at the spray tip, but ions must be sampled through a small orifice into a differentially pumped region at $\sim 1$ torr, then into an intermediate stage at $\sim 10^{-3}$ torr, and finally into the analyzer/detector region at $10^{-6}$ – $10^{-9}$ torr. Each stage is pumped separately (differential pumping) precisely because a single pump cannot economically bridge nine orders of magnitude of pressure in one step.

Figure

Differential pumping stages from atmosphere to high vacuum, ESI example

The core ideas

The single most important relationship governing the whole instrument is that mass analyzers do not measure mass; they measure mass-to-charge ratio:

\frac{m}{z}

where:

$m$ is the ion mass, conventionally in unified atomic mass units (u) or daltons (Da),
$z$ is the number of elementary charges on the ion (a dimensionless integer, sign indicates polarity),
the ratio $m/z$ is dimensionless in the strict IUPAC convention (mass in u divided by charge number), though many practitioners quote it in "Da" as shorthand.

Reading: a singly charged ion of mass 500 Da appears at $m/z = 500$ ; a doubly charged ion of mass 1000 Da (common for large peptides in ESI) appears at the same nominal $m/z = 500$ . Mass analyzers cannot distinguish these without additional information (isotope spacing, as covered in the treatment of the mass spectrum itself), which is precisely why charge-state determination is a distinct analytical step. The second governing relationship is the energy the ion source and accelerating optics impart to the ion, since nearly every mass analyzer's separation principle traces back to the kinetic energy equation:

zeV = \frac{1}{2}mv^2

where:

$z$ is the charge number and $e$ the elementary charge ( $1.602\times10^{-19}$ C),
$V$ is the accelerating potential difference (volts) the ion falls through,
$m$ is the ion mass (kg, for SI consistency) and $v$ its resulting speed (m/s).

Reading: every ion accelerated through the same potential $V$ acquires kinetic energy proportional to its charge, but its resulting speed depends on $\sqrt{z/m}$ — lighter ions (or more highly charged ones) end up moving faster for the same accelerating voltage. This single equation is the seed from which time-of-flight separation, magnetic-sector radius, and quadrupole stability all grow. The third relationship, essential to justifying the vacuum system rather than treating it as an afterthought, is the mean free path of a gas molecule (or ion) at pressure $P$ :

\lambda = \frac{k_BT}{\sqrt{2}\,\pi d^2 P}

where:

$k_B$ is Boltzmann's constant ( $1.381\times10^{-23}$ J/K),
$T$ is the absolute temperature (K),
$d$ is the effective collision diameter of the gas molecule (m), typically $\sim 3$ – $4\times10^{-10}$ m for N $_2$ or air,
$P$ is the pressure (Pa).

Reading: mean free path is inversely proportional to pressure — halving the pressure doubles the average distance an ion travels before hitting a gas molecule. Because analyzer path lengths run from tens of centimeters (quadrupoles) to meters (TOF, sectors, Orbitraps), the vacuum must be good enough to push $\lambda$ well beyond the analyzer length, or ions scatter and resolution collapses.

Derivation AMean free path sets the vacuum requirement

Goal: show quantitatively why an analyzer with a 1-meter flight path needs pressure near $10^{-6}$ torr, not just "some vacuum," and derive the pressure at which $\lambda$ matches a stated path length.

Step 1. Start from kinetic theory of gases. For a gas of identical hard-sphere molecules of diameter $d$ at number density $n$ (molecules/m $^3$ ), the mean free path is

\lambda = \frac{1}{\sqrt{2}\,\pi d^2 n}.

This comes from the collision cross-section $\sigma = \pi d^2$ and the relative-speed correction factor $\sqrt{2}$ that accounts for the fact that both the traveling molecule and the target molecules are moving (not just one recoiling off a fixed lattice).

Step 2. Convert number density to pressure via the ideal gas law. Since $P = nk_BT$ , we have $n = P/(k_BT)$ , so substituting:

\lambda = \frac{k_BT}{\sqrt{2}\,\pi d^2 P}.

This is the working formula. Note $\lambda$ depends on $T$ and $d$ (properties of the residual gas) and inversely on $P$ (the vacuum quality), but not on the identity or speed of the traveling ion, since the ion's own cross-section is normally much smaller than that of the background gas it collides with when the ion density is low.

Step 3. Plug in numbers for residual air (mostly N $_2$ /O $_2$ ) at room temperature. Take $T = 298$ K, $d = 3.7\times10^{-10}$ m (a standard kinetic-theory value for N $_2$ ), and solve for the pressure $P^{*}$ at which $\lambda$ equals a target path length $L$ :

P^{*} = \frac{k_BT}{\sqrt{2}\,\pi d^2 L}.

For $L = 1.00$ m (a modest analyzer flight path):

P^{*} = \frac{(1.381\times10^{-23}\ \text{J/K})(298\ \text{K})}{\sqrt{2}\,\pi (3.7\times10^{-10}\ \text{m})^2 (1.00\ \text{m})}

P^{*} = \frac{4.115\times10^{-21}}{\sqrt{2}\,\pi (1.369\times10^{-19})(1.00)} \ \text{Pa}

P^{*} = \frac{4.115\times10^{-21}}{6.083\times10^{-19}} \ \text{Pa} = 6.77\times10^{-3}\ \text{Pa}.

Converting to torr (1 torr = 133.32 Pa):

P^{*} = \frac{6.77\times10^{-3}\ \text{Pa}}{133.32\ \text{Pa/torr}} = 5.08\times10^{-5}\ \text{torr}.

Step 4. Interpret. To make the mean free path equal the path length is only marginal — collisions would still happen appreciably often along the way. In practice, instrument designers require $\lambda$ to exceed the path length by one to two orders of magnitude so that the probability of even a single collision is small (a collision at high kinetic energy can fragment or deflect the ion, corrupting the mass measurement). That drives working analyzer pressures down to $10^{-6}$ – $10^{-7}$ torr for quadrupoles and time-of-flight instruments, and lower still ( $10^{-9}$ – $10^{-10}$ torr) for very long path length or long-trapping-time instruments like Fourier-transform ion cyclotron resonance (FT-ICR) cells, where ions must survive for seconds without a single destabilizing collision.

Go deeper · Derivation B · 1 minIon acceleration, speed, and flight time — the physics every analyzer shares

Derivation BIon acceleration, speed, and flight time — the physics every analyzer shares

Goal: derive the flight time of an ion accelerated through a known potential and traveling a known distance, the calculation that underlies time-of-flight separation and is embedded implicitly in every other analyzer's ion optics.

Step 1. Energy conservation during acceleration. An ion of charge $ze$ starting at rest falls through potential difference $V$ (the source/extraction voltage). All of the electrical potential energy converts to kinetic energy:

zeV = \frac{1}{2}mv^2.

Step 2. Solve for the ion's speed. Rearranging,

v = \sqrt{\frac{2zeV}{m}}.

This shows the speed scales as the square root of charge-to-mass ratio — an ion with twice the charge (all else equal) moves $\sqrt{2}$ times faster; an ion with four times the mass moves half as fast.

Step 3. Relate speed to flight time over a field-free drift region of length $L$ . Since the ion moves at constant speed $v$ once it leaves the acceleration region (no further field acts on it in the drift tube),

t = \frac{L}{v} = L\sqrt{\frac{m}{2zeV}}.

Step 4. Rearrange to show mass-to-charge dependence explicitly. Squaring both sides and solving for $m/z$ ,

\frac{m}{z} = \frac{2eV t^2}{L^2}.

Step 5. Interpret. This is the master equation of time-of-flight mass spectrometry: for fixed accelerating voltage $V$ and drift length $L$ , $m/z$ is directly proportional to $t^2$ . Measuring the arrival time $t$ of an ion pulse — with sub-nanosecond timing electronics — is therefore equivalent to measuring $m/z$ . It also illustrates a universal fact used by every analyzer type: the accelerating voltage sets a known, common kinetic energy per unit charge for all ions leaving the source, and it is exactly this common starting condition that lets a downstream analyzer sort ions by mass without needing to independently know each ion's initial energy.

Worked examples

Example 1 (Foundational) — accelerating voltage, speed, and kinetic energy of a small ion. A singly charged ion of chlorobenzene, $\text{C}_6\text{H}_5\text{Cl}^{+\bullet}$ , formed by electron ionization, has monoisotopic mass 112.0 Da (using $^{35}$ Cl). It is accelerated through $V = 3.00\times10^3$ V in the source region of a sector instrument. Find its speed. First convert mass to kilograms using 1 Da $= 1.6605\times10^{-27}$ kg:

m = 112.0 \times 1.6605\times10^{-27}\ \text{kg} = 1.860\times10^{-25}\ \text{kg}.

With $z = 1$ , $e = 1.602\times10^{-19}$ C, $V = 3.00\times10^3$ V:

v = \sqrt{\frac{2(1)(1.602\times10^{-19}\ \text{C})(3.00\times10^3\ \text{V})}{1.860\times10^{-25}\ \text{kg}}}

v = \sqrt{\frac{9.612\times10^{-16}}{1.860\times10^{-25}}}\ \text{m/s} = \sqrt{5.168\times10^{9}}\ \text{m/s} = 7.19\times10^{4}\ \text{m/s}.

So the chlorobenzene molecular ion moves at about 71.9 km/s — roughly 210 times the speed of sound in air — after just 3 kV of acceleration. This illustrates why ion optics must be precisely aligned: at these speeds, an ion crosses a 1 cm lens aperture in about 140 nanoseconds, leaving essentially no time for stray fields to nudge its trajectory without being accounted for in the instrument's design.

Limiting-case check: if $V \to 0$ , then $v \to 0$ — an unaccelerated ion has no directed kinetic energy, consistent with the physical picture that acceleration is what gives the ion a defined trajectory into the analyzer.

Example 2 (Applied) — resolving two isobaric interferences in environmental GC-MS. A quadrupole GC-MS is used to quantify the pesticide breakdown product with nominal $m/z = 246$ in a soil extract, but a co-eluting hydrocarbon matrix interference also produces a fragment near $m/z = 246$ . High-resolution accurate-mass data (obtained on a benchtop Orbitrap-type instrument used for confirmation) show the analyte fragment at $m/z = 246.9891$ and the interference at $m/z = 246.9761$ . Determine the resolving power $R$ required to separate them, defined as

R = \frac{m}{\Delta m}

using $m$ as the average of the two masses and $\Delta m$ their difference.

\Delta m = 246.9891 - 246.9761 = 0.0130\ \text{Da}.

m_{\text{avg}} = \frac{246.9891 + 246.9761}{2} = 246.9826\ \text{Da}.

R = \frac{246.9826}{0.0130} = 1.90\times10^{4}.

A resolving power of about 19,000 (often quoted as "19k FWHM") is required to baseline-resolve these two ions — well beyond the roughly 1,000–2,000 unit-mass resolution of a standard quadrupole, which is why the confirmatory measurement must be made on a high-resolution analyzer. This example shows concretely why the choice of mass analyzer (a topic covered in its own dedicated treatment) is not a cosmetic detail: a quadrupole GC-MS could not have distinguished these two ions and would have reported a falsely inflated peak area for the pesticide residue, while the high-resolution analyzer resolves them cleanly.

Limiting-case check: if the two masses were identical ( $\Delta m \to 0$ ), $R \to \infty$ — no analyzer, however good, can separate truly isobaric species, which correctly predicts that some interferences require chromatographic (not just mass spectral) separation beforehand.

Example 3 (Exam/competition level) — full instrument budget: vacuum, transit time, and detector gain in series. A time-of-flight mass spectrometer has a 1.20 m field-free drift tube held at $2.40\times10^{-7}$ torr, accelerates ions through $V = 20.0$ kV, and uses an electron multiplier with gain $G = 1.0\times10^{6}$ per incident ion. A sample ion has monoisotopic mass 524.32 Da and charge $z=2$ (a doubly protonated peptide fragment, $[\text{M}+2\text{H}]^{2+}$ ). (a) Compute the ion's flight time. Using the master TOF equation solved for $t$ :

t = L\sqrt{\frac{m}{2zeV}}.

Convert mass: $m = 524.32 \times 1.6605\times10^{-27}\ \text{kg} = 8.706\times10^{-25}\ \text{kg}$ .

t = (1.20\ \text{m})\sqrt{\frac{8.706\times10^{-25}\ \text{kg}}{2(2)(1.602\times10^{-19}\ \text{C})(2.00\times10^{4}\ \text{V})}}

t = (1.20)\sqrt{\frac{8.706\times10^{-25}}{1.282\times10^{-14}}}\ \text{s} = (1.20)\sqrt{6.791\times10^{-11}}\ \text{s}

t = (1.20)(8.241\times10^{-6}\ \text{s}) = 9.89\times10^{-6}\ \text{s} = 9.89\ \mu\text{s}.

(b) Check that vacuum is adequate. Using the mean-free-path formula at $T = 298$ K, $d = 3.7\times10^{-10}$ m, and $P = 2.40\times10^{-7}\ \text{torr} = 2.40\times10^{-7}\times133.32\ \text{Pa} = 3.20\times10^{-5}\ \text{Pa}$ :

\lambda = \frac{(1.381\times10^{-23})(298)}{\sqrt{2}\,\pi (3.7\times10^{-10})^2 (3.20\times10^{-5})}

\lambda = \frac{4.115\times10^{-21}}{\sqrt{2}\,\pi (1.369\times10^{-19})(3.20\times10^{-5})}\ \text{m}

\lambda = \frac{4.115\times10^{-21}}{1.947\times10^{-23}}\ \text{m} = 211\ \text{m}.

Since $\lambda = 211$ m is nearly 180 times the 1.20 m drift length, the probability of a collision along the flight path is small (roughly $L/\lambda \approx 0.6\%$ ), confirming the vacuum is more than adequate for this instrument's path length. (c) Estimate the output charge pulse for a single ion arrival. With gain $G = 1.0\times10^6$ electrons produced per ion striking the multiplier, the charge in the output pulse is

Q = G \times e = (1.0\times10^{6})(1.602\times10^{-19}\ \text{C}) = 1.6\times10^{-13}\ \text{C}.

If the multiplier's output pulse has a characteristic width of about 5 ns, the peak current is roughly $Q/\Delta t \approx (1.6\times10^{-13}\ \text{C})/(5\times10^{-9}\ \text{s}) \approx 3\times10^{-5}$ A — readily measurable by fast preamplifier electronics, whereas the bare single-ion charge ( $1.6\times10^{-19}$ C) would be utterly undetectable without this multiplication. This calculation strings together all three governing relationships — kinetic energy/flight time, mean free path, and detector gain — exactly as a real TOF instrument budget would.

Limiting-case check: as $z$ increases at fixed $m$ (higher charge state, common for large biomolecules in ESI), $t \propto 1/\sqrt{z}$ decreases — multiply protonated ions arrive earlier, which is exactly why ESI mass spectra of proteins show a "ladder" of charge states spread across a range of $m/z$ rather than one clean peak at the molecular mass.

Limiting cases, scope & common traps

Scenario	What happens	Why
$z \to 0$ (neutral species)	Cannot be detected at all	Electric/magnetic fields exert no force on a neutral; only charged species can be steered and mass-analyzed
Vacuum degrades ( $P$ rises)	Peaks broaden, sensitivity drops, background rises	Mean free path shortens; ions scatter and lose energy before reaching the detector
Two ions with equal $m/z$ but different $m$ and $z$	Appear at the same nominal position in the spectrum	The analyzer only measures $m/z$ ; charge-state deconvolution needs isotope spacing or independent methods
Accelerating voltage $V$ drifts during a scan	Systematic mass calibration error	The TOF/sector equations assume $V$ is constant and precisely known; instruments actively regulate and calibrate against it
Ion source is at atmospheric pressure (ESI, APCI)	Requires multi-stage differential pumping to interface with the vacuum analyzer	A single pump cannot bridge 760 torr to $10^{-7}$ torr efficiently or affordably
High ion flux saturates the detector	Faraday cup used instead of electron multiplier for major/base peak measurements	Electron multipliers have finite dynamic range and can be damaged or saturate at high count rates

The chemist behind it

The architecture described here crystallized gradually, but its conceptual founder is usually identified as Francis William Aston, who built on the parabola spectrograph developed by his mentor J. J. Thomson at the Cavendish Laboratory. Aston's key insight, embodied in his 1919 mass spectrograph, was velocity focusing — arranging electric and magnetic fields so that ions of the same $m/z$ but slightly different initial velocities still converge to the same point on a photographic plate. This decoupled the resolving power of the instrument from the messiness of real ion sources, a problem every later analyzer design (sector, quadrupole, TOF) has had to solve in its own way. Aston used his instrument to demonstrate that many elements are mixtures of isotopes with nearly integer mass numbers, work that earned him the 1922 Nobel Prize in Chemistry and laid the empirical foundation for the modern nuclide chart. Arthur Jeffrey Dempster, working concurrently in Chicago, built the first instrument with true directional and velocity focusing sensitive enough for quantitative isotope abundance measurements, and it was Dempster's design lineage that fed into the electron-impact ion sources and magnetic-sector analyzers that dominated mass spectrometry for the next half-century.

Origins · 1 minFrancis William Aston and how this was discovered

Francis William Aston

1877–1945 · Cavendish Laboratory, Cambridge, early 20th century

Aston trained originally in chemistry and brewing science before being drawn into physics by Thomson's group studying "positive rays" — beams of positively charged gas ions. His mass spectrograph achieved resolving power sufficient to separate neon-20 from neon-22, settling a long argument about whether "neon" was a single element with fractional atomic weight or a mixture of isotopes. Aston's whole-number rule — that isotopic masses cluster near integers when referenced to oxygen-16 — became an essential clue in the emerging understanding of nuclear structure, predating the neutron's discovery by over a decade.

History

- 1897 — J. J. Thomson identifies the electron and begins positive-ray (canal ray) studies that motivate later ion-optics work.

- 1912 — Thomson's parabola spectrograph separates neon isotopes using combined electric and magnetic deflection, the direct ancestor of the mass spectrometer.

- 1919 — Aston builds the first true mass spectrograph with velocity focusing, measuring isotopic masses to about 1 part in 10,000.

- 1918–1922 — Dempster develops a magnetic-sector instrument with directional focusing, establishing the design lineage for electron-impact sector instruments.

- 1946 — Stephens proposes time-of-flight mass spectrometry, later realized practically by Wiley and McLaren in 1955.

Why does a mass spectrometer need a vacuum system if the ion source itself sometimes runs at atmospheric pressure?

Sources like ESI and APCI ionize the sample right at atmospheric pressure because the chemistry (droplet evaporation, corona discharge) depends on having a bath gas around. But once ions are formed, they must travel through the mass analyzer and reach the detector without colliding with air molecules, or their trajectories and energies get scrambled, ruining resolution. The instrument therefore uses a staged, differentially pumped interface: ions are sampled through a small orifice or capillary into a series of chambers at progressively lower pressure, each pumped by its own pump, until the analyzer region reaches $10^{-6}$ – $10^{-9}$ torr. This lets an atmospheric-pressure source coexist with a high-vacuum analyzer in the same instrument.

Is m/z the same thing as molecular weight?

Only when the ion carries a single charge ( $z=1$ ). In general $m/z$ is the ion's mass divided by its charge number, so a triply charged ion of a 3000 Da protein appears at $m/z \approx 1000$ , not 3000. Electrospray ionization routinely produces a whole distribution of charge states for large biomolecules, and reconstructing the true molecular weight requires combining several $m/z$ values from adjacent charge states algebraically — a calculation that belongs to the broader treatment of reading a mass spectrum rather than to the instrument architecture itself.

What is the practical difference between a Faraday cup and an electron multiplier detector?

A Faraday cup is simply a metal electrode that collects the ion's charge directly; it produces a current proportional to ion flux with essentially no amplification, but it is extremely linear and stable over a wide dynamic range, so it is favored for high-abundance ion beams such as isotope-ratio measurements. An electron multiplier instead uses the impact of a single ion to eject a cascade of secondary electrons from a series of dynodes (or a continuous resistive surface), amplifying a single-ion event into a measurable pulse of $10^5$ – $10^8$ electrons. Multipliers are far more sensitive — capable of true single-ion counting — but they age with use, have finite linear dynamic range, and require periodic gain recalibration.

Why do quadrupole mass spectrometers typically need lower vacuum quality than time-of-flight instruments?

Vacuum requirements scale with the ion's flight path length and dwell time. A quadrupole analyzer's ion path is typically only 10–20 cm, and ions transit it in tens of microseconds, so the mean free path only needs to exceed a short distance. A time-of-flight tube can be a meter or more, and Fourier-transform instruments (ICR, Orbitrap) require ions to orbit stably for hundreds of milliseconds to seconds, accumulating an enormous effective path length. Because mean free path is inversely proportional to pressure, instruments with longer effective flight paths or trapping times require correspondingly lower pressure to keep collision probability acceptably small.

Does every mass spectrometer need both an ion source and a separate mass analyzer, or can one module do both jobs?

Nearly all designs keep the two functions physically distinct because ionization and mass separation impose conflicting requirements — ion sources often need to be "messy" environments (collisional cooling in ESI, a plasma in ICP-MS, a solid matrix in MALDI) while mass analyzers need extremely clean, well-defined electric or magnetic fields. A few specialized designs blur the line, such as ion trap instruments where ions are both accumulated and mass-analyzed in the same physical trap by scanning the trapping field, but even then the initial ionization event (EI, ESI, etc.) is conceptually and often physically a separate step feeding into the trap.

Why is the mass spectrum reported as relative abundance rather than absolute ion counts?

Absolute ion counts depend on many instrument-specific factors — sample amount injected, ionization efficiency, transmission losses in the ion optics, detector gain and age — that vary from run to run and instrument to instrument. Normalizing each peak's intensity to the most abundant peak (the base peak, set to 100%) removes most of this run-to-run variability and lets spectra be compared meaningfully across instruments, operators, and even literature references, which is essential for library matching and structure elucidation.

Where does the mass calibration of an instrument actually come from?

Mass analyzers measure a physical quantity — flight time, magnetic field and radius, or RF frequency — that must be converted into $m/z$ via the appropriate equation (for example the TOF relation $m/z = 2eVt^2/L^2$ derived here). In practice, instruments are calibrated empirically by measuring known reference compounds of precisely known mass (such as perfluorotributylamine for EI instruments, or a mixture of well-characterized small molecules for ESI instruments) and fitting the instrument's raw signal to the known $m/z$ values, which corrects for small deviations from the idealized physics (fringing fields, contact potentials, timing electronics offsets).