The quadrupole is the workhorse detector behind most routine LC-MS and GC-MS systems, and its operating principle is far simpler than the mathematics used to describe it. Two ideas are enough: opposite charges attract while like charges repel, and a heavy ion changes direction more slowly than a light one. Everything else follows from combining those two facts with a voltage that flips sign millions of times per second.
Why it is a filter, not an analyser
The name causes genuine confusion. A quadrupole is routinely called a mass analyser, but it does not analyse anything in the sense that word suggests. At any instant it performs a single, brutally simple operation: it transmits one narrow band of m/z values and destroys everything else against the rods. It does not measure the ions it rejects. It does not sort them. They are neutralised and pumped away. The name given to the device by its inventors was, fittingly, a mass filter — a mass filter without a magnetic field [1][2].
A spectrum only exists because the window is swept. The instrument steps the transmitted mass across the range and the detector records how many ions arrive at each setting. The spectrum is assembled from hundreds of sequential yes-or-no measurements — it is never captured as a whole.
| Quadrupole (filter) | Time-of-flight (analyser) | |
|---|---|---|
| Ions handled at once | One m/z window; the rest are destroyed | All ions in the packet |
| How the spectrum forms | Sequentially, by sweeping the window | In parallel, from a single flight measurement |
| Cost of higher resolution | Fewer ions transmitted, lower sensitivity | Largely independent of transmission |
| Duty cycle in targeted mode | Improves — the window rests on one mass | No comparable gain from targeting |
This distinction is not pedantry; it explains the instrument’s behaviour at the bench. It is why a full scan is less sensitive than selected ion monitoring on the same instrument — while the window sits on mass 150, every ion of mass 300 in the source is thrown away. It is why adding transitions to a method dilutes each one. And it is why a quadrupole cannot retrospectively answer a question you did not ask: the data for the masses you skipped was never collected.
The hardware: four rods, two pairs, two planes
A quadrupole consists of four parallel metal rods arranged around a central axis. They are not four independent electrodes: opposite rods are wired together as a pair, and the two pairs sit in perpendicular planes. Ions enter along the axis, travel between the rods and either reach the detector at the far end or collide with a rod, pick up or lose charge, are neutralised and pumped away.
Each pair carries the same two voltages superimposed [3]:
- a direct current (DC) offset — positive on one pair, negative on the other
- a radiofrequency (RF) alternating voltage — oscillating between positive and negative, applied to both pairs but 180° out of phase between them
The instrument controls the amplitude of both. Scanning a spectrum means ramping them together while holding their ratio fixed.
The two principles that explain everything
1. Electrostatics
A positively charged ion is pushed away from a positively charged rod and pulled toward a negatively charged one. Nothing more subtle than that.
2. Inertia
When the field reverses, every ion is told to change direction. How quickly it complies depends on its mass-to-charge ratio. A light ion responds almost instantly; a heavy ion is still moving on its old trajectory by the time the field has flipped back. Think of a container ship versus a jet ski asked to make the same turn in the same time — the ship is committed to its course long after the order is given.
This is why the quadrupole separates by m/z rather than by mass alone. A doubly charged ion feels twice the force for the same inertia and therefore behaves like a much lighter singly charged one.
The positive rod pair: a high-pass filter
Consider the pair carrying a positive DC offset, with positive ions travelling between them. With DC alone, both light and heavy ions are simply repelled from both rods and pushed toward the axis. Everything gets through — useless as a filter.
Now add the RF component. During the part of the cycle where the RF is negative and exceeds the DC offset, the net potential on those rods turns momentarily negative and the ions are attracted rather than repelled.
| Ion | Behaviour during the negative half-cycle | Outcome |
|---|---|---|
| Light (low m/z) | Responds immediately, accelerates toward the rod before the field reverses | Collides, is neutralised and removed |
| Heavy (high m/z) | Too slow to respond before the field flips positive again; net motion stays close to the axis | Passes through |
The DC offset sets the threshold. Everything below a certain m/z is lost, everything above it survives — a high-pass mass filter.
The negative rod pair: a low-pass filter
The other pair carries a negative DC offset, so the resting tendency of a positive ion is to drift toward those rods. With DC alone nothing would reach the detector at all.
The superimposed RF rescues the light ions. Each time the RF drives those rods positive, the ion is pushed back toward the axis — and a light ion, being nimble, corrects its course before it can reach the rod. The heavy ion cannot: its drift toward the rod is already established, the brief positive phase barely deflects it, and it crashes.
Combining both: the stability window
Because the two pairs act in perpendicular planes, an ion does not travel in a simple zig-zag. Its trajectory is a spiral. Either the spiral stays bounded and the ion reaches the detector, or its amplitude grows until it strikes a rod.
Overlay the two filter curves and only a narrow band of m/z values satisfies both conditions at once. That band is the transmission window. In the formal treatment it emerges from the Mathieu equations and is drawn as the a–q stability diagram, which remains the standard way of visualising how the operating variables interact [3][4]. The physical content of that diagram is exactly the high-pass and low-pass behaviour described above: moderate RF amplitudes stabilise a trajectory, large ones destabilise it [4].
Why the rods have to be near-perfect
The field between the rods is defined by their shape and their alignment, so mechanical imperfections translate directly into performance. Recent simulation work shows that a defect in even a single rod can introduce non-linear resonances and produce anomalous peaks in the transmitted signal — artefacts that come from the geometry rather than from the sample [5]. This is the quadrupole’s counterpart to the manufacturing tolerances that limit Fourier-transform instruments, and it is one reason a quadrupole that has been mechanically disturbed rarely tunes back to its original specification.
How a spectrum is actually recorded
With the ratio fixed, the instrument ramps both amplitudes from low to high. The window sweeps across the mass range, and the detector counts ions at each position. Most positions yield nothing; where a fragment exists, a peak appears. Plot ion count against m/z and the mass spectrum is complete.
The timing is the impressive part. All of this has to finish before the next compound elutes from the column — a full scan across several hundred mass units in a fraction of a second.
Full scan versus selected ion monitoring
| Mode | What the quadrupole does | Use it for | Trade-off |
|---|---|---|---|
| Full scan | Sweeps the entire mass range continuously | Unknown identification, library searching, confirming spectra | Little time spent at any one mass, so sensitivity is limited |
| Selected ion monitoring (SIM) | Parks the window on a few chosen m/z values | Trace quantification of known targets | Anything not on the list is invisible |
The practical limit with narrow peaks
A quadrupole records one mass at a time, so a full spectrum is assembled sequentially rather than captured at once. With UHPLC peaks only a few seconds wide, the scan speed and dwell time must be set so that at least 12–15 data points define each chromatographic peak. Too slow, and peak shape and integration precision deteriorate; too many SIM transitions crowded into one window, and the dwell time per ion drops until the signal becomes noisy. This is the usual reason a method that worked on an HPLC system degrades after transfer to UHPLC — see our comparison of HPLC and UHPLC.
Reading the resulting spectrum
The peak corresponding to the intact ionised molecule is the molecular ion. It is often not the tallest one: fragmentation frequently makes a smaller piece dominate, and that most abundant signal is the base peak. Small satellite peaks a mass unit or two higher usually reflect natural isotope abundance rather than separate compounds — the carbon-13 pattern is a familiar example. Our article on nominal, average, exact and accurate mass explains which mass value you are actually looking at.
Identification from a chromatographic run therefore follows a fixed route: select the peak, extract its spectrum, then either interpret the fragmentation manually or run a spectral library search. A library match is a strong hint, not proof — confirmation still rests on retention time, qualifier-ion ratios and, where the answer matters, an authentic reference standard.
What this means at the bench
- Mass accuracy drifts because the relationship between voltage and transmitted m/z depends on stable electronics and clean surfaces. This is why regular tuning and mass calibration are not optional.
- Contaminated rods distort the field. Deposits from nonvolatile buffers, matrix or high sample loads degrade peak shape and sensitivity before they cause an outright failure — one more reason volatile ammonium buffers are mandatory in LC-MS.
- Sensitivity loss is often a resolution setting, not a broken instrument. If the tune has been narrowed, fewer ions get through by design.
- Unit resolution cannot separate isobars. Two compounds with the same nominal mass are indistinguishable to a quadrupole. Separating them requires either chromatographic separation, a tandem transition, or a high-resolution analyser such as a time-of-flight or Orbitrap system.
Summary
Two rod pairs, each acting as one half of a bandpass filter, define a narrow transmission window whose position is set by the voltage amplitude and whose width is set by the RF-to-DC ratio. Sweeping that window across the mass range produces the spectrum. Nothing is analysed along the way — everything outside the window is discarded, which is precisely why the device is a filter. That design rests on nothing more exotic than electrostatic attraction and the fact that heavy objects turn slowly, and it is why the quadrupole remains one of the most robust and affordable mass filters in routine analytical work.
References and further reading
The operating principle described here is standard textbook material. The sources below are the original papers that introduced the device and the reference works that treat its theory and behaviour in depth.
- Paul W, Steinwedel H. Ein neues Massenfilter ohne Magnetfeld. Zeitschrift für Naturforschung A. 1953;8:448–450 — the paper that introduced the quadrupole mass filter.
- Paul W. Electromagnetic traps for charged and neutral particles (Nobel Lecture). Angewandte Chemie International Edition in English. 1990;29:739–748.
- Miller PE, Denton MB. The quadrupole mass filter: basic operating concepts. Journal of Chemical Education. 1986;63(7):617–622 — the classic accessible treatment of the a–q stability diagram. doi:10.1021/ed063p617
- Understanding the quadrupole mass filter through computer simulation. Journal of Chemical Education. 1998;75:1049 — how moderate RF amplitudes stabilise and large ones destabilise a trajectory. doi:10.1021/ed075p1049
- Anomalous peak formation in the second stability zone of quadrupole mass filters: role of non-linear resonances induced by single rod defects. Preprint, 2026 — mechanical defects in a single rod as a source of artefact peaks. arxiv.org (PDF) (preprint, not yet peer-reviewed)
- Dawson PH. Quadrupole Mass Spectrometry and Its Applications. Elsevier, Amsterdam, 1976 — the standard monograph on quadrupole theory and design.
Leave a Reply