Quadrupole Theory Deep Dive

1. Scope and notation

The canonical quadrupole derivation is elegant, but by itself it can leave gaps when you try to explain why one instrument transmits better than another, why a scan line that should work does not, or why source behavior dominates observed performance. This page keeps the derivation explicit and then layers in the practical terms that working instrument scientists care about.

Throughout this page, the quadrupole is treated first as an ideal transverse field with infinitely long rods. That model gives the Mathieu equation and the standard stability diagram. Real instruments are finite, so transmission is a finite-time traversal problem with limited acceptance in transverse phase space and sensitivity to injection conditions.

x, y; s: Transverse coordinates x and y; s is the axial distance along the rods. All are in meters.
m, e, z: Ion mass m in kg, elementary charge e in coulombs, and integer charge state z. For a positive ion, charge is ze. Ion mass is not the neutral molecular mass.
r0: Field radius: minimum axis-to-electrode distance, not the radius of a cylindrical rod.
U: DC potential difference between the x and y rod pairs (pair-to-pair).
V; −V cos(Ωt): V is the zero-to-peak RF amplitude of the pair-to-pair voltage. The minus sign in U − V cos(Ωt) fixes the RF phase convention used below.
Ω: Angular RF drive frequency: Ω = 2πf. Use Hz for f and SI units in the equations.
ξ = Ωt / 2: Dimensionless time used to map the equations to Mathieu form.
a, q: Canonical Mathieu parameters. Formula coefficients must match the voltage reporting convention; a reporting change does not change the physical a or q.
L: Effective filter length (fringing fields make the exact boundary fuzzy).

Many derivations differ by factors of 2 or 4 because of voltage and potential definitions. The physics is unchanged, but the bookkeeping must be internally consistent.

Choose your route

For the full derivation, you need Newton's law, electric potential gradients, and second-order differential equations. For instrument intuition, read the field picture, then the scan-line explanation and finite-length examples. The guided explorer exercises provide a check on both routes.

What the first model deliberately leaves out

The ideal equations assume a pure sinusoidal drive, an ideal quadrupolar field without axial variation, and an isolated ion without collisions or space-charge forces. They contain no mechanical aperture, entrance or exit field, detector response, or chemical selectivity. Infinite-time stability answers a question about that differential equation; it is not a complete instrument-performance model.

2. Ideal quadrupole field from Laplace's equation

In an ideal quadrupole mass filter, the electrodes produce a transverse potential that satisfies Laplace's equation and has quadrupolar symmetry. Ignoring end effects and axial variation, the scalar potential can be written as a second-order harmonic in x and y.

$$\begin{aligned} \nabla^2 \Phi(x,y,t) &= 0 \\ \Phi(x,y,t) &= \frac{\phi_0(t)}{2r_0^2}(x^2-y^2) \\ \phi_0(t) &= U - V\cos(\Omega t) \end{aligned}$$

Here each rod pair is at ±[U − V cos(Ωt)]/2 relative to the electrical center. The factor of one half therefore agrees with the pair-to-pair definitions above. Choosing U + V cos(Omega t) instead flips the sign of q, which is equivalent to an RF phase shift for infinite-time stability. When comparing individual finite trajectories, keep the injection phase consistent. See the NIST Mathieu sign transformation and quadrupole voltage conventions.

Taking the gradient gives the electric field components:

$$\begin{aligned} E_x &= -\frac{\partial \Phi}{\partial x} = -\frac{\phi_0(t)}{r_0^2}x \\ E_y &= -\frac{\partial \Phi}{\partial y} = +\frac{\phi_0(t)}{r_0^2}y \end{aligned}$$

The symmetry is the key to quadrupole filtering: at a given instant the field focuses one transverse axis while defocusing the other. The alternating field can produce net bounded motion in both axes for selected ions when the operating point falls inside a stability region.

Exact equipotential contours from the ideal quadrupole field

Mathematically generated contour plot of the ideal quadrupole potential in the x-y plane.

These contours are computed from the ideal dimensionless potential proportional to x^2 - y^2, not hand-drawn curves.

Field-vector snapshot from the same potential (exact gradients)

Mathematically generated vector-field plot of the ideal quadrupole electric field at a fixed RF phase.

The arrows are computed from E = -grad(Phi) for a fixed field phase, showing the instantaneous focus/defocus symmetry explicitly.

Check the convention: what is Φ at x = r₀, y = 0, and at x = 0, y = r₀?

The first point is at +[U − V cos(Ωt)]/2 and the second at its negative. Their difference is U − V cos(Ωt). This boundary-value check is a quick way to catch a factor-of-two mismatch between a potential equation and an electronics voltage label.

3. Deriving the Mathieu equation

For an ion of charge +e and mass m, Newton's law gives the transverse equations of motion directly from the field above.

$$\begin{aligned} m\frac{d^2x}{dt^2} &= eE_x = -e\left[\frac{\phi_0(t)}{r_0^2}\right]x \\ m\frac{d^2y}{dt^2} &= eE_y = +e\left[\frac{\phi_0(t)}{r_0^2}\right]y \end{aligned}$$

Substitute phi0(t) = U - V cos(Omega t):

$$\begin{aligned} \frac{d^2x}{dt^2} + \left[\frac{e}{mr_0^2}\right]\left[U - V\cos(\Omega t)\right]x &= 0 \\ \frac{d^2y}{dt^2} - \left[\frac{e}{mr_0^2}\right]\left[U - V\cos(\Omega t)\right]y &= 0 \end{aligned}$$

Introduce dimensionless time xi = Omega t / 2. Since d^2/dt^2 = (Omega^2/4) d^2/dxi^2, multiply through by 4/Omega^2 to obtain:

$$\begin{aligned} \frac{d^2x}{d\xi^2} + \left[\frac{4eU}{mr_0^2\Omega^2} - \frac{4eV}{mr_0^2\Omega^2}\cos(2\xi)\right]x &= 0 \\ \frac{d^2y}{d\xi^2} - \left[\frac{4eU}{mr_0^2\Omega^2} - \frac{4eV}{mr_0^2\Omega^2}\cos(2\xi)\right]y &= 0 \end{aligned}$$

Compare to the canonical Mathieu equation

$$\frac{d^2u}{d\xi^2} + \left[a - 2q\cos(2\xi)\right]u = 0$$

which gives, for this convention:

$$\begin{aligned} a_x &= \frac{4eU}{mr_0^2\Omega^2} \\ q_x &= \frac{2eV}{mr_0^2\Omega^2} \\ a_y &= -a_x \\ q_y &= -q_x \end{aligned}$$

Convention difference	What changes	What does not
Use `U + V cos(Omega t)`	`q` sign flips	Stability physics (phase reference shifts)
Use peak-to-peak RF instead of zero-to-peak RF	The formula coefficient halves: `q = eVpp/(mr₀²Ω²)` for pair-to-pair Vpp = 2V	The physical field, calculated `q`, and canonical apex coordinates
Report per-rod U and zero-to-peak V instead of pair-to-pair values	Reported voltages halve; formula coefficients double to `a = 8eUrod/(mr₀²Ω²)` and `q = 4eVrod/(mr₀²Ω²)`	The physical field and calculated canonical `a` and `q`

In practice, the dominant source of "wrong equation" arguments is not physics but inconsistent use of electronics voltage calibration (especially RF amplitude reporting) against a design formula written in another convention.

Worked example: a correct calculation can describe a poor filter setting

Take a positive singly charged ion of mass 100 Da, r₀ = 5 mm, and f = 1 MHz. Set the pair-to-pair DC difference to U = 40 V and pair-to-pair zero-to-peak RF amplitude to V = 200 V. Convert before substituting:

m = 100 × 1.66053906892 × 10⁻²⁷ kg
r₀ = 0.005 m; Ω = 2π × 10⁶ s⁻¹
a = 0.156416129; q = 0.391040322; a/q = 0.400000

The same physical field is described by per-rod values Urod = 20 V and Vrod = 100 V, or by pair-to-pair RF Vpp = 400 V. Use the corresponding coefficients in the convention table; all three descriptions must give the same a and q. These numerical examples use CODATA 2022 constants.

Does that point have simultaneous first-region stability?

No. At this q, the mapped-Y upper boundary is −a₀(q) ≈ 0.075221, below the calculated a of 0.156416. X is stable, but Y is not. Also, the slope 0.4 exceeds the first-region apex slope derived below. Changing mass while holding these field settings fixed moves the point along the same line; it cannot create a simultaneous first-region interval on that line.

Predict the scaling: double the ion mass, double its positive charge, or double the RF frequency. What changes?

At fixed remaining quantities, doubling mass halves a and q; doubling positive charge doubles both; doubling frequency divides both by four. Doubling r₀ also divides both by four. Each change preserves a/q when U/V is unchanged. The dimensionless coordinates move, but the scan-line slope does not.

4. Stability regions and scan-line operation

The Mathieu equation has bounded and unbounded solutions depending on the pair (a, q). Shaded interiors indicate ideal infinite-time stability. Characteristic curves mark marginal boundaries; they do not guarantee bounded motion for every initial condition. For quadrupole mass filters, the first stability region (the one near the origin) is the one typically used for mass filtering.

Because a and q scale as 1/m for fixed U, V, r0, and Omega, ions of different masses occupy different points along the same scan line. The ideal stable interval requires both transverse axes to be stable. Actual transmission also requires survival within the aperture over the finite filter length.

Mass filtering mode (RF + DC)

Increasing U and V together at fixed U/V moves a given ion outward along a fixed line. A finite band of m/z values occupies the simultaneous stable interval at each setting. Moving the line closer to the apex narrows the ideal interval; the measured peak width and transmitted current also depend on finite-length and acceptance effects.

RF-only mode (ion guide behavior)

With U = 0, a = 0 and stability is determined by q. This behaves as a mass-dependent transmission guide with a low-mass cutoff set by the RF amplitude and frequency.

For a quadrupole mass filter, the physically relevant region is the simultaneous stability of the x and y equations. Using the phase-shift argument for the y-equation, the branch limits in the first quadrant can be written with Mathieu characteristic values as:

$$\begin{aligned} x\text{-axis stability: } & a_0(q) < a < b_1(q) \\ y\text{-axis stability (mapped): } & -b_1(q) < a < -a_0(q) \\ \Rightarrow\; \text{quadrupole mass-filter region: } & q>0,\quad 0 \le a < \min\!\left[-a_0(q),\, b_1(q)\right] \end{aligned}$$

The "apex" used in quadrupole scan-line discussions is the maximum-a point of this envelope, which occurs where the two branch limits cross: -a0(q) = b1(q). The upper bound must be positive. The a = 0 axis can lie in the stable interior of the full symmetric region, but the origin has no RF confinement: transverse velocity causes drift. See NIST DLMF on open stability regions and their boundaries.

$$\begin{aligned} a &= C_a \frac{U}{m} \\ q &= C_q \frac{V}{m} \end{aligned}$$

Here C_a and C_q collect the constants and convention choices (including e, r0, Omega, and RF voltage definition).

For the canonical Mathieu equation, the numerical apex is (q, a) = (0.705996069713371, 0.236993993112474). Its slope is a/q = 0.335687411416740, giving U/V = 0.167843705708370 when DC and RF use the same pair-to-pair or per-rod convention and RF is zero-to-peak. Voltage reporting does not change canonical apex coordinates.

Numerically evaluated first quadrupole region with scan lines

Numerically evaluated quadrupole mass-filter first stability region plot from Mathieu characteristic values with scan lines overlaid.

This sampled plot is generated from the characteristic branch functions a0(q) and b1(q) (via SciPy Mathieu characteristic values) and shows real scan lines intersecting the envelope.

Characteristic branch limits and envelope crossing condition

Numerically evaluated plot of minus a0(q), b1(q), their envelope, and the crossing function b1(q)+a0(q).

The lower panel samples the crossing function b1(q) + a0(q); its zero defines the apex where the upper envelope switches branches.

A positive scan line uses a = kq for both X and mapped Y. Its first-region passband is the open interval between −a₀(q) = kq (entry) and b₁(q) = kq (exit). At k = 0.33, this is 0.692840636 < q < 0.708700703. At k = 0.336, there is no simultaneous first-region passband. The apex itself is marginal, not a transmitting interval.

Finite-length consequence that textbooks often postpone

The Mathieu stability map describes infinite-time boundedness. A real mass filter has finite length L, so the experimentally relevant question is: does the ion remain within the mechanical aperture for the time it takes to traverse the rods? This is why acceptance and injection conditions matter so much in practice. A stable ion can still hit an electrode because of its initial position or velocity; an infinite-time unstable ion can survive a sufficiently short device.

Open interactive stability diagram

Two ways to move along the same line

Compare different ions at fixed voltages: for a fixed charge sign, increasing m/|z| decreases a and q, moving toward the origin. Scan the voltages for one ion: increasing both U and V by the same factor increases a and q, moving away from the origin. These are different experiments on the same geometry.

Worked passband: the q endpoints reverse when expressed as mass

Choose fixed field settings for which a singly charged 100 Da reference ion has qref = 0.700000, and choose k = a/q = 0.33. For other singly charged ions at those settings, q = 70 Da / m. The ideal q interval above therefore gives:

mmin = 70 / 0.708700703 ≈ 98.7723 Da
mmax = 70 / 0.692840636 ≈ 101.0333 Da
98.7723 Da < m < 101.0333 Da

The high-q exit is the low-mass edge; the low-q entry is the high-mass edge. For a common charge magnitude other than one, apply the same inverse relation to m/|z|. This is an ideal mathematical acceptance interval. Its width is not a measured FWHM and must not be presented as an instrument resolving-power specification.

If you double U and V together in this example, where does the mass window move?

It doubles in mass to approximately 197.5446 < m < 202.0667 Da for singly charged ions. The q boundaries and slope stay the same. The absolute ideal mass width doubles, while the fractional width is unchanged. Real scan laws may add offsets or other adjustments; this calculation is for a strictly fixed U/V ratio.

5. Secular motion, micromotion, and pseudopotential intuition

In the stable regime (especially for moderate q), the ion trajectory can be viewed as a slow secular oscillation with a superimposed RF micromotion. This decomposition is not exact everywhere, but it is a powerful design and diagnostic intuition tool.

$$\begin{aligned} x(\xi) &\sim A\cos(\beta \xi + \phi)\left[1 - \frac{q}{2}\cos(2\xi) + \cdots \right] \\ \omega_{\mathrm{sec}} &\sim \frac{\beta \Omega}{2} \end{aligned}$$

Here beta is the Mathieu characteristic exponent. Near the origin of the first stability region and for small-to-moderate q, one often uses approximate expressions for beta (their exact form depends on axis and convention). The point is conceptual: the RF field creates rapid modulation on top of a slower envelope.

In RF-only guides, the pseudopotential approximation provides a useful averaged picture of radial confinement. Near the mass-filter apex, however, where high selectivity is obtained by operating close to the stability boundary, pseudopotential intuition becomes less reliable and full trajectory calculations matter more.

Practical translation: pseudopotential thinking is excellent for building intuition about guides and transport optics, but high-resolution mass filtering performance is usually limited by boundary effects, finite transit time, fringing fields, and source-beam mismatch rather than by the simple averaged potential picture alone.

6. Phase-space acceptance and finite-length transmission

For a finite quadrupole, transmission is not just a yes/no function of (a, q). It is a phase-space problem. An ion entering the filter is described by position and angle (or transverse momentum) in both axes, plus axial velocity, kinetic-energy spread, and RF phase at injection. The filter transmits only the subset of those initial conditions that remain within the rod aperture for the traversal time.

This subset is the acceptance. In the most practical lab sense, people talk about a quadrupole's "acceptance" as if it were a single number, but in reality it is a multidimensional region that depends on operating point, RF phase, energy spread, and the end-field geometry.

Transverse phase space

At minimum, consider (x, x', y, y'), where x' = dx/ds and y' = dy/ds are trajectory slopes along axial coordinate s. They approximate angles for paraxial motion. We use s here to avoid confusing an axial coordinate with charge state z. Acceptance is the region in this space that survives the filter.

Why RF phase matters

The quadrupole field is time-dependent. Two ions with identical spatial coordinates but different injection RF phase can experience different early-cycle kicks and therefore different outcomes.

Phase-space acceptance and beam ellipse (computed)

Mathematically generated phase-space plot showing a quadrupole acceptance ellipse and a mismatched source beam ellipse with sampled points.

These illustrative ellipses and sampled points show a distribution-overlap calculation. They are not a measured source distribution or an acceptance prediction for a specified instrument; real acceptance need not be elliptical.

Finite-length trajectories from direct numerical integration

Numerically integrated stable and unstable quadrupole trajectories versus axial position for finite rod length.

These trajectories are obtained by directly integrating the canonical Mathieu-form equations for chosen (a,q) pairs over a finite number of RF cycles.

What shrinks or distorts acceptance in practice

Operating closer to the stability boundary for higher resolution.
Fringing fields at the filter entrance and exit (non-ideal axial variation of the field).
Energy spread and axial velocity spread (different transit times and RF phase histories).
Source-beam mismatch (beam emittance larger than or poorly aligned with acceptance).
Rod misalignment, RF imbalance, and field distortion from contamination or tolerances.
Space-charge effects at high ion current (beam self-fields and broadened distributions).

In other words, a quadrupole mass filter is not just a point on the Mathieu diagram. It is a dynamic acceptance system that must be matched to the source and transport optics.

Worked transit-time estimate

In a collisionless ideal field with constant axial speed vₛ, the number of RF cycles experienced is N = fL/vₛ. With axial kinetic energy Eₛ in joules, use vₛ = √(2Eₛ/m). Here axial energy is the kinetic energy along the filter, not the transverse energy or an MS/MS collision-energy setting.

For a 100 Da ion, L = 0.10 m, f = 1 MHz, and axial energy 10 eV, the speed is about 4393 m/s and the transit contains about 22.76 RF cycles. At 40 eV the speed doubles and the number of cycles halves to about 11.38. No universal “enough cycles” threshold follows from these inputs: rejection also depends on the operating point, aperture, and initial conditions.

Two beams have the same m/z and field settings but different axial energies. Must their transmitted fractions match?

No. Their ideal a and q are the same because axial energy is absent from those parameters. Their transit times differ, and their initial phase-space distributions may differ too. Bounded ideal motion does not guarantee that either beam fits inside a finite aperture. A transmission comparison needs trajectory and injection information, not just the stability coordinates.

7. Emittance vs "emissivity" (terminology and source matching)

In ion optics, the technically correct term for the beam's occupied phase-space area is emittance, not emissivity. In practice, people sometimes say "source emissivity" when they mean how broad or well-collimated the emitted ion beam is, or how much usable current the source produces. It is worth separating these ideas because they affect quadrupole performance in different ways.

Beam emittance (ion optics term)

Describes the position-angle spread of the beam (for example in x-x' and y-y' space). Large emittance or poor matching reduces the fraction of ions that fit within quadrupole acceptance.

Source output / brightness

Total ion current alone is not enough. What matters is how current is distributed through transverse phase space and energy, and how that distribution matches the entrance optics and operating point. Brightness definitions specify the phase-space and energy conventions; quote those conventions when comparing sources.

A common failure mode in method optimization is to maximize source current while ignoring the angular spread or energy spread of the beam. The measured current at the source or prefilter can increase while quadrupole transmission, stability near the scan apex, and spectral quality become worse because the beam no longer fits the acceptance.

Source parameters that strongly influence quadrupole transmission via emittance and energy spread

Extraction field geometry and voltages (beam divergence and initial phase-space ellipse orientation).
Source pressure / collisionality (cooling can reduce emittance; excessive collisions can broaden energy spread or cause clustering).
Plume geometry, capillary alignment, skimmer alignment, and source contamination.
Desolvation/thermal conditions in atmospheric-pressure interfaces (changes droplet/cluster survival and ion energy spread).
Space charge at high source currents (self-field expansion and nonlinear transport effects).

Quadrupole transmission ~= overlap(source distribution, quadrupole acceptance)

This overlap depends on:
  (x, x', y, y') distribution
  energy spread and axial velocity spread
  injection RF phase distribution
  end-field geometry and alignment

If you meant "ion source emissivity" in the sense of ion-emission efficiency or source output quality, the most quadrupole-relevant quantities are beam emittance, energy spread, and stability of those quantities over time.

8. Thermal emissivity of source and quadrupole rods

Thermal emissivity (radiative emissivity) is a different concept from beam emittance, but it can still matter in quadrupole systems. The ideal Mathieu derivation assumes perfect geometry and prescribed voltages; it does not model temperature, outgassing, adsorption/desorption, or surface-condition drift. Real instruments do.

Why source thermal emissivity can matter

Source wall temperature affects desorption, clustering, and contamination release/adsorption behavior.
Radiative heat balance influences warm-up time and thermal stability of nearby optics and insulators.
Thermal state can change outgassing load, which affects pressure and therefore collisional transport and background.
In hot sources, temperature-driven chemistry and ion-molecule processes can alter the distribution entering the quadrupole.

Why quadrupole rod thermal emissivity can matter (indirectly)

Rod emissivity usually does not enter the equations of motion directly, but it influences the rods' thermal equilibrium and therefore several second-order effects that become operationally important.

Warm-up behavior and thermal gradients (which affect dimensional stability and alignment at a small but nonzero level).
Surface contamination adsorption/desorption cycles, which alter patch potentials and field quality more directly than emissivity itself.
Outgassing and recovery after contamination or venting, via temperature-dependent surface processes.
Temperature of rod supports and nearby dielectrics, which can affect leakage, charging, and stability.

Category	Examples	Effect on quadrupole behavior
Direct field-defining parameters	`U`, `V`, `Omega`, `r0`, geometry symmetry	Set `a`, `q`, stability mapping, and ideal filtering behavior.
Transport and matching parameters	Source emittance, energy spread, alignment, end fields	Set effective acceptance overlap and therefore transmission/resolution in practice.
Thermal / surface parameters	Source/rod thermal emissivity, contamination films, outgassing state	Indirectly shift stability, noise, and reproducibility through temperature and surface condition.

This is a qualitative, indirect thermal argument. Do not assign rod emissivity a dominant or quantified share of routine drift without instrument-specific evidence. Surface cleanliness, RF phase/amplitude balance and other direct field or transport changes often matter more; emissivity can influence the temperature and surface history behind those effects.

9. Non-ideal fields and engineering realities

The ideal quadrupole is a foundational model, but performance is often limited by deviations from that model. Good quadrupole engineering is about controlling which non-idealities matter at the required resolution and sensitivity.

Common non-idealities with large practical impact

Electrode geometry approximation: circular rods approximate hyperbolic electrodes only for a chosen ratio of rod radius to field radius. Geometry errors introduce higher multipole components.
Rod alignment and spacing tolerances: asymmetry shifts stability boundaries, broadens peaks, and degrades abundance sensitivity.
RF amplitude and phase imbalance: imperfect drive symmetry changes the effective field and can reduce transmission or distort mass calibration behavior.
Fringing fields at the ends: ions see non-ideal entry/exit fields, so finite-length transmission differs from infinite-length predictions.
Electronic waveform purity: harmonic content and noise produce field perturbations and can widen peaks or shift apparent operating points.
Space charge: high ion current can broaden the beam distribution and alter trajectories, especially upstream of the filter.
Contamination and patch potentials: surface films can distort local fields and worsen long-term reproducibility.

Textbook-limited thinking

"The point is in the stability region, so transmission should be high." This misses finite length, injection phase, emittance mismatch, and end-field effects.

Instrument-limited thinking

"The rods are dirty, so the math is useless." Also wrong. The ideal model still defines the baseline behavior; the engineering problem is quantifying deviations from it.

10. Practical tuning and troubleshooting implications

The value of quadrupole theory is not only in deriving a and q, but in making better decisions when performance drifts. The checklist below translates the theory into an engineering workflow.

Start with the field-defining parameters. Verify RF amplitude calibration, frequency, DC/RF ratio, and scan electronics assumptions (including voltage convention).
Confirm finite-length and matching conditions. Evaluate source alignment, extraction optics, prefilter settings, and signs of emittance/acceptance mismatch.
Check energy spread and stability of the injected beam. A beam with high current but poor energy/angle distribution can collapse transmission near higher-resolution settings.
Inspect end-field and contamination clues. Look for changes after venting, source cleaning, rod cleaning, or maintenance; end effects and patch potentials often dominate unexplained behavior.
Trend performance, not just single runs. Mass accuracy, transmission, background, and peak-shape trends reveal thermal and surface-history effects that a single tune snapshot hides.
Write down the convention. Include whether RF amplitudes are zero-to-peak or peak-to-peak and which parameter formulas you used. This prevents future factor-of-two confusion.

Back to tutorials Method development section Troubleshooting section

Diagnostic case: low-resolution transmission is good, but raising selectivity causes an unexpectedly large signal loss. What does the stability plot establish?

It establishes whether the stated ideal operating points lie in the simultaneous region and how the ideal interval changes. It cannot tell you which hardware component failed. Compare the measured peak width and response at documented settings, check voltage conventions, and assess injection matching, energy spread, and recent maintenance. Do not infer rod emissivity or contamination as the cause from the plot alone.

Keep the tasks distinct: changes to transmission or selectivity are tuning questions; aligning reported peak positions is mass-axis calibration. The calibration decision cases show how to test the difference before changing a method.

11. Suggested references and next study topics

If you want to go deeper than this page, the next step is to study full Mathieu stability analysis, Floquet theory for periodic systems, trajectory simulation of finite-length quadrupoles with fringing fields, and ion-optical beam matching (Twiss-parameter style descriptions and emittance transport).

Eckhoff et al., quadrupole mass-filter research: an experimental example connecting electrode geometry, RF drive, measured peak width, and transmission. Its voltage convention must be converted before comparing numerical inputs here.
NIST DLMF: Mathieu stability regions, canonical equation and sign transformations, and SciPy characteristic values.
NIST CODATA 2022: elementary charge and atomic mass constant used in the worked examples.
Ion optics references on beam emittance, brightness, and acceptance matching.
Instrument papers and application notes on prefilters, rod geometry tolerances, and abundance sensitivity.
Vacuum/surface-science references for outgassing, contamination, and temperature-driven recovery behavior.

A useful exercise is to take one quadrupole method in your lab, write down its implicit assumptions (source conditions, beam stability, scan settings, acceptable drift), and map those assumptions onto the categories in this article: field-defining, transport/matching, and thermal/surface-state factors.

Quadrupole Theory: from the electric field to stability and transmission