mass-spec.tips

Learn by changing one thing

Build your intuition.
Test your prediction.

Three small models with guided experiments. Predict a result, change one thing, then explain what the result does and does not establish.

Pairs with Lesson 2 · One molecule can make many peaks

Experiment 1: change the charge, keep the molecule

  1. Reset the ion explorer. Record the m/z and isotope spacing for [M+H]+
  2. Predict both values for [M+2H]2+, then change charge magnitude to 2. Leave neutral mass and carbon count unchanged
  3. Switch the vertical axis from probability to relative signal. Did peak positions change? What does 100% now mean?
  4. Try 100 carbons, then lower the displayed ¹³C-count limit. Check the omitted probability before interpreting the tallest displayed peak
Worked comparison, available without the model

Using the model fixture M = 180.06338810418 Da and proton mass 1.0072764665789 Da, [M+H]+ is at m/z 181.070665. [M+2H]2+ is at 91.038971. The charge-2 value is not simply half the charge-1 value, because the proposed ion also contains another proton.

One ¹³C substitution gives a spacing of approximately 1.003355 at |z| = 1 and 0.501677 at |z| = 2. Changing vertical normalization changes neither position. Probability percentages describe a carbon-only distribution; relative 100% marks its tallest full-model peak. Neither is a concentration measurement.

At 100 carbons with the illustrative probability 0.0107, the most probable count is one ¹³C atom. The monoisotopic member need not be the largest. Truncating the plot never makes omitted probability disappear.

Limit: proton addition/removal and a carbon-only isotope model. Other elements, adduct chemistry and experimental response are omitted. For a complete worked ion-family investigation, use Lesson 2.

Pairs with Resolution is not mass accuracy

Experiment 2: change width, then move the center

  1. Reset. The first reference is m/z 500, resolving power is 50,000 under the FWHM criterion, and offset is +5 ppm
  2. Increase resolving power to 200,000. Predict the width before reading it. Does the first center move?
  3. Set the offset to zero. Now which quantity changes and which stays fixed?
  4. Use the “One FWHM apart” case, then reduce the second component height. Distinguish the known model components from what a real overlapping signal would reveal
Worked comparison, available without the model

FWHM = 500/50,000 = 0.0100 in m/z. At resolving power 200,000 it is 0.0025. A +5 ppm offset at m/z 500 is +0.0025, so the first component center stays at 500.0025 while its width changes.

Removing the offset moves that center to 500.0000. It does not change the width or reference separation. Two equal Gaussian components one FWHM apart are not baseline separated: at their midpoint each still contributes half its individual height.

The model specifies both component centers before constructing the sum. Real peak picking does not have that privilege. Overlap and processing can change a reported center even if the underlying component positions are unchanged.

Limit: equal-width Gaussian components, one additive m/z shift, and optional synthetic noise. The model does not give a pass/fail resolution or identity verdict. Read the full measurement-quality lesson for ppm, centroid and repeatability examples.

Pairs with Lesson 4 · How ions move through the instrument

Experiment 3: compare mass and charge

  1. Start with 500 Da at |z| = 1 and 1000 Da at |z| = 2. Explain why both have m/z 500 before calculating their ideal flight times
  2. Use the “Four times the m/z” case. Does time increase fourfold or twofold?
  3. Return to the initial case and increase accelerating voltage from 1000 to 4000 V. Compare the numerical times; the chart automatically rescales
Worked comparison, available without the model

For equal path and accelerating voltage, tB/tA = √[(mB/|zB|)/(mA/|zA|)]. Equal m/z gives equal ideal drift time even when the ion masses differ.

With a 1 m field-free path and 1000 V acceleration from rest, the 500 Da, charge-1 ion takes about 50.90 µs. A charge-1 ion of 2000 Da takes about 101.81 µs. Quadrupling the voltage halves both times; the ratio between them remains unchanged.

Input mass is the mass of the ion itself. Adding a proton mass again would change the stated model. Real instruments include extraction, focusing, packet spread and electronic timing that this model omits.