Soil dissolved organic matter FT-ICR MS is an ultrahigh-resolution mass-spectrometry approach for examining the elemental formulas associated with ions from dissolved organic matter, often abbreviated DOM. If you are trying to read a soil FT-ICR MS molecular formula table, van Krevelen diagram, H/C and O/C ratios, double bond equivalent or electrospray-ionisation spectrum, begin with the measured ion and the sample preparation. A molecular formula assigned to a detected ion is not yet a unique compound structure.
Fourier transform ion cyclotron resonance mass spectrometry distinguishes extraordinarily small mass differences, enabling investigators to propose thousands of elemental compositions in complex organic mixtures. Each formula can be plotted by hydrogen-to-carbon and oxygen-to-carbon ratios or summarised with calculated descriptors. However, not every soil molecule enters the instrument, ionises equally or produces a uniquely identifiable molecular structure. An ion-intensity percentage is not a directly measured percentage of soil organic carbon.
This guide teaches readers how to interpret soil DOM FT-ICR MS from sample collection through extraction, ion production, exact mass, calibration, elemental formula assignment, van Krevelen regions, double bond equivalents, Kendrick series and relative signal comparison. It includes worked calculations, scientific limits, checks and follow-up reading. The numerical exercises are hypothetical, not claims about a laboratory result, and the main purpose is to keep molecular evidence separate from structural guesses or unsupported ecological functions.
The immediate answer: a formula map, not a complete molecular inventory
FT-ICR MS is exceptionally useful for distinguishing ions whose mass-to-charge ratios differ by tiny amounts. A peak may support an elemental composition such as C, H, O, N and S in specified counts when a valid ion form, calibration and formula search are used. Researchers then use formula-derived ratios and indices to describe patterns in a complex DOM extract. The output is a selected ionisable portion of a mixture, not an unbiased count of all chemical compounds in soil.
A statement such as ‘The assigned CHO formulas shifted toward higher O/C’ is meaningful when processing conditions are comparable. ‘The soil has become mostly lignin’ is a stronger structural and source conclusion that exact masses alone rarely support. The diagram groups candidate compositions by chemical properties, but each point may represent several isomers and the entire map depends on ionisation, extraction and formula assignment choices.
Define dissolved organic matter before interpreting it
DOM commonly refers to operationally dissolved material separated from a soil or water matrix by a specified preparation, often involving extraction, centrifugation and sometimes filtration. The threshold separating dissolved from particulate material is operational rather than a sharp molecular boundary. Soil-water extracts and alkaline soil extracts can yield different organic pools even when sampled from the same soil.
Read whether the study examined soil porewater, water-extractable organic matter, dissolved organic carbon from a field lysimeter or an intentionally strong chemical extract. Those samples answer different questions. A formula found in a powerful laboratory extract may not describe a molecule freely moving through field porewater under natural conditions. The sample definition belongs beside every interpretation of molecular diversity.
Carbon concentration and formula richness are different
Dissolved organic carbon, abbreviated DOC, is often measured as a concentration of carbon in a liquid sample by an appropriate carbon analysis. FT-ICR MS produces mass spectral peaks and elemental formula assignments. A sample can have a high DOC concentration but fewer detectable peaks under one ionisation method, while another sample has lower DOC yet displays many well-resolved ion compositions.
A claim that molecular formula richness equals total carbon abundance therefore confuses two separate measurements. Compare DOC values on their own concentration units and the FT-ICR data on the instrument’s signal or formula basis. A valid interpretation can combine both observations, but neither should silently replace the other as a measurement of total carbon.
Why the cyclotron instrument can separate crowded peaks
In an ion cyclotron resonance analyser, charged molecules move in a strong magnetic field. Their motion generates a measurable electrical signal whose frequency is related to mass-to-charge ratio. Signal processing applies a Fourier transform to recover frequency components, after which calibration relates frequencies to mass-to-charge values. A complex mixture may yield many closely spaced spectral peaks.
High resolving power helps distinguish ions that nominal mass instruments would merge. But resolving power depends on measurement settings, signal strength, transient acquisition and instrument design. Merely naming the technique does not guarantee identical performance across studies. A publication should report sufficient acquisition and calibration information to support the precision claimed for its formula assignments.
Resolving power and mass accuracy are not the same
Mass resolving power concerns separating peaks near one another, often expressed as m divided by a relevant peak width. Mass accuracy concerns how close a measured mass is to a reference or calculated exact mass. A spectrum can contain narrow, well-separated peaks yet have a calibration error; a mass may also be accurately placed while some nearby signals remain unresolved.
Suppose two peaks occur near m/z 400 and are separated by 0.002 units. Whether the instrument resolves them depends on peak widths and acquisition performance, not simply whether the software prints six decimal places. A reliable report distinguishes precision, resolution and systematic calibration. Decimal length in an exported spreadsheet is not evidence of physical separation.
What mass-to-charge ratio means
The abbreviation m/z identifies the mass-to-charge ratio of an ion, using conventional instrument units. A molecule may be detected after losing or gaining a proton, forming another adduct, or appearing in a different charge state. The ion detected is therefore not always the neutral compound whose formula a reader imagines from the peak label.
Negative electrospray ionisation commonly detects deprotonated acidic molecules from DOM, although the exact mixture depends on conditions. A calculated neutral molecular formula requires correct consideration of the ion species. If an analyst forgets that the measured ion differs by a proton from the neutral compound, the proposed formula may be wrong even if the peak mass was measured accurately.
Ionisation selects molecules unequally
Electrospray ionisation favours particular chemical classes under a given solvent composition, pH, polarity and source setting. Molecules that ionise poorly may be underrepresented or invisible despite being abundant in soil. Some ions can also suppress others during mixed-sample ionisation, so peak intensities do not scale universally with the concentration of every underlying compound.
A negative-mode FT-ICR spectrum and a positive-mode spectrum may yield different formula families from the same sample. That is not automatically a contradictory result. They interrogate different ionisable subsets. When comparing samples, read whether ion source, solvent, extraction, instrument tuning and data processing stayed comparable.
Preparation can determine which molecules survive to the detector
Soil DOM may contain salts, metals, colloids and mixtures of organic molecules that affect ionisation. Laboratories sometimes desalinate or use solid-phase extraction to obtain an instrument-compatible fraction. These steps can selectively recover certain chemical components and lose others. A sample that passes through a resin is not chemically identical to the complete field solution.
Ask what recovery fraction was reported and whether recoveries differed across treatments. If half the original dissolved carbon failed to pass into the final extract, the molecular diagram characterises the recovered portion. It should not be described as a complete inventory of original porewater without a justification. The material entering the instrument is the immediate analyte.
Blank peaks should not become soil formulas
Laboratory solvents, extraction materials, tubing and instrument background can introduce peaks. A molecular formula present in a blank may appear in soil spectra without coming from the sampled ecosystem. Good processing considers procedural blanks and defined rules for retaining or subtracting signals, while protecting genuine low-intensity sample features from careless overcorrection.
A reader should identify the blank threshold and whether compounds just above that threshold were treated as secure identifications. Background subtraction can change the total formula count and the apparent percentage belonging to different chemical classes. A figure with a thousand points is no more reliable than its blank and signal-quality rules.
Internal and external calibration address different stages
External calibration measures standards separately to link instrument frequency with known masses. Internal calibration can use known reference peaks within the sample spectrum to correct remaining mass error. Each approach has assumptions, including the correctness of assigned calibration peaks and the stability of instrument response across mass range and signal intensity.
Check the stated mass error tolerance and whether error varies with m/z or peak intensity. An acceptable average error does not prove that every assigned formula is correct. Some spectral regions may be crowded or affected by local errors. Formula candidates should satisfy chemical constraints as well as mass agreement.
Parts-per-million error is a relative difference
Mass error in parts per million can be calculated as the measured mass minus theoretical mass, divided by theoretical mass, times one million. For an illustrative theoretical ion m/z of 500.0000 and measured value 500.0004, the error is positive 0.8 ppm. It is a relative mass difference, not a percentage of sample mass or a carbon content.
The same absolute difference of 0.0004 mass units represents a different ppm error at m/z 200 than at m/z 500. Keeping the denominator visible prevents the mistaken interpretation of a fixed mass window as a fixed mass-accuracy threshold. Mass error matters because several formulas may compete within the allowed search tolerance.
Formula assignment is a constrained inference
Software searches plausible combinations of elements whose calculated ion masses fall within a tolerance of each measured peak. A typical restricted search may permit specified ranges of carbon, hydrogen, oxygen, nitrogen and sulfur, with limits on double bond equivalents and elemental ratios. Such rules exclude some chemically implausible candidates but can also exclude real compounds outside the assumptions.
A formula result therefore reflects both instrument evidence and the chosen search space. Two labs analysing the same spectral peak with different element limits may assign different candidate formulas. A useful paper names allowed elements, charge/adduct conventions, calibration tolerance and final selection rules rather than presenting assignments as effortless direct observations.
Isomers are not separated by formula alone
Different organic structures can share the same elemental formula and exact monoisotopic mass. These are isomers; no mass accuracy can distinguish them merely by measuring their identical parent masses. FT-ICR MS formula tables are extraordinarily rich in elemental information, yet structural identification often requires chromatography, ion mobility, tandem fragmentation, NMR or authentic standards.
One assigned formula should not automatically be translated into the name of a specific lignin fragment, carbohydrate, lipid or environmental pollutant. The formula may be consistent with many arrangements. Compound-class labels inferred from formula properties are suggestive descriptors, not unique identifications. A reader who remembers this distinction will avoid many unsupported chemical conclusions.
A van Krevelen diagram maps elemental ratios
A van Krevelen plot commonly places oxygen-to-carbon ratio on the horizontal axis and hydrogen-to-carbon ratio on the vertical axis. Each point represents an assigned molecular formula, sometimes sized or coloured according to signal intensity or a chemical category. Ratios are computed from integer atom counts in the formula, not measured separately by another instrument.
For the illustrative formula C10H12O5, H/C equals 12 divided by 10, or 1.20, and O/C equals 5 divided by 10, or 0.50. Its plot position is therefore x equal to 0.50 and y equal to 1.20. That point does not prove a unique molecule, nor does it measure its original concentration in soil solution.
The meaning of class regions is approximate
Researchers often describe portions of a van Krevelen diagram with words such as carbohydrate-like, lipid-like, protein-like, lignin-like, tannin-like or condensed aromatic-like. Such labels are heuristic regions associated with elemental ratios, not perfect chemical structure detectors. Real molecules in different structural classes can occupy overlapping zones.
The article Mass Difference Matching Unfolds Hidden Molecular Structures of Dissolved Organic Matter illustrates why broad van Krevelen classes overlap and why further structural evidence is valuable. When translating a graph, use ‘formulas within the lignin-like region’ rather than ‘confirmed molecules of lignin’ unless additional chemical identification supports the stronger claim.
What high oxygen-to-carbon can suggest
An increased O/C ratio among detected formulas may indicate a shift toward more oxygen-rich elemental compositions within the selected ionised pool. Oxygen-rich compounds can contain a variety of functional groups, including carboxyl and hydroxyl groups, but the ratio alone does not specify which positions the oxygen atoms occupy or which reactions created them.
A high O/C position is not automatically evidence that the entire soil has become more oxidised in a thermodynamic sense. The instrument may preferentially detect different molecules under changed ionisation conditions. Interpret ratio shifts within the same extraction and detection framework, with appropriate sample controls and independent measures where oxidation chemistry is central.
Hydrogen-to-carbon and unsaturation are related indirectly
A lower H/C ratio often appears among more unsaturated or aromatic candidate compositions, whereas a higher H/C ratio can be consistent with more saturated formulas. But nitrogen, sulfur, oxygen, ion form and molecular architecture complicate the relationship. A plot region is not a direct measurement of aromatic-ring count in every candidate structure.
For example, two formulas with similar H/C can have quite different heteroatom content and possible structures. A good explanation connects H/C to formula composition without overstating molecular identity. If a study calculates double bond equivalent, inspect its assumptions and compare that descriptor alongside elemental ratios rather than treating one axis as an infallible structural label.
Double bond equivalent is an index, not a picture
For a neutral formula containing carbon, hydrogen and nitrogen with halogens accounted for where present, a conventional double-bond-equivalent calculation can be written DBE = 1 + C − H/2 + N/2 − X/2, where X represents monovalent halogens. Oxygen and sulfur do not enter that basic count. The result is a formula-derived count related to total rings and pi-bond unsaturation under assumptions.
For an illustrative neutral C10H12O5 formula with no nitrogen or halogens, DBE equals 1 + 10 − 6 = 5. A DBE of five does not prove a benzene ring, because different combinations of rings and double bonds may yield the same value. Interpretation must also respect ion form, valence and formula validity.
Negative or implausible DBE suggests trouble
If a proposed formula yields an impossible or strongly implausible DBE under the assumed valence conventions, the formula assignment or ion interpretation may be wrong. Search algorithms often filter such candidates, but users should still inspect calculation settings. Nitrogen and halogen counts influence DBE, and errors arise when a charged species is treated as a neutral formula.
A valid DBE is a necessary chemical plausibility check for many assignments, but not proof of a unique structure. Many candidate structures can satisfy the same formula, mass and DBE. Do not use a single descriptor as if it identifies a compound; use it to constrain possibilities and to compare patterns across consistently processed samples.
Kendrick mass helps reveal repeating units
Kendrick mass analysis rescales the mass axis around a chosen repeating chemical unit, commonly the CH2 group. Compounds differing by repeated units can align in a Kendrick mass defect display, revealing homologous series. This is especially useful for complex DOM spectra containing many related formulas separated by approximately the mass of CH2.
The key point is that a visual series indicates a recurring compositional difference, not necessarily a proven biosynthetic pathway. Two peaks aligned in Kendrick space may represent chemically related formulas, but environmental transformation and source attribution need further evidence. A diagram can reveal an orderly pattern in mass differences without reconstructing exactly how the soil produced the molecules.
Calculate a simple formula difference
Suppose two proposed neutral formulas are C10H12O5 and C11H14O5. The difference is C1H2, commonly described as a CH2 increment. This elemental difference can be represented in a Kendrick-series analysis. It does not show that one compound chemically converted into the other during the experiment.
A source claim would require time-resolved or otherwise causally informative evidence. Molecules that differ by CH2 could come from different organisms, plant inputs or transformations. The mass relationship is real at the formula level, but reaction history remains a hypothesis. This is a useful exercise in separating compositional pattern from process inference.
Formula count is conditional on peak picking
A reported count of 4,000 molecular formulas depends on signal-to-noise thresholds, instrument performance, chosen mass range, calibration, blank correction and elemental search settings. A second laboratory may report a different number from the same sample simply because its analytical pipeline differs. The formula count is not the total number of organic molecules in soil.
When comparing richness across treatments, verify that every sample was processed through the same peak-detection and formula rules. Technical replication helps reveal unstable low-intensity formulas. Some investigators restrict comparisons to formulas reproducibly detected across replicate analyses. This may improve robustness but also changes the set of formulas included.
Presence–absence matrices are sensitive to thresholds
A molecular formula may appear in one sample and not another because its ion signal moved just above or below a peak-picking threshold. Treating that difference as complete biological production or destruction can misrepresent the experiment. Peak absence can mean non-detection, not necessarily chemical absence.
When a study presents an UpSet diagram or Venn diagram of shared formulas, identify how detection was defined and whether samples had comparable DOC amounts and analytical sensitivity. Overlap percentages are percentages of retained assignments, not automatically percentages of total carbon molecules. The denominator should be named just as carefully as in a sequencing library.
Peak intensity is not compound concentration
In electrospray mass spectrometry, peak intensity depends on ionisation efficiency, matrix effects, source settings, transmission and detector response. Different molecules can produce very different signals at equal concentrations. Normalising a formula’s intensity to the sum of all detected signals can describe relative instrument response but does not correct every molecule-specific response factor.
Thus a group contributing 30% of summed signal intensity should be described as 30% of the defined spectral signal, not 30% of soil dissolved organic carbon. The latter requires independent quantitative calibration or a justified model. Accurate interpretation preserves the instrument denominator rather than replacing it with an appealing carbon-pool claim.
A compositional percentage can change without carbon changing
Imagine one formula’s signal is 200 arbitrary units in both samples. In sample A, the total retained signal is 1,000, making the formula’s share 20%. In sample B, other ions yield a total of 2,000, making its share 10%. Its measured intensity did not decrease; only its relative share did.
That difference is expected in normalised compositional data. A paper that interprets a percentage decline as evidence of degradation needs additional evidence. Even raw intensities can be affected by ion suppression and preparation variation. Combining consistent sample-normalisation practices with independent DOC measurements can improve the interpretation, but must still acknowledge the limits of ionisation response.
How negative and positive electrospray views differ
Negative electrospray can favour acidic oxygen-containing molecules, while positive electrospray may provide access to different components such as some nitrogen-containing species. Neither view is chemically neutral or exhaustive. A change in ion source polarity can produce an apparently different molecular universe from the same DOM extract.
A comparison across papers is weak if one used negative ESI and the other positive ESI without accounting for source selectivity. The most useful first question is not which plot has more coloured points, but what molecular classes were accessible to each experiment. Later integration can add complementary evidence when the ionisation conditions and compound coverage are clearly described.
Formula classes can be grouped by heteroatoms
One common summary separates CHO, CHON, CHOS and other formula families according to whether nitrogen, sulfur or additional allowed elements are present. This is a compact description of elemental compositions assigned to ions, not proof that all sulfur-bearing formulas are the same functional group or that every nitrogen formula comes from proteins.
Check the elements permitted in formula searching before interpreting the proportions. A dataset that did not allow phosphorus-containing formulas cannot establish phosphorus was absent from the underlying soil DOM. It simply did not assign such compositions under those rules. Group shares also change with ionisation response, making them conditional spectral descriptors rather than direct bulk elemental-analysis results.
Nitrogen and sulfur formula families need source caution
A formula containing N or S is not a uniquely identified biomolecule. Nitrogen may occur in several functional groups, and sulfur may appear in different oxidation or bonding environments. Soil organisms, plants, atmospheric inputs and chemical transformations can contribute to such compositional families. High-resolution parent-ion mass alone does not reveal which exact group is present.
If an article labels every CHON ion a peptide, it has crossed from composition to speculative structure. Likewise, an increase in CHOS features does not by itself prove a named pollutant source or sulfur-cycling pathway. A better discussion reports the change in assigned formulas and identifies targeted fragmentation or reference analysis needed to test a molecular identity.
Aromaticity indices are assumptions in equations
The modified aromaticity index, often abbreviated AImod, is calculated from elemental counts to help identify compositions consistent with certain aromatic structural possibilities under a chosen equation and applicability conditions. It can complement DBE and elemental ratios, yet it remains formula-based rather than a direct instrument image of aromatic rings.
A high AImod does not identify an individual polycyclic aromatic compound, and an inappropriate denominator or impossible formula can make the computed value misleading. When software returns an index, verify the exact equation, accepted elements and handling of edge cases. An index is a reproducible descriptor when the assumptions are documented; it is not a shortcut to compound identification.
Nominal oxidation state of carbon is a model descriptor
Some DOM studies compute a nominal oxidation state of carbon, often written NOSC, from molecular formula composition. This can contribute to hypotheses about energetics and chemical processing. But NOSC calculated from a formula does not directly measure the redox potential of a soil solution, nor does it specify every molecular oxidation state or prove a biochemical degradation rate.
Different compounds sharing a formula may react differently despite sharing a calculated elemental descriptor. If researchers connect NOSC with thermodynamic hypotheses, read the additional assumptions and reaction conditions. A formula table contributes useful constraints, but soil microbial energy yields depend on electron acceptors, concentrations, pH and the reaction network, not formula arithmetic alone.
An ion list may contain isotopic peaks
Natural carbon-13 and other isotopes produce predictable satellite signals near monoisotopic peaks. Software needs a method for recognising isotope patterns and avoiding the interpretation of every isotope peak as an entirely separate compound. Conversely, isotope envelopes may contribute evidence for elemental composition when the signals are measured with enough sensitivity.
If an apparent formula-rich dataset suddenly contains many suspicious near-duplicates, inspect isotope detection rules. High resolving power helps distinguish small differences, but peak annotation remains necessary. An isotope signal is chemically meaningful, yet counting it twice as two separate underlying molecules can inflate formula richness and mislead class comparisons.
Replicate spectra distinguish stable assignments from noise
Running technical replicates helps identify which low-intensity peaks appear consistently under the chosen acquisition and processing settings. Biological replicates answer a different question: whether the soil DOM pattern varies across independent samples, locations or treatment units. A study can have precise technical measurements but very uncertain ecosystem-level generalisation.
When comparing groups, check whether the authors retained formulas appearing in all replicates, some minimum fraction or any sample. Each rule balances sensitivity and reproducibility differently. Formula overlap and intensity shifts depend on this choice. State the filtering rule before interpreting unique formulas as molecules exclusively produced by one treatment.
A worked mass-error diagnostic
A peak assigned a theoretical ion m/z of 400.0000 is measured at 400.0006. The positive difference of 0.0006 divided by 400.0000, multiplied by one million, equals 1.5 ppm. If the study set an absolute tolerance of one ppm, that assignment would fail under the stated threshold. If it allowed two ppm, it might remain eligible but still needs chemical validation.
This simple exercise separates acceptance by a mass-error rule from proof of identity. A candidate within tolerance can share its window with another plausible formula, depending on element limits and resolving power. Do not call a mass match confirmed molecular structure; it is a constraint to be combined with other evidence.
A worked van Krevelen transfer problem
Consider two illustrative formulas, C10H12O5 and C10H16O5. Their O/C values are both 0.50, while H/C changes from 1.20 to 1.60. On a van Krevelen plot, the second point lies above the first, with no horizontal movement. The two candidate compositions differ by four hydrogen atoms, but the plot does not reveal which enzymes or reactions could connect them.
A reader should report the formula and ratio contrast, perhaps discussing a change in inferred saturation if chemically plausible. It would be unsupported to label the second formula a specific lipid solely from its diagram region. The exercise is transferable because it separates measured assignment and derived coordinates from a more demanding claim of chemical identity.
Formula abundance does not measure microbial activity
The presence or disappearance of a candidate DOM composition can reflect chemical supply, sorption, extraction recovery, ionisation and degradation. It does not directly quantify which soil microbes consumed the molecule. FT-ICR MS observes ions derived from the extracted organic mixture, not living cells or enzyme rates. A claim about biological uptake needs additional experimental evidence.
The soil 13C RNA-SIP guide provides a contrasting isotope-tracing method. Even RNA-SIP has indirect-transfer limits, but its tracer question differs fundamentally from unlabelled FT-ICR formula profiling. Combining methods can strengthen hypotheses when each observable is kept separate.
Contrast with untargeted LC-MS metabolomics
Untargeted liquid chromatography–mass spectrometry may separate molecules before detection and use fragmentation or reference databases to support candidate metabolite annotations. Direct-infusion FT-ICR profiling excels at detecting many exact-mass formula families but does not necessarily provide chromatographic separation or unique structure. The two techniques answer related but distinct analytical questions.
Read the soil untargeted LC-MS guide to compare feature tables, retention times and annotation confidence. A named metabolite identified with a reference standard is a different level of evidence from a plausible formula in a complex DOM spectrum. Both are valuable if presented accurately.
Contrast with fluorescence EEM-PARAFAC
Excitation–emission fluorescence measurements analyse light emitted by fluorescent components of DOM under controlled excitation. PARAFAC can help separate recurring patterns in multi-sample fluorescence datasets. FT-ICR MS instead characterises mass spectral ion formulas, including many species that may not be fluorescent. Neither measurement is a complete map of every molecule.
The soil DOM EEM-PARAFAC guide explains component validation and spectral interpretation. A fluorescence component described as humic-like should not be equated numerically with formulas in a lignin-like van Krevelen region. Their labels are drawn from different analytical spaces.
Read an apparent treatment shift critically
Suppose a paper reports a rise in the percentage of CHON formulas after a soil amendment. The first question is whether the proportion counts assigned formulas or sums normalised ion intensities. The second is whether total extracted carbon, ionisation conditions or peak thresholds changed. The third is how independent samples and statistical uncertainty were handled.
A valid observation may concern only relative detectable compositions. It does not automatically demonstrate greater nitrogen incorporation into soil organic carbon or an enhanced nitrogen cycle. Those stronger claims require mass balance, elemental analysis or isotope evidence. The analyst should write the evidence chain in the order it was actually measured, avoiding a dramatic ecological conclusion based on one formula class chart.
Source attribution requires more than formula resemblance
Soil DOM draws material from plants, microorganisms, roots, litter and other processes. Many sources can generate overlapping elemental compositions. A lignin-like diagram region is therefore not a unique fingerprint of fresh plant material, just as a protein-like region is not a direct cell-count measurement. Source inference needs complementary chemistry, sampling logic and sometimes labelled substrates.
A study comparing soils along an environmental gradient may find reproducible formula changes without identifying their sources. That is still useful. Phrase the conclusion as a compositional shift and explain possible processes separately. Overclaiming source identity can make a sound analytical discovery appear weaker when tested by another method.
Why time series are informative but not automatically causal
If a formula declines across successive sampling times, it could have been consumed, transformed, sorbed or lost during sample handling. Simultaneous appearance of another formula differing by one chemical unit does not prove a direct conversion. A time course offers temporal constraints, but complex mixtures contain many reactions and physical processes.
More convincing transformation evidence might involve isotope labelling, mass balance, targeted fragmentation or controlled reaction systems. Readers should ask what alternative explanations remain. It is appropriate to identify a possible molecular transformation network from mass differences, but not to present every network edge as a measured biochemical reaction.
Machine-learning classes are still conditional labels
Some studies apply clustering, ordination or predictive models to thousands of FT-ICR assignments. These methods can summarise complex patterns and sometimes distinguish sample groups. Their apparent success does not eliminate analytical bias, and a predictive class may depend on the training samples, peak filtering and chosen preprocessing.
When a model labels a DOM sample as ‘microbial’ or ‘terrestrial’, ask what independently verified training labels support that category. A classifier is not itself a structural detector. For readers, the safest explanation distinguishes measured ion features, calculated formula descriptors and statistical classifications as three separate layers of analysis.
How to audit a formula table
Begin with m/z, ion mode and mass error. Confirm the elemental limits and isotopic handling used in the formula assignment. Recalculate H/C and O/C for a small selection of formulas, and check DBE against the reported rules. Then identify how duplicates, adducts and low-signal peaks were handled. This makes the table interpretable before any environmental story is added.
For example, if one row lists C10H12O5 but gives H/C equal to 1.8, the ratio does not match the formula; it should be 1.2. A small independent recalculation can reveal a column mistake, formula-to-row mismatch or inconsistent convention. Good technical reading sometimes begins with the simplest possible arithmetic.
Questions to ask about a van Krevelen plot
What exactly does a plotted point represent: one formula, one detected feature or an aggregated family? Were ions matched across samples by identical m/z tolerances? Does colour encode heteroatoms, an inferred chemical class or sample membership? Does point size indicate uncalibrated intensity? Were formulas outside a defined ratio window discarded?
The answers determine what the reader can conclude. A cluster located in a carbohydrate-like region describes formulas with particular elemental ratios, not verified carbohydrates or microbial uptake rates. Reading axes and legend definitions first can prevent much larger errors in later discussion sections.
Frequently asked questions
Does each formula correspond to one chemical compound? No, multiple isomers can share one formula. Can FT-ICR MS measure all soil carbon? No, it observes selected ions from a prepared extract. Is a high peak intensity an absolute concentration? Not without suitable calibration and response considerations. What do H/C and O/C measure? Ratios calculated from formula atom counts. Does a van Krevelen region confirm lignin? No, it is a heuristic grouping.
Does low H/C automatically prove an aromatic ring? No. Does DBE identify a structure? No. Can two labs assign different numbers of formulas? Yes, because instruments and processing rules differ. Is formula richness equivalent to microbial richness? No. Can isotope patterns affect the peak count? Yes. Do coloured plot clusters prove sources? Not on their own.
Verified methodological sources and useful routes
The FT-ICR MS overview in Carbon Research explains cyclotron frequency, resolving power, formula assignment and van Krevelen interpretation. A soil-related DOM formula assignment study gives explicit calibration and formula selection rules. The mass-difference matching study examines why proposed structures overlap across formula regions. These sources document methods and limits, not a unique identity for every detected peak.
For neighbouring interpretation skills, read soil LC-MS metabolomics, soil EEM-PARAFAC, cross-linguistic vocabulary transfer and translation, context and meaning. Each helps a reader protect distinctions between an observation, its label and its interpretation.
Closing principle: preserve each evidentiary layer
FT-ICR MS is a powerful way to examine the elemental-composition complexity of ionisable molecules recovered from soil DOM. It measures mass spectral ions, helps assign chemical formulas and supports derived descriptors such as elemental ratios and unsaturation indices. It does not automatically identify every compound, quantify its true carbon abundance or prove the history of every molecular transformation.
A careful reader moves through sample definition, ion source, exact mass, calibration, formula assignment, ratio calculations and only then environmental interpretation. Keeping those steps visible makes a van Krevelen diagram genuinely informative rather than merely colourful. It also allows the result to be compared honestly with chromatography, fluorescence, isotope experiments and soil carbon measurements.