checkHNMR : an open-source tool for checking the consistency of 1 H NMR descriptions before publication

A 1 H NMR description in the Supporting Information is usually the only place where the values of a spectrum are stated. Such descriptions serve as models for students and researchers and as a reference for reviewers and editors, and they are used by databases, prediction programs and language models, so an error that passes peer review spreads. Here I describe checkHNMR , an open-source, rule-based tool that checks before publication whether the values of a 1 H NMR description of a small organic molecule agree with one another and with the proposed structure. It applies the items of a published checklist that can be checked from the text and the structure alone: the number of protons, a partner for every coupling constant, multiplicities consistent with the number of constants and of possible neighbors, and first-order notation only where it is valid. Each alert states its reason, and the decision is left to the author, the referee or the editor. The tool runs in a web browser at checkhnmr.org, where the data are not sent anywhere, and as a local version that reads Supporting Information files as published; the two versions give the same results. Each report carries a verification code. A reviewer can repeat a check by entering the entries saved with the report into the same version of the tool and comparing the code with the one in the report. The paper describes this procedure step by step, with an example. In stress tests with synthetic descriptions, the tool gave no false findings on correct descriptions, found most of the introduced errors, and misread 0.03% of descriptions written in the styles of published ones; these tests do not measure accuracy on new articles. A description that passes every check can still be wrong or invented: only the raw data (the free induction decay, FID) can show what was measured.

Authors

Publication Details

Published
2026-11-13
DOI
https://doi.org/10.3762/bxiv.2026.45.v1
Primary Topic
Molecular spectroscopy and chirality
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

checkHNMR : an open-source tool for checking the consistency of 1 H NMR descriptions before publication

Artur V. Eshtukov
Molecular spectroscopy and chirality
preprint

checkHNMR : an open-source tool for checking the consistency of 1 H NMR descriptions before publication

Artur V. Eshtukov
preprint en

Abstract

A 1 H NMR description in the Supporting Information is usually the only place where the values of a spectrum are stated. Such descriptions serve as models for students and researchers and as a reference for reviewers and editors, and they are used by databases, prediction programs and language models, so an error that passes peer review spreads. Here I describe checkHNMR , an open-source, rule-based tool that checks before publication whether the values of a 1 H NMR description of a small organic molecule agree with one another and with the proposed structure. It applies the items of a published checklist that can be checked from the text and the structure alone: the number of protons, a partner for every coupling constant, multiplicities consistent with the number of constants and of possible neighbors, and first-order notation only where it is valid. Each alert states its reason, and the decision is left to the author, the referee or the editor. The tool runs in a web browser at checkhnmr.org, where the data are not sent anywhere, and as a local version that reads Supporting Information files as published; the two versions give the same results. Each report carries a verification code. A reviewer can repeat a check by entering the entries saved with the report into the same version of the tool and comparing the code with the one in the report. The paper describes this procedure step by step, with an example. In stress tests with synthetic descriptions, the tool gave no false findings on correct descriptions, found most of the introduced errors, and misread 0.03% of descriptions written in the styles of published ones; these tests do not measure accuracy on new articles. A description that passes every check can still be wrong or invented: only the raw data (the free induction decay, FID) can show what was measured.

Yale University, University of Minnesota, University of Denver, University of Bristol, Russian Academy of Sciences
Molecular spectroscopy and chirality
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon