Ep 55 — Open Spectral Databases: NIST WebBook, SDBS, EPA, and More

Series: Infrared Spectroscopy Encyclopedia: From Principles to Practice
Chapter: Part 5 · Instruments and Tools — Open Data Ecosystem (Ep 46–55 latter part, concluding episode of Part 5)
Target Audience: Technical staff, graduate students, researchers, library/information managers, and anyone interested in open science who wish to improve data processing efficiency
Prerequisites: Ep 19 (Library Search), Ep 52 (Commercial Software), Ep 53–54 (Open Source Tools)
Reading Time: Approximately 45 minutes

Introduction: A "Ten-Thousand-Dollar" Free Gift

A doctoral student, Xiao Zhao, was working on a drug polymorphism study and needed an authoritative reference IR spectrum of caffeine. Her first reaction was to ask her advisor if they could purchase the Sadtler commercial spectral library (approximately 50,000 RMB/year subscription). The advisor smiled and said, "First check NIST WebBook, it's free."

Five minutes later, Xiao Zhao entered "caffeine" in the NIST Chemistry WebBook, opened the IR spectrum—gas phase + condensed phase data sets, downloadable in JCAMP-DX format, directly importable into SpectroChemPy (Ep 53). She also accessed AIST's SDBS from Japan, finding a more detailed caffeine KBr pellet IR spectrum with measurement conditions and peak list.

"Open spectral databases are one of the greatest 'free gifts' in analytical chemistry over the past 30 years. The free policies of NIST WebBook and SDBS have benefited hundreds of thousands of researchers and students worldwide."
— Adapted from Analytical Chemistry editorial comment [1]

This episode concludes the "Instruments and Tools" section and serves as the "data partner" for the open-source tools in Ep 53–54. We will systematically introduce major open spectral databases worldwide: NIST Chemistry WebBook (typically around 16,000+ compounds in public documentation, varying with updates), AIST SDBS, EPA/EMC related FTIR reference spectra (targeting hazardous air pollutants, etc., publicly stated as on the order of hundreds rather than "exactly 189 and permanently fixed"), as well as specialized libraries like PNNL, HITRAN, Caltech, USGS. Finally, we briefly introduce commercial libraries and explain search and comparison techniques.

💡 Positioning: Ep 52 covers commercial software with built-in commercial libraries; Ep 53–54 cover open-source tools with a few built-in libraries; this episode teaches you where to find free, authoritative, and downloadable standard spectra. Combining these three forms a complete data ecosystem.

1. Value and Ecosystem of Open Spectral Databases

1.1 Why Spectral Databases Are Needed

The "fingerprint region" principle of IR spectroscopy determines that by comparing unknown sample spectra with standard spectra, compounds can be identified [1][2]. This is the core logic of library search in Ep 19.

The quality of library search depends on two factors [1][2][3]:

• Algorithm: correlation coefficient, absolute difference, first derivative matching, etc. (see Ep 19);
• Library content: coverage, accuracy, and metadata completeness.

Commercial libraries (Sadtler, Aldrich, Hummel) have broad coverage and high quality, but they are expensive [1]. The rise of open databases makes "free and usable" possible.

1.2 Global Open Spectral Database Ecosystem

The table below provides an overview of major open FTIR spectral databases worldwide (compiled from [1][2][3][4]):

Table 1: Overview of major open FTIR/related spectral databases worldwide (compiled from [1][2][3][4])

Below we introduce the most important ones in detail.

2. NIST Chemistry WebBook: Authoritative "National Database"

2.1 Overview

NIST Chemistry WebBook is a chemical database maintained by the National Institute of Standards and Technology (NIST) in the United States. It is one of the most authoritative and widely used free chemical databases globally [1][2].

• Website: https://webbook.nist.gov/chem…
• Maintainer: NIST, USA
• Data size: IR spectra for 16,000+ compounds
• Data format: JCAMP-DX (.jdx), PERL, ASCII, images
• Access: Completely free, no registration required
• Coverage: Gas + condensed phase

Figure 1: NIST Chemistry WebBook homepage (source: https://webbook.nist.gov/chem…)

2.2 Core Features

① Data Authority [1][2]

• NIST is a national metrology institute in the U.S.; data undergo rigorous review;
• Each data entry includes source (literature reference) and measurement conditions;
• Suitable as a "gold standard" reference.

② Multi-Dimensional Data Integration [1][2]

• A compound page provides:
  • IR spectrum (gas + condensed phase)
  • Raman spectrum
  • UV-Vis spectrum
  • Mass spectrum
  • Thermochemical data (enthalpy of formation, entropy, heat capacity)
  • Ionization energy
  • Gas chromatography retention index

③ JCAMP-DX Format Download [1][2]

• JCAMP-DX is an international standard spectral exchange format (see Ep 52);
• Directly importable into SpectroChemPy, HyperSpy, OMNIC, OPUS;
• Contains complete metadata (compound name, CAS, measurement conditions).

④ Multiple Search Methods [1][2]

• By name (IUPAC, common, trivial);
• By CAS number;
• By molecular formula;
• By molecular weight;
• By structure (SMILES, InChI);
• By physical properties (ion energy, reaction).

2.3 Practical: Retrieving Caffeine IR Spectrum

The following demonstrates the complete workflow from search to download [1][2]:

Step 1 — Access Search

1. Visit https://webbook.nist.gov/chem…;
2. Select "Name" search;
3. Enter "caffeine" → Search.

Step 2 — Select Compound

• The system returns the caffeine entry (CAS 58-08-2);
• Click to enter the detailed page.

Step 3 — Select IR Data

• The left navigation has "Gas phase IR spectrum" and "Condensed phase IR spectrum";
• Condensed phase IR is more suitable for routine analysis (KBr pellet, ATR);
• Click "Condensed phase IR spectrum".

Step 4 — View and Download

• The webpage displays the spectrum visualization;
• Below there are options: "View: Image / JCAMP-DX / PERL / ASCII";
• Select "JCAMP-DX" → browser downloads caffeine.jdx.

Step 5 — Load with Python

# Method 1: Using SpectroChemPy (recommended)
import spectrochempy as scp

ds = scp.read('caffeine.jdx')
ds.plot(title='Caffeine IR (NIST)')

Method 2: using jcamp library

import jcamp
import matplotlib.pyplot as plt

with open('caffeine.jdx', 'r') as f:
data = jcamp.jcamp_read(f)
plt.plot(data['x'], data['y'])
plt.xlabel('Wavenumber (cm⁻¹)')
plt.ylabel('Absorbance')
plt.title('Caffeine IR (NIST)')
plt.show()

2.4 Limitations of NIST

• Mainly gas-phase data (high-temperature vaporization), relatively few condensed-phase data [1];
• Coverage still less than commercial libraries (Sadtler 250,000+);
• Some old data not updated;
• No batch download (only individual compounds);
• Limited search API, not suitable for large-scale automated comparison.

🔗 Extension: Caffeine contains characteristic groups such as C=O (amide carbonyl), aromatic ring, N-CH₃, etc. See ftir.fun functional group database for standard peak positions of various functional groups.

3. AIST SDBS: A "Treasure" for Organic Chemistry

3.1 Overview

AIST SDBS (Spectral Database for Organic Compounds) is maintained by the National Institute of Advanced Industrial Science and Technology (AIST) in Japan, and is a "treasure" of spectral databases for organic compounds [1][3].

• Official website: https://sdbs.db.aist.go.jp/
• Maintainer: AIST Japan
• Data size: 34,000+ organic compounds
• Data types: IR + Raman + ¹H NMR + ¹³C NMR + MS + ESR
• Access: Free (educational/research use only)
• Coverage: Mainly organic compounds

Figure 2: AIST SDBS homepage (Source: https://sdbs.db.aist.go.jp/)

3.2 Key Features

① Integration of multiple spectra [1][3]

• One compound has IR, Raman, NMR (¹H, ¹³C), MS, ESR simultaneously;
• Cross-validation: IR for functional groups + NMR for structure + MS for molecular weight;
• Suitable for teaching (let students see the correspondence of different spectra of the same compound).

② Detailed data [1][3]

• Each IR data includes:
  • Compound structure diagram;
  • Measurement conditions (KBr pellet, liquid film, etc.);
  • Instrument model;
  • Peak list (with assignments);
  • Spectrum visualization.

③ Free for educational use [3]

• Free for personal education/research use;
• Commercial use requires authorization;
• Batch query after account registration.

3.3 Practical: Searching Aspirin

Step 1 — Access Search

1. Visit https://sdbs.db.aist.go.jp/;
2. Click "Search" → "Compound Name";
3. Enter "aspirin" → Search.

Step 2 — Select Compound

• Returns the aspirin entry (CAS 50-78-2);
• Click to enter.

Step 3 — Select IR Data

• The top of the page has spectral type tabs: ¹H NMR | ¹³C NMR | IR | Raman | MS;
• Click "IR".

Step 4 — View and Download

• Display spectrum (image);
• Below is a peak list (with assignments);
• Note: SDBS does not support direct JCAMP-DX download; manual recording or screenshot is required.

Step 5 — Compare with NIST

• NIST usually only provides downloadable files;
• SDBS provides more detailed peak assignments;
• Best to use them complementarily.

3.4 Limitations of SDBS

• Only image + peak list, no JCAMP-DX download [1][3];
• Commercial use requires authorization;
• Some features require registration;
• Data updates are slow.

4. EPA FTIR Pollutant Spectral Library: A Tool for Environmental Monitoring

4.1 Overview

EPA FTIR Pollutant Spectral Library is maintained by the U.S. Environmental Protection Agency (EPA), designed for FTIR monitoring of hazardous air pollutants (HAPs) [1][4].

• Official website: https://www3.epa.gov/ttn/emc/…
• Maintainer: U.S. EPA
• Data size: Reference spectra for hazardous air pollutants (HAPs), etc. Public descriptions often state the order of hundreds, and it changes with updates; the number of HAPs listed in CAA Title III is not necessarily equal to the number of spectra in the library.
• Data type: Gas-phase FTIR
• Access: Completely free, no registration required
• Coverage: HAPs of concern in environmental monitoring

Figure 3: EPA FTIR Reference Spectral Library homepage (Source: https://www3.epa.gov/ttn/emc/…)

4.2 Key Features

① Specifically for atmospheric monitoring [1][4]

• Entries mostly correspond to volatile/semi-volatile species of concern in environmental monitoring (e.g., benzene, toluene, formaldehyde), with the specific list subject to the current EPA page;
• Gas-phase, high-temperature measurements (consistent with flue gas conditions);
• Data can be directly used for EPA Method 320 (FTIR measurement of stationary source pollutants).

② High data quality [4]

• Measured uniformly by EPA laboratory;
• Standardized conditions (path length, temperature, pressure);
• Includes absorption cross-sections (for quantification).

③ Suitable for open-path monitoring [4]

• Standard spectra are essential for open-path FTIR (OP-FTIR) monitoring of atmospheric pollutants;
• The EPA library is an industry standard.

4.3 Practical: Searching Formaldehyde

Step 1 — Visit Homepage

1. Open https://www3.epa.gov/ttn/emc/…;
2. The list is sorted alphabetically.

Step 2 — Find Target Compound

• Find "Formaldehyde" (CAS 50-00-0);
• Click the link.

Step 3 — Download

• Provides multiple formats (JCAMP-DX, ASCII, CSV);
• Select JCAMP-DX to download.

Step 4 — Load with Python

import spectrochempy as scp
formaldehyde = scp.read('formaldehyde.jdx')
formaldehyde.plot(title='Formaldehyde Gas-Phase IR (EPA)')

🔗 Extension: Formaldehyde contains C=O aldehyde group, characteristic peak at 1720–1740 cm⁻¹. See ftir.fun carbonyl functional group page for the difference between aldehyde and ketone groups.

5. Other Specialized Open Databases

5.1 PNNL FTIR Library: Gas-Phase Organic Compounds

PNNL (Pacific Northwest National Laboratory) FTIR Library is maintained by the Pacific Northwest National Laboratory in the U.S., focusing on gas-phase organic compounds [1][5].

• Size: Gas-phase FTIR of 480 organic compounds;
• Features:
  • High-quality gas-phase data (25 °C standard conditions);
  • Includes absorption cross-sections (for quantification);
  • Suitable for remote sensing, atmospheric chemistry;
• Access: Free, requires registration (https://secure2.pnl.gov/nsd/f…);
• Format: JCAMP-DX, ASCII.

5.2 HITRAN: High-Resolution Gas-Phase Small Molecules

HITRAN (High-Resolution Transmission) database is the "gold standard" for high-resolution infrared spectra of gas-phase small molecules [1][6].

• Official website: https://hitran.org/

• Maintainer: Harvard-Smithsonian Center for Astrophysics;

• Scale: 47 small molecules (CO₂, H₂O, CH₄, O₃, etc.);

• Features:
  • High resolution (0.0001 cm⁻¹ level);
  • Suitable for atmospheric remote sensing, astrophysics;
  • Contains parameters such as line intensity, half-width, energy levels;
• Access: Free registration;
• Format: HITRAN format, HAPI interface.

⚠️ Note: HITRAN is a "line parameter" database, not direct spectra. The HAPI (HITRAN Python interface) must be used to simulate spectra:

from hapi import *
db_begin('data')
fetch('CO2', 2, 1, 600, 7000)
nu, coef = absorptionCoefficient_Lorentz(SourceTables='CO2', OmegaStep=0.01)

5.3 Caltech Mineral Spectroscopy Server

Caltech Mineral Spectroscopy Server is maintained by the California Institute of Technology, focusing on mineral reflectance spectra [1][7].

• Website: http://minerals.gps.caltech.e…
• Scale: Hundreds of minerals;
• Features:
  • Reflectance IR + NIR;
  • Suitable for geology, planetary science;
  • High-quality data;
• Access: Free web.

5.4 USGS Spectral Library

USGS Spectral Library is maintained by the U.S. Geological Survey [1][8].

• Website: https://speclab.cr.usgs.gov/s…
• Scale: ~700 spectra of minerals + plants + man-made materials;
• Wavelength range: 0.4–2.5 μm (visible + near infrared);
• Features:
  • Suitable for remote sensing, mineral exploration;
  • Contains detailed sampling information;
• Access: Free.

5.5 RRUFF Mineral Raman Database

RRUFF project maintained by the University of Arizona [1][9].

• Website: https://rruff.info/
• Scale: Thousands of minerals;
• Data types: Raman primarily, some IR, XRD;
• Features: De facto standard for mineral identification;
• Access: Free.

6. Introduction to Commercial Spectral Libraries

Although commercial libraries are paid, their coverage and quality are still irreplaceable by open libraries [1][10]. A brief introduction follows:

6.1 SpectraBase (Bio-Rad / Wiley)

• Website: https://spectrabase.com/
• Maintained by: Bio-Rad (Sadtler library) + Wiley;
• Scale: Hundreds of thousands of IR, Raman, NMR, MS, UV-Vis spectra;
• Features:
  • Largest commercial spectral library globally;
  • High quality, reviewed;
  • Includes sub-libraries for commercial products, polymers, pharmaceuticals, etc.;
• Access: Free web query (basic functions), paid download / API.

6.2 NICODOM IR Library

• Website: https://www.ir-spectra.com/
• Maintained by: NICODOM (Czech Republic);
• Scale: Tens of thousands of IR spectra;
• Features:
  • Focus on polymers, food, pharmaceuticals;
  • Provides both ATR and transmission data;
  • Suitable for specific industries;
• Access: Paid (one-time purchase + upgrades).

6.3 Sigma-Aldrich / Merck IR Library

• Maintained by: Sigma-Aldrich (Merck group);
• Features:
  • One-to-one correspondence with reagent catalog;
  • Suitable for "what reagent you buy, what spectrum you check";
• Access: Included with reagent purchase or purchased separately.

6.4 Hummel Polymer Library

• Maintained by: Hummel (now part of Sadtler / Bio-Rad);
• Features:
  • Gold standard for polymers and additives;
  • Essential for industrial analysis;
• Access: Paid.

💡 When to buy a commercial library [1][10]:

• QC production environment (requires stability, compliance, traceability);
• Professional fields (polymers, pharmaceuticals, drugs);
• Library searching is a core analysis method;
• Sufficient budget (annual subscription 50,000–200,000 RMB).
  Otherwise, open libraries + self-built libraries can meet most needs.

7. How to Efficiently Use Spectral Libraries for Searching and Comparison

7.1 Review of Search Algorithms

Mainstream library search algorithms (see Ep 19 for details) [1][2]:

Table 2: Comparison of mainstream library search algorithms (synthesized from [1][2])

7.2 Meaning of HQI Value

HQI (Hit Quality Index) is the "match quality" index returned by library search, 0–1 or 0–100 [1][2]:

• HQI > 0.95 (95): Excellent match;
• HQI 0.90–0.95: Good match;
• HQI 0.80–0.90: Possible match, requires human confirmation;
• HQI < 0.80: Low match, may be a different compound or mixture.

Important principles [1][2]:

• HQI is not a "probability", it is a similarity measure;
• High HQI does not necessarily mean it is that compound (homologues, isomers also have high HQI);
• Low HQI does not necessarily mean it is not (baseline, sample preparation differences can cause low HQI);
• Must be combined with manual spectral interpretation.

7.3 Efficient Search Workflow

Standard workflow [1][2][3]:

Unknown sample spectrum
   │
   ▼
1. Preprocessing (baseline, smoothing, normalization)
   │
   ▼
2. Search NIST WebBook for candidates (gas phase + condensed phase)
   │
   ▼
3. Search SDBS for candidates (with peak assignments)
   │
   ▼
4. Multiple candidates? Rank by HQI
   │
   ├─ HQI > 0.95 → High confidence identification
   │
   ├─ HQI 0.85–0.95 → Re-evaluate with first derivative method
   │
   └─ HQI < 0.85 → Possibly mixture, need MCR-ALS decomposition
   │
   ▼
5. Manual verification: key peak positions, shape, relative intensity
   │
   ▼
6. Cross-validation: IR + NMR + MS (if available)

7.4 Library Search for Mixtures

Searching pure compound libraries is relatively simple; mixtures are the challenge [1][2]:

Method 1: Spectral Subtraction [1]

• Assume the main component is A, take the standard spectrum of A from NIST;
• Subtract the scaled spectrum of A from the unknown spectrum and examine the residual;
• Identify the second component B in the residual;
• Repeat until the residual is close to zero.

import spectrochempy as scp

unknown = scp.read('unknown.spa')
component_a = scp.read('a_from_nist.jdx')

# Scaling factor (minimize residual)
from scipy.optimize import minimize_scalar
res = minimize_scalar(lambda scale: np.sum((unknown.data - scale * component_a.data)**2))
scale_a = res.x

residual = unknown - scale_a * component_a
residual.plot(title='Residual after subtracting A')

Method 2: MCR-ALS Decomposition [1][3]

• Suitable for complex mixtures with 3+ components;

• Built-in in SpectroChemPy (see Ep 53);

• Perform library search after decomposing pure component spectra.

Method 3: Self-built library [1]

• Build a small library for compounds commonly measured in the lab;
• More accurate than generic commercial libraries (same instrument, same sample preparation);
• Common practice in industrial QC laboratories.

7.5 ftir.fun as a Chinese-friendly tool supplement

🔗 Extension: ftir.fun is a Chinese-friendly online spectroscopy tool providing:

• Spectrum upload and multi-format parsing: JCAMP-DX, CSV and other common formats;
• Peak identification and functional group query: automatic peak identification, linked to functional group database;
• Spectrum search: linked to open databases for retrieval;
• Visualization and comparison: directly view and compare spectra on the web;
• Cross-device access: works on mobile, tablet, and PC.

Workflow of ftir.fun with open databases [11]:

1. Download JCAMP-DX standard spectra from NIST WebBook;
2. Upload to ftir.fun for comparison with unknown samples;
3. ftir.fun automatically identifies unknown sample peaks and associated functional groups;
4. Export comparison results as a report.

This collaborative model of "NIST providing data + ftir.fun providing tools" enables Chinese users to efficiently utilize open spectral data [11].

8. Practice of Building a Self-built Spectral Library

Open databases cover common compounds, but specific samples frequently measured in the lab (self-made catalysts, specific polymer formulations, internal QC standards) are not included [1][2]. In such cases, self-built libraries are important.

8.1 Elements of a Self-built Library

A high-quality self-built library requires [1][2]:

Table 3: Essential metadata for self-built libraries (compiled from [1][2])

8.2 Storage Format of Self-built Library

Recommended format: JCAMP-DX [1][2]

• International standard;
• Contains metadata;
• Cross-software compatibility.

Example of a self-built library (JCAMP-DX header) [1]:

##TITLE=Catalyst_Pt_Al2O3_Fresh
##JCAMP-DX=5.01
##DATA TYPE=INFRARED SPECTRUM
##ORIGIN=My Lab - Catalyst Group
##OWNER=Bob
##CAS NAME=Pt/Al2O3 catalyst (fresh)
##CAS REGISTRY NO=NA (mixture)
##NPOINTS=1769
##XUNITS=1/CM
##YUNITS=ABSORBANCE
##XFACTOR=1.0
##YFACTOR=1.0
##FIRSTX=4000
##LASTX=900
##FIRSTY=0.012
##END HEADER
4000 0.012
3999 0.013
...

8.3 Management of Self-built Library

• Use SpectroChemPy's NDDataset (Ep 53) to manage metadata [2];
• Use version control (Git) to track changes;
• Regular backups (cloud storage + local);
• Document (which compounds, when added, who is responsible).

8.4 Integration with Open Libraries

Combining self-built libraries with open libraries forms the laboratory's "complete spectral resource" [1][2]:

import spectrochempy as scp

# Load open library (downloaded from NIST)
nist_compounds = [scp.read(f) for f in nist_files]

# Load self-built library
inhouse_compounds = [scp.read(f) for f in inhouse_files]

# Merge
all_library = nist_compounds + inhouse_compounds

# Search against unknown sample
unknown = scp.read('unknown.spa')

best_match = None
best_hqi = 0
for ref in all_library:
    hqi = scp.cosine_similarity(unknown, ref)
    if hqi > best_hqi:
        best_hqi = hqi
        best_match = ref

print(f"Best match: {best_match.name}, HQI={best_hqi:.3f}")

9. Open Data and Open Science

9.1 FAIR Data Principles

Modern scientific data management follows the FAIR principles [1][12]:

• Findable;
• Accessible;
• Interoperable;
• Reusable.

Open databases such as NIST WebBook and SDBS are exemplars of FAIR principles [1]:

• Findable: standard URLs, DOIs, CAS number search;
• Accessible: free, no registration required (NIST);
• Interoperable: JCAMP-DX standard format;
• Reusable: clear usage licenses, source attribution.

9.2 Citation and Acknowledgment

When using open databases, proper citation is a scientific norm [1][12]:

• NIST WebBook: Cite Mallard W G, et al. NIST Chemistry WebBook, NIST Standard Reference Database 69. https://webbook.nist.gov/chem…
• SDBS: Cite SDBSWeb: https://sdbs.db.aist.go.jp/ (National Institute of Advanced Industrial Science and Technology, date of access)
• EPA FTIR: Cite EPA Method 320 document and database URL;
• HITRAN: Cite Gordon I E, et al. J Quant Spectrosc Radiat Transfer 2022.

9.3 Contributing Back to the Open Ecosystem

Laboratories can contribute their own high-quality spectra back to the open ecosystem [1][12]:

• Publish spectra alongside articles (e.g., Figshare, Zenodo);
• Upload to open databases (some accept external contributions);
• Use open-source tools to process data and open-source the code;
• Write blog posts in Chinese and English to share experiences.

10. Pitfalls in Database Use

10.1 Pitfall 1: Confusing Gas Phase with Condensed Phase

Scenario: Using NIST gas-phase benzene spectrum to compare with KBr pellet benzene sample, HQI only 0.7 [1].

Reason: IR spectra differ greatly between gas and condensed phases (peak shifts, peak shape changes).

Solution:

• For condensed-phase samples, use NIST condensed-phase spectra;
• For gas-phase samples, use NIST/EPA gas-phase spectra;
• Do not mix.

10.2 Pitfall 2: Inconsistent Sample Preparation Methods

Scenario: Comparing self-built ATR library sample with NIST KBr pellet spectrum, the relative intensity of the 1740 cm⁻¹ peak differs by 30% [1].

Reason: ATR and transmission have different peak intensity rules (see Ep 18 ATR correction).

Solution:

• Perform ATR correction before comparison;
• Or build a library using the same sample preparation method;
• Use SpectroChemPy's atrcorrect function.

10.3 Pitfall 3: Over-reliance on HQI

Scenario: Library search HQI = 0.92, identified as A, but actually it was a homolog B of A [1].

Solution:

• Homologs and isomers all give high HQI;

• Must manually verify key peaks;

• Cross-validate with NMR, MS.

10.4 Pitfall 4: Ignoring Metadata

Scenario: Retrieved a spectrum from SDBS without noting measurement conditions, used it to compare with room-temperature samples, resulting in a 5 cm⁻¹ peak shift [1].

Countermeasures:

• Always check measurement conditions when retrieving;
• Note temperature, phase, sample preparation method;
• Avoid using data inconsistent with sample conditions.

10.5 Pitfall 5: Copyright Risks of Commercial Libraries

Scenario: A company uploaded a purchased Sadtler commercial library to internal cloud sharing, which was audited and found in breach of contract [10].

Countermeasures:

• Commercial library licenses typically prohibit redistribution;
• Internal use must purchase licenses according to quantity;
• Self-built libraries can be shared freely;
• Open libraries can be shared freely (with proper citation).

Summary

Questions

1. You need to identify an unknown organic compound (white crystal, melting point 135–136 °C). Design a complete library search workflow using NIST WebBook + SDBS, describing what to do and what to look for at each step. If the search returns candidate A with HQI 0.85 and candidate B with HQI 0.82, how do you decide which one it is?

2. Compare NIST WebBook and AIST SDBS in terms of data format, coverage, and metadata completeness. If you are identifying an organic synthesis product, which one would you prioritize and why?

3. You are responsible for an environmental monitoring project using open-path FTIR (OP-FTIR) to monitor formaldehyde, benzene, and acetone in stack gas. Explain which open database should be used to obtain standard spectra and how to use these data to build a quantitative model. Hint: Refer to ftir.fun carbonyl functional group page.

4. Your lab has 200 self-made polymer formulation samples, each with an FTIR spectrum. Design a self-built library plan including data format, metadata fields, storage method, and integration with the NIST open library.

5. Discuss the application of FAIR data principles in spectral databases. How do NIST WebBook and SDBS perform across the four FAIR dimensions? If you were a lab manager, what policies would you implement to make your lab's data FAIR-compliant?

References

[1] mallard W G, Linstrom P J, eds. NIST Chemistry WebBook, NIST Standard Reference Database Number 69. National Institute of Standards and Technology, Gaithersburg MD, 2024.
https://webbook.nist.gov/chem…

[2] Linstrom P J, Mallard W G. "The NIST Chemistry WebBook: A Chemical Data Resource on the Internet." Journal of Chemical & Engineering Data, 2001, 46(5): 1059–1063. DOI:10.1021/je000960e.

[3] AIST. "Spectral Database for Organic Compounds (SDBS)." National Institute of Advanced Industrial Science and Technology, Japan.
https://sdbs.db.aist.go.jp/

[4] US EPA. "FTIR Open Path Measurements Method 320 & Reference Spectra." EPA Technology Transfer Network Emission Measurement Center.
https://www3.epa.gov/ttn/emc/…

[5] Sharpe S W, et al. "Gas-Phase Databases for Quantitative Infrared Spectroscopy." Applied Spectroscopy, 2004, 58(12): 1452–1461. DOI:10.1366/0003702042641281.
PNNL FTIR Library: https://secure2.pnl.gov/nsd/f…

[6] Gordon I E, et al. "The HITRAN2020 Molecular Spectroscopic Database." Journal of Quantitative Spectroscopy & Radiative Transfer, 2022, 277: 107949. DOI:10.1016/j.jqsrt.2021.107949.
https://hitran.org/

[7] Calvin W M, King T V V, Clark R N. "Caltech Mineral Spectroscopy Server." California Institute of Technology.
http://minerals.gps.caltech.e…

[8] Clark R N, et al. "USGS Digital Spectral Library 06." U.S. Geological Survey Open File Report 03-395, 2007.
https://speclab.cr.usgs.gov/s…

[9] Lafuente B, Downs R T, Yang H, Stone N. "The RRUFF Project: An Integrated Study of Minerals." IMA General Meeting, 2015.
https://rruff.info/

[10] Bio-Rad. "SpectraBase: Spectral Database." KnowItAll Informatics System.
https://spectrabase.com/

[11] ftir.fun. "Online Infrared Spectroscopy Tool."
https://ftir.fun

[12] Wilkinson M D, et al. "The FAIR Guiding Principles for Scientific Data Management and Stewardship." Scientific Data, 2016, 3: 160018. DOI:10.1038/sdata.2016.18.

[13] IUPAC. "JCAMP-DX: Standard for Infrared Spectral Data Exchange." Pure and Applied Chemistry, 1995, 67(12): 1955–1984. DOI:10.1351/pac199567121955.

[14] Travert A, et al. "SpectroChemPy: A Python Framework for Processing Spectral Data." Journal of Open Source Software, 2022, 7(71): 3904. (Ep 53)

Preview of Next Episode: Ep 56 — Using ftirfun: Tips and Practical Cases
This episode concludes the "Instruments and Tools" section. Starting from the next episode, we enter the "Practice and Expansion" section. In Ep 56, we will systematically introduce ftir.fun, a Chinese-friendly online FTIR tool: from spectral data upload and multi-format parsing (JCAMP-DX, SPA, etc.), peak identification and management, spectral search and database query, visualization and comparative analysis, to integration with open databases such as NIST WebBook, and finally a practical case — from uploading an unknown sample spectrum to completing the full identification process. This serves as the "Chinese-friendly gateway" to the open-source tools and open databases introduced in Ep 53–55.

This article is licensed under CC BY-NC-SA 4.0. Images are from the public domain or online resources with credited sources, copyrights belong to their respective owners.