interpret-ir-spectrum
О программе
Этот навык Claude анализирует инфракрасные спектры для идентификации функциональных групп путем изучения диагностических и пальцевых областей. Он обнаруживает эффекты водородных связей и составляет перечень функциональных групп с указанием уровней достоверности. Используйте его для первичного скрининга соединений или для подтверждения структурных особенностей в неизвестных образцах.
Быстрая установка
Claude Code
Рекомендуетсяnpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/interpret-ir-spectrumСкопируйте и вставьте эту команду в Claude Code для установки этого навыка
Документация
Interpret IR Spectrum
Read IR absorption spectra. Identify functional groups. Check hydrogen bonding. Compile inventory of structural features in sample.
When Use
- Identify functional groups in unknown compound as first screen
- Confirm presence or absence of specific functional groups (e.g., verify reaction turned alcohol to ketone)
- Monitor reaction progress by tracking appearance/disappearance of characteristic absorptions
- Distinguish between similar compounds differing in functional group content
- Complement NMR + mass spec data with vibrational info
Inputs
- Required: IR spectrum data (absorption frequencies in cm-1 with intensities, as %Transmittance or Absorbance plot)
- Required: Sample prep method (KBr pellet, ATR, Nujol mull, thin film, solution cell)
- Optional: Molecular formula or expected compound class
- Optional: Known structural fragments from other spectroscopic data
- Optional: Instrument parameters (resolution, scan range, detector type)
Steps
Step 1: Check Spectrum Quality and Format
Verify spectrum suitable for interpretation before analyzing peaks:
- Check y-axis format: %Transmittance (%T, peaks point down) or Absorbance (A, peaks point up)? All analysis assumes consistent convention
- Verify wavenumber range: Covers at least 4000-400 cm-1 for standard mid-IR? Note any truncation
- Assess baseline: Good baseline relatively flat, near 100%T (or 0 Absorbance) in regions with no absorption. Sloping or noisy baselines cut reliability
- Check resolution: Adjacent peaks closer than instrumental resolution cannot be distinguished. Typical FTIR resolution = 4 cm-1
- Identify prep artifacts: KBr pellets may show broad O-H band from moisture (~3400 cm-1). Nujol mulls obscure C-H stretches. ATR spectra show intensity distortion at low wavenumbers. Note any artifacts limiting interpretation
Got: Spectrum confirmed suitable. Format, range, artifacts documented.
If fail: Severe baseline problems, saturation (flat-bottomed peaks from too-concentrated samples), or prep artifacts obscuring critical regions? Note limitation. Flag affected spectral regions unreliable.
Step 2: Scan Diagnostic Region (4000-1500 cm-1)
Systematic analysis of high-frequency region where most functional groups make characteristic absorptions:
- O-H stretches (3200-3600 cm-1): Look for broad absorptions. Sharp peak near 3600 cm-1 = free O-H. Broad band centered at 3200-3400 cm-1 = hydrogen-bonded O-H (alcohols, carboxylic acids, water)
- N-H stretches (3300-3500 cm-1): Primary amines show two peaks (symmetric + asymmetric). Secondary amines show one. Typically sharper than O-H bands
- C-H stretches (2800-3300 cm-1):
| Frequency (cm-1) | Assignment |
|---|---|
| 3300 | sp C-H (alkyne, sharp) |
| 3000--3100 | sp2 C-H (aromatic, vinyl) |
| 2850--3000 | sp3 C-H (alkyl, multiple peaks) |
| 2700--2850 | Aldehyde C-H (two peaks from Fermi resonance) |
- Triple-bond region (2000-2300 cm-1):
| Frequency (cm-1) | Assignment | Notes |
|---|---|---|
| 2100--2260 | C triple-bond C | Weak or absent if symmetric |
| 2200--2260 | C triple-bond N | Medium to strong |
| ~2350 | CO2 | Atmospheric artifact, disregard |
- Carbonyl region (1650-1800 cm-1) — most diagnostic single region in IR:
| Frequency (cm-1) | Assignment |
|---|---|
| 1800--1830, 1740--1770 | Acid anhydride (two C=O stretches) |
| 1770--1780 | Acid chloride |
| 1735--1750 | Ester |
| 1700--1725 | Carboxylic acid |
| 1705--1720 | Aldehyde |
| 1705--1720 | Ketone |
| 1680--1700 | Conjugated ketone / alpha-beta unsaturated |
| 1630--1690 | Amide (amide I band) |
- C=C and C=N stretches (1600-1680 cm-1): Alkene C=C = 1620-1680 cm-1 (weak to medium). Aromatic C=C = multiple peaks near 1450-1600 cm-1. C=N (imine) = 1620-1660 cm-1
Got: All absorptions in diagnostic region identified, with functional group assignments and confidence levels (strong, tentative, absent).
If fail: Carbonyl region obscured (e.g., water absorption in KBr, atmospheric CO2)? Note gap. Expected functional group absorption absent? Confirm with second prep method before concluding truly absent.
Step 3: Analyze Fingerprint Region (1500-400 cm-1)
Examine lower-frequency region for confirmatory and structural detail:
- C-O stretches (1000-1300 cm-1): Ethers, esters, alcohols, carboxylic acids make strong C-O stretching. Esters show characteristic strong band near 1000-1100 cm-1 in addition to carbonyl
- C-N stretches (1000-1250 cm-1): Amines and amides. Overlap with C-O makes assignment tentative without other evidence
- C-F, C-Cl, C-Br stretches:
| Frequency (cm-1) | Assignment |
|---|---|
| 1000--1400 | C-F (strong) |
| 600--800 | C-Cl |
| 500--680 | C-Br |
- Aromatic substitution pattern (700-900 cm-1): Out-of-plane C-H bending reveals substitution:
| Frequency (cm-1) | Pattern |
|---|---|
| 730--770 | Mono-substituted (+ 690--710) |
| 735--770 | Ortho-disubstituted |
| 750--810, 860--900 | Meta-disubstituted |
| 790--840 | Para-disubstituted |
- Overall fingerprint compare: Fingerprint region unique to each compound. Reference spectrum available? Overlay and compare this region for identity confirmation
Got: Confirmatory assignments for functional groups from Step 2, plus additional structural detail (substitution patterns, C-O/C-N assignments).
If fail: Fingerprint region inherently complex and overlapping. Assignments ambiguous? Flag as tentative. Rely on diagnostic region and other spectroscopic data for final conclusions.
Step 4: Assess Hydrogen Bonding and Intermolecular Effects
Evaluate how sample state and intermolecular interactions affect spectrum:
- Hydrogen bonding broadening: Compare width and position of O-H and N-H bands. Free O-H sharp, near 3600 cm-1. Hydrogen-bonded O-H broad, shifted to 3200-3400 cm-1. Carboxylic acid dimers show very broad O-H from 2500-3300 cm-1
- Concentration and state effects: Solution spectra at different concentrations distinguish intramolecular (concentration-independent) from intermolecular (concentration-dependent) hydrogen bonds
- Fermi resonance: Two overlapping bands interact to split into doublet. Classic example: aldehyde C-H pair near 2720 and 2820 cm-1. Recognize Fermi resonance to avoid misassigning extra peaks as separate functional groups
- Solid-state effects: KBr pellets and Nujol mulls reflect solid-state packing. Broadens bands. Shifts frequencies 10-20 cm-1 relative to solution spectra. ATR spectra closest to neat liquid state
Got: Hydrogen bonding state characterized. Prep-method artifacts accounted for. Any anomalous band shapes explained.
If fail: Hydrogen bonding effects cannot be resolved (e.g., overlapping O-H and N-H bands)? Note ambiguity. D2O exchange experiment or variable-temperature study can help, but need additional data.
Step 5: Compile Functional Group Inventory
Assemble findings into structured report:
- List confirmed functional groups: Strong, unambiguous absorptions in diagnostic region (e.g., sharp C=O at 1715 cm-1 = ketone or aldehyde)
- List tentative assignments: Weaker evidence or overlapping absorptions that could be explained by more than one functional group
- List absent functional groups: Characteristic strong absorptions clearly missing (e.g., no broad O-H band = no free alcohol or carboxylic acid)
- Note discrepancies: Absorptions not fitting proposed functional group set, or expected absorptions missing
- Cross-reference: Compare IR-derived inventory with info from other techniques (NMR, MS, UV-Vis) if available
Got: Complete functional group inventory by confidence level, with specific frequencies and intensities cited as evidence for each assignment.
If fail: Inventory incomplete or contradictory? Identify which additional experiments (ATR vs. KBr compare, variable concentration, D2O exchange) would resolve ambiguities.
Checks
- Spectrum quality assessed (baseline, resolution, artifacts, y-axis format)
- Solvent, prep-method, atmospheric artifacts identified and excluded
- All absorptions in diagnostic region (4000-1500 cm-1) assigned or flagged
- Carbonyl region analyzed with specific sub-type assignment where possible
- Fingerprint region examined for confirmatory evidence
- Hydrogen bonding effects evaluated, influence on peak shape/position documented
- Functional group inventory compiled with confidence levels
- Absent functional groups explicitly noted (negative evidence informative)
- Assignments cross-referenced with other available spectroscopic data
Pitfalls
- Ignoring prep artifacts: KBr moisture (broad 3400 cm-1), Nujol C-H (2850-2950 cm-1), ATR intensity distortion at low wavenumbers all mimic or obscure real absorptions. Always consider prep method.
- Over-interpret fingerprint region: Region below 1500 cm-1 complex and overlapping. Use for confirmation, not primary ID. Avoid assigning every peak.
- Confuse atmospheric CO2 with sample peaks: Sharp doublet near 2350 cm-1 almost always atmospheric CO2, not sample absorption. Background subtraction should remove, but verify.
- Neglect band intensity and width: Strong broad absorption differs from weak sharp peak at same frequency in diagnostic value. Report intensity (strong/medium/weak) and shape (sharp/broad) alongside frequency.
- Single-peak assignments: Never identify functional group from single absorption alone. Carbonyl groups should be supported by additional bands (C-O for esters, N-H for amides, C-H for aldehydes).
- Assume absence from weak absorption: Some functional groups make inherently weak IR absorptions (symmetric C=C, triple bonds in symmetric alkynes). Absence of peak does not always mean absence of group.
See Also
interpret-nmr-spectrum— determine detailed connectivity and hydrogen environmentsinterpret-mass-spectrum— establish molecular formula and fragmentation patterninterpret-uv-vis-spectrum— characterize chromophores complementing IR functional group datainterpret-raman-spectrum— complementary vibrational data for IR-inactive modesplan-spectroscopic-analysis— select and sequence spectroscopic techniques before data acquisition
GitHub репозиторий
Похожие навыки
llamaguard
ДругоеLlamaGuard — это модель от Meta с 7–8 миллиардами параметров для модерации входных и выходных данных больших языковых моделей по шести категориям безопасности, таким как насилие и разжигание ненависти. Она обеспечивает точность 94–95% и может быть развернута с помощью vLLM, Hugging Face или Amazon SageMaker. Используйте этот навык, чтобы легко интегрировать фильтрацию контента и защитные механизмы в ваши ИИ-приложения.
cost-optimization
ДругоеЭтот навык Claude помогает разработчикам оптимизировать облачные расходы за счет правильного подбора ресурсов, стратегий тегирования и анализа затрат. Он предоставляет framework для сокращения облачных расходов и внедрения управления затратами в AWS, Azure и GCP. Используйте его, когда вам нужно проанализировать расходы на инфраструктуру, оптимизировать ресурсы или уложиться в бюджетные ограничения.
quantizing-models-bitsandbytes
ДругоеЭтот навык выполняет квантизацию LLM до 8-битной или 4-битной точности с использованием библиотеки bitsandbytes, обеспечивая сокращение использования памяти на 50-75% при минимальной потере точности. Он идеально подходит для запуска больших моделей при ограниченной памяти GPU или для ускорения вывода, поддерживая форматы INT8, NF4 и FP4. Навык интегрируется с HuggingFace Transformers и позволяет использовать обучение QLoRA и 8-битные оптимизаторы.
dispatching-parallel-agents
ДругоеЭтот навык Claude распределяет нескольких агентов для исследования и устранения трёх и более независимых проблем параллельно. Он предназначен для сценариев с несвязанными сбоями, которые можно устранить без общего состояния или зависимостей. Ключевая возможность — параллельное решение проблем, где за каждую независимую предметную область назначается отдельный агент для максимальной эффективности.
