정보
이 스킬은 샘플 내 작용기를 식별하기 위해 적외선 스펙트럼을 체계적으로 분석합니다. 수소 결합 효과를 고려하여 진단 영역(4000-1500 cm⁻¹)과 지문 영역(1500-400 cm⁻¹)을 모두 해석합니다. 개발자는 이를 통해 신뢰도 수준이 표시된 작용기 목록을 작성하여 초기 단계의 화합물 스크리닝에 활용할 수 있습니다.
빠른 설치
Claude Code
추천npx 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-spectrumClaude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요
문서
IRスペクトルの解釈
Analyze infrared absorption spectra to identify functional groups, assess hydrogen bonding, and compile a comprehensive inventory of structural features present in the sample.
使用タイミング
- Identifying functional groups in an unknown compound as a first screening step
- Confirming the presence or absence of specific functional groups (e.g., verifying a reaction converted an alcohol to a ketone)
- Monitoring reaction progress by tracking the appearance or disappearance of characteristic absorptions
- Distinguishing between similar compounds that differ in functional group content
- Complementing NMR and mass spectrometry data with vibrational information
入力
- 必須: IR spectrum data (absorption frequencies in cm-1 with intensities, either as %Transmittance or Absorbance plot)
- 必須: Sample preparation method (KBr pellet, ATR, Nujol mull, thin film, solution cell)
- 任意: Molecular formula or expected compound class
- 任意: Known structural fragments from other spectroscopic data
- 任意: Instrument parameters (resolution, scan range, detector type)
手順
ステップ1: Establish Spectrum Quality and Format
Verify that the spectrum is suitable for interpretation before analyzing peaks:
- Check y-axis format: Determine whether the spectrum is plotted in %Transmittance (%T, peaks point down) or Absorbance (A, peaks point up). All subsequent analysis assumes consistent convention.
- Verify wavenumber range: Confirm the spectrum covers at least 4000--400 cm-1 for a standard mid-IR analysis. Note any truncation.
- Assess baseline: A good baseline should be relatively flat and near 100%T (or 0 Absorbance) in regions with no absorption. Sloping or noisy baselines reduce reliability.
- Check resolution: Adjacent peaks separated by less than the instrumental resolution cannot be distinguished. Typical FTIR resolution is 4 cm-1.
- Identify preparation artifacts: KBr pellets may show a broad O-H band from absorbed moisture (~3400 cm-1). Nujol mulls obscure C-H stretches. ATR spectra show intensity distortion at low wavenumbers. Note any artifacts that limit interpretation.
期待結果: Spectrum confirmed as suitable for analysis, with format, range, and artifacts documented.
失敗時: If the spectrum has severe baseline problems, saturation (flat-bottomed peaks from too-concentrated samples), or preparation artifacts obscuring critical regions, note the limitation and flag affected spectral regions as unreliable.
ステップ2: Scan the Diagnostic Region (4000--1500 cm-1)
Systematically analyze the high-frequency region where most functional groups produce characteristic absorptions:
- O-H stretches (3200--3600 cm-1): Look for broad absorptions. A sharp peak near 3600 cm-1 indicates free O-H; a broad band centered at 3200--3400 cm-1 indicates hydrogen-bonded O-H (alcohols, carboxylic acids, water).
- N-H stretches (3300--3500 cm-1): Primary amines show two peaks (symmetric and asymmetric stretch); secondary amines show one peak. These are 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) -- the 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 appears at 1620--1680 cm-1 (weak to medium). Aromatic C=C shows multiple peaks near 1450--1600 cm-1. C=N (imine) appears at 1620--1660 cm-1.
期待結果: All absorptions in the diagnostic region identified, with functional group assignments and confidence levels (strong, tentative, absent).
失敗時: If the carbonyl region is obscured (e.g., water absorption in KBr, atmospheric CO2), note the gap. If an expected functional group absorption is absent, confirm with a second preparation method before concluding it is truly absent.
ステップ3: Analyze the Fingerprint Region (1500--400 cm-1)
Examine the lower-frequency region for confirmatory and structural detail:
- C-O stretches (1000--1300 cm-1): Ethers, esters, alcohols, and carboxylic acids produce strong C-O stretching absorptions. Esters show a characteristic strong band near 1000--1100 cm-1 in addition to the 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 comparison: The fingerprint region is unique to each compound. If a reference spectrum is available, overlay and compare this region for identity confirmation.
期待結果: Confirmatory assignments for functional groups identified in Step 2, plus additional structural detail (substitution patterns, C-O/C-N assignments).
失敗時: The fingerprint region is inherently complex and overlapping. If assignments are ambiguous, flag them as tentative and rely on the diagnostic region and other spectroscopic data for final conclusions.
ステップ4: Assess Hydrogen Bonding and Intermolecular Effects
Evaluate how sample state and intermolecular interactions affect the spectrum:
- Hydrogen bonding broadening: Compare the width and position of O-H and N-H bands. Free O-H is sharp and near 3600 cm-1; hydrogen-bonded O-H is broad and shifted to 3200--3400 cm-1. Carboxylic acid dimers show a very broad O-H from 2500--3300 cm-1.
- Concentration and state effects: Solution spectra at different concentrations can distinguish intramolecular (concentration-independent) from intermolecular (concentration-dependent) hydrogen bonds.
- Fermi resonance: Two overlapping bands can interact to split into a doublet. The classic example is the 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, which broadens bands and can shift frequencies by 10--20 cm-1 relative to solution spectra. ATR spectra are closest to the neat liquid state.
期待結果: Hydrogen bonding state characterized, preparation-method artifacts accounted for, and any anomalous band shapes explained.
失敗時: If hydrogen bonding effects cannot be resolved (e.g., overlapping O-H and N-H bands), note the ambiguity. A D2O exchange experiment or variable-temperature study can help, but these require additional data.
ステップ5: Compile Functional Group Inventory
Assemble all findings into a structured report:
- List confirmed functional groups: Groups with strong, unambiguous absorptions in the diagnostic region (e.g., sharp C=O at 1715 cm-1 = ketone or aldehyde).
- List tentative assignments: Groups with weaker evidence or overlapping absorptions that could be explained by more than one functional group.
- List absent functional groups: Groups whose characteristic strong absorptions are clearly missing from the spectrum (e.g., no broad O-H band means no free alcohol or carboxylic acid).
- Note discrepancies: Any absorptions that do not fit the proposed functional group set, or expected absorptions that are missing.
- Cross-reference: Compare the IR-derived functional group inventory with information from other techniques (NMR, MS, UV-Vis) if available.
期待結果: A complete functional group inventory categorized by confidence level, with specific frequencies and intensities cited as evidence for each assignment.
失敗時: If the inventory is incomplete or contradictory, identify which additional experiments (ATR vs. KBr comparison, variable concentration, D2O exchange) would resolve the ambiguities.
バリデーション
- Spectrum quality assessed (baseline, resolution, artifacts, y-axis format)
- Solvent, preparation-method, and atmospheric artifacts identified and excluded
- All absorptions in the 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 and their influence on peak shape/position documented
- Functional group inventory compiled with confidence levels
- Absent functional groups explicitly noted (negative evidence is informative)
- Assignments cross-referenced with other available spectroscopic data
よくある落とし穴
- Ignoring preparation artifacts: KBr moisture (broad 3400 cm-1), Nujol C-H (2850--2950 cm-1), and ATR intensity distortion at low wavenumbers all mimic or obscure real sample absorptions. Always consider the preparation method.
- Over-interpreting the fingerprint region: The region below 1500 cm-1 is complex and overlapping. Use it for confirmation, not primary identification. Avoid assigning every peak.
- Confusing atmospheric CO2 with sample peaks: The sharp doublet near 2350 cm-1 is almost always atmospheric CO2, not a sample absorption. Background subtraction should remove it, but verify.
- Neglecting band intensity and width: A strong, broad absorption has different diagnostic value than a weak, sharp peak at the same frequency. Report intensity (strong/medium/weak) and shape (sharp/broad) alongside frequency.
- Single-peak assignments: Never identify a functional group from a single absorption alone. Carbonyl groups, for example, should be supported by additional bands (C-O for esters, N-H for amides, C-H for aldehydes).
- Assuming absence from weak absorption: Some functional groups produce inherently weak IR absorptions (symmetric C=C, triple bonds in symmetric alkynes). Absence of a peak does not always mean absence of the group.
関連スキル
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-- obtain complementary vibrational data for IR-inactive modesplan-spectroscopic-analysis-- select and sequence spectroscopic techniques before data acquisition
GitHub 저장소
Frequently asked questions
What is the interpret-ir-spectrum skill?
interpret-ir-spectrum is a Claude Skill by pjt222. Skills package instructions and resources that Claude loads on demand, so Claude can perform interpret-ir-spectrum-related tasks without extra prompting.
How do I install interpret-ir-spectrum?
Use the install commands on this page: add interpret-ir-spectrum to Claude Code as a plugin, or clone its repository into your skills directory, then restart Claude so it picks up the skill.
What category does interpret-ir-spectrum belong to?
interpret-ir-spectrum is in the Other category, tagged general.
Is interpret-ir-spectrum free to use?
Yes. interpret-ir-spectrum is listed on AIMCP and free to install. It runs inside Claude, so no separate service account is required to use the skill itself.
연관 스킬
LlamaGuard는 폭력 및 혐오 발언 등 6가지 안전 범주에서 LLM 입력과 출력을 조정하기 위한 Meta의 70-80억 파라미터 모델입니다. 94-95% 정확도를 제공하며 vLLM, Hugging Face 또는 Amazon SageMaker를 사용해 배포할 수 있습니다. 이 기술을 사용하여 AI 애플리케이션에 콘텐츠 필터링 및 안전 가드레일을 손쉽게 통합하세요.
이 Claude Skill은 리소스 적정화, 태깅 전략, 지출 분석을 통해 개발자들이 클라우드 비용을 최적화할 수 있도록 지원합니다. AWS, Azure, GCP에서 클라우드 비용을 절감하고 비용 거버넌스를 구현하기 위한 프레임워크를 제공합니다. 인프라 비용을 분석하거나, 리소스를 적정화하거나, 예산 제약을 충족해야 할 때 사용하세요.
이 Claude Skill은 스프레드, 오버/언더, 프로프 베트를 포함한 스포츠 베팅 시장을 분석합니다. 역사적 추이와 상황별 통계를 검토하여 가치 베트를 발견하고, 교육적 목적으로 실행 가능한 권장 사항이 담긴 구조화된 마크다운 결과를 제공합니다. 개발자는 이 기능을 스포츠 베팅 분석 도구에 활용할 수 있으며, 단순히 엔터테인먼트/교육 목적으로만 설계되었음을 유의해야 합니다.
이 스킬은 bitsandbytes를 사용하여 LLM을 8비트 또는 4비트 정밀도로 양자화하며, 최소한의 정확도 손실로 50-75%의 메모리 감소를 달성합니다. 제한된 GPU 메모리에서 더 큰 모델을 실행하거나 추론을 가속화하는 데 이상적이며, INT8, NF4, FP4와 같은 형식을 지원합니다. 이 스킬은 HuggingFace Transformers와 통합되어 QLoRA 학습 및 8비트 옵티마이저를 가능하게 합니다.
