Ep 50 — Instrument Daily Maintenance and Performance Verification (ASTM Standards)

Series: Infrared Spectroscopy Encyclopedia: From Principles to Practice
Chapter: Part 5 · Instruments and Tools — Brand Selection and Purchasing (Pre-Final Episode)
Target Audience: Laboratory managers, instrument responsible persons, QC supervisors, GMP compliance personnel, technical leads needing to establish instrument maintenance SOPs
Prerequisites: Ep 12 (Michelson Interferometer and KBr Beamsplitter), Ep 16 (Common Sample Preparation Mistakes), Ep 17 (Basic Instrument Operation Procedure), Ep 48 (Selection and Acceptance Checklist), Ep 49 (Accessory Systems)
Reading Time: Approximately 55 minutes

Introduction: A "Save-Use" FTIR Breaks Down the Fastest

In 2018, a Bruker Tensor II in a university laboratory began frequent errors in its third year of operation: baseline drift, decreased signal-to-noise ratio, and a persistent strong water peak at 3400 cm⁻¹. After an engineer disassembled the instrument, it was discovered that the KBr beamsplitter crystal had become severely hygroscopic and turbid, nearly ruined. The cause: to "save purge gas," the lab had long ceased nitrogen purging, and desiccant was not replaced on schedule [1].

Replacing the beamsplitter cost $4,000, equivalent to the lab's six-month consumables budget. This is a classic case of "saving money but costing more." This episode systematically explains FTIR daily maintenance checklists and performance verification standard methods (ASTM E1421), helping laboratories establish a robust maintenance system.

"Routine maintenance and periodic performance verification are not optional expenses — they are the cheapest insurance against catastrophic failure and the foundation of defensible analytical data."
—— ASTM E1421 Standard Practice [2]

1. Daily Maintenance Checklist: Weekly, Monthly, Quarterly

1.1 Maintenance Frequency Classification

Table 1: FTIR Daily Maintenance Frequency Classification (Compiled from ASTM E1421 [2] and manufacturer maintenance manuals [3][4])

1.2 Desiccant and Purge Gas Management

KBr beamsplitter is the most fragile component of FTIR — KBr (potassium bromide) is highly hygroscopic; after moisture absorption, the crystal becomes turbid, transmittance decreases, ultimately leading to beamsplitter failure [3][5].

🔗 Extended Reading: Monitoring KBr moisture absorption is critical, focusing on absorption of water molecules at 3400 cm⁻¹ (O–H stretch) and 1640 cm⁻¹ (H–O–H bend) — see ftir.fun water functional group page. A healthy instrument, after adequate purging, should show very weak water peaks at 3400/1640 cm⁻¹ (absorbance < 0.05); if water peaks persist > 0.3, it suggests insufficient purging or beamsplitter moisture absorption.

Key Desiccant Management Points [3][4]:

• The instrument's built-in desiccant container (usually filled with silica gel or molecular sieve) should be checked every 1–2 weeks; color-changing silica gel turning pink indicates need for regeneration/replacement;
• Desiccant regeneration: bake at 120°C for 4 hours (silica gel) or 300°C in vacuum for 6 hours (molecular sieve);
• Do not bake molecular sieve above 600°C — this destroys pore structure.

Purge Gas Management [3][4]:

• Use dry, oil-free nitrogen or clean air, pressure 0.2–0.4 MPa, flow rate 5–10 L/min;
• After starting up, purge for at least 15 minutes before collecting background to stabilize humidity in the optical path;
• When not in use for extended periods, maintain low-flow purge (2 L/min) or seal the instrument and place desiccant inside;
• Never blow directly with compressed air — oil and water contamination will pollute the optical path.

1.3 Beamsplitter Moisture Prevention

KBr beamsplitter's "life-or-death" line is humidity [3][5]:

Table 2: KBr Beamsplitter Humidity Risk Classification (Source: Harrick [5]; Bruker Maintenance Manual [3])

⚠️ Special Reminder for Humid Regions: During the rainy season in southern China (humidity often > 80%), FTIR laboratories must use 24-hour continuous purging or dehumidification. A university in Guangdong once shut down for 2 months during summer break; upon return, the beamsplitter was directly ruined [1].

1.4 ATR Crystal Cleaning and Inspection

Cleaning of ATR crystals (especially diamond, ZnSe) directly affects reproducibility [4][6]:

• After each use: Wipe the crystal surface gently with a lint-free tissue soaked in ethanol or isopropanol to remove sample residue;
• Stubborn stains: Can be gently wiped with a mild detergent (neutral pH), avoid strong acids and bases (especially for ZnSe);
• Regular inspection for scratches: Check the crystal surface with a magnifying glass or microscope; scratches can degrade signal reproducibility;
• Diamond ATR: Extremely hard, almost impossible to scratch, but still requires cleaning;
• ZnSe ATR: Low hardness (Mohs 3–4), do not use abrasive tissues, otherwise the crystal will be ruined.

1.5 Source Lifetime Monitoring and Replacement

FTIR sources (Globar silicon carbide rods or ceramic sources) have limited lifetimes and require monitoring [3][4]:

• SiC rod lifetime: Approximately 2–3 years (about 15,000 hours of continuous use);
• Aging signs: Interferogram peak drop > 30%, continuous decrease in signal-to-noise ratio, weakened single-beam spectrum intensity;
• Replacement trigger: Peak energy of single-beam background drops below 50% of a new source, or SNR fails to meet requirements;
• Replacement notes: Must be performed by an engineer; after installation, optical alignment and performance verification are required.

💡 Practical Experience: It is recommended that labs keep a spare source on hand. Sources often "die suddenly" just before failure; having a spare can avoid a 1–2 week downtime waiting for a replacement [1].

2. Performance Verification (OQ/PQ): ASTM E1421 Standard Method

2.1 Concepts of OQ and PQ

Instrument verification consists of three phases (IQ/OQ/PQ) [2][7]:

• IQ (Installation Qualification): After delivery, confirm hardware, software, and accessories match the contract;
• OQ (Operational Qualification): Confirm that instrument performance indicators meet manufacturer specifications;
• PQ (Performance Qualification): Confirm that the instrument performs satisfactorily under actual analytical methods.

The core standard for OQ/PQ is ASTM E1421: Standard Practice for Describing and Measuring Performance of Fourier Transform Mid-Infrared (FT-MIR) Spectrometers [2].

2.2 Five Verification Items in ASTM E1421

ASTM E1421 specifies five core performance indicators for FTIR verification [2][7]:

A. Wavenumber Accuracy

• Reference material: Polystyrene film (thickness ~75 μm, NIST SRM 1921);
• Calibration peaks: 1601.4, 1583.1, 1494.5, 1181.8, 1154.5, 1028.0, 906.7 cm⁻¹, etc.;

• Acceptance criteria:

• Regular instruments: deviation < ±0.05 cm⁻¹;

• Research-grade instruments: deviation < ±0.01 cm⁻¹;
• Test method: 4 cm⁻¹ resolution, 16 scans, measure characteristic peak positions, compare with standard values.

🔗 Extended reading: For aromatic ring-related assignments of polystyrene film calibration peaks, refer to ftir.fun aromatic functional group page. 1601 cm⁻¹ is often used as a reference for aromatic C=C, and peaks like 1494 cm⁻¹ are also commonly used for wavenumber verification—subject to ASTM E1421 / NIST SRM specifications.

B. Wavenumber Repeatability

• Test method: Continuously acquire 5 spectra of polystyrene film, measure the 1601.4 cm⁻¹ peak position;
• Acceptance criteria: Standard deviation < 0.01 cm⁻¹ (regular) / < 0.005 cm⁻¹ (research-grade) for the 5 measurements;
• Significance: Poor repeatability indicates unstable interferometer moving mirror motion, He-Ne laser reference failure, or optical path misalignment.

C. Transmittance Accuracy

• Test method: Use a set of standard filters with known transmittance (e.g., 10%, 50%, 90% T);
• Acceptance criteria: Deviation between measured and nominal values < 0.5% T;
• Alternative method: Use polystyrene films of different thicknesses and compare the intensity ratio of two peaks to theoretical values.

D. Resolution Verification

• Test method:
  • Gas method: Measure ammonia (NH₃) vapor, check if the doublet around ~950 cm⁻¹ can be resolved at 0.5 cm⁻¹ resolution;
  • Film method: Polystyrene 1601/1583 cm⁻¹ doublet should show a "valley depth > 8% T" resolution at 0.5 cm⁻¹;
• Acceptance criteria: At nominal resolution, characteristic doublets/multiplets of standard materials should be clearly resolved [2][7].

E. Signal-to-Noise Ratio Test

• Test conditions: 4 cm⁻¹ resolution, 2200–2000 cm⁻¹ (no sample absorption window), 1-minute scan, single-beam background;
• Measurement method: Peak-to-Peak method—take the peak-to-peak noise of the 100% line in the 2200–2000 cm⁻¹ range;
• Acceptance criteria: Meet contract specifications (e.g., conventional 50,000:1, research-grade 100,000:1);
• Key reminder: Must use Peak-to-Peak rather than RMS, as RMS artificially inflates values by 5–8 times [2][7].

2.3 Frequency of Performance Verification

Table 3: Recommended performance verification frequency (combined from ASTM E1421 [2] and USP <852> [7])

III. Instrument Logs and Compliance Documentation

3.1 Essential Contents of Instrument Logs

GMP/ISO 17025 compliant laboratories must maintain instrument usage logs[7][8]:

   ┌─────────────────────────────────────┐
   │ FTIR Instrument Log (one per instrument) │
   ├─────────────────────────────────────┤
   │ □ Basic info: model, serial number, installation date │
   │ □ Daily usage record: operator, number of samples, anomalies │
   │ □ Maintenance record: date, item, performer │
   │ □ Performance verification record: date, result, pass/fail │
   │ □ Failure and repair record: fault, cause, repair │
   │ □ Part replacement record: source, beamsplitter, detector │
   │ □ Calibration certificate: validity, calibration body │
   │ □ Software version change record │
   └─────────────────────────────────────┘

3.2 Special Requirements for GMP Compliance

Pharmaceutical GMP (21 CFR Part 11) imposes additional requirements on FTIR [7][8]:

• Audit Trail: Software automatically records who did what and when, non-modifiable;
• Electronic Signatures: Data review requires electronic signatures;
• Access Levels: Separation of operator, reviewer, and administrator permissions;
• Data Integrity: Raw data cannot be deleted and must be traceable.

⚠️ Compliance reminder: Only "Compliance Edition" versions of OMNIC, OPUS support the above features; standard versions do not meet GMP requirements. When purchasing FTIR for pharmaceutical use, confirm the software version [4][8].

IV. Common Maintenance Misconceptions and "Penny Wise, Pound Foolish" Cases

4.1 Five Common Misconceptions

Myth 1: "Saving purge gas saves money"
Consequence: KBr beamsplitter absorbs moisture and fails, replacement cost $3000–5000.
Correct: Purge nitrogen costs < $100/month, which is 1/30 of the beamsplitter [1][3].

Myth 2: "Desiccant can still be used, so don't replace it yet"
Consequence: Saturated desiccant fails, humidity in optical path uncontrolled, water peaks interfere with all spectra.
Correct: Replace silica gel when it turns pink; better early than late[3].

Myth 3: "Just wipe the ATR with a tissue"
Consequence: Tissue abrasion scratches ZnSe crystal, reducing repeatability, eventually replacing crystal costs $800–1500.
Correct: Use lint-free wipes (Kimwipes) + ethanol and wipe gently [4][6].

Myth 4: "The source is still on, so don't replace it"
Consequence: Source aging reduces intensity, SNR fails, quantitative analysis errors increase.
Correct: Monitor single-beam background peak value; replace when it drops to 50% [3].

Myth 5: "Performance verification is too troublesome; once a year is enough"
Consequence: Instrument drift not detected in time, data unreliable, GMP audit fails.
Correct: Monthly wavenumber calibration, quarterly full verification [2][7].

4.2 "Penny Wise, Pound Foolish" Case Collection

Table 4: "Penny Wise, Pound Foolish" case comparison (based on lab feedback [1])

V. Maintenance SOP Template

To help labs quickly establish standards, a simplified FTIR daily maintenance SOP template is provided [2][3][7]:

   FTIR Daily Maintenance SOP (Simplified)

   [Daily Startup]
   1. Check instrument appearance, desiccant color
   2. Turn on nitrogen purge, flow rate 8 L/min, purge for 15 min
   3. Turn on instrument, run self-test
   4. Collect background, check water peaks at 3400/1640 cm⁻¹ < 0.05 A
   5. If water peaks exceed limit, extend purge to 30 min; if still exceeding, check desiccant/beamsplitter

   [Daily Shutdown]
   1. Clean ATR crystal (lint-free wipe + ethanol)
   2. Record usage log
   3. Maintain low flow purge (2 L/min) or seal + desiccant

   [Weekly]
   1. Check desiccant, regenerate/replace if color changes
   2. Visually check source stability (single-beam background peak)
   3. Clean instrument exterior and accessories

   [Monthly]
   1. Collect polystyrene film spectrum, verify wavenumber accuracy at 1601.4 cm⁻¹
   2. Check ATR crystal surface for scratches
   3. Backup software and data

Quarterly

1. Perform full ASTM E1421 performance validation
2. Check optical alignment (contact engineer if necessary)
3. Evaluate remaining light source寿命

Troubleshooting

• Sudden increase in water peaks: Check purge, desiccant, beamsplitter
• Decrease in signal-to-noise ratio: Check source, detector, optical alignment
• Wavenumber drift: Check He-Ne laser reference, interferometer
• Any abnormality: Log and notify instrument responsible person

Summary of This Episode

Review Questions

1. An FTIR in your lab shows increased water peak at 3400 cm⁻¹ for two consecutive weeks. Enhanced purging has not improved it. List at least four possible causes and the sequence to check them one by one.

2. According to ASTM E1421, why is polystyrene film used for wavenumber accuracy verification instead of water vapor or CO₂? Explain from three aspects: stability of reference material, traceability, and characteristic peak distribution.

3. A GMP pharmaceutical company’s FTIR was cited in an annual audit for “not saving audit trail records”. Explain the meaning of audit trail, why standard software does not comply, and how to rectify.

4. The manufacturer claims a signal-to-noise ratio of 50,000:1, but your actual measurement is only 20,000:1. List at least five possible reasons for the lower measured value, and explain how to rule out “non-instrument failure” factors.

5. After replacing the light source on an FTIR, the signal-to-noise ratio returns to 45,000:1, but the wavenumber accuracy deviation increases from 0.02 cm⁻¹ to 0.08 cm⁻¹. Why does replacing the light source affect wavenumber accuracy? How should this be handled?

References

[1] China Instrumentation Society Analytical Instrument Branch. White Paper on Procurement and Application of Fourier Transform Infrared Spectrometers. 2023.

[2] ASTM E1421-22. Standard Practice for Describing and Measuring Performance of Fourier Transform Mid-Infrared (FT-MIR) Spectrometers: Level Zero and Level One Tests. ASTM International, 2022.
https://www.astm.org/e1421-22…

[3] Bruker Optics. Tensor II / INVENIO Maintenance Manual. 2023.

[4] Thermo Fisher Scientific. Nicolet iS20 / iS50 User Maintenance Guide. 2024.

[5] Harrick Scientific Products. Care and Maintenance of KBr Beamsplitters and ATR Crystals. Technical Note TN-2021, 2021.
https://www.harrick.com/techn…

[6] Milosevic M A. "Diamond ATR for FTIR Spectroscopy." Spectroscopy, 2012, 27(7): 26–31.

[7] USP General Chapter <852>. Spectrophotometry and Light-Scattering — Infrared Spectroscopy. United States Pharmacopeia, 2024.

[8] European Medicines Agency. Guideline on the Use of Near-Infrared Spectroscopy by the Pharmaceutical Industry. EMA/CHMP/QWP/177604/2009, 2014.

[9] ISO/IEC 17025:2017. General Requirements for the Competence of Testing and Calibration Laboratories.
https://www.iso.org/standard/…

[10] Griffiths P R, de Haseth J A. Fourier Transform Infrared Spectrometry. 2nd ed. Wiley, 2007. Ch. 5: Instrument Performance.

[11] ftir.fun Online Infrared Spectrum Analysis Tool. https://ftir.fun

[12] ftir.fun Water Molecule Functional Group Page. https://ftir.fun/ir/group/wat…

[13] ftir.fun Aromatic Ring Functional Group Page (aid in understanding PS film peaks). https://ftir.fun/ir/group/aro…

Next Episode Preview: Ep 51 — Common Troubleshooting and "Pitfall" Real Cases
So far, the front part of Instruments and Tools (Ep 46–50) concludes: cross-comparison of six major camps, purchasing methodology, detailed accessory systems, and establishment of maintenance and validation system. Next episode enters the “troubleshooting” battlefield: use fault tree methods to resolve issues like flat spectra, poor signal-to-noise, baseline tilt, water peak interference, abnormal peaks; compile classic “pitfall” cases (misidentifying CO₂ peaks, ATR scratches, uneven pellet pressing, false peaks from atmospheric compensation); and provide preventive maintenance strategies to keep instruments “trouble-free and minor issues.”

This article is licensed under CC BY-NC-SA 4.0. Images are from public domain or attributed online sources, copyright belongs to the original authors.