Ep 16 — Sample Preparation Practice and Common Mistakes

Series: Infrared Spectroscopy Encyclopedia: From Principles to Practice
Chapter: Part II · Beginner Level — Entering the Lab
Audience: High school/undergraduate/graduate students, lab technicians new to the lab
Prerequisites: Ep 13 (Transmission), Ep 14 (ATR), Ep 15 (ATR vs Transmission)
Reading Time: ~40 minutes

Introduction: The Birth of a "Ghost" Spectrum

One Friday afternoon, a novice named Xiao Li rushed into the spectroscopy room with a freshly made KBr pellet. The spectrum appeared—"Huh, why is there a big peak at 3400 cm⁻¹?" He thought it was a hydroxyl group from the sample, but upon comparing with the standard, it didn't match at all. After half a day of repeated experiments, he finally realized—the KBr had absorbed moisture. The "ghost peak" at 3400 cm⁻¹ was actually the O-H stretching of water vapor from the air [1][2].

Similar "ghost" spectra occur almost daily in infrared labs:

• Too much sample → flat-top peaks, absorbance > 2, spectrum "clipped"
• Too little sample → poor SNR, small peaks drowned in noise
• Insufficient grinding → slanted baseline, severe scattering
• Poor ATR contact → full spectrum signal as weak as background
• Liquid cell path too long → solvent peaks completely mask the solute

The root of these errors is not the instrument, but sample preparation. There's an old saying in spectroscopy: "Garbage in, garbage out" — no matter how good the instrument, it cannot save poor sample preparation [1][3].

In this episode, we will systematically explain practical sample preparation techniques for solids, liquids, and aqueous samples, analyze 10+ typical erroneous spectra, and finally provide a "mine-clearing checklist." After reading this, you will be able to obtain a usable spectrum on your first experiment.

1. Solid Sample Preparation Practice

Solid samples are the most common form in infrared analysis, with main methods being the KBr pellet method and the ATR method. This section focuses on the KBr pellet method (transmission); ATR preparation tips are in Section 4.

1.1 KBr Pellet Method: The Classic

The KBr pellet method was proposed by Stimson and O'Donnell in the 1950s [4] and remains the mainstream method for solid sample transmission measurements. The principle is to mix the sample with dry KBr powder, grind, and press into a transparent thin disk through which infrared light passes directly.

Standard Procedure [1][3][4]:

1. Dry KBr → 2. Weigh sample → 3. Grind and mix
   105°C vacuum      1–2 mg        Agate mortar
   overnight        (1:100–1:200)   Clockwise 2–5 min
        ↓
4. Load die → 5. Vacuum press → 6. Remove pellet
   3 mm depth       8–10 ton       Transparent disk
                     30 s–2 min     Thickness 0.5–1 mm

Figure 1: Schematic of the standard KBr pellet method (refer to [1][4])

1.2 Drying KBr: The Crucial First Step

KBr is highly hygroscopic. Insufficiently dried KBr will show "water ghost peaks" at the following positions [1][2][5]:

Table 1: Positions of "ghost peaks" from KBr moisture absorption (data from [1][2])

🔗 Further reading: Detailed infrared absorption peaks of water molecules can be found at ftir.fun water molecule functional group page.

Drying methods [1][3][5]:

• Standard: 105°C for 2–4 hours, store in a 130°C desiccator
• Strict: 150°C vacuum overnight (recommended for high-precision analysis)
• Emergency: Commercially available "spectroscopic grade KBr" is pre-dried; use immediately after opening

Operating environment [3][5]:

• Humidity < 30% RH is optimal (requires dehumidifier or dry glove box)
• Humidity 30–50% RH is acceptable, but must work quickly
• Humidity > 50% RH: KBr pellet method is not recommended

⚠️ Common beginner mistake: After drying, KBr is left in an open glass dish and reabsorbs moisture within 2 hours. Correct practice: Store in a sealed desiccator and take out only when needed.

1.3 Sample-to-KBr Ratio: The 1:100 Golden Rule

The classic ratio is sample : KBr = 1:100 to 1:200 (by mass) [1][3][4].

Why 1:100? [1][4]

• Too little sample → absorbance < 0.1, poor SNR, small peaks invisible
• Too much sample → absorbance > 2, flat-top peaks, quantitative distortion
• At 1:100, the main peak absorbance is about 0.5–1.0, just in the optimal range

Example: 1 mg sample + 100 mg KBr = 200 mg total (pressed into 13 mm disk, ~0.7 mm thick)

Adjustment strategy [1][3]:

• Strongly absorbing samples (e.g., containing C=O, C-O-C): 1:200 or 1:500
• Weakly absorbing samples (e.g., saturated hydrocarbons, sulfur compounds): 1:50 or 1:20
• Rule of thumb: First try 1:100, then adjust based on peak height

1.4 Grinding Particle Size: The Hard Requirement of < 2 μm

Grinding particle size directly determines spectral quality. Rayleigh scattering law tells us [1][3][6]:

$$I_{\text{scatter}} \propto \frac{d^3}{\lambda^4}$$

where $d$ is the particle diameter and $\lambda$ is the light wavelength. The larger the particle diameter $d$, the more severe the scattering [1][6].

Empirical thresholds [1][3][6]:

• $d < 2\,\mu\text{m}$: scattering negligible, spectrum normal
• $d = 2$–$10\,\mu\text{m}$: scattering noticeable, baseline tilted, spectrum distorted
• $d > 10\,\mu\text{m}$: scattering severe, full spectrum distorted, unusable

Grinding tips [3][5]:

1. Grind the sample alone first into fine powder (e.g., hard crystals need to be crushed first)
2. Add about 1/3 of the KBr, grind for 1 minute
3. Add the remaining KBr, continue grinding for 2–5 minutes
4. Alternate clockwise and counterclockwise grinding to ensure uniformity
5. The ground powder should be as fine as flour, with no gritty feel

Choosing an agate mortar [3]:

• Diameter 50–75 mm (suitable for 100–300 mg total)
• Must use agate (high hardness, chemically inert, no contamination)
• Clean with ethanol before and after use to avoid cross-contamination

💡 Pro tip: When grinding, do not "press" but "grind"—move the pestle in circles against the mortar wall so that particles are refined by friction. Pressing can destroy crystal form (if crystal form analysis is needed, ATR must be used) [3].

1.5 Pressing Operation

Die [3][4]:

• 13 mm diameter (most common)
• 7 mm diameter (for small sample amounts)
• Material: High-carbon steel, hardened

Pressing steps [3][4]:

1. Clean the die: Wipe the die and plunger with ethanol
2. Load powder: Evenly spread the ground powder in the die and level it
3. Evacuate (if vacuum port available): Evacuate for 30 seconds to remove air between particles
4. Apply pressure: 8–10 ton (10–14 kN force), hold for 30 seconds–2 minutes
5. Release pressure: Slowly release (to avoid pellet cracking)
6. Remove: Carefully push out the pellet, avoid touching with fingers

Criteria for a good pellet [3][4]:

• Transparent or translucent (opaque = poor pressing)
• No cracks, no holes
• Uniform thickness (~0.5–1 mm)
• Intact edges

⚠️ Common beginner mistakes:

• Pressing too fast → pellet cracks
• Releasing pressure too fast → internal stress in pellet, cracks during measurement
• Die not cleaned → contamination from previous sample

1.6 Other Solid Sample Preparation Methods

① Film method [1][3]

• Suitable for soluble polymers (e.g., polystyrene, PMMA)
• Prepare a 1–5% solution, drop onto a KBr window, and allow solvent to evaporate to form a film
• Advantages: Uniform film, accurate quantification
• Note: Solvent must be completely removed (avoid solvent peak interference)

② Nujol Mull method [1][3]

• Suitable for samples where KBr ion interference is undesired (e.g., some inorganic salts)
• Grind sample with a few drops of mineral oil into a paste

• Spread on a KBr window for measurement

• Disadvantages: Mineral oil has strong C-H absorptions at 2800–3000, 1460, 1375 cm⁻¹

③ Hot Pressing Method [3]

• Suitable for thermoplastic polymers
• Heat and pressure melt the polymer into a film
• Note: Temperature must not degrade the sample

II. Liquid Sample Preparation in Practice

The core of liquid sample preparation is selecting the appropriate path length cell. Improper path length leads to two typical problems: too long causes flat-top (absorbance > 2), too short causes poor SNR [1][3][7].

2.1 Cell Types

① Demountable Cell [1][3][7]

Structure: Two windows + spacers (lead, PTFE)

   ┌──────────┐
   │   Window  │  ← KBr / NaCl / CaF₂ / ZnSe
   │  Spacer   │  ← Thickness 0.025–1 mm
   │  Sample   │
   │  Spacer   │  
   │   Window  │
   └──────────┘

Figure 2: Schematic of a demountable liquid cell (adapted from [7])

Advantages:

• Easy to disassemble and clean
• Spacers replaceable, flexible path length

Disadvantages:

• Low path length accuracy (±10%)
• Prone to leakage
• Windows easily deliquesce

② Sealed Cell [1][7]

Fixed path length (e.g., 0.1 mm, 0.5 mm), filled by syringe.

Advantages:

• Precise path length (±1 μm)
• Suitable for quantitative analysis
• Suitable for volatile liquids

Disadvantages:

• Difficult to clean
• High cost

③ Micro Cavity Cell [7]

• Volume < 1 μL
• For precious samples
• Path length 0.05–0.5 mm

2.2 Path Length Selection: Based on Sample and Concentration

Pure liquids (e.g., ethanol, acetone) [1][3]

• Recommended: 0.025–0.1 mm
• High molar concentration (~10 mol/L), short path length sufficient

Concentrated solutions (10%–50% solute) [1][3]

• Recommended: 0.1–0.2 mm
• Solvent peak may be flat-top, but solute peaks in optimal range

Dilute solutions (1%–10% solute) [1][3]

• Recommended: 0.5–1 mm
• Brings solute peaks into 0.3–1.0 absorbance range

Very dilute solutions (< 1% solute) [1][3]

• Recommended: 1–10 mm (multiple reflection cell)
• Solvent peak may be flat-top, but can be subtracted

Empirical formula for initial path length selection [1]:

$$b_{\text{optimal}} \approx \frac{1}{\varepsilon_{\text{max}} \cdot c}$$

where $\varepsilon_{\text{max}}$ is the molar absorptivity at the strongest peak (L·mol⁻¹·cm⁻¹), $c$ is concentration (mol/L).

2.3 Window Material Selection

Table 2: Comparison of common liquid cell window materials (data from [1][7])

⚠️ KBr/NaCl must not contact water! They are water-soluble; aqueous solutions will instantly dissolve the windows [1][7].

2.4 Liquid Injection and Cleaning

Injection Steps [3][7]:

1. Draw sample with a syringe (1 mL)
2. Insert needle into cell inlet
3. Slowly push sample, observe liquid reaching the opposite port
4. Close valves (sealed cell) or attach cover plate (demountable)
5. Wipe off spilled liquid with lens paper

Cleaning Steps [3][7]:

1. Inject solvent (e.g., ethanol, chloroform) with syringe
2. Rinse 3–5 times
3. Blow dry with compressed air
4. Check for no residue (collect blank spectrum)

Precautions [3][7]:

• Use different syringes for different samples to avoid cross-contamination
• Cool volatile samples before injection
• Viscous samples need heating or dilution

III. Aqueous Samples: Challenges and Strategies for Water Peaks

Water is one of the biggest "enemies" in infrared analysis—it has extremely strong absorptions at 3400 cm⁻¹ and 1640 cm⁻¹, with absorptivity far exceeding most organics [1][5][8].

3.1 Infrared Absorption of Water

Water molecule has three vibrational modes [1][5][8]:

Table 3: Main infrared absorptions of water molecule (data from [1][5][8])

🔗 Extension: Detailed mechanism of water infrared absorption at ftir.fun water functional group page.

3.2 Sample Preparation Strategies for Aqueous Samples

Strategy 1: ATR Method (Preferred) [3][5][8]

ATR path length is only μm level, water peaks relatively controllable:

• Single-reflection diamond ATR: water peak absorbance ~0.3–0.8
• Multiple-reflection ATR: water peak absorbance can reach 1.5+

Procedure: Place 1–2 drops of aqueous solution on ATR crystal, measure directly.

Strategy 2: Transmission Short Path Length Cell [1][7]

• CaF₂ cell, path length 6–10 μm
• Water peak absorbance ~0.5–1.5 (can be subtracted)
• Solute peaks must avoid 3400, 1640 cm⁻¹

Strategy 3: Background Subtraction [1][3]

• First measure pure water (background)
• Then measure sample aqueous solution
• Software automatically subtracts water peaks

But water subtraction has limitations [1][8]:

• At water peak centers (3400, 1640) where absorption is complete, cannot subtract
• Temperature changes affect water absorption, leaving residuals after subtraction
• Solute-water interactions (hydrogen bonding) alter water peak shape, causing inaccurate subtraction

3.3 "Forbidden Zones" for Aqueous Samples

Absolute Forbidden Zones [1][5][7][8]:

• 3300–3500 cm⁻¹: Completely masked by water O-H; alcohols/acids/amide N-H invisible
• 1620–1660 cm⁻¹: Masked by water H-O-H bending; amide I band invisible
• Below 550 cm⁻¹: Water rotational peaks

Observable Regions [1][8]:

• 2800–3000 cm⁻¹: C-H stretch
• 1700–1800 cm⁻¹: C=O stretch (if not near 1640)
• 1000–1500 cm⁻¹: C-O, C-N, etc.
• 800–1000 cm⁻¹: Aromatic CH out-of-plane

3.4 Alternative Methods for Aqueous Samples

① Dry Before Measurement [3][5]

• Freeze-dry or vacuum dry to remove water
• Measure solid sample (KBr pellet or ATR)
• Note: Drying may alter sample state (e.g., protein denaturation)

② Extraction [3]

• Use organic solvent to extract solute from aqueous solution
• Measure organic phase (avoid water peaks)
• Note: Extraction efficiency affects quantification

③ Solvent Replacement [3][8]

• Substitute H₂O with D₂O
• D₂O O-D stretch at 2500 cm⁻¹ (avoids 3400)
• Common method for protein secondary structure studies

IV. Common Mistakes in ATR Sample Preparation

Although ATR "requires no sample preparation", some pitfalls can severely degrade spectral quality [3][9][10].

4.1 Poor Contact

Symptom: Very weak full-spectrum signal (< 0.1), extremely poor SNR [3][9]

Cause:

• Sample particles too coarse (> 10 μm), small contact area with crystal
• Sample irregular shape (e.g., spherical particles), only point contact
• Insufficient pressure (spring not tightened)

Remedy [3][9]:

• Grind powder sample to < 5 μm
• Cut a flat surface for solid samples
• Fully tighten pressure screw (manual ATR) or confirm pneumatic pressure is applied (automatic ATR)

4.2 Crystal Contamination

Symptom: Residual peaks after background subtraction, poor reproducibility [3][9][10]

Cause:

• Previous sample not cleaned thoroughly
• Fingerprint or oil on crystal surface

Remedy [3][9][10]:

• Wipe crystal with ethanol or isopropanol 2–3 times between measurements
• Confirm crystal is clean before collecting background (background spectrum shows no residual peaks)
• Wipe different samples with separate cotton swabs

4.3 Excess Sample Amount

Symptom: ATR spectrum normal but sample wasted [3]

Cause: When sample is too thick, the evanescent field attenuates exponentially with depth; beyond the effective penetration depth (typically several micrometers, varying with wavelength/crystal/angle), deeper layers contribute very weakly, not strictly "no signal"; effective pathlength $d_e$ is also not equivalent to penetration depth $d_p$.

Remedy: 1–5 mg powder or 1–2 drops of liquid is sufficient

4.4 Improper Measurement of Volatile Liquids

Symptom: Spectrum changes continuously during measurement [3][9]

Cause: Liquid evaporates, concentration changes

Remedy [3][9]:

• Use a cover (available on some ATR setups)
• Measure quickly (< 30 s)
• Dilute with low-volatility solvent

5. Top 10 Common Error Spectrum Case Studies

Below are 10 most common mistakes made by beginners, each with error spectrum illustration, cause analysis, and remedy [1][3][5][9].

Case 1: KBr Not Fully Dried → 3400/1640 Ghost Peaks

Absorbance
   ↑
1.0│       ▓▓▓                    
0.8│       ▓▓▓          ▓▓        
0.6│       ▓▓▓          ▓▓        
0.4│   ▓▓  ▓▓▓  ▓▓▓▓  ▓▓  ▓▓▓▓▓ 
0.2│   ▓▓  ▓▓▓  ▓▓▓▓  ▓▓  ▓▓▓▓▓ 
0.0│────────────────────────────→ ν
   3000  3400  1640  1000   500  
              ↑     ↑
            O-H ghost peak  H₂O ghost peak

Cause: Inadequate KBr drying or high operating humidity [1][5]

Remedy: Dry KBr in vacuum oven at 150°C overnight; operate at humidity < 30% RH; subtract KBr blank pellet background

Case 2: Too Much Sample → Flat-top Peak

Absorbance
   ↑
2.5│       ███              ← strong peak "clipped" at A=2
2.0├───────███──────────────  ← instrument limit
1.5│       ███              
1.0│   ▓▓  ███  ▓▓▓▓  ▓▓  ▓▓▓ 
0.5│   ▓▓  ███  ▓▓▓▓  ▓▓  ▓▓▓ 
0.0│────────────────────────────→ ν
            ↑
       C=O flat-top (quantitative distortion)

Cause: Sample concentration too high (e.g., 1:10 instead of 1:100) [1][3]

Remedy: Dilute sample to 1:100–1:200; for strong absorbers, try 1:500

Case 3: Too Little Sample → Poor SNR

Absorbance
   ↑
0.5│   ··  ··  ···  ··  ···  ··  ← high noise
0.3│   ··  ··  ···  ··  ···  ··  
0.1│   ··  ··  ···  ··  ···  ··  
0.0│────────────────────────────→ ν
       ↑
   Weak peak buried in noise, indistinguishable

Cause: Sample concentration too low (e.g., 1:1000) [1][3]

Remedy: Increase sample ratio to 1:50–1:100; increase number of scans (16 → 64)

Case 4: Insufficient Grinding → Baseline Tilt, Scattering

Absorbance
   ↑
1.5│ ▓▓▓                 ▓▓▓▓▓ ← baseline high at low wavenumber, low at high wavenumber
1.0│  ▓▓▓              ▓▓▓▓▓   
0.5│   ▓▓▓  ▓▓▓▓▓   ▓▓▓▓▓      
0.0│────────────────────────────→ ν
   3000             1000   500
        ←─────tilt──────

Cause: Particles > 5 μm, Rayleigh scattering $I \propto d^3/\lambda^4$ [1][6]

Remedy: Grind in agate mortar for 5–10 minutes; visually check powder "as fine as flour"

Case 5: Pellet Not Transparent → Full Spectrum Distortion

Symptom: Pellet appears milky white, high baseline (absorbance > 1.5), no clear peaks

Cause [3][4]:

• Insufficient grinding, large particles
• Insufficient pressure
• Pellet cracked or has fissures

Remedy: Regrind; increase pressure to 10 ton; check die

Case 6: Poor ATR Contact → Very Weak Signal

Absorbance
   ↑
0.3│ ··························  ← full spectrum < 0.3
0.2│ ··  ··  ···  ··  ···  ··  
0.1│ ··  ··  ···  ··  ···  ··  
0.0│────────────────────────────→ ν
       ← signal too weak, nearly only noise

Cause [3][9]:

• Sample particles too coarse
• Pressure not fully applied
• Sample irregular shape

Remedy: Grind powder; cut flat surface; confirm pressure is applied

Case 7: Liquid Cell Pathlength Too Long → Solvent Flat-top

Symptom: Solvent peak (e.g., ethanol C-H at 2900) flat-top > 2, solute peaks obscured

Cause: Wrong pathlength chosen (e.g., 1 mm for pure liquid) [1][7]

Remedy: Use short-pathlength cell (0.025–0.1 mm)

Case 8: Air Bubble in Liquid Cell → Spectrum Abrupt Changes

Symptom: Localized abnormal "dips" or sudden noise increase

Cause: Air bubble introduced during filling, blocking the optical path [3][7]

Remedy: Refill slowly; check cell for bubbles

Case 9: Improper Background Collection → Full Spectrum Drift

Symptom: Non-zero absorbance when measuring pure KBr pellet (blank)

Cause [1][3]:

• Environmental changes (e.g., temperature, humidity) after background collection
• Background collected with sample still in beam path (e.g., not removed)
• Background collected before instrument stabilization

Remedy: Let instrument stabilize for 30 min before collecting background; re-collect background before each measurement

Case 10: CO₂ Interference → Doublet at 2350

Absorbance
   ↑
0.5│                       ▓▓   ← CO₂ antisymmetric stretch
0.4│                       ▓▓   ← 2349 cm⁻¹
0.3│                       ▓▓   
0.2│                       ▓▓   ← CO₂ bending combination
0.1│                       ▓▓   ← doublet near 2360
0.0│────────────────────────────→ ν
                       2350

Cause: CO₂ concentration changes in the optical path, inconsistent between background and sample [1][3]

Remedy: Purge instrument for 10–15 min; keep time between background and sample < 1 min; seal sample compartment

6. Error Spectrum vs. Correct Spectrum Comparison

Case A: KBr Pellet

| Sample:KBr | 1:10 | 1:100 |
| KBr Drying | Not dried | 150°C vacuum overnight |
| Grinding | 30 s | 5 min |
| 3400 cm⁻¹ | Strong peak (water) | Almost no peak |
| 1640 cm⁻¹ | Medium peak (water) | Almost no peak |
| Main peak shape | flat-top | Normal bell shape |
| Baseline | Slanted | Flat |
| SNR | Poor | Good |

Case B: Liquid cell measurement

7. Demining checklist: Must-check before measurement

Before each measurement, check against the following demining checklist one by one [1][3][5][9]:

7.1 Instrument status

• [ ] Instrument has been on ≥ 30 min, source stable
• [ ] Desiccant effective (silica gel not turned pink)
• [ ] Instrument purged ≥ 15 min (if CO₂ and water vapor need to be subtracted)
• [ ] Detector pre-cooled (MCT detector needs liquid nitrogen)

7.2 Sample preparation

• [ ] KBr dried at 150°C vacuum overnight
• [ ] Sample ground to < 2 μm (like flour)
• [ ] Sample:KBr ratio = 1:100–1:200
• [ ] Pellet transparent, crack-free, uniform thickness (0.5–1 mm)
• [ ] Liquid cell path length appropriate (avoid flat-top)
• [ ] Liquid cell window material matches sample (use CaF₂ for aqueous)

7.3 ATR measurement

• [ ] Crystal wiped with ethanol 2–3 times
• [ ] Crystal clean before background collection (no impurity peaks)
• [ ] Sample amount moderate (1–5 mg or 1–2 drops)
• [ ] Pressure applied (manual tightening / pneumatic)
• [ ] Sample fully covers crystal center

7.4 Background collection

• [ ] Background collection conditions consistent with sample (temperature, humidity)
• [ ] Collect background with empty beam path (transmission) or clean crystal (ATR)
• [ ] Background spectrum no CO₂ doublet (2350), no water ghost peaks (3400, 1640)
• [ ] Interval between background and sample collection < 5 min

7.5 Data acquisition

• [ ] Parameter settings appropriate (resolution 4 cm⁻¹, scans 16–64)
• [ ] Spectral range coverage (4000–400 cm⁻¹)
• [ ] Maximum peak in sample spectrum < 1.5 (avoid flat-top)
• [ ] SNR > 1000:1 (weakest target peak identifiable)

💡 Professional advice: Before starting a batch of samples, first measure a quality check sample (e.g., polystyrene film) to confirm the instrument is working properly, then start measuring unknown samples [3][10].

8. Special requirements for quantitative sample preparation

If quantitative analysis (e.g., content determination) is to be performed, sample preparation requirements are higher [1][3][11]:

8.1 KBr pellet quantitative

• Internal standard method: Add a known amount of internal standard (e.g., KSCN at 2050 cm⁻¹) to eliminate pellet thickness error
• Weight method: Accurately weigh sample (accuracy 0.01 mg) and KBr (accuracy 0.1 mg)
• Peak area method: Use peak area instead of peak height to reduce particle size effects
• Multiple replicates: At least 3 pellets per sample, average

8.2 Liquid cell quantitative

• Precise path length measurement: Use interference fringe method to measure actual path length [1]
• Constant temperature: Temperature affects density, thus concentration
• Multiple injections: At least 3 replicates, RSD < 2%

8.3 ATR quantitative

• Constant pressure: Pneumatic pressure or fixed spring pressure
• Constant crystal temperature: Avoid thermal expansion affecting path length
• Consistent sample amount: Same amount each time (e.g., 2 mg)
• Relative intensity method: Use peak height ratio or peak area ratio to avoid influence of absolute path length

⚠️ Golden rule of quantitative analysis: All conditions must be reproducible. Any uncontrollable factors (e.g., manual pressing force) will introduce error [1][11].

9. Sample preparation tools and consumables list

9.1 Basic tools

• Agate mortar (diameter 50–75 mm)
• Stainless steel spatula
• Analytical balance (accuracy 0.01 mg)
• Pellet press (10–15 ton)
• KBr pellet die (13 mm, 7 mm)
• Vacuum pump (connect to die for vacuum)

9.2 Liquid cell consumables

• Demountable liquid cell + spacers of different thicknesses (25, 50, 100, 200, 500 μm)
• Fixed liquid cells (path length 0.1, 0.5, 1 mm)
• Syringe (1 mL, dedicated)
• Lens paper, absorbent cotton
• Solvents (ethanol, chloroform, acetone, ethyl acetate)

9.3 ATR consumables

• Absorbent cotton or lint-free wipes
• Isopropanol, ethanol (crystal cleaning)
• Sample spatula (PTFE or wooden, avoid scratching crystal)

9.4 Drying consumables

• Desiccator (with silica gel or P₂O₅)
• Vacuum drying oven (105°C + vacuum)
• Desiccants: silica gel, P₂O₅, CaCl₂

10. Objective evaluation of sample preparation quality

10.1 Spectrum quality check indicators

A good infrared spectrum should meet [1][3]:

Table 4: Objective indicators of spectrum quality (compiled from [1][3])

10.2 Quality assessment workflow

Acquire spectrum → Check main peak → Check SNR → Check baseline → Check water/CO₂ → Evaluate
   ↓        ↓          ↓          ↓            ↓           ↓
 1-2 min  0.3-1.5?  >1000:1?   <0.1?       <0.05?       Pass?

If any item fails, analyze the cause and reprepare the sample [1][3].

Summary of this section

Review questions

1. Novice Xiaoming made a KBr pellet for the first time, and the spectrum showed a large broad peak at 3400 cm⁻¹. List 3 possible reasons and explain how to identify which one it is.

2. You need to determine the ethanol content in a 5% ethanol-water solution. Design a sample preparation plan and explain why the KBr pellet method is not used.

3. Using ATR to measure a powder sample, the full spectrum signal is very weak (< 0.2) and reproducibility is poor. Analyze possible causes and give troubleshooting steps.

4. When pressing a pellet, the KBr disc is milky white and opaque, and the spectrum baseline is severely tilted. What is the problem? How to solve it?

5. A liquid sample was measured in a 0.5 mm liquid cell, and the C=O peak at 1700 cm⁻¹ was found to be flat-topped. Quantitative analysis: If the sample is a 5% acetone-CCl₄ solution (ε_C=O ≈ 500 L·mol⁻¹·cm⁻¹), what path length (in mm) should be used instead?

References

[1] Griffiths P R, de Haseth J A. Fourier Transform Infrared Spectrometry. 2nd ed. Wiley, 2007. Chapter 11–13. ISBN: 978-0-471-19404-0.

[2] Socrates G. Infrared and Raman Characteristic Group Frequencies. 3rd ed. Wiley, 2001. ISBN: 978-0470093078.

[3] Coates J. "Interpretation of Infrared Spectra, A Practical Approach." In: Encyclopedia of Analytical Chemistry, Meyers R A (Ed.). John Wiley & Sons, 2006. DOI:10.1002/9780470027318.a5607.

[4] Stimson M M, O'Donnell J F. "The Infrared Spectra of Some Solid Biological Compounds Using Potassium Bromide." Journal of the American Chemical Society, 1952, 74(7): 1805–1808. DOI:10.1021/ja01127a504.

[5] Smith B C. Fundamentals of Fourier Transform Infrared Spectroscopy. 2nd ed. CRC Press, 2011. Chapter 4. ISBN: 978-1420069297.

[6] Bohren C F, Huffman D R. Absorption and Scattering of Light by Small Particles. Wiley, 1998. Chapter 3. ISBN: 978-0471293408.

[7] Thermo Fisher Scientific. "Infrared Liquid Cells Selection Guide." Application Note AN-011.
https://www.thermofisher.com/…

[8] Max J J, Chapados C. "Infrared Spectroscopy of Aqueous Solutions: Influence of Water and Solute Concentration." Journal of Chemical Physics, 2009, 131(18): 184505. DOI:10.1063/1.3258320.

[9] Harrick Scientific. "ATR Sampling Techniques." Application Notes.
https://harricksci.com/applic…

[10] Bruker. "Sample Preparation for Infrared Spectroscopy." Technical Note.
https://www.bruker.com/en/pro…

[11] Chinese Pharmacopoeia Commission. Chinese Pharmacopoeia 2020 Edition, Part IV, General Rule 0402 Infrared Spectrophotometry. China Medical Science Press.

[12] ftir.fun. "Water Molecule Infrared Functional Groups."
https://ftir.fun/ir/group/wat…

Next Episode Preview: Ep 17 — Basic Instrument Operation Procedure (General FTIR)
We will start with power-on, purge, stabilization, and background acquisition, and systematically explain the basic operation procedure of FTIR instruments. This includes parameter settings (resolution, number of scans, spectral range), data file formats (SPA, OPUS, SP, JCAMP-DX), shutdown and daily maintenance, using Thermo Nicolet and Bruker Tensor as examples.

This article is licensed under CC BY-NC-SA 4.0. Images are from public domain or web sources with credited attribution; copyright belongs to the original authors.