Ion Chromatography (IC)
Quality & Inspection Standards
PRODUCTS CENTER
Get Custom Quote
PRODUCTS CENTER
Ion Chromatography (IC) – PCB Ionic Contamination Analysis, Method 2.3.28 & IPC Limits
Ion chromatography (IC) is the only test method that tells you what ionic residues are on your PCB, not just how much. ROSE gives you a single number (NaCl equivalent). IC separates and quantifies individual anions (chloride, bromide, sulfate, fluoride) and cations (sodium, potassium, ammonium, calcium). This specificity is critical for root cause analysis. We have seen boards pass ROSE at 0.8 µg/cm² but fail SIR due to high bromide from a poorly cured no‑clean flux. ROSE missed it; IC caught it. Per IPC‑6012E §3.9.1, IC is the recommended method for verification when ROSE results are borderline or when specific ion limits are required. This guide covers IC sample preparation, calibration, data interpretation, and how to use IC in high‑reliability PCB qualification.
Table of Contents
- Why IC – Beyond ROSE: Specific Ion Identification
- Sample Preparation – Extraction, Solvents & Blank Control
- Calibration – Multi‑Point Standards & Linearity Requirements
- Analysis Procedure – Anions and Cations
- Interpreting Results – Typical Ion Profiles & Failure Modes
- Acceptance Criteria – IPC Limits & OEM Specifications
- IC Testing Checklist
- FAQ
1. Why IC – Beyond ROSE: Specific Ion Identification
ROSE (resistivity of solvent extract) measures total conductivity and reports as NaCl equivalent. But not all ions are equally harmful. For example:
- Chloride (Cl⁻): Highly corrosive, promotes electrochemical migration. Limit typically <0.5 µg/cm².
- Bromide (Br⁻): Often from no‑clean flux activators or brominated flame retardants. Can cause CAF even at low levels.
- Sulfate (SO₄²⁻): Less aggressive but can form insoluble precipitates.
- Organic acids (acetate, formate, succinate): Weakly ionic, may not be fully detected by ROSE but can cause corrosion under humidity.
- Ammonium (NH₄⁺): Can react with chloride to form corrosive residues.
IC separates these ions by passing the extraction solution through a column that retains different ions at different rates. A conductivity detector measures each ion as it elutes. The result is a chromatogram with peaks corresponding to specific ions.
In one medical device program, boards passed ROSE at 1.2 µg/cm² but caused intermittent failures in high‑humidity testing. IC revealed chloride at 1.1 µg/cm² – the ROSE reading was dominated by chloride, which is far more aggressive than the NaCl equivalent suggests. The cleaning process was adjusted, and chloride dropped to 0.2 µg/cm². Failures stopped. ROSE would never have identified the root cause.
2. Sample Preparation – Extraction, Solvents & Blank Control
Per IPC‑TM‑650 2.3.28 (for general IC) and 2.3.28.2 (for bare boards), sample preparation is critical. The standard extraction uses a sealed bag (like ROSE) but with a different solvent ratio.
- For bare boards (2.3.28.2): 10/90 v/v IPA/DI water (10% isopropanol, 90% deionized water). Extraction at 80±2°C for 60 minutes.
- For assembled boards or general IC (2.3.28): 75/25 IPA/DI water is also permitted, but 10/90 is preferred for better ion solubility.
- Blank sample: Always extract an empty bag under identical conditions. Subtract blank values from board results.
- Surface area calculation: Same as ROSE: length × width × 2 for bare boards. For assemblies, estimate component contribution.
Critical: Use ultra‑pure DI water (≥18 MΩ·cm). Contaminated water adds ions (e.g., chloride from tap water) and invalidates results. We have seen labs use 16 MΩ·cm water and measure 0.1 µg/cm² chloride in the blank – enough to cause false failures for sensitive products.
After extraction, the solution is filtered (0.45 µm) to remove particles that could clog the IC column. The sample must be analyzed within 4 days, stored refrigerated.
3. Calibration – Multi‑Point Standards & Linearity Requirements
IC calibration requires a set of standard solutions containing known concentrations of target ions. Per IPC‑TM‑650 2.3.28.2 §5.2, a minimum of three calibration points per ion is required, covering the expected range (typically 0.05 to 5 ppm). The correlation coefficient R² must be ≥0.98.
Typical anion standard mixture (per 2.3.28.2):
- Fluoride (F⁻): 0.05‑0.2 ppm
- Acetate (CH₃COO⁻): 0.25‑1.0 ppm
- Formate (HCOO⁻): 0.5‑2.0 ppm
- Chloride (Cl⁻): 0.5‑2.0 ppm
- Bromide (Br⁻): 0.5‑2.0 ppm
- Nitrate (NO₃⁻): 0.2‑0.8 ppm
- Sulfate (SO₄²⁻): 0.2‑0.8 ppm
- Phosphate (PO₄³⁻): 0.2‑0.8 ppm
Cation standards: Lithium (0.05‑0.2 ppm), sodium (0.5‑2 ppm), ammonium (0.5‑2 ppm), potassium (0.5‑2 ppm), magnesium (0.5‑2 ppm), calcium (0.5‑2 ppm).
Prepare fresh daily. Organic acids (acetate, formate) degrade quickly – do not store standards for more than 24 hours.
In one lab, we found that the acetate peak was missing because the standard solution was a week old. The lab had been reporting "acetate not detected" while actual boards had high acetate residues. Switching to daily standards resolved the issue.
4. Analysis Procedure – Anions and Cations
IC analysis typically runs in two modes: anion (using a hydroxide or carbonate/bicarbonate eluent) and cation (using methanesulfonic acid eluent). Each run takes 15‑30 minutes per sample.
Key parameters per IPC‑TM‑650 2.3.41 (similar principle):
- Flow rate: 1.0‑2.0 mL/min
- Injection volume: 25‑100 µL
- Column temperature: ambient or controlled (typically 30‑40°C)
- Suppressor: required to reduce background conductivity (e.g., anion self‑regenerating suppressor)
After each sample, a "rinse" or "regeneration" step ensures no carryover. For high‑concentration samples (e.g., >10 ppm), dilute and re‑run – column contamination can take hours to flush.
Example chromatogram interpretation: A typical no‑clean flux residue shows a large peak at retention time corresponding to chloride (often 0.5‑2 ppm) and sometimes bromide. If acetate or formate peaks appear, the flux may be under‑cured. If sulfate appears, check for acidic cleaner residues.
5. Interpreting Results – Typical Ion Profiles & Failure Modes
The IC result is reported in µg/cm² for each ion. Typical limits (industry practice, not always in IPC):
| Ion | Typical Limit (µg/cm²) | Failure Mode if Exceeded | Common Source |
|---|---|---|---|
| Chloride (Cl⁻) | ≤0.5 | Corrosion, CAF, dendrites | Flux activators, tap water, fingerprints |
| Bromide (Br⁻) | ≤0.5 | CAF, corrosion (less than Cl⁻) | Flux, brominated flame retardants |
| Sulfate (SO₄²⁻) | ≤1.0 | Less corrosive, but forms white deposits | Cleaner residues, paper handling |
| Fluoride (F⁻) | ≤0.2 | Glass etching, corrosion | Flux, from some cleaning chemistries |
| Acetate / Formate | ≤1.0 | Mild corrosion, SIR reduction | Under‑cured no‑clean flux |
| Sodium / Potassium | ≤0.5 | Less aggressive but can combine with halides | Fingerprints, residues |
| Ammonium (NH₄⁺) | ≤0.5 | Corrosive with chloride | Flux activators |
Table 1: Typical IC limits. Not in IPC‑6012E; defined by OEMs based on reliability requirements.
In an aerospace program, IC revealed chloride at 0.8 µg/cm² and bromide at 1.2 µg/cm². The ROSE reading was 1.4 µg/cm² – below the 1.56 limit, so the board would have passed ROSE. But the customer's internal spec required chloride <0.3 and bromide <0.5. The lot was rejected. The supplier switched to a different flux and cleaning process, and both ions dropped below 0.2 µg/cm².
6. Acceptance Criteria – IPC Limits & OEM Specifications
IPC‑6012E does not specify numerical limits for individual ions. It only references the cleanliness requirement (≤1.56 µg/cm² NaCl equivalent) and recommends IC for verification. However, many OEMs have their own limits:
- Telecom / server (major OEMs): Chloride <0.5 µg/cm², Bromide <0.5 µg/cm², Sulfate <1.0 µg/cm².
- Aerospace (IPC‑6012ES): Chloride <0.2 µg/cm², total halides (Cl+Br+F) <1.0 µg/cm².
- Medical implantable: Chloride <0.1 µg/cm², often require IC on every lot.
- Automotive (high‑reliability): Chloride <0.3 µg/cm², no individual ion >0.5 µg/cm² except possibly sulfate.
When to use IC instead of ROSE:
- ROSE result is borderline (1.2‑1.8 µg/cm²).
- Field failures with suspected ionic contamination (SIR, CAF, corrosion).
- Qualification of a new flux or cleaning process.
- Customer specification requires specific ion limits.
- For assembled boards where component bodies may trap residues that ROSE cannot fully extract.
In one automotive program, the customer required both ROSE (<1.56) and IC (chloride <0.3, bromide <0.3). The supplier passed ROSE at 1.2 but failed IC with chloride 0.6. The root cause was poor rinsing after cleaning. IC saved a potential field failure.
Ion Chromatography Testing Checklist
- Specify IC method and ion limits in PO. Do not rely on IPC default (none exist).
- Use fresh, ultra‑pure DI water (≥18 MΩ·cm) for extraction. Test the blank before each run.
- Prepare fresh standards daily (especially for organic acids). Verify R² ≥0.98.
- Extract at 80±2°C for 60 minutes (bare boards) or as specified.
- Analyze within 4 days; store samples refrigerated.
- Subtract blank values from sample results. Blank should be ≤0.05 µg/cm² for all ions.
- Report results in µg/cm² for each ion detected. Include detection limits.
- If ions exceed limits, perform root cause analysis (flux, cleaning, handling).
7. FAQ
Q: What is the advantage of IC over ROSE?
A: IC identifies specific ions and their concentrations. ROSE only gives a total conductivity equivalent to NaCl. IC is required for root cause analysis and for verifying that harmful ions (e.g., chloride, bromide) are within acceptable limits. IC is also more sensitive (ppb vs ppm).
Q: How much does IC testing cost compared to ROSE?
A: ROSE testing is $20‑$50 per panel. IC is $150‑$300 per sample (anions + cations). However, IC is typically run on a sampling basis (e.g., first article, quarterly validation), not on every lot. The cost is justified for high‑reliability programs.
Q: Can IC detect non‑ionic residues like rosin?
A: No. IC only detects ions that dissociate in water. For non‑ionic residues (rosin, oils, particles), use other methods like visual inspection, FTIR, or ion chromatography with suppressed conductivity is not effective. However, rosin itself is not corrosive – it is the ionic activators in the rosin that cause problems.
Q: What is a typical chloride limit for Class 3 PCBs?
A: IPC does not specify. However, many OEMs require chloride <0.5 µg/cm². For aerospace and medical, limits can be as low as 0.1‑0.2 µg/cm². Always confirm with your customer's specification.
Continue Reading – Cleanliness & Contamination Series
Need ion chromatography analysis for your high‑reliability PCBs? Our lab offers full IC services per IPC‑TM‑650 2.3.28, with rapid turnaround and expert interpretation.
Request IC Analysis →References: IPC‑TM‑650 2.3.28, 2.3.28.2; IPC‑6012E §3.9.1; IPC‑WP‑008 (Setting Up IC Capability). IPC‑TM‑650 Official Method.
Get Custom Quote
PRODUCTS CENTER


