How Are Migration Levels in Plastic Packaging Actually Calculated: 4 Common Migration Tests + Which Ones Food / Cosmetics / Toy Companies Should Test

A sauce maker in Suzhou sent a batch of composite pouches for GB 4806.7 testing. When the report came back, the migration column read "not detected." He was happy, figured everything was fine, and put the product on the shelf right away.

Three months later, a regulatory spot-check found that the executive standard printed on the label was wrong. The "not detected" result in the migration column was actually tested using a 10% ethanol simulant — but his product was high-salt soy sauce. Under GB 5009.156, the simulant should have been 4% acetic acid. The "not detected" on the report was real, but the "not detected" in the correct simulant column had never been tested at all.

This is not an isolated case. Over the past year we have seen at least 7 factories pull out a migration report stamped with a CMA seal — only to discover the wrong simulant had actually been used. It's an unfortunate mistake: the report is genuine, the standard is real, but the customer's product category (high-oil / high-salt / acidic / alcoholic) doesn't match the report.

Migration Isn't a Single Test — It's a Combination of 4

Many factory owners hear "migration test" and think, "Oh, they're just measuring a content level." In fact, it's a combination test — and 4 variables all have to be correct: simulant type, contact temperature, contact time, and contact area.

Variable 1: Simulant

There isn't just one simulant — there are 4 categories:

• 10% ethanol — simulates alcoholic food contact
• 4% acetic acid — simulates acidic foods (vinegar, soy sauce, jam, juice)
• Vegetable oil — simulates high-fat foods (pastries, nuts, chocolate, fried foods)
• Distilled water — simulates neutral foods (rice, noodles, freshwater fish, infant complementary foods)

There's also a "chemical alternative solvent" — 95% ethanol or isooctane. Because actual oil simulant tests run painfully slowly (10 days), many labs substitute 95% ethanol. Customers need to check the report carefully. "95% ethanol substitute" is not the same as "oil simulant" — whether it matches the customer's actual product needs separate confirmation.

Variable 2: Temperature

Migration temperatures range from 4°C (refrigerated) to 100°C (boiling/steam cooking) to 121°C (retort pouches) to 175°C (oven applications). Each temperature corresponds to a different food scenario.

Variable 3: Time

Contact times range from 10 minutes (instant contact, e.g., hot-drink cups), 2 hours (short-term), 24 hours (room temperature), 10 days (room-temperature long-term simulation), to 30 days (room-temperature long shelf life). 10-day and 30-day tests are usually conducted under accelerated conditions at 40°C.

Variable 4: Contact Area and Ratio

GB 5009.156 specifies that contact area be calculated based on actual use area. But lab sample area is limited, so the test uses an "area / simulant volume" ratio — typically 6 dm² / kg of food or 6 dm² / L of simulant. If this ratio isn't stated in the report, ask about it.

Which Combinations the 4 Types of Companies Should Test

Case 1: Fried-chip inner pouch — A customer went 2 months thinking everything was fine. Then oil in the bag migrated into the ink layer, creating off-odors, and only then did they investigate. This product category: vegetable oil simulant + 40°C + 10 days + 6 dm²/kg. Focus the report review on "total migration" and "aromatic amine migration."

Case 2: Infant complementary food packaging — A new brand making rice-cereal packaging submitted samples; the report only tested 10% ethanol and distilled water, and skipped 4% acetic acid. Reason: the cereal contained tomato powder (acidic). These products must run all 4 simulants, because GB 4806.1 tightens limits for infant products (factor of 0.5).

Case 3: Cosmetics packaging (plastic bottle/jar) — Cosmetics don't fall under the GB 4806 system; they follow GB/T 29680 (shampoo/body wash) and EC 1223/2009 (EU export). The migration test differs, but the principle is similar. For the lotion bottle submitted, the contact surface at the bottle opening is the thread; the migration test's "contact area" should be calculated from the inner wall of the bottle mouth, not the whole bottle. The inner liner of the cap also needs to be tested separately.

Case 4: Children's toy packaging (plastic bag containing toys) — Toys themselves have EN 71-3 heavy-metal migration; the packaging bag is a separate matter. If the bag is PVC, EU 2019/904 (Single-Use Plastics Directive) has specific requirements for plastic packaging in children's contact. Migration isn't the core concern for toy bags — the core concern is the 3 phthalates (DEHP/DBP/BBP) total content. This report needs to be done by a specialized laboratory.

3 Types of Common Pitfalls

Pitfall 1: Only testing overall migration, not specific migration

Overall Migration Limit (OML) is "the total of everything that could come out" — a macro-level conclusion reflecting "did anything come out at all." But Specific Migration Limit (SML) tells you "what came out, and is it toxic." You need both.

Some cheap tests only measure OML — a few hundred bucks and done. Single-item SML tests cost more; for example, a single heavy-metal migration test runs 1,500–3,000 RMB. Customers try to save money by only testing OML, and only discover too late that they have no data.

Pitfall 2: Temperature tested at "room temperature 25°C" — but the product is actually filled at 80°C

A jam factory using PET jars fills at 80°C, but the submitted report used 25°C / 10 days. Data tested this way reflects "shelf life" — not "the moment of filling." The GB 4806 series explicitly requires testing under the "most severe contact conditions." An 80°C fill needs to be tested at 100°C / 2 hours or the corresponding reflux time.

Pitfall 3: Third-party lab methods aren't updated

GB 4806.7 was revised in 2024, and some test methods changed. One customer kept using a report from 2023 through 2024, but the methods in the report were already the old version. We recommend customers re-test every 18–24 months, or at least have the lab confirm whether the currently effective version is still being used.

What to Prepare Before Sending Samples

Before sending samples, the customer must clearly tell the lab 4 things:

1. Product type: food category (oily / aqueous / acidic / alcoholic / infant / special diet)
2. Contact temperature: filling temperature, storage temperature, use temperature
3. Contact duration: how long at room temp, how long refrigerated, how long heated
4. Contact area: the whole bag, or just the inner wall (inner wall vs. outer wall of a bottle — migration data is completely different)

If a lab starts the test without asking these questions, it's time to switch labs.

How LeXiang Packaging Handles This

Starting last year, when a customer comes in for a quote, our sales manager asks 5 questions: "Product type + filling temperature + storage conditions + whether it contains oil / alcohol / acid." If the customer can't answer clearly, we don't quote first — instead we send a Migration Test Requirements Confirmation Form for the customer's product manager or QA to complete.

This one step lifted our migration report first-pass approval rate from 60% to over 90%. Customers didn't understand at first why we were asking all this — but later, during regulatory inspections, our reports passed directly while competitors couldn't produce matching reports.

At its core, migration testing is about "whether the report matches your product." The "not detected" on the report is real — but it's "not detected" under 4 specific test conditions. When customers get a report, they should first check whether those 4 variables match their product, then look at the results. That order cannot be reversed.

Further reading:

• GB 4806.1 to 4806.11 Demystified: How to Actually Use the 5 Core Standards for Food Contact Packaging
• GB 4806 Testing Projects for Food Contact Packaging in Practice: 5 Commonly Failed Items + 3 Remediation Paths
• 4 EU Compliance Requirements for Cosmetics Packaging: REACH / EC 1223/2009 / 94/62/EC / EN 13427 Explained
• Dual-Track Compliance for Children's Product Packaging: CPSIA and EN 71-3 — Who Should Issue the Ink Heavy-Metal Report?