Contact angle measurement in food packaging

Five datasets follow. The first four are published or manufacturer-published measurements and are cited in place; the fifth is a synthetic capability study, labelled as such. Together they cover the questions a packaging technologist actually asks: how far the substrate is from the process, what treatment does to the angle, what it does to the seal, whether the dyne number predicts adhesion, and how long the effect lasts.

The natural surface energy of commodity packaging polymers sits far below what any printing, coating or laminating process requires. The values below are the ones used in equipment specification for corona and plasma treaters [2].

Material or process | Surface energy (mN/m) | Comment

PP, untreated | 29 | Natural dyne level of the base resin

PE-LD, untreated | 31 | Natural dyne level

PE-HD, untreated | 32 | Natural dyne level

BOPP, untreated | 32–33 | Natural dyne level; the most commonly treated film

Printing, solvent-based ink | 40–42 | Requirement on the treated side

Printing, water-based ink | 46–48 | Higher because the ink itself is more polar

Coating (functional, barrier, heat-seal lacquer) | 44–54 | Range spans the coating chemistry

Lamination (adhesive wet-out) | 46–56 | Highest requirement; solventless systems at the top

Table 5 — Natural surface energies and process requirements for packaging films, after Vetaphone [2]. The gap between the supply rows and the demand rows is what the treater has to close, and it is typically 10–25 mN/m.

Figure 2 — The same data as a gap chart. No untreated polyolefin reaches even the least demanding process; the treater has to add between 8 and 24 mN/m depending on the application.

Two consequences follow for specification. First, because the gap is large compared with the measurement uncertainty, a treated-versus-untreated check is robust even with a dyne pen — this is why the pen survives in an industry that otherwise measures everything. Second, because the gap differs by process, a single "treated" specification is meaningless: a film treated to 38 mN/m is adequate for some solvent inks, marginal for coatings, and unacceptable for solventless lamination.

Published data: six commercial packaging films treated on a laboratory corona unit at two power levels (100 W and 300 W, approximately 99 and 297 W·min/m²), with static water contact angle measured by drop-shape analysis (2 μL, 5 s after deposition, six replicates) and surface free energy calculated by OWRK from water and diiodomethane [1].

Film | Untreated | 100 W | 300 W

PE | 93.9 ± 3.1° | 62.0 ± 4.5° | 55.7 ± 2.2°

OPP | 100.7 ± 2.2° | 53.1 ± 3.4° | 56.8 ± 2.6°

Table 6 — Static water contact angle versus corona power, mean ± SD, n = 6, after Żołek-Tryznowska et al. [1]. The same study reports a minimum at 100 W followed by a rise at 300 W for the other four films (OPP2, BOPP, BOPP2 and PET).

Figure 3 — The two behaviours visible in one plot. PE keeps improving up to 300 W, while OPP reaches its minimum at 100 W and then reverses — the signature of over-treatment.

The PE curve is the textbook result: 94° falling to 56° as polar groups are incorporated. The OPP curve is the more instructive one. It falls to 53° at 100 W, then rises to 57° at 300 W, and the authors attribute the reversal to surface degradation and crosslinking in the uppermost monolayer rather than to a loss of oxygen. Total surface free energy still rises — from about 29 mN/m untreated to more than 52 mN/m at 300 W — but the wetting behaviour no longer follows it, because the layer the drop touches has been mechanically and chemically altered rather than simply oxidised.

What this means for a specification

A one-sided specification ("angle below 65°") cannot detect over-treatment, and over-treatment is the failure mode that damages seals. Write the window on both sides: for OPP-based films, 45–65° measured at 2 s is a defensible starting window, with 40° as the lower alarm. The same dataset shows why "more corona is better" is a dangerous operating habit: the treater can be turned up until the angle stops improving and the seal is already compromised. The seal test, not the angle, is the arbiter.

The same study measured seal strength after heat sealing the treated faces together (120 °C for PE, 130 °C for PP, 5 s, 2 mm bar) and T-peel testing after seven days [1]. The result is the most important caution in this note.

Film | Untreated (N/mm) | 100 W (N/mm) | 300 W (N/mm) | Direction

PE | 0.549 ± 0.080 | 0.609 ± 0.073 | 0.513 ± 0.034 | Up at 100 W, down at 300 W

BOPP | 0.231 ± 0.068 | 0.267 ± 0.041 | 0.202 ± 0.034 | Up at 100 W, down at 300 W

Table 7 — Seal strength versus corona power, mean ± SD, after Żołek-Tryznowska et al. [1]. Polypropylene-based films other than this grade lost strength at every treatment level: the study reports −29 % and −23 % for two OPP grades at 100 W, and −33 % for the most sensitive BOPP grade at 300 W.

Three findings matter. First, the effect is material-dependent: PE and one BOPP grade gain strength at the lower power, while the other polypropylene films lose it at every level. Second, the effect is non-monotonic, so an optimum exists and can be found. Third, and most important for anyone hoping to use wettability as a proxy: across the six films and three treatment levels, the correlation between surface free energy and seal strength was r = −0.186 (Pearson) and ρ = 0.109 (Spearman). Surface energy did not predict seal strength, and if anything the relationship ran slightly the wrong way.

The physical reason is that sealing is a rheological process. What matters is how the seal layer flows, wets and interdiffuses under heat and pressure — governed by molecular weight, comonomer content, the presence of a low-melting sealing layer, and the thickness and brittleness of any oxidised skin. A heavily oxidised surface is a barrier between two polymer melts, not a help to them. That is why the seal test cannot be replaced by a wetting measurement, and why an over-treated film can pass the dyne pen and fail the seal.

Manufacturer data: BOPP treated to the same 40 dyne/cm by four different methods, then peel tested with two tapes against an untreated control [2]. The dyne reading is identical in four of the five rows. The adhesion is not.

Treatment | Dyne level | Peel, Tesa 7445 (N/m) | Peel, 3M 610 (N/m)

Untreated | 30 | 281 | 162

Corona | 40 | 321 | 254

Plasma | 40 | 369 | 284

Plasma, advanced gas 1 | 40 | 359 | 301

Plasma, advanced gas 2 | 40 | 358 | 292

Table 8 — Peel strength on BOPP at equal dyne level, after Vetaphone [2]. Tapes differ by a factor of 1.2–1.4 on the same surface, which is why peel results are only comparable within one tape and one test method.

The lesson is that a dyne level is a threshold, not a predictor. It tells you that a liquid of that surface tension will wet; it says nothing about the chemistry of the layer that will form the bond. Plasma in a controlled atmosphere deposits amine, amide and imide functionality that corona in air cannot, and those groups react with the adhesive or ink. Two surfaces can therefore wet identically and bond differently.

Figure 4 — The same data plotted. Between corona and plasma at an identical 40 dyne, peel rises by 15 % (Tesa) and 12 % (3M); between the two plasma recipes the two tapes rank the surfaces differently.

The same source reports a second, related effect: reaching 42 dyne on PE required twice the dose with plasma (20 W·min/m²) as with corona (10 W·min/m²), yet the plasma-treated film delivered higher peel (291–376 N/m against 265–323 N/m). Energy per square metre, not dyne level, is the process variable that should be logged.

Corona treatment decays. The mechanism is a combination of reorientation of polar groups away from the surface, migration of low-molecular-weight species from the bulk, and reaction of the radicals left by the discharge. Manufacturer data for BOPP illustrates the practical range [2].

Treatment | Initial (dyne/cm) | After storage (dyne/cm) | Behaviour

Corona | ≈ 47 | ≈ 37 | Falls quickly, then plateaus below the print threshold

Plasma, medium power | ≈ 57 | ≈ 47 | Falls, then stable for many months

Plasma, high power | ≈ 60 | ≈ 60 | Stable for more than a year in the reported case

Table 9 — Ageing of BOPP treatment, after Vetaphone [2]. The decay of a corona-treated surface can take it from comfortably printable to marginal within weeks.

The operational consequence is a shelf-life limit that must be written into the specification. Two practices make it manageable. Measure and record the angle at a defined age (commonly 24 h after treatment, once the initial reorientation has settled), and set a second, lower limit for the age at which the film will actually be printed. A film specified at 55–65° at 24 h and printed four weeks later needs a demonstrated value at four weeks, not an assumption.

For recycled content the decay is often faster and more variable, because the recyclate brings its own low-molecular-weight species and surface-active contaminants. Section 8.3 treats this case.

Anti-fog performance is the one application where the relevant angle is the receding angle, and where the target is set by appearance rather than by adhesion. When water condenses on a surface with a high receding angle, the droplets pin and grow until they scatter light; when the receding angle is low, the condensate retracts and sheets into a continuous film that remains transparent.

The targets below are engineering rules derived from that mechanism, not values taken from a single published study. They should be validated on the finished pack with a hot-fog test: film suspended over a water bath at 60 °C, assessed at 20 cm for the time to first fog and for droplet size, at 4 °C and at 23 °C.

Property | Target | Why it matters

Static water angle | < 20° | Indicates a hydrophilic, high-energy surface

Receding angle | < 20° (ideally < 10°) | Governs whether condensate retracts or pins

Hysteresis | < 15° | Large hysteresis means droplets pin even at low static angle

Angle after 7 days at 23 °C | Within 5° of day 1 | Allows for additive bloom; checks the additive is still available

Angle after 24 h at 4 °C | Still < 25° | Cold storage is the service condition for chilled retail

Hot-fog test | Clear film, no droplets > 1 mm at 60 s | Functional confirmation on the finished pack

Table 10 — Anti-fog design targets. These are engineering rules, not literature values; they must be confirmed with the hot-fog test on the actual pack and product.

Two practical notes. Anti-fog additives work by blooming to the surface, so the angle improves over the first days after extrusion and then degrades as the additive is consumed or washed away — measure at two ages, not one. And because the additive is a surfactant, it can interfere with printing and with seal strength if it reaches the outside face; the treated side and the anti-fog side are usually different faces, and confusing them is a common source of unexplained print defects.

The dataset in this section is synthetic. It demonstrates how contact angle data become a release decision for a printed film; the numbers are not measurements of any real product. Specification: 60° ± 10° on the treated side, measured at 2 s, n = 10 per lot, three web positions.

Lot | Mean (°) | SD (°) | Cpk | Decision | Comment

A | 58.2 | 2.4 | 1.36 | Release | Centred; normal production

B | 61.9 | 3.1 | 0.87 | Release, monitor | Drifting toward the upper limit

C | 64.8 | 4.2 | 0.38 | Investigate | High mean and high spread: treater or additive issue

D | 52.1 | 3.8 | 0.69 | Release, monitor | Low mean: approaching over-treatment

E | 59.4 | 1.9 | 1.41 | Release | Recovered after electrode cleaning

Table 11 — Illustrative capability study. Cpk = min(USL − mean, mean − LSL) / 3σ, with LSL 50° and USL 70°.

Lot D is the interesting one. Its mean is comfortably inside the specification, so a pass/fail check would release it — yet at 52° it is close to the over-treatment boundary where seal strength begins to suffer, and its spread of 3.8° suggests the treater is not stable. A capability index computed against a two-sided window catches this; a single-sided limit does not.

A three-layer BOPP laminate is corona treated on the print side, printed flexo with a solvent-based ink, and examined for ink adhesion by tape test at the end of the line. The historical control is a dyne pen at the treater exit, read once per roll by the operator.

The failure that motivates a change is intermittent: one or two rolls per shift fail the tape test, always toward the drive side, and always on rolls that have been in storage for more than three weeks. A pen reading taken at the treater cannot see a decay that happens in the warehouse.

The fix is a two-point protocol. At the treater, the pen remains the operator's check. In the laboratory, a water angle is measured at 2 s on three web positions, on a sample retained from each roll, at day 1 and at the age of actual printing. The specification is written as 45–65° at day 1 and 50–70° at the age of use, with the edge positions required to be within 8° of the centre. The tape test is retained as the functional confirmation, run at the same frequency as before. Within a quarter the intermittent failures were traceable to two causes: an electrode that had drifted out of gap on the drive side, and a batch of film whose slip agent bloomed faster than the standard grade.

A PET/PE lidding film for tomato punnets carries an anti-fog masterbatch in the sealing layer. The customer complaint is condensation obscuring the fruit in chilled display, on packs that passed the supplier's incoming checks.

The incoming check is a static angle on the outside face — the wrong face, measured in the wrong condition. The relevant measurement is the receding angle on the inside face, immersed, after the film has been stored long enough for the additive to bloom. A captive-bubble measurement at 20 °C, after 7 days at 23 °C and after 24 h at 4 °C, separates good from bad lots where the static angle does not.

The corrective action has three parts: specify the inside face and the receding angle; add a hot-fog test on the finished pack as the functional gate; and control the masterbatch loading by lot, because the additive is consumed at the surface and a low loading passes on day 1 and fails on day 30. The static angle remains useful as a quick check, but only as a screen.

A mono-PE laminate for a dry-food pouch uses 30–50 % post-consumer recyclate in the core layer. Print adhesion and seal strength both become more variable, and the customer asks for a single incoming test that predicts both.

No single test does. What works is a small set: water angle on the treated print face (predicts ink wetting), seal strength on a laboratory sealer (predicts the seal), and a visual check for gels and contamination. The recyclate raises the angle variability because it carries surface-active residues and because its molecular weight distribution broadens the bloom behaviour; the angle trend by lot is therefore a useful early warning, while the seal test remains the release criterion.

The specification that emerges is asymmetric: an upper limit on the angle (printability) and a lower limit set by the seal result rather than by the angle, because the correlation between the two is weak (Section 7.3). Where a supplier proposes a single number, the right question is which failure it is intended to prevent.

Symptom | Probable cause | Corrective action

Angle rises steadily over days with no process change | Additive bloom; reorientation of polar groups | Fix the sample age in the method; measure at day 1 and at the age of use

Every sample reads 90–105° | Untreated face measured | Confirm the side with a dyne pen or water-break test; mark the film

Drop jumps or sits asymmetrically | Static charge on corona-treated film | Ionise for 2 s; ground the stage; repeat the measurement

Angle varies by more than 10° across the web | Treater gap or electrode wear; edge effects | Map drive side, centre and operator side; service the electrode

Dyne pen says pass, tape test fails | Equal dyne, different surface chemistry (Section 7.4) | Add a peel test; specify the treatment method as well as the level

Seal strength falls while the angle looks ideal | Over-treatment: brittle oxidised skin on the seal layer | Set a lower angle limit; reduce power; re-run the seal test

Values fall continuously during capture | Evaporation from a 2 μL drop; surfactant migration | Capture at 2.0 s; enclose the sample; use 4 μL

Anti-fog film fogs although the static angle is low | High receding angle; additive not bloomed or depleted | Measure receding angle or captive bubble; run a hot-fog test

Results differ from the film supplier's certificate | Different method: dyne versus angle, or different dwell time | Compare method statements; re-measure both samples in one laboratory

Sessile drop disappears on paper or board | Absorption into a porous substrate | Use Washburn or a captive-bubble configuration

Table 12 — Diagnostic table for packaging substrates. The first four rows account for most of the disagreement between a converter's laboratory and a film supplier's certificate.

Three rules make a contact angle specification workable in a converting plant. Set the window at least six times the expanded uncertainty wide — with U = 3.4°, that is about 20° for single-drop testing or 7° if ten drops are averaged. Make the window two-sided, because over-treatment is a real failure mode (Section 7.2). And anchor the target to the process liquid rather than to a generic "treated" state: the requirement for solventless lamination is not the requirement for a water-based ink.

Where the customer insists on a dyne specification, keep it — but correlate it locally. Measure both on twenty production rolls, plot angle against dyne, and record the relationship. It will not be a constant offset, and the exercise is the evidence that the two specifications describe the same process.

Treat the method like any other test in a quality system: accuracy against a reference film, precision by repeated measurement of one film, repeatability within operator, reproducibility between operators, and robustness against drop volume, capture time, humidity and fit model. A Gage R&R with ten samples, three operators and three replicates is the standard design; a gauge contribution below 20 % of total variation is generally acceptable, and 20–30 % acceptable with justification.

One packaging-specific validation is worth adding: the same-roll repeatability across a week. Cut ten samples from one roll, measure five on day 1 and five on day 7, and record the drift. If the drift is comparable to the specification window, the method is measuring ageing rather than treatment, and the sampling plan needs to change.

Standard | Subject | Use it for

ISO 19403-2 | Wettability — surface free energy of solid surfaces from contact angle | SFE reporting; liquid set and model

ASTM D7334 | Surface wettability by advancing contact angle | Advancing angle on flat substrates

ASTM D5946 / ISO 15989 | Water contact angle of corona-treated polymer film | Incoming and in-process checks on film and web

ASTM D2578 / ISO 8296 | Wetting tension of polyethylene and polypropylene film | Dyne-pen method; the contractual language

ASTM D7541 | Critical surface tension by the Zisman method | Ranking low-energy surfaces with a liquid series

ASTM D8597 | Contact angle with portable goniometers | Line-side measurement on large or installed parts

Table 13 — Method standards. Cite the one that matches the measurement actually made: a D2578 wetting-tension result is not an ISO 19403-2 surface free energy, and the two are not interchangeable in a specification.

Contact angle sits beside the food-contact framework, not inside it. The relevant instruments in the European Union are the framework Regulation (EC) No 1935/2004, the plastics Regulation (EU) No 10/2011 as amended — including by (EU) 2020/1245 — and the good manufacturing practice Regulation (EC) No 2023/2006. Compliance with 10/2011 is demonstrated by composition checks, by specific migration testing against the substance list, and by overall migration testing against a limit of 10 mg/dm² of surface area (conventionally 60 mg/kg of food) using the food simulants defined in the regulation.

A contact angle measurement has no role in those determinations. What it can do is support the process validation around them: demonstrating that a coating is continuous, that a surface treatment is stable over shelf life, or that a change of supplier has not altered the interface. Written that way in a design or validation file, it is useful evidence. Written as a compliance claim, it is not defensible.

In the United States the corresponding instruments are the Federal Food, Drug and Cosmetic Act and the Title 21 CFR parts covering adhesives (175), coatings (175.300) and polymers (177), with the appropriate food-contact notifications where required. The same distinction applies: wettability is a process characteristic, not a regulatory status.

Contact angle reports the outermost layer, averaged over the drop footprint, with no chemical specificity and no mechanical information. Add the following when the question moves beyond wetting:

• Peel or bond strength whenever the claim is adhesion — always, if the answer will be used to release product.

• Seal strength and hot-tack whenever the claim involves sealing, since Section 7.3 shows the two are uncorrelated.

• OTR and WVTR whenever the claim is barrier performance.

• XPS or ToF-SIMS when the question is what the treatment actually deposited, particularly when comparing corona with plasma or two plasma recipes.

• Migration testing whenever the claim is food-contact compliance.

A contact angle result is only as useful as its method statement. The following items should appear in any report, specification or supplier certificate that quotes one.

• Angle type — static, advancing, receding or captive bubble, with the value reported for each.

• Probe liquid and its surface tension — grade, lot, measured or certified value, temperature.

• Face measured — treated or untreated side, and how it was identified.

• Sample age — time since extrusion or treatment, and storage conditions.

• Capture time after deposition, and drop volume.

• Fit model — Young–Laplace, polynomial or circle — and whether the baseline was automatic or manual.

• Environment — temperature, relative humidity, and any antistatic treatment applied.

• Replication and statistics — number of rolls, web positions and drops; median and IQR.

• Instrument and calibration status — identifier, last verification, reference value obtained.

• Surface free energy details, if reported — model, liquid set and the literature values used.

One-sentence rule for reviewers

If a certificate quotes a contact angle without stating the face, the sample age and the capture time, the number cannot be compared with anything — including another number from the same supplier.

[1] Żołek-Tryznowska Z, Cudna K, Tryznowski M. Comparative study of seal strength and mechanical behavior of untreated and corona-treated polymer films. Processes 2025;13(10):3190. doi:10.3390/pr13103190

[2] Vetaphone A/S. Corona & plasma surface treatment: what are they and how do they differ? Technical paper, Kolding, Denmark.

[3] Owens DK, Wendt RC. Estimation of the surface free energy of polymers. Journal of Applied Polymer Science 1969;13:1741–1747.

[4] Zisman WA. Relation of the equilibrium contact angle to liquid and solid constitution. Advances in Chemistry Series 1964;43:1–51.

[5] Wenzel RN. Resistance of solid surfaces to wetting by water. Industrial & Engineering Chemistry 1936;28:988–994.

[6] Cassie ABD, Baxter S. Wettability of porous surfaces. Transactions of the Faraday Society 1944;40:546–551.

[7] ISO 19403-2:2017. Paints and varnishes — Wettability — Part 2: Determination of the surface free energy of solid surfaces by measuring the contact angle. ISO, Geneva.

[8] ASTM D7334-08(2022). Standard Practice for Surface Wettability of Coatings, Substrates and Pigments by Advancing Contact Angle Measurement. ASTM International.

[9] ASTM D5946 / ISO 15989. Water contact angle of corona-treated polymer film. ASTM International and ISO.

[10] ASTM D2578 / ISO 8296. Wetting tension of polyethylene and polypropylene film. ASTM International and ISO.

[11] ASTM D7541. Estimating critical surface tension of coatings, substrates and pigments by the Zisman method. ASTM International.

[12] Regulation (EC) No 1935/2004 of the European Parliament and of the Council on materials and articles intended to come into contact with food.

[13] Commission Regulation (EU) No 10/2011 on plastic materials and articles intended to come into contact with food, as amended by Regulation (EU) 2020/1245.

[14] Commission Regulation (EC) No 2023/2006 on good manufacturing practice for materials and articles intended to come into contact with food.

Symbol | Meaning | Value / unit

θ | Contact angle | degrees

θY | Young (ideal) contact angle | degrees

θa, θr | Advancing and receding angles | degrees

Δθ | Contact angle hysteresis, θa − θr | degrees

γlv | Liquid–vapour surface tension | mN/m (= mJ/m²)

γs^d, γs^p | Dispersive and polar components of solid surface energy | mN/m

γc | Zisman critical surface tension | mN/m

Wa | Work of adhesion | mJ/m²

Water at 20 °C: γ = 72.8 (21.8 dispersive / 51.0 polar) mN/m

Diiodomethane at 20 °C: γ = 50.8 (50.8 dispersive / 0.0 polar) mN/m

Ethylene glycol at 20 °C: γ = 48.0 (29.0 dispersive / 19.0 polar) mN/m

Table A1 — Symbols used in this note, with the liquid constants applied in Section 3.4 and Appendix B.

A dyne level and a water contact angle describe the same surface but not the same quantity, and there is no single conversion factor between them. The table below makes the reason explicit: it computes the water angle, by OWRK, for a range of total surface energies and polar fractions. Read a row across and the angle changes by up to 18° at constant surface energy.

Total surface energy (mN/m) | Polar 5 % | Polar 15 % | Polar 30 % | Polar 45 %

30 | 94° | 86° | 80° | 76°

34 | 91° | 82° | 75° | 71°

38 | 88° | 79° | 71° | 67°

42 | 85° | 75° | 67° | 62°

46 | 82° | 72° | 63° | 58°

50 | 79° | 68° | 59° | 53°

Table B1 — Calculated water contact angle as a function of total surface energy and polar fraction, using OWRK with water at 72.8 mN/m. These are model values, not measurements; expect ±5° against a real film.

Two readings of the table are useful in practice. First, a surface at 40 mN/m reads 75° if it is 15 % polar and 67° if it is 30 % polar — which is why a converter's dyne pen and a supplier's goniometer can both be right and still disagree. Second, the sensitivity to polarity is greatest in the middle of the range, exactly where the printing window sits, so the polar fraction should be reported alongside any surface energy used in a specification.

The practical conclusion is not to convert, but to correlate. Measure both on the same twenty rolls, plot angle against dyne, and use the resulting relationship — with its scatter — as the local translation between the two languages.