04 — Coming out

Why molecules leave a solid.

+ The questionWhat decides how much of an additive moves into blood, saline or air?

Nothing holds the additive in place

Blood stored in a soft plastic bag picks up some of the bag. In 1970 Jaeger and Rubin, two researchers at Johns Hopkins University, reported that blood extracted phthalate plasticizers, the softeners of flexible PVC, from plastic tubing and from the bags used to store it. Two years later they measured the uptake: human and dog blood kept at 4 °C in PVC bags gained about 0.25 mg of plasticizer per 100 mL each day. Nothing in the bag had failed. Flexible PVC gives some of its plasticizer to the liquids it touches, and three questions follow for every plastic in a device: what lets an additive leave, where it goes, and how much of it arrives.

The first answer comes from Chapter 3: additives are, in nearly all cases, not chemically bound to the polymer chains. An additive sits between chains in constant thermal motion, and that motion keeps opening and closing small gaps around it. The molecule can move only when a gap large enough for it opens beside it, and over many such steps it wanders through the network at random, until a step happens to carry it out through the surface.

That random wandering is diffusion. It carries molecules from where they are crowded to where they are scarce, because more random steps lead out of a crowded region than into it, so a plastic touching a liquid free of its additive loses the additive through the surface. Size sets the first limit on the pace. Molar mass is the chemist's measure of how heavy a molecule is. It runs from about 200 to 2,000 g/mol for additive molecules and from about 300 to 600 g/mol for plasticizers, in the figures of the 2018 review by Hahladakis and colleagues (Chapter 3). The heavier, larger ones migrate more slowly. A larger molecule needs a larger gap, and the rate of diffusion follows the chance that a gap of the needed size forms beside it, in the free-volume theory that Duda reviewed in 1985.

Temperature and the polymer's stiffness also set the pace, and one number gathers all three effects: the diffusion coefficient, which measures how fast one kind of molecule spreads through one material at one temperature. It rises with temperature, falls as the migrating molecule gets heavier, and is higher in soft, flexible polymers such as low-density polyethylene than in polymers with stiff chains, such as polyesters. A 2010 report of the European Commission's Joint Research Centre, edited by Simoneau, sets out these dependences for plastics in contact with food. Heat adds to the free space between the chains, as Duda's review notes, and so to the chance that a large enough gap opens.

In slides on estimating what patients receive from devices, an FDA scientist plotted upper-bound diffusion coefficients for molecules of 100 to 1,000 g/mol in polymers from silicone rubber to polycarbonate, PVC among them. The values spanned about nine tenfold steps, so the fastest was about a billion times the slowest. Release grows only with the square root of this coefficient, as a later section of this chapter shows, so the same span moves the amount released by about 30,000 times, the square root of a billion. Diffusion decides how fast an additive reaches the surface. Whether the additive crosses it depends on what lies on the other side.

The additive goes where it dissolves better

At the surface the additive can stay in the plastic or dissolve in the liquid outside, and it spreads between the two until as many molecules cross back as cross out. That sharing is partitioning, and a molecule ends up mostly where it dissolves better. The partition ratio measures that sharing: the concentration in the plastic divided by the concentration in the liquid once both have stopped changing, as the 2010 Commission report defines it. A ratio of 1 means the liquid holds the substance at the plastic's own concentration, and a ratio of 1,000 means it holds a thousandth of that.

The fat in a liquid decides much of a plasticizer's release. Plasticizers in general leach more into fat-loving, or lipophilic, solutions. SCENIHR, the European Commission's scientific committee on emerging health risks, makes this point in its 2015 opinion on DEHP, di-2-ethylhexyl phthalate, a plasticizer of PVC. In the food studies that Hahladakis and colleagues reviewed in 2018, plasticizer migration rose with fat content. Blood carries fat in the lipids of its plasma and saline carries none, so blood draws more plasticizer from the same tubing than saline does. FDA acted on that contrast in its 2008 guidance on intravascular administration sets, which asked makers to label DEHP in the fluid path of PVC tubing because it can leach on contact with lipids.

The running device's foam and hose meet air, and air is not a solvent. A molecule passes from a plastic into air only by evaporating, so the molecules that reach a breathing circuit's gas are those small and volatile enough to evaporate at the temperature of use. These volatile organic compounds have a part of their own in the breathing-gas standard, ISO 18562, and Chapter 19 takes them up.

11 — Diffusion and partitioning at one surface
PLASTIC — ADDITIVE NOT BOUND TO THE CHAINS DIFFUSION: RANDOM STEPS BETWEEN CHAINS faster when warm faster in a soft polymer faster for a small molecule SURFACE BLOOD plasma lipids take up a fat-loving additive SALINE no fat: takes up a fat-loving additive poorly AIR not a solvent: only molecules that evaporate (Chapter 19) same plastic: blood draws more of a fat-loving additive than saline does Schematic; dot counts illustrative. Partition defaults set for foods (European Commission report edited by Simoneau, 2010): 1 in a fatty liquid, 1,000 in a watery one; not measured for blood.
Device materialReleased substanceTissue and cellsGas or fluid pathFocal detail
Plate 11 — An unbound additive diffuses to the surface, and partitioning then decides how much each medium takes. On the default ratios that a 2010 European Commission report set for foods, a fatty liquid can reach 1,000 times the concentration of a watery one, and air takes only molecules that evaporate.

Release slows as the surface empties

Diffusion and partitioning together decide how release grows with time. The molecules nearest the surface leave first, molecules from deeper in move up to replace them, and a layer just under the surface runs low. Every later molecule must cross that depleted layer, which deepens as contact goes on. The depth that diffusion empties grows with the square root of time, so release slows although nothing in the plastic has changed. The amount released therefore rises in proportion to the contact area and the starting concentration, and with the square roots of the contact time and the diffusion coefficient, in the release model the FDA scientist presented for devices.

Because release follows the square root of time, four times the contact time releases twice as much, and a hundred times the contact time ten times as much. Area enters in direct proportion, so twice the wetted surface releases twice as much. The factors listed for leaching in drug and device guidance, from temperature to flow rate, follow from these mechanisms (Chapter 12).

12 — Release grows with the square root of time
0 2 4 6 0 1 4 9 16 25 30 AMOUNT RELEASED, RELATIVE TO ONE DAY CONTACT TIME (DAYS) if release grew in proportion to time: a steady daily rate 4 days: twice one day square root of time warmer: faster diffusion, same shape (schematic) 30 days, the end of prolonged contact: about 5.5 times one day, not 30 1 DAY 4 DAYS depleted layer: 4 times the time, twice the depth Schematic. Shape: the square-root-of-time model presented by an FDA scientist, for release limited by diffusion in the plastic (early contact, plastic far from empty). 30 days: upper end of prolonged contact, FDA draft guidance, 2024.
Device materialReleased substanceReadoutFocal detail
Plate 12 — Release grows with the square root of time while diffusion in the plastic limits it, because each later molecule must cross a deeper depleted layer. Thirty days of contact release about 5.5 times what one day releases, not 30 times; a steady daily rate would be the straight line, and a warmer plastic follows the same curve higher up.

The square-root rule describes release that diffusion in the plastic limits. It holds early in a contact, while the plastic keeps most of its additive and the liquid takes up all that arrives. Saline against a plastic that holds a fat-loving additive reaches its small share quickly, and partitioning then stops the transfer before diffusion runs short. Chapter 15 shows how such models are built to overestimate what a patient receives.

Solved example: how far the liquid and the time move DEHP's release from PVC
  • The liquid. The Commission report's default partition ratio is 1 for a substance that dissolves readily in the liquid and 1,000 for a fat-loving substance, such as a plasticizer, in a watery liquid; the report set both for foods. For the same concentration left in the plastic, a fatty liquid can therefore reach 1,000 times the concentration of a watery one: three tenfold steps.
  • The time. FDA's 2024 draft guidance on chemical analysis counts contact of 1 to 30 days as prolonged (Chapter 2). The square root of 30 is about 5.5, so 30 days of contact release about 5.5 times what one day releases, not 30 times.
  • The result. On these defaults, the liquid moves the concentration at which release stops by a factor of 1,000, while 30 days instead of one day move the amount released by about five and a half times. Which liquid touches the PVC therefore matters more than how long it touches.
  • Assumptions. The ratios are screening defaults built to overestimate, not values measured for DEHP or for blood. The ratio sets where release stops; the square root sets how release grows before it stops, and it holds only while diffusion in the plastic limits release and the plastic is far from empty.

What can come out, and what does

ISO 10993-18, the 2020 standard on the chemical characterization of device materials, defines the two terms that laboratories use for what these mechanisms release. An extractable is a substance released from a device or its material when it is extracted under laboratory conditions and in laboratory liquids, and a leachable is a substance released from the device during its clinical use. The first names what can come out, and the second what does come out where the patient is. ICH is the international council in which drug regulators and industry harmonize technical requirements. On 1 August 2025 it endorsed a draft guideline on extractables and leachables in drug products for public consultation. The draft ties leachables to manufacture and labeled storage instead of clinical use, and it states that they "generally represent a subset of the extractables".

Leachables are generally a subset because an extraction pushes each mechanism of this chapter past use (Chapter 12). Extraction solvents are chosen to dissolve more than body fluids would, so they shift partitioning toward the liquid. Temperatures above body temperature speed diffusion, and times longer than the contact let the depleted layer reach deeper. One reason for the word "generally" follows from Chapter 5: a substance that forms only during use, as a material breaks down or reacts with what it carries, is not there to be extracted beforehand.

13 — Extractables contain the leachables
EXTRACTABLES — RELEASED UNDER LABORATORY EXTRACTION solvents of three polarities: water; ethanol or isopropanol; hexane above body temperature: 50 °C generally used longer than the contact each pushed past use PARTITIONING DIFFUSION TIME LEACHABLES — RELEASED DURING CLINICAL USE the liquid or gas of use the temperature of use the contact time forms only in use, by breakdown or reaction (Chapter 5) generally a subset Definitions: ISO 10993-18:2020, terms 3.16 and 3.22; ICH Q3E, draft (2025). Conditions: examples from an FDA staff presentation, 2026, not policy.
Released substanceFocal detail
Plate 13 — Extraction pushes partitioning, diffusion and time past use, so what leaves in use generally leaves in the laboratory too. The exception is a substance that forms only during use.
An extract's amounts are an upper estimate for what was there before use

An extractable is only a potential leachable, as ICH's draft puts it. An extract's amounts come from conditions built to release more than use does, so for substances present before use they are an upper estimate, not the patient's dose. Chapters 12 to 15 turn extract amounts into a dose.

DEHP in flexible PVC

DEHP brings every mechanism of this chapter into one material. PVC softened with DEHP holds about 30 % to 40 % of it by weight, according to the 2024 guidelines of SCHEER, a scientific committee of the European Commission. SCENIHR's 2015 opinion gives a typical 30 %. In SCENIHR's words, plasticizers "are not chemically bound to PVC and may therefore leach", and SCHEER calls the interaction of phthalates with their polymer weak. By weight, about a third of a bag made of such PVC is therefore a fat-loving molecule held by weak attractions alone. The DEHP has also made the PVC soft, and the chain mobility that makes the plastic flexible is the same mobility that lets the plasticizer move through it.

Jaeger and Rubin also followed the plasticizers beyond the bag. In their 1970 study, an isolated rat liver kept working by perfusion broke one plasticizer down while another accumulated unchanged. The second plasticizer was identified in tissue from patients transfused with bag-stored blood. In their 1972 paper, the plasticizer they had measured in stored blood was detected in seven of 12 samples of lung tissue from transfused patients, and one patient excreted a metabolite of it in urine after open-heart surgery.

DEHP passes from PVC into blood components during storage, which can last up to 6 weeks after collection, SCENIHR's opinion states. In the opinion's list, the rate of leaching depends on the mechanisms of this chapter: how fat-loving the plasticizer and the stored material are, the storage time and temperature, the contact area and, in some cases, agitation. The same opinion records "a severe lack of data" on the solubility, water/oil partition coefficients and leaching of the plasticizers it assessed, DEHP among them, under conditions relevant to their use. For that reason the solved example rests on default ratios.

SCENIHR's opinion gathers estimates of exposure in micrograms for each kilogram of body weight, each day. They run up to 8,000 to 10,000 for transfusion in trauma patients and for adults on ECMO, the machine that oxygenates blood outside the body. Newborns in intensive care reach up to 6,000, chronic hemodialysis at most 2,200, and apheresis donors up to 38, more than 200 times below the highest. Adults on hemodialysis are nevertheless the most exposed group in SCENIHR's summary, because the treatment is chronic, and newborns are prone to high exposure because they weigh so little.

The rate Jaeger and Rubin measured, 0.25 mg per 100 mL a day, holds for blood at 4 °C. A steady daily rate would mean release in proportion to time, the straight line of the release plate, while the square-root curve of the FDA scientist's slides describes release that diffusion in the plastic limits. One daily figure gives neither the storage period nor the shape of the uptake curve, so the 1972 measurement neither confirms the square-root rule nor contradicts it. Applied to tubing at body temperature, the measured rate would also ignore the faster diffusion of a warm plastic. Each release figure holds for the liquid, the temperature and the time that produced it.

Chapter 14 shows how a tolerable intake is set and Chapter 15 how an exposure is weighed against it. Chapter 11 gives DEHP's listing under REACH, the EU's chemicals law, in its account of endocrine disruption. Chapter 27 sets out the EU rule under which a device of the kinds it names may contain DEHP above 0.1 % by weight only with a justification, and the same holds for any other substance that causes cancer, damages genes, harms reproduction or disrupts hormones.

Particles break off rather than dissolve

Everything so far concerns molecules that dissolve. A solid can also release pieces of itself, and a piece follows none of the rules above: it does not diffuse between the chains or partition into a liquid, but breaks away when whatever holds it to the rest of the material gives way. The slippery coating of an intravascular catheter or wire may separate in some circumstances, by peeling, flaking, shedding, delaminating or sloughing off, FDA's 2019 guidance on such coatings states. The guidance links separated coating to pulmonary and myocardial embolism, adverse tissue reactions and thrombosis. Chapter 5 shows how a degrading foam crumbles into particles, and Chapter 9 how a particle's size decides where it goes.

In practice

Before choosing a material for a part, write down the liquid or gas it touches, the temperature, the contact time and the wetted area; the four entries also become the starting conditions of the chemical characterization in Chapter 12. The running device meets no blood: its humidifier's water chamber holds heated water, its hose and sound-damping foam sit in the air path, and its mask cushion rests on skin. Check the water chamber for what heated water draws out, since heat speeds release; the foam and hose for volatile substances and for pieces that break off (Chapter 5); and the cushion for what skin takes up (Chapter 11). On another device, check a part that meets blood or a fatty liquid for unbound, fat-loving additives.

+ What this chapter established
  • Additives leave a plastic because, in nearly all cases, they are not bound to its chains, and diffusion brings them to the surface faster when warm, in a soft polymer and for a small molecule.
  • Partitioning decides how much crosses: blood draws more of a fat-loving additive than saline does, and air takes only what evaporates.
  • Release grows with contact area and with the square root of time while diffusion in the plastic limits it, and extraction pushes each mechanism past use, so leachables are generally a subset of extractables.
  • DEHP makes up about a third of the PVC it softens and is not bound to it, so it enters stored blood, and the highest estimated exposures come with transfusion, ECMO, hemodialysis and newborn intensive care.
Created with Claude (Anthropic).
Educational reference material, not engineering, regulatory or clinical advice. Product and company names referenced are the property of their respective owners.