aHoffmann (1996)
^b The Danish Plastics Federation (1996)
^c Hansen and Havelund (2000)
^d At the moment increasing globally and constant on the Nordic market, but decreasing a little on the Danish marked.
^e According to SFA (Hoffmann, 1996) consumption is decreasing but according to the Danish Plastics Federation (1996) it might be increasing.
^f The application "other applications" under non-PVC -products in Hoffmann (1996) is estimated to cover as a maximum 50 tons.
^g Inventory for the Consumption in 1994 made by FDLF for The Danish EPA.

The trend shown in Table 4.1 for the non-PVC-products, is a decline in the consumption of phthalates. Concerning the PVC-products the general trend is difficult to deduce from Table 4.1. According to the suppliers the overall consumption is at the moment increasing but within the near future it is ex-pected to decline.

The consumption of phthalates is slightly increasing in the EU as a whole, stagnant in northern Europe, and decreasing slowly in Denmark (Hansen and Havelund, 2000).

4.2 Selection of substitute substances

In the following the background for the selection of the 11 substitutes for phthalates is described. In Table 4.2 a total of 18 compounds are listed that, in variable degree, all are potential substitutes for phthalates.

• 5 chemical compounds (substitutes), and
• 6 groups of substances.

Within each of the 6 groups, one specific substance has been selected as marker for the group.

The primary source of information is the industry and the initial information from The Danish Product Register.

To get an impression of how the substitution will take place selected indus-trial organisations have been contacted. Suppliers and users of phthalates and/or have been contacted to give an estimate of how a complete substitu-tion of phthalates 5 years from here can be predicted.

The same substance will not substitute phthalates in all applications. The substitution will within the different applications take place by a distribution of substitutes. Estimates of this distribution are given in the substitution matrix in Table 4.5 and Table 4.6.

In Table 4.2, a short description of the selection of substitutes for phthalates for various applications is given.

Table 4.2 The plasticiser substitutes and suggestions for example substances in the groups of plasticisers. Other possible substitutes are shown in italics. 

The Danish Product Register has conducted a search on these CAS No.s and a general search to identify which CAS No.s are registered in Denmark as plasticisers.

The result of the general search was that approx. 180 different substances are registered as plasticisers. These were screened with respect to their uses, and only a minority was found to be relevant substitutes for phthalates.

Reduction of the list in Table 4.2 has been performed based on information from the industry, the result of comprehensive information on use patterns from The Danish Product Register, and assessment of the data availability regarding toxicological and ecotoxicological information necessary for the assessment.

4.2.1 Assessed substitutes for phthalates - substances 

The plasticisers assessed are those for which most information is expected to be available for the environmental and health assessment and which have a use pattern involving high PVC volume and/or expected high exposure of humans and/or the environment. The substances are listed in Table 4.3.

Table 4.3 Substances used for the environmental and health assessment.

No specific substance to be used as a marker for polyester-substitutes has been identified in PR or from suppliers. The industry has emphasised that these substances may become important and the branch organisation has suggested polyadipate as an example substance. However, no information on health or environmental properties has been identified on this substance or group during the present project.

4.2.2 Assessed substitutes for flexible PVC - materials

Polyethylene (PE) and polyurethane (PU) are both materials which are identified as possible substitutes for flexible PVC in a number of products and they thereby contribute to the overall substitution of phthalates. The two materials polyurethane and polyethylene substitute the PVC polymer as such and not only the plasticiser additive. Both materials are polymers.

The two alternative materials PE and PUR
In this study low density polyethylene (LDPE) rather than high density polyethylene (HDPE) is selected for the environmental and health assess-ment, since it is expected that LDPE will substitute plasticised PVC in the main uses in toys and garden hoses.

PU is expected to substitute plasticised PVC in waterproof cloths, shoes, boots and waders, and PUR based on the diphenylmethane-4,4'-diisocyanate (MDI) monomer is selected for the environmental and health assessment.

LDPE and HDPE
Industrial polyethylenes are thermoplastics, which exist in different ver-sions. Low-density versions (LDPE and Linear LDPE) are produced in branched forms in a structure with long and short branches respectively. LDPE is therefore only partly crystalline and the polymer is highly flexible. Principal uses include packaging film, waste bags and soft type plastic bags, tubes, agricultural mulch, wire and cable insulation, squeeze bottles, house-hold items, and toys. LDPE has already substituted flexible PVC in the majority of household packaging products, and the potential for substitution is therefore greater for other product groups like toys.

High density versions (HDPE) are produced in linear forms which allow the polymer chains to pack closely together. This structure results in a dense and highly crystalline material of high strength and moderate stiffness. Principal uses include bottles, pails, bottle caps, packaging, household appliances, and toys. Because of the strength and stiffness HDPE is more commonly used for industrial products compared to household products and consumers in general are more likely to be exposed to LDPE than HDPE. LDPE and HDPE are assumed comparable with regard to effects on environment and health.

HDPE is used for toys of rigid materials, but the proportion of toys manu-factured from LDPE compared to HDPE is not known. It has been suggested that the two PE polymers have an equal share of toy market, and it is as-sumed that LDPE may lead to higher exposures than HDPE.

Polyurethanes
Polyurethanes cover a broad range of synthetic resinous, fibrous, or elasto-meric compounds belonging to the family of organic polymers made by the reaction of diisocyanates with other difunctional compounds such as glycols (polyols). Polyurethanes are one of the most versatile of any group of plas-tics, capable of an almost infinite number of variations in chemistry, struc-ture and application. Polyurethanes can be produced as a foam, in solid form, as an elastomer, coating, adhesive or binder. Foamed polyurethanes form about 90% by weight of the total market for polyurethanes, but there is also a wide range of solid polyurethanes used in many diverse applications.

By itself, the polymerisation reaction produces a solid polyurethane. Polyu-rethane foams are made by forming gas bubbles in the polymerising mixture, which is achieved by using a blowing agent.

MDI
MDI is one of the most important raw materials to make polyurethane. MDI can be grouped into polymeric MDI, which in the form of foam is being used for several heat protection materials, motor car seats, etc., and mono-meric MDI, being used for shoe soles, coating materials, synthetic leather, etc.

Because of the health risks ascribed to monomeric diisocyanates, much at-tention is paid to the physico-chemical and toxicological properties of the monomer in situations where substitution with a polyurethane is required.

MDI has a particularly low vapour pressure compared to TDI (toluene diiso-cyanate), HDI (hexamethylene diisocyanate) and IPDI (isophorone diiso-cyanate ) and is under European legislation classified as harmful whereas the other mentioned diisocyanates are classified as toxic. This often makes MDI a better choice where it is technologically feasible.

MDI is already used in the production of PU for waterproof clothes, shoes, boots and waders, application areas, which are suggested for substitution of flexible PVC in the substitution matrix, and is therefore selected to the health and environmental assessment.

4.3 Proposed use pattern for substitutes

The Danish Product Register (PR) has been used to establish an overview of the function of phthalates in chemicals until now. The Register mainly con-tains information about chemicals and to a lesser extent about materials.

The historical main use of phthalates in Denmark, summarised in Table 4.1, forms the basis for the search for relevant alternatives in the PR.

The most important function of phthalate has until now been plasticising, but besides this function phthalates had have the function of being denatur-ants in cosmetics (Hoffmann, 1996).

The PR has conducted a search to identify the plasticisers used in selected types of chemical products or materials. The result is illustrated in Table 4.4 for the first 10 that are the substances selected for the assessment. The Poly-ester was not included due to lack of CAS no.

Table 4.4 The registered use of the selected substances as plasticisers in the selected product groups. Data from the Danish Product Register. The polyester plasticiser (polyadipate) was not included due to lack of CAS no. Materials such as cables, profiles, floor and wall covering are not covered by the PR.

a Not found in the Product Register.
b Unclear whether 'plast' is PVC

Not all substances are registered as plasticisers in the selected products. This has to be seen in the light of the fact that the register only contains informa-tion about substances classified dangerous to the environment or the health. The result of the search is therefore as mentioned earlier supplemented with industrial information about the development of plasticisers not containing phthalates.

4.3.1 Substitution matrix for the 11 substances in tons

By using the amounts found within the different applications in the sub-stance flow analyse (Hoffmann, 1996) and the proposed %-distribution of the alternatives in Table 4.5, the following "amount-substitution matrix" shown in Table 4.6 within the different applications can be established.

Table 4.5 Substitution matrix for the 11 substances with anticipated share given in %.

Note: It is only relevant to ad up the figures horizontal and not vertically because each row describe the substitution within one application. One column rep-resents non-comparable figures.

Based on information from the industry, Table 4.6 represents the best pres-ent estimate on substitution of phthalates within different types of products. The dominating amount is for each substance marked in bold. The informa-tion is primarily based on interviews with industry sources rather than trade bodies, since only little overview information is available.

This best present estimate has to be seen in the light of the situation in which all phthalates in a product are substituted by only one substitute.

The actual substitution five years from now, will presumably not be exactly as illustrated in Table 4.6, but the information indicates in which areas the substances might be used extensively and in which areas the use is expected to be negligible.

It should be emphasised that a large portion of the expected use is placed in "Other applications of PVC" (e.g. toys).

Another scenario is that one substance substitutes the phthalates 100% within an application area (a 'worst worst case').

Table 4.6 Substitution matrix for the 11 substances (in tonnes)

 4.3.2 Substitution matrix for the two materials

In view of the general phase out policy for PVC the substitution of phtha-lates may obviously take place exchanging the PVC-material by other mate-rials that do not need to be plastified with phthalates. However, PE and PU cannot substitute flexible PVC across-the-board, but as seen in substitution matrix Table 4.7 PE and PU are possible substitutents for flexible PVC in different kinds of products:

• PE will mainly substitute flexible PVC in toys
• PU will mainly substitute flexible PVC in waterproof clothes, shoes, boots and waders.

Using the same procedure as for the 11 substances, but with the 1994-inventory from The Danish Plastics Federation of the consumption of plasti-cised PVC, a substitution matrix for the PVC-substituting materials can be established as in Table 4.8

Table 4.7 Substitution matrix for selected flexible PVC products in % of the total amount plasticised PVC (tonnes).

*: The vertical sum across different applications of PVC is not relevant to cal-culate. It is only the horizontal sum within the same application, which is relevant to calculate because it is describing the situation within one spe-cific application and has to add up to 100%.

Table 4.8 Substitution matrix for alternative materials to flexible PVC. The unit is tons.

As for the 11 substances, the information from the industry in Table 4.8 rep-resents the most likely substitution of phthalates within different types of PVC-products. The dominating amount for each material is marked in bold.

Again, the substitution five years from now, will presumably not be exactly as illustrated in Table 4.8, but the information indicates in which areas the materials might be used extensively and in which areas the use is expected to be negligible.

The worst case scenario is when one material substitutes the plasticised PVC 100% within an application area.

As seen in Table 4.8, polyethylene is most likely going to substitute 339 tons flexible PVC in toys. With an average concentration of phthalates in soft-PVC toys, similar to 34%, the 339 tons PVC represent 115 tons phthalates.

Polyurethane is, as shown in Table 4.8, the mayor substitute for PVC in waterproof clothes. It is therefore of special interest to undertake an EUSES-calculation on the 208 tons flexible PVC, which may contain up to 100 tons phthalates.

4.4 Assessment of emission and exposure

Data compilation for substitution matrix
The qualitative information from suppliers of phthalates and the alternative substances is based on an assumed complete substitution of phthalate in the mentioned applications in the near future (a five year perspective).

In Table 4.5 and Table 4.7 the qualitative information is transferred to quantitative figures in percent. These figures form a possible scenario for how the complete substitution can take place. It will probably not corre-spond to the real situation in five years time, but it illustrates where the use of a substance might be extensive and where the use might be negligible. This overview is useful in connection with evaluation of results from calcu-lations in EUSES and in connection with priority of efforts of the environmental regulating authorities.

The 11 substances and the 2 materials in this project are regarded as the main basis for the complete substitution for phthalates.

According to discussion with the organisations listed in the Appendix, the most likely way to substitute phthalates is illustrated in the following sub-stitution matrixes Table 4.5 and Table 4.7.

Substitution matrix for the 11 substances in %
The consumption of phthalates within the relevant applications is based on substance flow analyses covering the situation in 1992 (Hoffmann, 1996). In Table 4.5, the share of 11 substances for the substitution of the phthalates within each application is estimated in %.

According to the Danish Plastics Federation, flexible PVC is not used in packaging, today. The actual consumption of phthalates for this purpose is therefore estimated to be of minor importance.

The first five substances are expected to substitute phthalates directly. The next six are selected as markers for chemical groups from which substitutes are expected to be identified in the near future. Meanwhile the six markers are used to calculate a scenario for substitution of phthalate.

Other substances and materials cover less important substitutions, conducted by other means than the 11 substances. Examples could be substances not covered by the 11 substances in Table 4.6, or new technology in the produc-tion of the mentioned products. The new technology could be the use of new materials without the need for plasticising with phthalates.

The point of origin of Table 4.5 is the Danish consumption of phthalates in 1992 shown in Table 4.1 (Hoffmann, 1996). These data are selected because they are the result of a comprehensive survey, and the studies conducted later, confirm the amounts and the indicated development trends within the different applications. For the non-PVC products a decline in the use of phthalates has been identified (Hansen and Havelund, 2000). For the PVC-products the suppliers expect a decline in the near future.

The background for the ratios in Table 4.5 is information gathered in con-nection with one of the substitution projects initiated by the Danish Envi-ronmental Protection Agency. It is the general impression among suppliers and users of phthalates that a complete substitution will be possible for both PVC and non-PVC products. Available substitutes for non-PVC products have been identified earlier (Hansen and Havelund, 2000).

Table 4.9 Estimated use of the substitutes. These volumes are used for consumer expo-sure

4.4.1 Considerations regarding specific uses of phthalates/substitutes

Industrial processes
There is no synthesis of phthalates or substitutes for phthalates in Denmark.

The synthesis of phthalates for the Danish market is at the moment mainly conducted in Sweden. The identified substitutes are expected also to be synthesised in countries outside of Denmark.

The main source to emissions and exposures in Denmark is expected to be from formulation of products containing plasticisers such as plasticised PVC, printing inks, adhesives and fillers. For paints and lacquers there is also an emission and exposure from the professional use of the products.

Based on the substitution matrix focus in this investigation has been set on the use of the eleven substances with known potential application the fol-lowing process:

• Use of di(2-ethylhexyl) phosphate (CAS No. 298-07-7), tri(2-ethylhexyl) phosphate (CAS No. 78-42-2), tri-2-ethylhexyltrimellitate (CAS No. 3319-31-1) and alkylsulfonic acid ester (toluene sulphona-mide, CAS No 88-19-7) in the production of cables

• Use of butane ester diethylhexyl adipate (CAS No. 103-23-1), (2,2,4-trimethyl 1,3-pentanediol diisobutyrate, CAS No 6846-50-0) in the pro-duction of floor and wall covering

• Professional use of epoxidised soybean oil (CAS No. 8013-07-8) con-taining lacquer and paint products

• Use of o-acetyl tributyl citrate (CAS No. 77-90-7) and dioctyl sebacate (CAS No. 122-62-3) in the production of printing inks.

• Use of polyester and dipropylene glycol dibenzoate (CAS No. 27138-31-4) in the production of fillers.

In general, uses of the products are regarded as diffuse and as minor sources to emission, but concerning consumer exposure of the 10 substances there are relevant scenarios that are described in Section 5.4.

Human exposure is estimated mainly to take place in connection with:

• Formulation process.
• Uses of the products

To illustrate the potential human exposure from the 10 well defined sub-stances a calculation in EASE, which is based on the principles in the TDG has been conducted.

EASE calculates a theoretical exposure of humans in the working environ-ment and private consumers.

The input data takes point of origin in the scenarios in Table 4.10, which is assessed to represent the most extensive exposure.

Table 4.10 Exposure scenarios of the 11 substances

To approach a situation five years from now the exposure calculation in EUSES is therefore conducted for both the most likely situation represented in bold figures and the 100%-scenario.

4.4.2 Worker and consumer exposure

Assessment of the exposure of humans in the working environment and con-sumers has been conducted using the model Estimation and Assessment of Substances Exposure Physico-chemical Properties (EASE), which is a part of the model European Union System for the Evaluation of Substances (EUSES).

Limitations and uncertainties of the EASE
The equations in EASE are intended to provide a simple description of con-sumer exposure. Most equations give a worst case estimation of exposure, by assuming that all of the compound in the product is at once available for intake and uptake. Intake and uptake themselves are modelled as simple fractions.

Workers exposure
EASE provides a general-purpose predictive model for exposure assessment in the workplace. The model predicts external exposure only: it does not take into account absorption and bioavailability.

If reliable and representative measured data are available these can be used to overwrite the model results. The general-purpose model is called EASE (Estimation and Assessment of Substance Exposure) which is described in details in Section 2.2 of the TGD.

EASE was specifically developed for the purpose of modelling inhalation and dermal workplace exposure across a wide range of circumstances.

EASE is an analogue model, i.e. it is based on measured data which are as-signed to specific scenarios. The user can build scenarios by choosing be-tween several options for each of the following variables: physical proper-ties during processing (tendency to become airborne, potential for dermal contact), use pattern and pattern of control. Numerical ranges have been as-signed using measured data contained within the UK National Exposure Database for inhalation exposure, and experimental data and expert judge-ment for dermal exposure.

The data used to assign ranges within the model are all 8-hour time weighted averages and the numbers generated by the model are only valid when the exposures being assessed can be related to such averages.

The output of EASE is numerical ranges of concentrations or ppm. These are converted from ppm to kg.m-3 and can be used as input for risk charac-terisation in which the exposure estimates are compared to the results of the human effects assessment.

Consumer-exposure
The consumer, i.e. a member of the general public who may be of any age, either sex, and in any stage of health may be exposed to chemical substances by using consumer products.

A consumer product is one, which can be purchased from retail outlets by members of the general public and may be the substance itself, or a prepara-tion, or an article containing the substance.

The EASE equations for consumer exposure can be used to estimate exter-nal exposure substances used as or in consumer products. Absorption or bioavailability is not taken into account by the equations implemented in EASE.

The focus in EASE is on substances used indoors for a relatively short pe-riod of time per event (such as e.g. a carrier/solvent in a cosmetic formula-tion; a powder detergent).

The equations in the model apply to both volatile substances and airborne particulates. It is assumed the substance is released as a vapour, gas, or air-borne particulates, and the room is filled immediately and homogeneously with the substance. Ventilation of the room is assumed to be absent.

The equations can also be adapted to estimate exposure arising from 'rea-sonably foreseeable misuse', i.e. when products are not used according to the instructions, but as if they were other, allied products.

To adapt the equations, the values for the parameters used in the equations are changed to reflect values foreseen in "reasonably foreseeable misuse". For example, the volume of product or the area of application is set to a dif-ferent value, reflecting reasonable foreseeable misuse.

If a substance is released relatively slowly from a solid or liquid matrix (e.g. solvent in paint, plasticiser or monomer in a polymer, fragrance in furniture polish), the equation in EASE acts as a worst case estimation, estimating the maximum possible concentration.

Dermal
The calculation in EASE has in this investigation been operating with two scenarios concerning dermal exposure: A and B

Dermal A: a substance contained in a medium. This dermal scenario also applies to

• a non-volatile substance in a medium used without further dilution (set dilution D=1), and
• a non-volatile substance in a volatile medium.

The assumption behind the equations in the calculations is that all of the substance on the skin is potentially available for uptake. This is the case when the medium is well mixed or only present as a thin film on the skin. The dermal equations apply for:

• a non-volatile substance in a diluted product, 
• a non-volatile substance in a medium used without further dilution, and,
• a non-volatile substance in a volatile medium.

Dermal B: a non-volatile substance migrating from an article (e.g. dyed clothing, residual fabric conditioner, dyestuff/newsprint from paper).

The assumption behind the equation is that only part of the substance will migrate from the article (e.g. dyed clothing, residual fabric conditioner, dye-stuff/newsprint from paper) and contact the skin. The migration is assumed to be slow enough to be represented by a constant migration rate multiplied by the time of contact.

The exposure calculation will involve estimating the amount of substance which will migrate from the area of the article in contact with skin during the time of contact. Dyestuff amounts in fabrics and paper are usually given as weight of product per unit area (e.g. mg/m2).

Oral
The calculation concerning oral exposure has also been operating with two scenarios: A and B.

Oral A: a substance in a product unintentionally swallowed during normal use (e.g. toothpaste).

The exposure equations may also be used to estimate exposures arising from ingestion of the non-respirable fraction of inhaled airborne particulates.

The equations may also be used to estimate exposures arising from ingestion of the non-respirable fraction of inhaled airborne particulates.

Oral B: a substance migrating from an article into food or drink (e.g. plastic film, plastic-coated cups/plates).

It is assumed that the substance in a layer of thickness of article (e.g. plastic film, plasticcoated cups/plates) in contact with the food will migrate to the food. The migration rate is assumed to be constant, and the migration rate multiplied by the contact duration is the fraction of substance that is migrated to the food. The equation can be used to give a conservative estimate of substance uptake by a defined volume of food. The value of the migration rate will be influenced by the type of food (e.g. fatty/dry/moist), the period of exposure and the temperature at which it occurs. Consumer exposure level will also be influenced by the proportion of contaminated food eaten.

Use pattern scenarios
Based contacts to the Danish industry and the substance flow analyse (Hoffmann, 1996) relevant scenarios has been identified and are described in section 4.3.

The background for choosing scenarios is the most likely way of substitu-tion described by industrial actors and rendered in the substitution matrix in Table 4.5. Within the substitution matrix the application representing the largest estimated consumption of each substitute is selected as the most relevant scenario. This application is marked in bold figures in the substitu-tion matrix.

The input data for the calculation are the substitution matrix and scenarios with the largest estimated consumption substitute for phthalate plasticisers.

With point of origin in the substitution matrix the substances are distributed in the following most relevant application areas.

Plasticisers in the "Cables"-application are expected to be:

• Di(2-ethylhexyl) phosphate, CAS No. 298-07-7
• Tri(2-ethylhexyl) phosphate. CAS No. 78-42-2
• Tri-2-ethylhexyltrimellitate, CAS No. 3319-31-1
• Alkylsulfonic acid ester (o-toluene sulphonamide, CAS No 88-19-7).

For the production of "floor and wall covering" the following plasticisers are chosen:

• Diethylhexyl adipate, CAS No. 103-23-1
• Butane ester (2,2,4-trimethyl 1,3-pentanediol diisobutyrate, CAS No 6846-50-0).

Concerning the production of lacquer and paint the focus is on:

• Epoxidised soybean oil (CAS No. 8013-07-8).

In connection with printing ink the relevant substances are estimated as:

• O-acetyl tributyl citrate, CAS No. 77-90-7
• Dioctyl sebacate (CAS No. 122-62-3).

The production of "fillers" is assumed to include:

• Polyester
• Dipropylene glycol dibenzoate (CAS No. 27138-31-4).

The scenarios in EASE for the consumer exposure are based on the amounts for these uses and the exposure characteristics for the application.

4.4.3 Exposure in environment

Substance parameters
For each substance the required input parameters molecular weight, octanol-water partition coefficient, water solubility, vapour pressure and physical state were fed to the model in accordance with the data search and evaluation.