METALS AND ALLOYS IN CONTACT WITH FOODMETALS AND ALLOYS IN CONTACT WITH FOOD How to establish compliance with European regulations and recommendations 6-7 NOVEMBER2014, LJUBLJANA,

METALS AND ALLOYS IN CONTACT WITH FOOD

How to establish compliance with European regulations and recommendations

6-7 NOVEMBER 2014, LJUBLJANA, SLOVENIA

SESSION III

1

Resolution on Metals and Alloys of the Council of Europe:

Innovative approach to product testing

Rue Juliette Wytsmanstraat 14 | 1050 Brussels | BelgiumT +32 2 642 51 11 | F +32 2 642 50 01 | www.wiv-isp.be

Dr. Ing. Fabien [email protected]

7 November 2014

SRL

Technical guide

1. New simulants

2. New concept for repeated use

3. Problem of measurement of utensils

Conclusions

Resolution on Metals and Alloys of the Council of Europe:

Innovative approach to product testing

2

SRL vs SML

a) Release vs migration: electrochemical influence of O2, simultaneous use of 2 or more metalsresults in galvanic effect, passivation…

b) No diffusionc) Definition of SRLs: unlike substances in many

plastics (except for phthalates) metals are present in the environment

d) Use of allocation factor or ALARA (dependingon the metal concerned)

3

• Migration (such as from plastics): diffusion

• Ceramics, enamelled cast iron: release (diffusion followed by dissolution)

• Release from metals and alloys: electrochemical

4

Different types of release:

3

Example of multiple exposure: Aluminium

Exposure for a 60-kg adult Al (mg/day)

From food (P50) 1.8

From food (P95) 4.7

From drinking water (P50) 0.027

From drinking water (P90) 0.091

From ingestion of 10 mg/day of dust/soil (71/29) (P50) 0.30

From ingestion of 20 mg/day of dust/soil (71/29) (RME) 0.60

From inhalation (16.7 m³/day) (P50) 0.0017

From inhalation (22.6 m³/day) (RME) 0.020

From food contact materials (P50) 0.18

From food contact materials (P95) 0.49

Sum for median scenario 2.31

Sum for RME scenario 5.90

TRV (TDI) 8.6

% TRV: median scenario 26.9%

% TRV: RME scenario 68.6%

Different concepts used in definition of an SRL

CriteriaAppropriate toxicological reference value

Appropriate oral exposure estimate

Proportion of toxicological reference value allocated to metallic FCM

Comment

1 Yes Yes VariableAllowance fills the gap between worst-case oral intake and the toxicological reference values (e.g. copper, molybdenum & zinc)

2 YesYes, but exceeds toxicological reference value

ALARA or if no data available 10% (e.g. aluminium)

3 Yes No 20%(e.g. silver, cobalt, manganese & nickel) Allowance is in agreement with the WHO “Drinking Water Guidelines”

4 No Yes Not applicableSince no toxicological reference value is available, an SRL was derived from intake data. (e.g. chromium, iron & vanadium)

5 Yes Not applicable 10%Only applicable for impurities. Independent of oral intake; for impurities a general allowance of 10% is applied

6 Not applicable Not applicable Not applicable Setting of a SRL was considered unnecessary (e.g. magnesium & titanium)

4

SRLs and transitional SRLs (3 years)

Symbol

Name SRL [mg/kg food]

Maximum tolerable concentration during

transitional period(f: multiplication factor)

Ag Silver 0.08Al Aluminium 5Co Cobalt 0.02 0.1 (f=5)

Cr Chromium 0.250 1.0 (f=4)Cu Copper 4Fe Iron 40Mg Magnesium -Mn Manganese 1.8Mo Molybdenum 0.12 0.6 (f=5)Ni Nickel 0.14 0.7 (f=5)Sn* Tin 100Ti Titanium -V Vanadium 0.01 0.05 (f=5)Zn Zinc 5

For the transposition of quality specifications described in the Guide into national legislation, the Committee recommended that transitional periods be considered by national authorities.

During the transitional period (3 years), a tolerable deviation from the SRL for certain metallic elements could be indicated.

Symbol Name SRL [mg/kg food]

Maximum tolerable concentration during

transitional period(f: multiplication factor)

As Arsenic 0.002 0.01 (f=5)Ba Barium 1.2Be Beryllium 0.01 0.05 (f=5)Cd Cadmium 0.005 0.02 (f=4)Hg Mercury 0.003 0.015 (f=5)Li Lithium 0.048Pb Lead 0.010 0.04 (f=4)Sb Antimony 0.04 0.2 (f=5)Tl Thallium 0.0001 0.0005 (f=5)

SRL

Technical guide

1. New simulants



Conclusions

I. Resolution on Metal and Alloys of the Council of Europe

5

New concepts:1. New simulants: citric acid, artificial tap water…2. Double limits for repeated use3. New methodology of measurement of utensils

Technical guide: Chapter 3 Analytical methods

SRL

Technical guide

1. New simulants



Conclusions

I. Resolution on Metal and Alloys of the Council of Europe

6

1. New simulants

a) Artificial tap water:

DIN 10531 (2011-06) Food hygiene - Production and dispense of hot beverages from hot beverage appliances - Hygiene requirements, migration test.Ion concentrations: calcium 16.4 mg/L, magnesium 3.3 mg/L, sodium 16 mg/L, hydrogencarbonate 44 mg/L, chloride 28.4 mg/L, sulfate 13 mg/L.

More realistic simulant than distilled water

b) Citric acid:

more realistic simulant than acetic acid for a few reasons

The reasons for the choice and change supported by CoE

a) ATW: release in distilled water vs ATW

12

-50,0

0,0

50,0

100,0

150,0

200,0

250,0

300,0

0 200 400 600 800 1000 1200 1400 1600

Rele

ase i

n µ

g/d

m²

Time in min.

Release of nickel from nickel-plated pieces in water MQW and ATW at 20 °C, 40°C and 70°C

MQW 20°C

MQW 40°C

ATW 20°C

ATW 40°C

Distilled water underestimates the release, in this case at 20°C and 40°C. And, in any case, no-one drinks distilled water …

7

b) Citric acid

1. Choice of acetic acid was conventional: One of the reasons was that acetic acid permitsdetermination of global migration by evaporation/gravimetryOther experts mention the common use of this acid…

2. As release in metals and alloys iselectrochemical, the choice of acid is critical

Example: for aluminium with acetic acid the results are totally unrealistic… (dissolution of the material)

13

Ad hoc group work on choice of simulant

Food Composition and Nutrition Tables 2000: SW Souci, W Fachmann, H Kraut mbH, Stuttgart 2000 1182 pp (English, German, French).

14

Acid Acetic Butyric Citric Lactic Malic Oxalic

average (mg/kg) 2.4 89.5 46.4 41.3 39.8 3.5

minimum (mg/kg) 0.35 12 1.19 0.3 0.55 0

maximum (mg/kg) 7.7 260 492 121 569 44.2

Food containing acid 4 6 29 12 26 21

pKa 4.76 4.86 3.13 3.86 3.46 1.25

pH at average [ ] 4.58 3.93 3.37 3.6 3.49 2.83

pH at maximum [ ] 4.33 3.69 2.86 3.37 2.92 2.28

Total of considered food

(with acid content)106 106 106 106 106 106

8

2

3

4

5

6

7

8

Biscuit

s cra

ckers

Camembert

Homard cu

it

Moules

Jus de c

arottes

Cacao

Beurre

Veau

Petits

pois en

conse

rve

Pain co

mplet

Carottes e

n conse

rve

Banane

Jus d

e tom

ates

Poires e

n conse

rve

Ketchu

p

Sorbet

framboise

Moutard

e

Myrtille

s

Citron ve

rt

Vinaigr

e

Jus d

e citro

n

pH

Typical acidity of different foods

Under pH 4.5: only fruits and vegetables (except vinegar)

4.5

Simul. B

2.5

Apples: malic, quinic, aketoglutaric, oxalacetic, citric, pyruvic, fumaric, lactic and succinic acids.

Apricots: malic and citric acids.

Avocados: tartaric acid.

Bananas: malic, citric, tartaric and traces of acetic and formic acids.

Bilberry: citric acid.

Blackberries: isocitric, malic, lactoisocitric, shikimic, quinic, and traces of citric and oxalic acids.

Blueberries: citric, malic, glyceric, citramalic, glycolic, succinic, glucuronic, galacturonic, shikimic, quinic, glutamic and aspartic acids.

Boysenberries: citric, malic, and Isocitric acids.

Cherries: malic, citric, tartaric, succinic, quinic, shikimic, glyceric and glycolic acids.

Crabapple: malic acid.

Cranberries: citric, malic and benzoic acids.

Currants: citric, tartaric, malic and succinic acids.

Elderberries: citric, malic, shikimic and quinic acids.

Figs: Citric, malic and acetic acids.

Gooseberries: citric, malic, shikimic and quinic acids.

Grapefruit: citric, tartaric, malic and oxalic acids.

Grapes: malic and tartaric (3:2), citric and oxalic acids.

Kiwifruit: citric acid.

Kumquat: citric acid.

Lemons: citric, malic, tartaric and oxalic acids (no isocitric acid).

Limes: citric, malic, tartaric and oxalic acids.

Loganberry: malic, citric acids.

Nectarine: malic acid.

Orange peel: malic, citric and oxalic acids.

Orange: citric, malic and oxalic acids.

Passionfruit: malic acid.

Peaches: malic and citric acids.

Pears: malic, citric, tartaric and oxalic acids.

Pineapples: citric and malic acids.

Plums: malic, tartaric and oxalic acids.

Raspberry: citric acid.

Rosehip: malic acid.

Quinces: malic acid (no citric acid).

Salad: citric and malic acids.

Strawberries: citric, malic, shikimic, succinic, glyceric, glycolic and aspartic acids.

Tangerine: citric acid.

Tomatoes: citric, malic, oxalic, succinic, glycolic, tartaric, phosphoric, hydrochloric, sulphuric, fumaric, pyrrolidinonecarboxylic and galacturonic acids.

Youngberries: citric, malic and isocitric acids.

n = 37

85% contain citric acid!

Acids in fruit (1)

9

Beans: citric, malic and small amounts of succinic and fumaric acids.

Broccoli: malic and citric (3:2) and oxalic and succinic acids.

Carrots: malic, citric, isocitric, succinic and fumaric acids.

Mushrooms: lactarimic, cetostearic, fumaric and allantoic acids.

Peas: malic acid.

Potatoes: malic, citric, oxalic, phosphoric and pyroglutamic acids.

Rhubarb: malic, citric and oxalic acids.

A majority of acidicvegetables containcitric acid

Acids in vegetables (2)

Natural occurrence of citric acid in fruits and vegetables

Plant Citric Acid, wt %

Lemons 4.0 - 8.0

Grapefruit 1.2 - 2.1

Tangerines 0.9 - 1.2

Oranges 0.6 - 1.0

Blackcurrant 1.5 - 3.0

Red Currant 0.7 - 1.3

Raspberries 1.0 - 1.3

Strawberries 0.6 - 0.8

Apples 0.008

Potatoes 0.3 - 0.5

Tomatoes 0.25

Asparagus 0.08 - 0.2

Turnips 0.05 - 1.1

Peas 0.05

Corn Kernels 0.02

Lettuce 0.016

Median: 0.57 % or 5.7 g/L

10

Soft Drinks: citric acid, adds a tangy flavour; sequesters harmful heavy metal ions.

0.25 - 0.4%

Soft Drinks: sodium citrate, gives an agreeable, cooling, saline taste; assists in retainingcarbonation.

0.1%

Wine: citric acid, prevents or reduces turbiditycaused by tannin-iron or phosphate-iron complexes.

0 - 0.3%

Candies: citric acid, enhances flavour of fruits; promotes sucrose inversion.

0.5%

Canned fruits and vegetables: citric acid, allowsreduction of retorting temperature; preserves flavour, appearance and consistency.0.1%

Jams and jellies: citric acid/sodium citrate, pH adjustment for optimal pectin gelling. 0 - 0.3%

Gelatin desserts: citric acid/sodium citrate pH adjustment for optimal gelatin setting; contributes to tartness and refreshing properties.2.5%

Processed cheese: sodium citrate, stabilises emulsified fat; complexes calcium ions; improvesmicrobiological stability.3 – 4%

Antioxidants: citric acid, acts as synergist to antioxidants due to complexing of heavy metal ions.0.02%

Acidic food additives (3)

Choice of typical pH (1)

List of n = 488 types of foods → pH < 4.5: n = 143

Median: pH = 3.45

P95: pH = 2.69

P99: pH = 2.00

11

Simulant: 3% acetic acid: pH = 2.5

Citric acid: 1 g/L: pH = 2.7 Citric acid: 5 g/L: pH = 2.35

Choice of typical pH (2)

Effects on alloy or metal

a. Quicker passivation of re-usable stainless steel utensils when citric acidis used as simulant: more realistic.

Citric acidMalic acid

0,00000,01000,02000,03000,04000,05000,06000,07000,0800

0 200 400 600 800 1000 1200 1400 1600Rele

ase in

mg

/dm

²

Time (min)

Release in different acid simulant at pH 2,5 and 70°C

Acetic acid

12

b. Citric acid is a more realistic simulant than acetic acid for aluminium

0

0,5

1

1,5

2

2,5

3

0 50 100 150

Rele

ase (

mg

/dm

2)

Time (min)

Comparison of Aluminium release in lemon juice and citricacid at same pH

Unrealistic

Real situation: lemon juice

Dissolution of aluminium in aceticacid 4%

Laboratory practice

• Purities of citric acid and acetic acid for metalanalysis are comparable

• Citric acid is not more expensive than acetic acid• Availability: ok for both• Handling of citric acid is easier, particularly at hot

temperatures (vapour of acetic acid)

13

SRL

Technical guide

1. New simulants



Conclusions

Resolution on Metals and Alloys of the Council of Europe

2) New concept for repeated use (1):

Sum first and second release

14

• “Where a material or article is intended to come into repeated contact with foodstuffs, the release test(s) shall be carried out successively three times. Its compliance shall be checked on the basis of the level of release recorded from the third test.

Of the release of the first and second tests: the sum of the results of the first and second tests shall not exceed an exposure of one week.

tests shall not exceed an exposure of 7 times SRL.

• JECFA uses the term PTWI, or provisional tolerable weekly intake, for contaminants that may accumulate in the body. The weekly designation is used to stress the importance of limiting intake over a period of time for such substances.

New concept for repeated use (2):

SRL

Technical guide

1. New simulants



Conclusions


15

3) Problem of measurement of utensils

• The plastics regulation was essentially designedfor packaging.Example: concept of 6 dm²/kg → problems for non-packaging food-contact articles: typicallyutensils.

• The procedure requires that the surface area of the utensils must be measured: Therefore in practical terms this means measuring

the surface areas of objects such as whisks or sieves!!

Exposure is not related to the surface area of the utensil

Relation Surface- typical food wheight in

contact

0

100

200

300

400

500

600

0,00 2,00 4,00 6,00 8,00 10,00

Surface (dm²)

typic

al fo

od w

eig

ht in

conta

ct

Problem very large scatter!

Another system is necessary: see nextpresentation by Ingo Ebner

16

SRL

Technical guide

1. New simulants



Conclusions


Conclusions

1) New, more suitable concepts have been introduced by MSs after consultation withindustry representatives.

2) The Technical Guide is already applied by manylabs in EU and implemented in surveillance testing by some authorities of member states.

3) Some conventional concepts have been replaced by scientifically or technologicallybased concepts.

17

Thank you for your attention!

18

FE

DE

RA

L I

NS

TIT

UT

E

FO

R R

ISK

AS

SE

SS

ME

NT Envelope volume,

a new concept

Dr. Ingo Ebner

Ingo Ebner, 2014-11-07 Page 2

To check the compliance it is necessary to know the surface-to-volume ratio• either by applying the real surface to volume ratio,• or by using a convention for articles that cannot be filled

(„articles for which, due to their form it is impracticable to estimate the relationship between the

surface area … and the quantity of food in contact therewith“)

How to determine the volume coming in contact with an article

Reg. (EU) No 10/2011

• surface-to-volume ratio: 6 dm2 / kg food

• example:1.3 dm2 0.22 kg

Metals and alloys

• Envelope volume on a 5 cm - scale

• example:Z = 12.6 cmX = 6.0 cmY = 2.9 cm

15 x 10 x 5 = 750 cm3

0.75 kg

19


Envelope volume: How to determine the height (Z)

Measure total length of sample (Htotal) ↔

Define length of handle (Hhandle)• measure ↔• if not clearly seperated: Hhandle = ⅓ Htotal ↔

Calculate height reasonably in contact with food (Hr) ↔Hr = Htotal - Hhandle

Measure part necessarily in contact with food (Hn) ↔

Calculate part possibly in contact with food (Hp) ↔Hp = Hr - Hn

Compare Hp and Hn

Hp ≤ 0.5 Hn

Z = Hr

Hp > 0.5 Hn

Z = ⅔ Hr↔ ↔

Hp ≤ 0.5 Hn

Z = ⅔ Hr Z = HrZ = ⅔ Hr

Photos: www.ikea.com


• After calculating Z, measure the two other dimensions (X,Y)• Round every dimension-value to the next 5 cm (to a max. of 30 cm!)• Multiply all dimensions: Envelope volume (EV) = X × Y × Z (Result in cm3)• Reference weight (kg) = EV / 1000

Envelope volume: How to determine the Reference weight

example

• Htotal = 28.4 cmHhandle = ⅓ Htotal = 9.5 cmHr = Htotal - Hhandle = 18.9 cmHn = 10.5 cmHp = Hr – Hn = 8.4 cm

Hp > 0.5 Hn

Z = ⅔ Hr = 12.7 cm

• X = 6.0 cm• Y = 2.9 cm

15 x 10 x 5 = 750 cm3

0.75 kg

20


Results of an inter-laboratory comparison: Hf

• Inter-laboratory comparison 2013 of the EURL on FCM:ILC 003 2013 – Food Contact Surface Area of Kitchen Utensils (EUR 26477 EN)

• 5 plastic Kitchen Utensils • Define the height foreseeably in contact with food (Hf analogous to Z)

• Measure/calculate the Food Contact Surface Area • Determine the Envelope volume on a 2-cm- and a 5-cm-scale

Sample A Sample B Sample C Sample D Sample E

Htotal (cm) 31.0 ± 1.6 27.8 ± 1.2 28.6 ± 1.1 28.8 ± 1.1 29.0 ± 1.0

|zU|≤2 41 65% 49 78% 47 75% 54 86% 45 71%

Hf=Hr 13 21% 2 3% 2 3% 3 5% 1 2%

unassigned error 9 14% 12 19% 14 22% 6 10% 17 27%

Results of the ILC 003 2013, Htotal calculated from Hf and No of results


Results of an inter-laboratory comparison: Envelope volume

• Inter-laboratory comparison 2013 of the EURL on FCM:• Determine the Envelope volume on a 2-cm- and a 5-cm-scale

Sample A Sample B Sample C Sample D Sample E

correct results (2cm) 35 66% 19 36% 32 60% 19 36% 23 43%

error plausible 8 15% 23 43% 11 21% 25 47% 23 43%

error not plausible 10 19% 11 21% 10 19% 9 17% 7 13%

thereof no correct raster (X, Y) 6 11% 3 6% 3 6% 6 11% 7 13%

correct results (5cm) 32 62% 27 52% 47 90% 45 87% 30 58%

error plausible 9 17% 21 40% - - 17 33%

error not plausible 11 21% 4 8% 5 10% 7 13% 5 10%

thereof no correct raster (X, Y) 1 2% 3 6% 1 2% - 2 4%

Results of the ILC 003 2013, No of results

21


3D-Scan

• Low-cost 3D scanner on the principle of structured light scanning (DAVID® Vision Systems GmbH)• Scanning from different (overlapping) angles, aligning of the scans, merging into 3D model


EV 1-cm-scale y = 1,3xR2 = 0,24

0100200300400500600700800

0 100 200 300 400contact volume 10/2011 [cm3]

enve

lope

vol

ume

[cm3 ]

y = x

EV 2-cm-scale y = 1,8xR2 = 0,31

0

200

400

600

800

1000


enve

lope

vol

ume

[cm3 ]

y = x

EV 5-cm-scale y = 4,1x

R2 = 0,27

0

500

1000

1500

2000


enve

lope

vol

ume

[cm3 ]

y = x

3D-Scan-Data: Correlation between Envelope volume and „EU-cube“

• Computer calculation of the dimensions (bounding box) and surface of the 3D-model • Comparison of the Envelope volume and the contact volume derived from the surface area

Correlation between contactvolume derived from thesurface area (6 dm2/dm3, Reg.(EU) No 10/2011) and theEnvelope volume on a 1 cm-,a 2 cm- and a 5 cm-scale.(red line: slope=1,corresponding to an perfectagreement of the results ofboth methods)

22


Summary and Outlook

• The Envelope volume provides a quick and elegant method to determine the surface-to-volume ratio

• BUT: the results of a first inter-laboratory comparison showed significant deviations

• To overcome this, sources of error should be identified (Follow-up to ILC003 2013)

• The description of the method should be clarified according to the outcome of the error discussion

• The envelope volume should always be reported along with the specific release

• The length of the scale may be discussed (1 cm-, 2 cm- or 5 cm-scale)

• The German NRL for FCM will provide more data from 3D-scans for a comparison next year

• We will happily accept samples for that task

FE

DE

RA

L I

NS

TIT

UT

E

FO

R R

ISK

AS

SE

SS

ME

NT Thank you for your attention

Dr. Ingo Ebner

Federal Institute for Risk AssessmentMax-Dohrn-Str. 8-10 10589 Berlin, GERMANYTel. +49 30 - 184 12 - 12 11 Fax +49 30 - 184 12 - 47 [email protected] www.bfr.bund.de

23

EURL-FCM Interlaboratory comparisons on surface area measurements for kitchen utensils and impact on migration results

Anja Mieth, Eddo Hoekstra, Catherine Simoneau

127 October 2014

Institute for Health and Consumer ProtectionChemical Assessment and Testing Unit

Serving societyStimulating innovationSupporting legislation

Who are we and what do we do?

JRC is the European Commission’s in-house science service. It

provides the science for policy decisions, with a view to ensuring

that the EU achieves its Europe 2020 goals for a productive

economy as well as a safe, secure and sustainable future.

The JRC plays a key role in the European Research Area and

reinforces its multi-disciplinarity by networking extensively with

leading scientific organisations in the Member States, Associated

Countries and worldwide.

27 October 2014 2

24

Role as JRC and EU Reference Laboratory

Risk management

(SANCO)

Official controls

Member State Authorities and

Enforcement Laboratories

As FCM activities since 1995 Serving sectorial policies on release of substances from

food contact materials

As EU Reference Laboratory nominated in 2004

Methods, migration data, scientific support for FCM

legislation

Ad-hoc contributions to EFSA for exposure assessment

Work

Supporting Regulation 882/2004 on official food controls Member State

authorities and enforcement Laboratories (NRLs)

Work

Official control laboratories

Article 32.1 of Regulation (EC) No 882/2004 on official controls

• Providing NRLs with details of analytical methods, including reference methods

• Co-ordinating application by NRLs of the methods, in particular by organising comparative testing and by ensuring an appropriate follow-up of such comparative testing in accordance with internationally accepted protocols

• Co-ordinating practical arrangements needed to apply new analytical methods and informing NRLs of advances in this field

• Conducting initial and further training courses for the benefit of staff from NRLs and of experts from developing countries

• Providing scientific and technical assistance to the Commission, especially in cases where Member States contest the results of analysis

• Collaborating with laboratories responsible for analysing feed and food in third countries

Roles of EURL

25

Background

ILC01 2012 Formaldehyde and Melamine in 3% acetic

acid migration solution

additional task for 19 volunteer NRLs:

determination of migration from a melamine spoon

NRLs reported:

• concentration of HCHO in the migration solution

• volume of simulant used to perform the test

• estimation of the contact surface area

(results: 0.73-1.99 dm²)

what were the reasons for the broad distribution of results in

ILC01 2012?

(unacceptable performance?, approaches unsuitable?)27 October 2014 5

exposed during

migration testing

Aims of the ILC03 2013

• to assess the capability of laboratories to measure the surface

area of kitchen utensils

• to compare and assess the most popular approaches for the

determination of the surface area (method validation)

627 October 2014

26

Design of the ILC03 2013

• selected methods for the determination of the surface area:

calculation

wrapping in paper

wrapping in aluminium foil

drawing the outline on paper

• samples: 5 different kitchen utensils purchased from the market

• requested results (single fold determinations, following the given

instructions):

sample height Hf with foreseeable food contact

food contact surface area (as defined by Hf) using each of

the 4 methods

voluntary exercise: determination of envelope volume

• follow-up exercise in 2014 (questionnaire)

727 October 2014

Sample height with foreseeable food contact (Hf)

827 October 2014

photos: www.ikea.com

Htotal total length

Hhandle handle

Hn food contact necessary

Hp food contact possible

Hr food contact reasonable

Hf food contact foreseeable

handle clearly defined

measured values

Htotal, Hhandle, Hn

handle not defined

measured values

Htotal, Hn

Hhandle ≝ 1/3 Htotal

Hp = Htotal-Hhandle-Hn

if Hp>½ Hn: Hf = (Hp+Hn)

if Hp≤½ Hn: Hf = Hp+Hn

in accordance with:EDQM "Metals and Alloys used in food contactmaterials and articles, 1st edition, 2013)

27

Sample A

927 October 2014

• 63 values for Hf

• Two groups (1: Hf < 16 cm, 2: Hf ≥ 19 cm)

PROLab Plus

Laboratory

LC

0044

LC00

48LC

0031

LC00

06LC

0045

LC00

49LC

0054

LC00

56LC

0009

LC00

10LC

0014

LC00

16LC

0017

LC00

24LC

0025

LC00

26LC

0036

LC00

37LC

0038

LC00

60LC

0062

LC00

04LC

0005

LC00

11LC

0015

LC00

20LC

0021

LC00

22LC

0029

LC00

32LC

0040

LC00

61LC

0063

LC00

30LC

0035

LC00

47LC

0055

LC00

02LC

0012

LC00

19LC

0058

LC00

23LC

0003

LC00

59LC

0067

LC00

33LC

0057

LC00

66LC

0028

LC00

42LC

0018

LC00

27LC

0041

LC00

13LC

0007

LC00

08LC

0051

LC00

50LC

0065

LC00

46LC

0052

LC00

53LC

0064

cm

20.520

19.519

18.518

17.517

16.516

15.515

14.514

13.513

12.5

Sample: A (spatula)Measurand: height foreseeable in food contactMethod: DIN 38402 A45No. of laboratories: 63

Assigned value: 13.8 cm (Empirical value)Reproducibility s.d.: 0.3 cmReproducibility (R): 1.0 cmRel. target s.d.: 2.50% (Empirical value)Range of tolerance: 13.1 - 14.5 cm (|Zu score| <= 2.00)

Assign

ed va

lue

Mean

Limit of tolerance

Limit of tolerance

Alarm limit

Alarm limit

Assessment = DIN38402 A45; Assigned value = M; Target s.d. = S; |Zu score| <= 2

10

Hf

Hf

Hf

measured diagonalHtotal = 31.0 cm

Hn = 13.9 cm

Hp = 6.8 cm

measured along curved shapeHtotal = 31.5 cm

Hn = 13.9 cm

Hp = 7.1 cm

smaller value for HnHtotal = 31.0 cm

Hn = 13.6 cm

Hp = 7.1 cm

Hp < ½ Hn

Hf = 20.7 cm

Hp > ½ Hn

Hf = 14.0 cm

Hp > ½ Hn

Hf = 13.8 cm

Hp ≈ ½ Hn

Sample A

27 October 2014

28

1127 October 2014

Determination of Hf - Sample A

measurement method Htotal measurement method Hhandle measurement method Hn

1c

1b

1a others1

1b = 1c

1a

others1

others2

1227 October 2014

+ Hf calculated by EURL from respective values for Htotal, Hhandle and Hn reported by the laboratories during thefollow up exercise

Hf reported in ILC03 2013, no further information available

Determination of Hf - Sample A

Hn > 13.5 cm

handle = mark on the tool

29

Sample E

1327 October 2014

• 63 values for Hf

• one group (Hf < 15 cm), some outliers

PROLab Plus

Laboratory

LC

0048

LC00

67LC

0026

LC00

06LC

0028

LC00

29LC

0016

LC00

02LC

0044

LC00

57LC

0061

LC00

03LC

0008

LC00

45LC

0007

LC00

21LC

0040

LC00

09LC

0010

LC00

38LC

0049

LC00

52LC

0004

LC00

05LC

0011

LC00

12LC

0013

LC00

14LC

0017

LC00

18LC

0019

LC00

20LC

0023

LC00

25LC

0027

LC00

30LC

0031

LC00

32LC

0033

LC00

35LC

0036

LC00

37LC

0041

LC00

46LC

0047

LC00

50LC

0051

LC00

53LC

0054

LC00

55LC

0056

LC00

58LC

0060

LC00

62LC

0063

LC00

64LC

0065

LC00

66LC

0022

LC00

15LC

0059

LC00

42LC

0024

cm

19181716151413121110987654

Sample: E (tongs)Measurand: height foreseeable in food contactMethod: DIN 38402 A45No. of laboratories: 63

Assigned value: 12.9 cm (Empirical value)Reproducibility s.d.: 0.2 cmReproducibility (R): 0.6 cmRel. target s.d.: 1.74% (Empirical value)Range of tolerance: 12.4 - 13.3 cm (|Zu score| <= 2.00)

Assign

ed va

lue

Mean

Limit of tolerance Limit of toleranceAlarm limit Alarm limit

Assessment = DIN38402 A45; Assigned value = M; Target s.d. = S; |Zu score| <= 2

1427 October 2014

Determination of Hf - Sample E

+ Hf calculated by EURL from respective values for Htotal, Hhandle and Hnreported by the participants during the follow up exercise

Hf reported in ILC03 2013, no further information available

handle = groove

1c

1b

1a

others1

1c

1a

others1

others2

others1

others3

others4

1b

30

1527 October 2014

Main difficulties in determination of Hf

mistakes in calculation of Hf itself or in the calculation of the

default value for Hhandle

important which sample part was considered for measurement

of Htotal, Hhandle and Hn

way of measurement of Htotal, Hhandle and Hn (i.e. straight or

along the curved shape) was of minor influence

NOTE: Hf procedure is relevant for the determination of EV, but not for compliance testing according to EU 10/2011

1627 October 2014

Determination of the surface area

Calculation Wrap in paper Wrap in Al foil Draw the shape*

foo

d c

on

tact

su

rface a

rea

Hf(h

eig

ht

in

foo

d c

on

tact)

Hf

*with/without cutting

31

1727 October 2014

surface area reported in ILC03 2013, no further information available

Surface area measurement – sample A

again a broad distribution of results

1827 October 2014

Reasons for surface area values outside of tolerance limits

"calculation mistakes"

• surface area measured only of one side of each sample

• mistake in weighing of paper

• mistake in determining the surface area by calculation

(e.g. use of a formula)

very high or very low value for Hf (not of relevance for

migration results!)

for sample A and B: slotted parts not removed/subtracted

‼ in several cases reasons remain unclear

Hf

32

1927 October 2014

Comparison of methods-3

,9

0,5 9,

0

-8,1

5,5

1,1 3,

8

10,9

7,0 7,8

4,1 5,8 19

,0

13,6

9,2

0,3

-2,0

-2,2

0,1 5,1

-40

-20

0

20

40

A(Hf 13.6-13.9 cm)

B(Hf 12.2-12.4 cm)

C(Hf 12.5-12.8 cm)

D(Hf 12.7-12.9 cm)

E(Hf 12.8-12.9 cm)

bias

to la

ser s

cann

er [%

]

calculation wrap paper wrap Al foil draw shape

Relative robust reprod. s.d. [%]Method A B C D E

calculation 11.7 12.3 10.2 11.4 6.5wrap paper 8.3 13.2 6.6 5.4 5.7wrap Al foil 8.7 10.5 6.8 5.8 7.0draw shape 8.1 13.9 16.1 13.9 10.6

NOTE: only those values selected that were based on the same Hf

(n=28-40)

• high rel. robust reprod. s.d.: 5.4-16.1%

• comparison of results vs. laser scanner:

sample A+B: no significant difference to laser scan

sample C, D, E: surface area is overestimated with “wrap in paper” and “wrap in Al foil”

“wrap in Al foil”: overestimation up to (19.0±8.1)% (crinkles!)

• overall and specific migration:

• total systematic bias of the migration result:

if surface area is systematically overestimated, OM and SM result will be underestimated by the same percentage

‼ PROBLEM: bias unknown in practice → correction of results impossible → bias has to be kept as low as possible

Effect of bias on migration results

2027 October 2014

OM: overall migrationSM: specific migrationx: migrated amount [mg]S: food contact surface area [dm2]

and ⋅

0

or : relative systematic bias for overall or

specific migration [%]

: relative systematic bias for the determination

of the migrated amount [%]

: relative systematic bias for the surface area

measurement [%]

33

• overall and specific migration:

• uncertainty of migration results:

• Example: Migration of HCHO (SML 15 mg/kg food), draw shape, sample B

uncertainties of surface area measurement similar to those of migration measurement → resulting combined uncertainty still acceptable

Effect of measurement uncertainty on migration results

2127 October 2014

OM: overall migrationSM: specific migrationx: migrated amount [mg]S: food contact surface area [dm2]

and ⋅

uOM or SM/OM or SM: combined relative uncertainty (rel.

s.d.) of the OM or SM resultsR(x)/x or sR(S)/S: relative reproducibility standard

deviation for the determination of x or S

ILC01 2012:rel. reproducibility s.d. for determination of HCHO: 12.0%(HCHO-level 9.1 mg/kg)

12.0% 13.9% 18.4%9.1 1.7

2227 October 2014

Summary and conclusions – Surface area measurement uncertainties are acceptable

uncertainty + bias of methods depend on the sample shape

suitable method should be selected based on the sample shape

some difficulties in surface area measurement could be identified

• taking into consideration one/both sides of an article

• subtraction of slotted parts

will be addressed in a guideline on surface area measurement

34

Voluntary exercise – Envelope Volume

Envelope Volume

= estimation for amount of food in contact

with the sample

Determination procedure

• actual dimensions of width (x) and

depth (y) measured (0.1 cm accuracy)

• height z = Hf

• values rounded on a 2-cm- or 5-cm-

scale (= assigned values xass, yass, zass)

• Envelope Volume [cm³] = xass · yass · zass

2327 October 2014

z = Hf

x

15

5

10

z

x

y

see: EDQM "Metals and Alloys used in food contactmaterials and articles, 1st edition, 2013)

2427 October 2014

Main difficulties in the determination of the envelope volume

determination of z• z = Htotal (instead of z = Hf)• problems in determination of Hf as discussed before

determination of x and y• width not measured at widest point but at the very edge• sample E: width measured for compressed tongs (even

more strict!)• depth = material thickness

"calculation mistakes"

recording of values

35

2527 October 2014

Envelope volume – sample B

Hf5a

5b

others1

depth = material thickness

two groups of results due to measurement of depth

+ EV calculated by EURL from respective values for Hf, depth and widthreported by the participants during the follow up exercise

EV reported in ILC03 2013, no further information available

2627 October 2014

Envelope volume – sample E

Hf

Hf

4a

others1

5a

others2

others1

others2

5b

+ EV calculated by EURL from respective values for Hf, depth and widthreported by the participants during the follow up exercise

EV reported in ILC03 2013, no further information available

width = compressed tong

Hf

36

2727 October 2014

Summary and conclusions – Envelope volume

determination of EV is convenient

main difficulties in determination of Hf and EV

• which sample part needs to be considered for measurement of Htotal, Hhandle and Hn?

• when to consider a handle as clearly separated?• how to measure the width of compressible items?

clarification/discussion needed

Thank you for your attention!

2827 October 2014

37

TÜV SÜD Product Service GmbH Folie 103.11.2014

Release testing according to the Practical Guide, Council of Europe 2013

Markus Jahns – Senior Product Specialist Chemical

Content


Introduction

Experiences

Challenges

Summary

38

TÜV SÜD Product Service

TÜV SÜD Product Service GmbH Slide 303.11.2014

Employees: 3.691

Sales Revenue: € 317 million

Employees: 690Sales Revenue: € 99 million

Employees: 448Sales Revenue: € 64 million

• Global

Europe

Germany

Introduction


What we do in the area of food contact materials

• TÜV SÜD provides testing and certification in several different schemes

(CB, GS, etc.)

• A large number of house hold appliances with food contact is certified with

GS-Mark and TÜV SÜD own mark

• The products are sold in many countries with different requirements

39

Content

TÜV SÜD Product Service GmbH Folie 503/11/2014

Introduction

Experiences

Challenges

Summary

Experiences


Evaluation of current projects

• 51 projects in total

• Household appliances including mixers, food processors and baking

ovens

• All projects met the SRL from EDQM guideline!

• Simulant used was Citric acid acc. to EDQM guideline

40

Eperiences



0

5

10

15

20

25

30

35

40

Metal and Alloy Components

Element detected in Project but below SRL

Eperiences



0

2

4

6

8

10

12

14

16

18

Arsenic Barium Beryllium Cadmium Lead Lithium Mercury Thallium

Contaminants and Impurities

Elements detected but below SRL

41

Experiences



• 4 contaminant detectable: As (max. 0.002 mg/kg), Pb (max. 0.016 mg/kg),

Ba (max. 0.028 mg/kg) and Li (max. 0.001 mg/kg)

• All below SRL

• For household appliances no major fail results in the area of common part

like blades, graters and cooking ware -> materials are well developed

Element SRL [mg/kg]

As 0,002

Ba 1.2

Pb 0.010

Li 0.048

Experiences



• Comparison of Migration in Acetic Acid versus Citric Acid for 3 projects

• One Silver Coated Steel Sample and 2 Silver Coated Brass Samples

• Sampling according to EDQM Guidance

Document

• Migration conditions: 70°C, 2 hours

42

Experiences



20,3

7,7

39,8

13,5

0

5

10

15

20

25

30

35

40

45

Migration of Nickel in mg/kg

Migration of Nickel from Brass with Silver Coating

1st + 2nd Migration Citric Acid 3rd Migration Citric Acid

1st + 2nd Migration Acetic Acid 3rd Migration Acetic Acid

0,36 0,13

7,5

3,25

0

1

2

3

4

5

6

7

8


Migration of Nickel from a Steel Plate with Silver Coating



Experiences



48

24

87

41,2

0

10

20

30

40

50

60

70

80

90

100


Migration of Nickel from Brass with Silver Coating 2



43

Experiences



0

0,01

0,02

0,03

0,04

0,05

0,06

0,07

0,08

0,09

1st + 2nd MigrationCitric Acid 3rd Migration Citric Acid

1st + 2nd MigrationAcetic Acid 3rd Migration Acetic Acid

<0.010<0.010

0,05

0,03<0.010

<0.010

0,05

<0.010

0,0490,047

0,081 0,084

Migration of Lead from Certain Samples in mg/kg

Steel Plate with Silver Coating Brass with Silver Coating Brass with Silver Coating



Sample Migration with Citric Acid Migration with Acetic Acid

Steel Plate with Silver Coating Pass Fail for Pb & Ni

Brass with Silver Coating Fail for Ni Fail for Ni

Gbrass with Silver Coating Fail for Ni & Pb Fail for Ni & Pb

• Results for Nickel and Lead in Citric Acid significantly lower than in Acetic Acid

• 1 case of 3 shifts from Fail to Pass result

• Further investigations are ongoing to establish a profound database

44

Content


Introduction

Experiences

Challenges

Summary

Challenges


Plastics -> PIM

• European Regulation with

clear legal status and

European applicability

Ceramics ->

• Directive with the need to

be transposed into

national law of the

European member states

Metals and Alloys ->

• EU-Resolution CM/Res

(2013)9 on metals and

alloys used in food contact

• Linked to EDQM Guidance

Document

45

Technical Challenges


Metals and alloys versus

• EDQM guide requires citric acid as a simulant

Verification tests on the final product

with material combinations

DOC for metals and alloys

recommended

Plastics in contact with food

• 10/2011/EU requires acetic acid as simulant

Legally required by the Regulation

DOC for Plastics required

Challenges


Implementation

• According to the Resolution the Guidance Document should be implemented into national law of European member states -> No harmonized approach

• European Single Market?

Applicability

• Guide line is not applicable to enamel articles in contact with food

• During the past it was applied to enamel

Transition Period

• A transition period for certain elements was granted but not announced officially

46

Challenges


Legal implementation

Challenges


Certification – Harmonized Conditions

• GS-Mark requires the fulfilment of all applicable laws etc.

• Experience Exchange Circles develop test programs for „non-harmonized“ testing

• Small baking ovens need to comply with legal requirments from 1935/2004/EG etc.

• In several cases the inner surface released Zn, Fe, Ni and Tl above SRL, but some

institutes considered the inner surface as a non-food contact surface

• Condenser effects considered the related working group decided to evaluate inner

surface for the GS-Marking as a food contact surface

• Small baking ovens are in the most cases advertised with a nearly completely

stuffed interior -> reasonable decision

47

Challenges


Household Appliances with Combined Materials

DirectiveRegulation

Guideline

Content


Introduction

Experiences

Challenges

Summary

48

Summary


• A large amount of materials is able to fulfil EDQM guideline‘s requirements

• Citric Acid seems to be less severe for certain materials and elements -> further

investigation necessary

• Applicability of EDQM guide for certain items like house hold appliances needs

to be addressed more clearly

• Legal status of guideline needs to be strengthened

• Difference in legislation with parallel regulation, directive, guidance and partial

agreements on the single European market has improvement potential

• A guideline for complex products with combined materials should be established

Discussion


Thank you very much!

49

1

Application of CoE CM/Res(2013)9

November 6/7, 2014Metals and Alloys in Contact with Food, Ljubljana, Slovenia

Dr. Ulrich NehringINSTITUT NEHRING GmbH - Braunschweig

Test conditions for hermeticallysealed containers and „pack tests“

2

Background

The edqm/CoE-Practical Guide contains several remarksregarding specific testing conditions for food cans

Any metal and alloy used in food contact materials andarticles has to comply with Art. 4 of CM/Res (2013) 9. This, of course, applies also for metals and alloys usedin cans.

Why ? How ?

50

3

Why is release testing for metals from metalpackaging different from other types of foodcontact articles and requires specifictesting conditions?

Why ?

4

Corrosion is different from migration Why ?

• Electrochemical reaction between metal and liquid reactants (food)

• Influenced by redox potential of the metals, pH value ofthe reactant, presence of oxygen, composition of thereactant (food)

51

5

Corrosion is different from migration Why ?

• Corrosion is restricted to the metal surface

• No or limited influence of temperature

6

Cans are different Why ?

52

7

Metals used for cans and closures

• Tin plate steel plated with tincoated and plain (not coated)

• Aluminium coated

• Tin free steel steel plated with chromiumcoated

• Steel coated

Why ?

8

Properties of canned food Why ?

• Hermetical sealing prevents the access of oxygen (hotfilling and/or steam injection)

• High temperatures (up to 145 °C) for short time (up to30 min) during processing

• Long shelf life (up to 5 years, storage at ambienttemperature)

• Low residual oxygen in the can due to vacuum

53

9

The effect of oxygenMandarin oranges in light syrup, 3pc tinplate can, plain body

Why ?

10

The effect of oxygenMandarin oranges in light syrup, 3pc tinplate can, plain body

Why ?

54

11

Properties of canned food Why ?

• Hermetical sealing prevents the access of oxygen (hotfilling and/or steam injection)

• High temperatures (up to 145 °C) for short time (up to30 min) during processing

• Long shelf life (up to 5 years, storage at ambienttemperature)

• Low residual oxygen in the can due to vacuum

• In most cases internal surface protected by a polymericcoating

12

Use of partly coated or plain tinplate cans

• light fruits in fruit juice, water or syrup (e.g.pineapplaes, pears, peaches, apricots, mandarin-oranges)

• tomato products (e.g. tomato concentrate, peeledtomatoes in tomato juice, tomato sauce)

• mushrooms in brine (e.g.whole mushrooms, mushroomslices)

• dehydrated food (e.g. milk powder, instant baby food, coffee)

• vegetable oils (e.g. olive oil)

Why ?

55

13

Release testing with food simulants (1) Why ?

• If aluminium or tinplate cans are brought into contactwith aqueous food simulants or foodstuffs without hotfilling and without proper seaming due to oxygenaccess the behaviour of the metal packaging will significantly differ from real food applications

• Back in 1994 CEN TC194 has demonstrated thatmigration testing for coatings on metal substrates like tinplate or aluminium does not give reliable and usefullresults because migration is interferred by excessivecorrosion (see CEN TC194 SC1 WG5 Doc15/Rev1)

14

Release testing with food simulants (2) Why ?

• Excessive corrosion has also to be expected ifaluminium or tinplate cans are tested with 0,5 % citricacid or other acidic food simulants by filling the canswithout seaming

• Different from all types of acidic foodstuffs acidic foodsimulants like 0,5 % citric acid do not have anybuffering effect

• The metal release from a cans or closures into cannedfoodstuff throughout a shelf life of up to five yearscannot be appropriately simulated within short testingtime under laboratory conditions (forced conditions).

56

15

How is appropriate release testing formetals from metal packaging carried out?

How ?

16

Standards for metallic substrates How ?

• GMP: For metal packaging only plate according tostandards is used

• Standards set maximum limits for toxic metals andother impurities in the plate

Tinmill productsEN 10202, EN 10295, EN 10333, EN 10334, EN 10335Aluminium EN 602, EN 541, EN 573-1 and EN 573-3EN 14287, EN 14392

57

17

DIN EN 10334 Packaging steel How ?

18

Standards for metallic substrates How ?

• GMP: For metal packaging only plate according tostandards is used

• Standards set maximum limits for toxic metals andother impurities in the plate

Tinmill productsEN 10202, EN 10295, EN 10333, EN 10334, EN 10335Aluminium EN 602, EN 541, EN 573-1 and EN 573-3EN 14287, EN 14392

• Restrictions for impurities prevent from release ofelements in harmfull concentrations in the case ofunintended corrosion

58

19

Internally coated metal packaging

• The major purpose of internal protection lacquers formetal packaging is the prevention of chemicalinteraction between the metal surface and the foodstuff.

• Cans and closures with a completely coated internal surface do not release significant amounts of metalsinto the foodstuff because- metal substrates comply with standards- pack tests prove the absence of significant corrosion

How ?

• For internally coated metal packaging according toArt. 2 para. 2.2 the application of CM/Res(2013)9 can be excluded without further release testing

20

Pack tests

• As part of product development and qualification metalpackaging manufacturers and fillers usually carry out pack tests.

How ?

59

I. Preliminary Tests

II. Industrial Trial

III. Validation

IV. Industrial Development (Scale up)

V. Final Approval

Pack performance approval

Manufacturing approval

Canmaker qualification procedure

a few

hundreds

thousands

millions

full production

21

How ?

22

Pack tests


• The conditions of pack tests are selected in order torepresent worst case conditions of industrialmanufacture and storage within the supply chain

How ?

60

23

Process of canning

empty canfilling

(warm or cold)

hermeticalseaming of can

heat treatment(sterilisation pH >4,5,

pasteurisation pH <4,5)

cooling(ambient temperature)

final product(preserve)

How ?

24

Procedure of pack tests How ?

Picture: Tata Steel, NL

1. Filling and seaming

Laboratory filling device forbeverage cans

Cans, glass jars or bottles arefilled with representativefoodstuff under simulated orreal industrial conditions.

The filling conditions areselected in order to achievelow O2 concentrations(vacuum or CO2 pressure)

61

25

Procedure of pack tests

2. Thermal processing (retort)

How ?

Laboratory retort forpasteurisation and sterilisation


The filled containers aresterilised or pasteurised usinglaboratory retorts or industrialequipment.

Time, temperature andpressure have to be controlledaccording to the real industrialretort process.

26

Procedure of pack tests How ?

Storage under controlled thermal and humidity conditions


3. Storage and regular testingThe cans are stored undercontrolled conditions for thetime of the entire intendedshelf life.

Samples are regularly drawnand tested for performance ofthe metal packaging andcondition of the foodstuff. Metal release is tested in caseof indication of corrosion.

62

27

Pack tests


• The conditions of pack tests are selected in order torepresent worst case conditions of industrialmanufacture and storage within the supply chain

• Accelerated tests under forced conditions (elevatedtemperature, shaking) can replace long term pack testsonly after full specific validation

How ?

• Results of pack tests prove technical suitablity of themetal packaging in contact with the respective foodstuffas well as the shelf life of the product

28

Conclusions

• Internally completely coated cans are excluded fromCoE CM/Res(2013)9

• Metal packaging is different from other food contactarticles and require specific test conditions in order tomeasure metal release into foodstuff

• It is important that exemptions and specific test conditionsmentioned in the Practical Guide for metal packaging andcanned food are regarded

• Release testing has to be carried out in real food andafter realistic contact conditions (seamed and retorted cans,long term storage)

• Release testing can be restricted to those metals whichpotentially can exceed the SRL

63

29

Thank you very much for your attention !

INSTITUT NEHRING GmbHHeesfeld 17 – D38112 Braunschweig

fon +49-531-238990 fax +49531-2389977info(at)institut-nehring.de www.institut-nehring.de

64

–Metals and Alloys Nordic Guidance for industry and trade Grímur Ólafsson

Nordic cooperation on FCM

• The Nordic Council of Ministers

• Long tradition for cooperation in

• the field of food and nutrition• and later on FCM

• Participants: Denmark, Finland, the Faroe Islands, Iceland, Norway and Sweden

65

Nordic cooperation on FCM

Previous projects include:

1. Paper an board – 2008

2. In-house documentation – 2008

3. Printing inks – 2012

Ongoing projects:

• Checklist for FCM (an update of In-house doc) • Metals and alloys

The purpose of the project

• M&A are not regulated as FCM in the EU

• Contaminates regulation 1881/2006 (Pb, Cd, Hg, Sn)

• General requirements in EU regulation 1935/2004 and 2023/2006 apply

• Guidance document

• trade and industry, • official food inspection

66

Contents

Chapter 1 Food contact materials - use of metals and alloys

Chapter 2 Legislation

Chapter 3 Guidance limits for migration

Main component: Al, Sb, Cr, Co, Cu, Fe, Mg, Mn, Mo, Ni, Ag, Sn, Ti, V, Zn

Impurity: As, Ba, Be, Cd, Pb, Li, Hg, TlChapter 4 Proposals for future work

Annex I Links to legislation and guidance: EU and Nordic webpages

Annex II Analytical testingAnnex III Alloys and stainless steel

SML and risk assessment

• Collates, evaluates and recommends risk based guidance limits for migration

• The report is based:

Council of Europe resolution EFSA risk assessments, and JECFA risk assessments

67

Migration of metals from FCM

Considerations when setting limits:

• Toxicological reference dose (TDI)• How much of the TDI should be allocated?• Does regulation 1881/2006 set maximum levels in

food?• Does the ALARA principle apply? • What is the analytical quantification level?

The Council of Europe SML

Metal Council of Europe SRLa

(mg/kg)Aluminiumb 5Antimony 0.04Arsenic 0.002Bariumb 1.2Beryllium 0.01Cadmiumb 0.005Chromium 0.250Cobaltb 0.02Copperb 4Ironb 40Leadb 0.010Lithiumb 0.048Magnesium -cManganeseb 1.8Mercury 0.003Molybdenumb 0.12Nickelb 0.14Silverb 0.08Thallium 0.0001Tinb 100Titanium -cVanadium 0.01Zincb 5

68

Guidance limits

Intended to assist:

• Industry and trade

• Food inspectors

For guidance only – no legal effect

FCM exceeding limits:

evaluate case by case

The tables summarize information on:

• Source and dietary intake• Toxicological effects• Risk assessment • Co E limits• limits set by other bodies• Nordic comments• Analysis• Which FCM• Food sources• Special recommendations.

69

Tables - Al Aluminium (Al) - Main component What and where Aluminium is the third most abundant element in the earths crust and occurs in many in minerals. In addition to FCM, aluminium is used in some food additives, especially in colours and also in antiperspirants, which can contribute to the total exposure of aluminium. Typical dietary intakes range from 0.2 to 1.5 mg/kg bodyweight/week. Toxicological effects Aluminium is excreted by the kidneys, and only a small amount is absorbed. The tolerable daily intake set by EFSA is based on effects on the developing brain, neuro and embryotoxicity. Some consumers show sensitivity towards aluminium. Risk assessments EFSA has in 2008 set a PTWI (provisional tolerableweekly intake) of 1 mg/kg body weight per week. Intake may exceed the PTWI. JECFA has in 2011 set a PTWI of 2 mg/kg bodyweight/week Limit values set by Council of Europe The Council of Europe has in 2013 set a specific release limit of 5 mg/kg based on the ALARA principle. Limit values set by other bodies EU Drinking water directive (98/83) has set a limit of 200 microgram/liter. Nordic comments 10 % of the EFSA PTWI corresponds to 1 mg/kg. However, the natural levels found in foodstuffs can be high compared to migration and can complicate a low migration limit. Analysis The limit of quantification by ICP-MS is approximately 0.1 mg/kg in food. Contamination during the analysis is very common. FCM Aluminium is used for e.g. sauce pans, coffee potsand cans, and it is often a constituent in alloys Processed food Industrially processed foods with food additives containing aluminium. Recommendations Storage of acidic, alkaline or salty food, especially as liquids, in uncoated aluminium should be avoided. Uncoated aluminium should be labelled for users. An example of such a label could be: “User information: do not use this utensil for storage and processing of acidic, alkaline or salty food”.

Tables - Hg Mercury (Hg) - Impurity What and where? Mercury contamination originates primarily from vulcanic and industrial activity. Mercury accumulates in the food chain with highest levels found in big predatory fish such as swordfish, tuna and marine mammals. Typical dietary intakes range from 0.02 to 0.13 microgram/ kg bodyweight/day. Toxicological effects The most sensitive organ is the developing brain of fetuses and small children. Mercury is toxic to the nervous system and causes a risk for slow brain development and learning for children. Methyl mercury is considered as one of the most dangerous chemicals for humans. Risk assessments EFSA has in 2012 set a tolerable weekly intake (TWI) for methylmercury at 1.3 microgram/kg body weight/week expressed as mercury. Limits set by Council of Europe

The Council of Europe has in 2013 set a specific release limit for mercury at 0.003 mg/kg allocating 10% of the toxicological safety limit of JECFA of 0.004 mg/kg bodyweight/week and assuming that a 60 kg person consumes 1 kg of food. Limit values set by other bodies EU maximum limits for mercury in fish and fishery products have been set in regulation 1881/2006. The levels vary from 0.5 to 1.0 mg/kg. For vegetables, fruits and meat EU maximum levels are set in the legislation for pesticide residues (regulation 396/2005). EU drinking water directive (98/83) 1.0 microgram/liter.Nordic comments No comments

Analysis The limit of quantification by ICP-MS is approximately 0.01 mg/kg in food. FCM Mercury has no known use in food concact materials. Mercury forms alloys (amalgam) with almost all other metals except iron. One example is dental amalgam, which contains tin and silver dissolved in mercury. Dental amalgam is banned in Denmark, Norway and Sweden and the use is discouraged in the other EU countries.

70

How will this publication be used?

• The situation in the Nordic countries varies

- Denmark and Finland have prioritized FCM• National regulations on FCM have not been set

• Co E resolutions and Nordic guidelines have not been transposed into national regulations

How will this publication be used?

• A short practical guideline (54 pages)

• For guidance only – no legal effect

• If a product exceeds guideline value it is considered case by case

• In Norway all FCM have to issue a declaration of compliance and supporting documents

71

The Roadmap 2012

• Roadmap for future action in this area of non-plastic food contact materials

• General wish from the MS to continue this work

• Priorities – printing inks, coatings, m&a?

• Co E work on non-plastic materials is necessary and important

Proposals for future work

• Knowledge on actual migration is often limited

• Migration values are to a large extent based on theoretical estimations and,

• Calculations based on toxicological threshold values

• Lower limits could possibly be achievable

72

Proposals for future work

• Many metal FCM’s coated – fluoropolymers

• Knowledge on impurities in the metals used and

• Migration when the coating is no longer intact is limited

The Nordic guideline

• The Nordic guideline will be published in English

• The guideline can be downloaded for free from the NCM homepage:

http://www.norden.org/en/search?SearchableText=food+contact+material

73

Thank you for your attention !!

74

Testing of metallic food contact materialsin an NRL

Andreja Zorič, National Laboratory of Health, Environment and FoodViviana Golja, National Institute of Public Health

National Laboratory of Health, Environment and Food

• main public health laboratory in Slovenia

• food contact materials:

- mainly compliance testing of official control samples, safety andrisk assessment of those samples

- cooperation in preparation of national inspection surveyprogrammes

- expert support to the competent authority and inspectionservices

- participation in working groups on european level

- involvement in public information and expert educationalwork in the field.

Testing of metallic food contact materials in an NRL 2

75

National Laboratory of Health, Environment and Food

• accredited according to EN ISO/IEC 17025:2005 for:

15 methods on FCM testing

accredited method for Cr and Ni release

into 3% acetic acid with ICPMS

• carries out tasks of Slovenian NRL for food contact materials

regulation EC/882/2004 (Art. 33)


Legislation on metalic FCM

• framework regulation 1935/2004/EC (art.3) – general requirements

• national legislation until september 2012

stainless steel Cr, Ni, Mn < 0,1 mg/l; compositionalrequirements, impurities (Pb, As, Cd, Zn, Al, Cu), provisions for cans, tubes, alu foil, coatings

• september 2012 - 2013

old legislation as guidance, risk assessment when limitsexceeded

• y. 2014

Technical guide on metals and alloys used in food contact materials and articles (Council of Europe)


76

Metals release testing

• national legislation until y.2010

test conditions (T, t, simulant) defined for stainlesssteel: ambient temperature vs. boiling contents

other materials: actual conditions of use

• since y.2010

Guidelines on Testing Conditions for Articles In Contact With Foodstuffs - A CRL-NRL-FCM Publication, 1st

Ed. 2009

unified reference consensus guidelines for NRLs



• y.2014

starting to use Technical guide on metals and alloys

NEW: food simulants

articles for repeated use (1st and 2nd test resultrequirement)

envelope volume approach

for articles that cannot be filled


77


Analytical procedures

• testing is carried out under the resonable worst-caseconditions with respect to time, temperature, foodcomposition

• successive tests for repeated

use articles

• pretreatment of samples before

release testing (cleaning according

to the manufacturer‘s instruction)

• release testing into foodstuffs (tomato, dough)



Quality control procedures

• callibrated equipment

• control samples (spiked simulant)

• proficiency testing (avaliabiliy?)


78

Results stainless steel 2009-2012

TEST CONDITIONS Cr Ni TEST CONDITIONS Cr NiSAMPLES (y.2009) t / h T / °C mg/l mg/l SAMPLES (y.2009) t / h T / °C mg/l mg/l

1 Knife * 24 20 13.47 0,11 20 Knife 8 cm 24 20 < 0,01 < 0,012 Knife * 24 20 13,49 0,09 21 Knife 15 cm 24 20 0,07 < 0,013 Knife * 24 20 0,96 0,02 22 Knife 19 cm 24 20 0,04 < 0,014 Pizza knife * 24 20 0,02 < 0,01 23 Coffee pot 0,5 100 0,01 < 0,015 Knife * 24 20 0,20 6,55 24 Knife 24 20 < 0,01 0,086 Thermo flask * 24 20 < 0,01 < 0,01 25 Spoon 24 20 < 0,01 0,077 Baking accessories * 0,5 100 0,03 < 0,01 26 Fork 24 20 < 0,01 0,078 Knife * 24 20 4,98 0,04 27 Pocket knife 24 20 0,08 9,339 Bowl 0,5 100 0,05 0,23 28 Masher * 24 20 < 0,01 < 0,01

10 Spoon 24 20 < 0,01 < 0,01 29 Blade 24 20 < 0,01 < 0,0111 Knife 24 20 < 0,01 < 0,01 30 Stainless steel plate 4 20 < 0,01 < 0,0112 Fork 24 20 < 0,01 < 0,01 31 Knife 24 20 1,85 0,0413 Knife 8 cm 24 20 0,01 < 0,01 32 Food processor 0,5 40 < 0,01 < 0,0114 Knife 12 cm 24 20 0,13 < 0,01 33 Juicer 0,5 40 < 0,01 < 0,0115 Scissors 24 20 0,57 < 0,01 34 Stainless steel lids 24 20 0,02 < 0,0116 Knife 20 cm wide 24 20 10,90 0,10 35 Knife 0,5 40 < 0,01 < 0,0117 Knife 20 cm 24 20 5,68 0,08 36 Thermo flask 24 100, then 25 < 0,01 < 0,0118 Thermo flask * 24 20 < 0,01 < 0,01 37 Peeler 0,5 40 < 0,01 < 0,0119 Flask * 24 20 < 0,01 5,95 38 Knife 0,5 40 < 0,01 < 0,01

2

Notes: Results marked with * are from 1st test. Food simulant was 3% acetic acid.





1 Garlic press 0,5 40 < 0,01 < 0,01 22 Fork 2 70 < 0,01 < 0,012 Flask * 240 40 < 0,01 < 0,01 23 Potatoe peeler 0,5 40 < 0,01 < 0,013 Jar 0,5 100 0,04 < 0,01 24 Strainer 2 70 < 0,01 < 0,014 Coffee pot 0,5 100 < 0,01 < 0,01 25 Strainer 2 70 < 0,01 < 0,015 Bowl 2 70 0,01 < 0,01 26 Pot 4 100 0,05 0,026 Plate 2 70 0,01 < 0,01 27 Cup 2 70 < 0,01 < 0,017 Pot 4 100 0,04 0,02 28 Plate 2 70 < 0,01 < 0,018 Pot 4 100 0,04 0,01 29 Salad bowl 2 70 < 0,01 < 0,019 Scraper 0,5 40 0,02 < 0,01 30 Kiwi knife 0,5 40 < 0,01 < 0,01

10 Ice cream spoon 0,5 20 < 0,01 < 0,01 31 Knife 2 70 < 0,01 < 0,0111 Mug 0,5 100 0,03 < 0,01 32 Flask 2 70 < 0,01 < 0,0112 Potatoe press 0,5 70 < 0,01 < 0,01 33 Metalic cups 0,5 40 < 0,01 < 0,0113 Garlic press 0,5 40 < 0,01 < 0,01 34 Knife 2 70 < 0,01 /14 Sugar box 24 40 < 0,01 < 0,01 35 Knife 2 70 2,82 /15 Garlic press 0,5 40 < 0,01 < 0,01 36 Scraper 0,5 40 0,01 < 0,0116 Scraper 0,5 40 < 0,01 < 0,01 37 Pot 4 100 0,08 < 0,0117 Spoon 2 70 < 0,01 < 0,01 38 Knife - sample 1 2 70 1,30 0,0118 Cheese scraper 0,5 40 < 0,01 < 0,01 39 Garlic press 0,5 40 < 0,01 < 0,0119 Scraper 0,5 40 < 0,01 < 0,01 40 Whisk 0,5 100 < 0,01 < 0,0120 Knife 2 70 1,85 0,02 41 Strainer 2 70 < 0,01 < 0,0121 Spoon 2 70 < 0,01 < 0,01 42 Masher 2 70 < 0,01 0,04

Notes: Food simulant for all samples except* was 3% acetic acid. Simulant for sample* was 40% ethanol. All results are from 3rd test.

79




1 Steel plate 240 40 0,06 0,06 15 Scraper 0,5 40 < 0,01 < 0,012 Pizza pan 0,5 100 < 0,01 < 0,01 16 Electric barbecue 2, then 24 70, then 40 < 0,01 0,073 Pizza baking spatula 0,5 100 0,01 < 0,01 17 Bowl 2, then 24 70, then 40 < 0,01 < 0,014 Knife 2 70 0,06 < 0,01 18 Ladle 2 70 < 0,01 < 0,015 Mug 2 70 < 0,01 < 0,01 19 Coffee pot 0,5 100 0,01 < 0,016 Spoon 2 70 < 0,01 < 0,01 20 Salad bowl 2, then 24 70, then 40 0,01 < 0,017 Knife 2 70 < 0,01 < 0,01 21 Pot 4 100 < 0,01 < 0,018 Scraper 0,5 40 < 0,01 < 0,01 22 Pot 2 100 0,01 < 0,019 Ice cream spoon 0,5 5 < 0,01 < 0,01 23 Pot 4 100 0,02 0,01

10 Pot 1 100 0,30 < 0,01 24 Pot 2 100 < 0,01 < 0,0111 Spoon 2 70 < 0,01 < 0,01 25 Baking spatula 2 100 < 0,01 < 0,0112 Butcher's axe 0,5 40 < 0,01 < 0,01 26 Preparation knife 0,5 40 < 0,01 < 0,0113 Spoon 2 70 < 0,01 < 0,01 27 Pot for cream 24 40 < 0,01 < 0,0114 Ladle 2 100 < 0,01 < 0,01 28 Electric barbecue * 1 100 < 0,07 <0,07

Notes: All results are from 3rd test. Food simulant was 3% acetic acid, except for sample *which was tested with food-tomato.


TEST CONDITIONS Cr Ni TEST CONDITIONS Cr Ni

SAMPLES (y.2012) t / h T / °C mg/l mg/l SAMPLES (y.2012) t / h T / °C mg/l mg/l1 Knife 0,5 40 0,72 <0,01 18 Pot 0,5 100 <0,01 <0,012 Grilling grate * 1 100 <0,06 0,11 19 Big spoon 2 70 <0,01 <0,013 Food chopper 0,5 40 <0,01 <0,01 20 Food mixer 0,5 70 <0,01 <0,014 Food chopper 0,5 40 <0,01 <0,01 21 Inox tray 24 40 0,02 0,015 Food mixer 0,5 40 <0,01 <0,01 22 Grater 0,5 40 <0,01 <0,01

6 Food mixer 0,5 40 <0,01 <0,01 23 Spoon 2 70 <0,01 <0,01

7 Mini oven * 1 100 <0,06 <0,06 24 Pot 0,5 100 <0,01 <0,018 Inox tap 24 40 <0,01 0,01 25 Baking tray 2 100 <0,01 <0,019 Pot 2 100 <0,01 <0,01 26 Cup 2 70 0,01 <0,01

10 Coffee pot 0,5 100 <0,01 <0,01 27 Pot 2 100 0,01 <0,01

11 Spoon 2 70 <0,01 <0,01 28 Pizza server ** pizza serving <0,09 /12 Meat fork 2 70 <0,01 <0,01 29 Pizza knife ** pizza cutting <0,09 /13 Fork 2 70 <0,01 <0,01 30 Pot 2 100 0,01 <0,0114 Spreading spatula 2 70 3,95 0,03 31 Food chopper 0,5 70 <0,01 <0,0115 Coffee pot 0,5 100 0,01 <0,01 32 Skimmer 2 100 <0,01 <0,0116 Pot 2 70 0,01 <0,01 33 Baby bottles 0.17, then 2 100,then 70 <0,01 <0,0117 Pizza knife 2 70 2,70 0,04

Notes: All results are from 3rd test. Food simulant was 3% acetic acid, except for samples * which were tested with food-tomato and samples ** which were tested with pizza.


80


- %of samples is not decreasing

- more exceeding results for Cr

- majority of tests with 3% acetic acid

% of samples exceeding SRL Cr 0,25mg/kg (transit.1,0)

Ni 0,14mg/kg (transit. 0,7)


Conclusions

• test results exceeding limit values – testing is important

• Technical guide is essential guidance document andexpected to be recognised on national level

• harmonised approach to testing is crucial for officialcontrol laboratories


81

Documents

METALS AND ALLOYS IN CONTACT WITH FOODMETALS AND ALLOYS IN CONTACT WITH FOOD How to establish compliance with European regulations and recommendations 6-7 NOVEMBER2014, LJUBLJANA,