FEATURES limit ofthe "nickel directive". But the directive does not apply to coins, and rightly so: interpolatingthe release rates obtained, tak­ Euro coins and the ing into account that the manipulation ofa coin typically takes less than three seconds ratherthan a week, one finds numbers that are several orders of magnitude lower than what is actually potential risk of nickel observed in tests that simulate the daily manipulation ofcoins [11,7]. The ENI811 test is relevantto long-term solvation upon contactwith the skin, butbecause itdoes notaccountfor the fric­ allergy tion that characterises the manipulation of coins, it is not representative ofthe amountofnickel transferredto thefingers by l 3 Paul-Guy Fournier , Thomas R. Goversl, and Anne Brun the handling of coinage. As anyone can verify when taking a 1 Laboratoire de Spectroscopie de Translation, Universite Paris­ shower, friction is much more efficient intransferringcontamina­ Sud,Orsay tion than frictionless solvation. 2 Aecono Consulting, Paris 3 Service Medical du Travail, Universitti Paris-Sud, Orsay Manipulation tests To take friction into account,we have evaluatedmetal contamina­ tion from coins by means of a simple test that simulates daily The concern about nickel in coins handling [7]; participating volunteers count a set ofcoins by HE choice ofdesign and material to be used in the manufac­ transferring them from one polyethylene container to another, Tture ofthe European common coinage has been the subject and metal contamination sampledbywiping the fingers with cel­ ofmuch debate [1]. The incorporation ofnickel, inparticular,was lulose cloths is analysed by inductively-coupled plasma emission questioned because of its possible contribution to contact der­ spectroscopy (ICP-OES). Details ofthe procedure canbefound in matitis: nickel allergy is reported to affect more than 10% of references [7] and [12]. women and several % ofmen in industrialised countries [2]. A Table 1 summarizes the results ofsuch tests carried out by 3 compromise was reached by limiting nickel to the two highest people countingthree sets of25 used coins each: bi-metallic euros denominations: the 1- and 2€ coins. (12 pieces of 1€ and 13 of2€), Cu75Ni25 copper-nickel coins The surface of both combines a (2SFr), and pure-nickel2FF pieces. The first two sets were taken white copper-nickel alloy from circulation in a non-industrial environment early 2003, and nickel allergy is (Cu75Ni25) with a yellow nickel­ the last in the beginning of2002. These 2FFcoins had since been brass (Cu75Zn20Ni5), the 1€ stored in a closed polyethylene container. All coins were counted having a yellow outer ring and a as collected. reported to affect white centre, the 2€ having a white Ambient [71 1 &2€ Cu75Ni25 Ni100 outer ring and a yellow centre. The Ni 3.1 t 0.8 0.31 ± 0.08 0.50tO.16 0.55 to.23 more than 10% of six other euro coins do not contain Cu 11.4±3.5 1.95 ± 0.54 1.77 t 0.54 0.16 ± 0.09 nickel. Zn 18.8±5.8 0.29 ± 0.09 0.04 ± 0.03 0.09 ± 0.06 2 women and Several recent articles dealing with Area 2 faces (cm \ 8.5 & 10.4 11.8 11.0 leaching experiments or patch tests several % of men have renewed attention to the , ... Table 1: Average contamination levels resulting from the potential allergy risk ofthe 1- and manipulation ofa single used coin.The second data column lists in in dustria Iised 2€pieces [2-4]. The work published numbers obtained by counting a set of 12 1€ and 13 2€ pieces; by Nestle et a1. in the September the third column was obtained with 25 2SFr coins, and the last with th 12 , 2002, issue of Nature [4], in 25 2FF pieces. All data are expressed in I-Ig. Uncertainties are 90% countries particular, has been given wide­ confidence limits.The first column recalls the average spread public coverage, claiming, contamination levels found on three fingers as a result of daily for instance, that the "euros break l activities~[7].__. __1 EU allergy directive" [5] and that"anybodywho is sensitive to nickel may also wish to handle the coins with care" [6]. At time of Clearly, handlingthe two nickel-containing euros does transfer writing, an internet query with the keywords "euro + nickel + significant amounts of nickel to the fingers, but the amounts allergy" brought up close to 900 web pages using a popular involved are about 40% lower than those resulting from the search engine. VIrtually all ofthese claim that the euro coins rep­ manipulation ofpure-nickel or copper-nickel coins ofcompara­ resent a danger in regard to nickel allergy, ignoring the fact that ble size. The comparison between the euros and the copper-nickel the only two nickel-containing euro denominations release less coins agrees well with the compositions averaged over their nickel upon manipulation than pure-nickel or nickel-alloy coins respective face areas: Cu75Ni15ZnlO for both bi-metallic euros of comparable size [7]. and Cu75Ni25 for the single-alloy coins. Note that a significant The 1994 EU"nickel directive" [8] concerns objects designed to amount ofcopper is transferred when manipulating pure-nickel come into direct and prolonged contact with the skin. The corre­ francs. We ascribe it to the contamination ofthe pure-nickel sponding normalised test procedure, ENI811 [9], measures the pieces upon contact with copper-containing coins in daily usage. amount of nickel dissolved when the object concerned is The contamination transferred to the fingers consists mainlyof immersed in synthetic sweat during one week. When one applies pre-existing species, rather than compounds generated during this procedure to the 1- and 2€ coins one detects nickel in manipulation [12]. This is illustrated by fig. 1which shows the amounts comparable to those obtained with other copper-nickel decrease in nickel contamination observed upon repeated manip­ coins andhigher than those measured for pure-nickel pieces [2,4, ulation ofthe same used copper-nickel coin (a Swiss 2SFr) or the 10,11]. These amounts do indeed exceed the 0.5 fig!cm 2!week same used 2€. Handling ofa used coin is seen to quite effectively europhysics news SEPTEMBER!OCTOBER 2003 195 Article available at http://www.europhysicsnews.org or http://dx.doi.org/10.1051/epn:2003506 FEATURES remove pre-existing contamination by transferring it to the fin­ separatelystoredin closedpolyethylene containers.This sampling gers. This transfer becomes less rapid as the number of procedure was carriedout at the University health centre, under manipulations increases, because the remaining nickel species are medical supervision. those which are strongest bound to the coin's surface. Figure 2 illustrates the imprints observed after 72 h skin con­ The amount ofpre-existing metallic species accessible to cont­ tact offour different coins, identifiedinthe photograph. There are aminate the fingers can be evaluated by rubbing the coins to a no obvious visual differences between the various pieces, except, shiny polish and determining the amount ofmetal recovered on perhaps, for the stronger mark left by the copper-nickel Swiss the"wipes" used to do so [12]. Such polishing reduces contami­ 2SFr as compared to that due to the 2€ coin. The markleft bythe nation upon subsequent handling by more than a factor 10, nickel-free 0.50€ piece (Cu89Al5ZnSSnl) is visually quite similar indicating that it removes more than 90% ofaccessible pre-exist­ to that ofthe three other coins, made ofnickel-alloys, illustrating ing contamination [12]. The results obtained for several types of thatmuch ofthe observedcoloration is due to copper ratherthan nickel containing coins are summarized in table 2. to nickel. There is no resemblance at all to the spectacular visual difference between the Swiss franc and the euro observed upon 2€ Cu75Ni25 Ni100 long-term immersion in artificial sweat [4, 15]. Ni 0.48 t 0.07 0.97 t 0.37 1.1 t 0.1 In the laboratory, ICP-OES analysis was applied to the"wipes" Cu 3.7 t 0.5 4.0 t 1.1 1.2 to.7 used to rub the imprints on the arm and also to those employed Zn 0.8 ± 0.2 0.4tO.2 0.2tO.2 for sampling metals released on the surfaces ofthe coins that had .--_._-------_._------------, been in contact with the skin. The results are summarised in i ... Table 2: Metal amounts pre-existing on the surface of used fig.3, where the sum ofthe metals collected on the skin and on I coins and accessible to contamination ofthe fingers by the coins are reported as release rates per cm2 and per week. i manipulation.The data are expressed in IJg per cm2 and were Expressed in this fashion, the results from the 24 h experiment, I obtained by dividing the amounts collected when polishing the indicated byan asterisk"", and from the 72 h patch test showlittle I coins by the geometric surface area of their two faces. The central difference, indicating that there are no important saturation I data column is the average offive different used coins made from effects in spite ofthe limited amount ofsweat available to dis­ I Cu75Ni25 copper-nickel alloy.The 2€and pure-nickel (2FF) data solve into. j are the average of 60 different coins in each case. Uncertainties are The data demonstrate a remarkable consistency between ! 90% confidence limits. L. --. --l coins ofsimilar composition anda close correlation with the sur­ face-averaged proportion ofthe metals examined. The Swiss, In this table, the accessible contamination for a single coin has American, and Swedish coins, as well as for the British lOp are all been divided bythe geometric surface area ofits two faces. The 2€ made ofCu75Ni25 copper-nickel, and for both constituents the and the pure-nickel2FF data are averages pertainingto samples of release rates per cm2 too are verysimilar.