Forensic Science International

170 Pages·2007·21.822 MB·English

FSI IFC 167(2-3).qxd:INNERCOVER 3/19/07 1:16 PM Page 1
Forensic Science International is an international journal publishing original contributions in the many different scientific disciplines pertaining to the
forensic sciences. Fields include forensic pathology and histochemistry, chemistry, biochemistry and toxicology (including drugs, alcohol, etc.), biology Author information
(including the identification of hairs and fibres), serology, odontology, psychiatry, anthropology, the physical sciences, firearms, and document examination,
as well as investigations of value to public health in its broadest sense, and the important marginal area where science and medicine interact with the law. Submission of manuscripts:Manuscripts should be submitted online by using the Elsevier Editorial System at http://ees.elsevier.com/fsi. Submission implies
Review Articles and Preliminary Communications (where brief accounts of important new work may be announced with less delay than is inevitable that the material is original, that it is not being considered for publication elsewhere and that all authors concur with the submission. Multiple submission is
with major papers) may be accepted after correspondence with the appropriate Editor. Case Reports will be accepted only if they contain some important new not acceptable to the Editor of this journal.
information for the readers.
Forensic Science Internationalhas no page charges
Submission of Articles:Manuscripts prepared in accordance with Instructions to Authors should be sent to the Editor-in-Chief or one of the Associate Editors
according to their area of expertise as listed below. If there is any doubt, a preliminary letter, telefax or telephone enquiry should be made to the Editor-in-Chief
or his Assistant. Online electronic submission:Authors can upload their article as a LaTeX, Microsoft®(MS) Word®, WordPerfect®, PostScript or Adobe®; Acrobat®PDF
document via the ‘Author Gateway’ page of this journal at www.authors.elsevier.com, where they will also find a detailed description on its use. The
system generates an Adobe Acrobat PDF version of the article, which is used for the reviewing process. Authors, Reviewers and Editors send and receive
all corr spondence by e-mail and no paper correspondence is necessary. If you encounter any problems while submitting online, please e-mail:
[email protected].
EDITOR-IN-CHIEF SECRETARY TO THE EDITOR-IN-CHIEF
English language help service: Upon request, Elsevier will direct authors to an agent who can check and improve the English of their paper (before
P. Saukko– (for: Experimental Forensic Pathology, Traffic A. Saarenpää– Address as for P. Saukko.
submission). Please contact [email protected] for further information.
Medicine and subjects not listed elsewhere)Department of Forensic Tel: [phone]; Fax: [phone];
Medicine, University of Turku, SF-20520 Turku, Finland E-Mail: ansaare@utu.fi All questions arising after acceptance of the manuscript, especially those relating to proofs, should be directed to Author Support at: authorsupport@
Tel.: [phone]; Fax: [phone]; elsevier.com
E-mail: psaukko@utu.fi
For a complete Guide for Authors, please use the link provided on
ASSOCIATE EDITORS the home page of Forensic Science International,
at http://www.elsevier.com/locate/forsciint
A. Carracedo – ( for: Forensic Genetics) P. Margot– (for: Questioned Documents, with the assistance of A. Khanmy
Instituto de Medicina Legal, and W. Mazzela; and for Physical Science: ballistics, tool marks, contact
Advertising information: Advertising orders and enquiries can be sent to: USA, Canada and South America: Mr Tino DeCarlo, The Advertising
Facultad de Medicina, traces, drugs analysis, fingerprints and identification, etc.)
Department, Elsevier Inc., 360 Park Avenue South, New York, NY 10010-1710, USA; phone: (+1) [phone]; fax: (+1) [phone]; e-mail:
15705 Santiago de Compostela, Galicia, Spain Ecole de Sciences Criminelles (ESC),
[email protected]. Japan:The Advertising Department, Elsevier K.K., 4F Higashi-Azabu, 1-Chome Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo
Fax: [phone] UNIL-BCH, CH-1015 Lausanne, Switzerland
106-0044, Japan; phone: (+81) (3) 5561 5037; fax: (+81) (3) 5561 5047; e-mail: [email protected]. Europe and ROW:Evelina Eurén, Elsevier BV.,
Fax: [phone]
Rodarweg 29, 1043 NX Amsterdam, Netherlands, Tel.: [phone]; E-mail: [email protected]
O.H. Drummer – ( for: Toxicology)
Department of Forensic Medicine, G. Willems – ( for: Odontology) Publication information:Forensic Science International (ISSN 0379-0738). For 2007, volumes 164–173 are scheduled for publication. Subscription prices
Victorian Institute of Forensic Medicine Katholieke Universiteit Leuven, School of Dentistry, Oral Pathology and are available upon request from the Publisher or from the Regional Sales Office nearest you or from this journal’s website
57–83 Kavanagh Street, Maxillo-Facial Surgery, Departments of Orthodontics and Forensic (http://www.elsevier.com/locate/forsciint). Further information is available on this journal and other Elsevier products through Elsevier’s website:
Southbank 3006, Victoria, Australia Odontology, Kapucijnenvoer 7, B-3000 Leuven, Belgium (http://www.elsevier.com). Subscriptions are accepted on a prepaid basis only and are entered on a calendar year basis. Issues are sent by standard mail (surface
Tel.: [phone]; Tel: [phone]; within Europe, air delivery outside Europe). Priority rates are available upon request. Claims for missing issues should be made within six months of the date of
Fax: [phone] Fax: [phone] dispatch.
Cristina Cattaneo– ( Anthropology and Osteology) Orders, claims, and journal enquiries:please contact the Customer Service Department at the Regional Sales Office nearest you:
Instituto de Medicina Legal, Universit`a degli Studi Orlando:Elsevier, Customer Service Department, 6277 Sea Harbor Drive, Orlando, FL 32887-4800, USA; phone: (+1) (877) 8397126 [toll free number for US
Via Mangiagalli 37, 20133 Milano, Italy customers], or (+1) (407) 3454020 [customers outside US]; fax: (+1) (407) 3631354; e-mail: [email protected]
Tel: [phone]; Amsterdam:Elsevier, Customer Service Department, PO Box 211, 1000 AE Amsterdam, The Netherlands; phone: (+31) (20) 4853757; fax: (+31) (20) 4853432;
Fax: [phone] e-mail: [email protected]
Tokyo:Elsevier, Customer Service Department, 4F Higashi-Azabu, 1-Chome Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; phone: (+81) (3)
5561 5037; fax: (+81) (3) 5561 5047; e-mail: [email protected]
Singapore: Elsevier, Customer Service Department, 3 Killiney Road, #08-01 Winsland House I, Singapore 239519; phone: [phone]; fax: (+65)
EDITORIAL BOARD
67331510; e-mail: [email protected]
A. Amorim (Porto, Portugal) A.W. Jones (Linköping, Sweden) K. Püschel (Hamburg, Germany)
J. Buckleton (Auckland, New Zealand) H. Kalimo (Helsinki, Finland) G. Quatrehomme (Nice, France) Printed and Published in Ireland
B. Budowle (Quantico, VA, USA) B. Kneubuehl (Thun, Switzerland) J. Robertson (Canberra, Australia)
J. Clement (Melbourne, Australia) S. Leadbeatter (Cardiff, UK) P.M. Schneider (Cologne, Germany)
S.D. Cohle (Grand Rapids, MI, USA) P.J. Lincoln (Surrey, UK) S. Seta (Tokyo, Japan)
S. Cordner (South Melbourne, Australia) A. Luna Maldonado (Espinardo (Murcia), Spain) J. Simonsen (Copenhagen, Denmark)
P. Dickens (Buxton, UK) B. Madea (Bonn, Germany) P. Sótonyi (Budapest, Hungary)
M. Durigon (Garches, France) N. Morling (Copenhagen, Denmark) M. Steyn (Pretoria, South Africa)
A. Eriksson (Umeä, Sweden) B. Olaisen (Oslo, Norway) F. Tagliaro (Verona, Italy)
J.A.J. Ferris (Auckland, New Zealand) V. Pascali (Rome, Italy) T. Takatori (Chiba, Japan)
M.C. Fishbein (Encino, USA) S. Pollak (Freiburg i. Br., Germany) S. Tsunenari (Kumamoto, Japan)
P. Gill (Solihull, UK) M.S. Pollanen (Toronto, Canada) D.N. Vieira (Coimbra, Portugal)
C. Henssge (Essen, Germany) D. Pounder (Dundee, UK) X. Xu (Shantou, People’s Republic of China)
M.A. Huestis (Baltimore, MD, USA) O. Prokop (Berlin, Germany) J. Zhu (Guangzhou, People’s Republic of China)
ForensicScienceInternational167(2007)93
www.elsevier.com/locate/forsciint
Introduction
On behalf of the scientific committee of EAFS2006 whichpapersweresuitableforpublicationandthemaincriteria
welcome to this special edition of Forensic Science Interna- was to have a high quality special edition.
tional which is specially produced to commemorate the Members of the Scientific Committee are gratefully
Conference held in Helsinki, 13–16 June, 2006. acknowledged for their efforts and contribution to this special
I think it is important to outline the review process for this issue:Dr.Bramley(UK);Dr.Broeders(The Netherlands),Dr.
special edition as it differs from the normal system. Houck (USA), Prof. Kopp (Sweden), Prof. Piekoszewski
AllauthorsfromtheConferencewereinvitedtosubmittheir (Poland), Dr. Sippola (Chair and Organiser of EAFS2006;
work for consideration to the Journal. All members of the Finnland), Dr. Willis (Chair of the Scientific Committee) and
scientific committee were given access to all submissions. Prof. Pierre Margot.
During the Conference the committee met and reviewed their
decisions as a group. Any member who had contact with a Sheila Willis, Chair,
particular institute abstained from the discussion when papers EAFS Standing Committee,
from that institute was being discussed. EAFS2006 Scientific Committee
Any paper where two or more of the committee rejected
the submission were eliminated from discussion and the Special Issue Editor,
remaining papers were read again and included for Associate Editor
publication. The process was of necessity quite harsh Pierre Margot*
because papers were submitted directly for publication and Universite de Lausanne, Criminologi UNIL-BCH,
werenotaltered.Ibelievethis processfavouredauthorswho Institut de Police Scientifique et,
weremorefamiliarwithpublicationrequirements.Anumber 1015 Lausanne, Switzerland
of the papers reviewed contained good quality material and
their authors were encouraged to rewrite or carry out some *Corresponding author. Tel.: [phone];
additional work and resubmit to the editor for consideration fax: [phone]
at another time. E-mail address: [email protected]
The selection of papers in this special edition may not be (P. Margot)
fully representative of the range of material presented at the
Conference. There was much soul searching in relation to Available online 7 September 2006
0379-0738/$–seefrontmatter#2006ElsevierIrelandLtd.Allrightsreserved.
doi:10.1016/j.forsciint.2006.07.019
ForensicScienceInternational167(2007)94–101
www.elsevier.com/locate/forsciint
The difference between drug money and a ‘‘lifetime’s savings’’
Karl A. Ebejera, Jane Winna, James F. Cartera,*, Richard Sleemana,
Jill Parkerb, Fritjof Ko¨rberb
aMassSpecAnalyticalLtd.,Building20F,GolfCourseLane,P.O.Box77,Filton,BristolBS347QS,UK
bFacultyofAppliedScience,UniversityoftheWestofEngland,FrenchayCampus,ColdharbourLane,BristolBS161QY,UK
Received8June2006;accepted14June2006
Availableonline4August2006
Abstract
Inmanycountries,moniessuspectedofbeingassociatedwithdrugtraffickingcanbeseizedbytheauthorities.Oneofthewaysofinvestigating
this association isthroughtheanalysis ofseized banknotesfortracesof controlled drugs.
Wereportthreestudieswhichmayassisttheexpertinassessingwhetherbanknotescontaminatedwithdiamorphinearepartofthegeneral
population ofnotesin circulation orwhethertheyshowunusualcontamination patterns which require explanation.
Study1isbasedonthreeplausiblecontaminationscenariosastheymayoccurduringthevariousstagesofanillicitdrugtransactionandseizure.
Itshowsthatnoteswhichhavebeenindirectcontactwithvisibletracesofdiamorphineshowsignificantlyhighercontaminationtothoseinmore
indirect contactwith thedrug.
Study2investigatesthetransferofdiamorphinefromonehighlycontaminatednotetoothernotesinabundleoveraperiodof10weekswithand
withoutagitation.Itwasfoundthatthetotalamountofdrugtransferredwassmallerthan6%andnomorethan4outofabundleof10previously
clean notesbecame lightly contaminated.
Basedonextensivebackgrounddata,study3proposesaprobabilisticmodeltoassesswhetheranobservedproportionofdiamorphinebearing
banknotesislikelytohavebeencontaminatedbychance.Themodelpredictsthatthereisonlya0.3%chancethatabundleof100notesfromthe
general banknotepopulationcontains more thansixcontaminated specimens.
Jointly,thethreestudiesgiveusefulindicationsforthespreadofcontaminationthroughoutasampleandtheamountsofheroinwhichmaybe
expectedgiven plausible contamination scenarios.
#2006Elsevier Ireland Ltd.Allrightsreserved.
Keywords: Diamorphine;Banknotes;Transfer
1. Introduction circumstantial evidence. However, this requires that the
banknotes be distinguishable from those typically in general
The Financial Action Task Force (FATF) lists 40 recom- circulation [3].
mendations, which member countries are committed to Banknoteanalysisbygaschromatographymassspectrometry
implement [1]. The first of these recommendations is to (GC/MS) has been performed successfully by a number of
criminalise money laundering on the basis of the United workers[3–7].AmajordrawbackofanymethodbasedonGC/
NationsConventionagainstIllicitTrafficinNarcoticDrugsand MS is the long preparation and analysis time required, in
Psychotropic Substances, 1988 (the Vienna Convention) [1]. comparison with thermal desorption atmospheric pressure
TheUnitedKingdomisoneofthe33membersofFATFand,in chemical ionisation tandem mass spectrometry (APCI MS/
the UK, the Proceeds of Crime Act allows for the seizure of MS)[8–10].Thelatteriscurrentlythemethodofchoiceforlaw
cashreasonablysuspectedtobegainedunlawfully[2].Where enforcementagenciesintheUK.Thespeedofthistechniquehas
illegal drug activity is suspected, the results of the analysis of allowedahugenumberofbanknotesfromgeneralcirculation,so
the banknotes for traces of illicit substances can be used as called ‘‘background’’ banknotes, to be analysed individually.
This background database of illicit drug traces detected on
banknotes from the UK is maintained and routinely used for
comparisonwithdrugtracesdetectedonbanknotesseizedfrom
* Correspondingauthor.Tel.:[phone];fax:[phone].
E-mailaddress:[email protected](J.F.Carter). suspectsthoughttobeinvolvedinillegaldrugactivities.
0379-0738/$–seefrontmatter#2006ElsevierIrelandLtd.Allrightsreserved.
doi:10.1016/j.forsciint.2006.06.039
K.A.Ebejeretal./ForensicScienceInternational167(2007)94–101 95
Sincethebackgrounddatabasecontainsdatafromindividual shown to be free from diamorphine, prior to making contact
banknotes, the contamination of bundles of banknotes can be with the banknotes.
described in terms of both the quantity of drug contamination Theadvantageofthedescribedanalyticaltechniqueisthata
[11]andtheproportionofbanknotesinabundlewithdetectable singlebanknotecanbeanalysedtwice—onceforeachend.The
traces of a drug [12]. The advantage of having two resultsfromtwoendsofanindividualbanknotemayvary,due
complementary methods for data interpretation is that either to localised deposits. However, average results from the two
orbothcanbeemployedasappropriatetothedrugofinterest. ends are found to converge when large numbers of banknotes
Cocaine,forinstance,occursonvirtuallyallBankofEngland are analysed (unpublished results). In forensic casework,
banknotesingeneralcirculation[10].Therefore,anyattemptto exhibits that have been analysed once can be reanalysed if
differentiate on the basis of contaminated proportion is futile. required.
Such an approach, however, is valid for diamorphine
contamination, where the drug occurs on much smaller 2.1. Study 1—primary, secondary and tertiary transfer to
numbers of banknotes in the background. banknotes
The proportion of banknotes with diamorphine contamina-
tionhasbeenusedasthediscriminatingfactorinalargenumber British banknotes, being composed of a mixture of cotton
of legal proceedings, but this forms only one aspect of the and linen, were shown to be simulated best by cotton paper,
evidence.Defendantscommonlyclaimthatlargequantitiesof ratherthanwoodpulpbasedproductsaspreviouslyused[13].A
banknotes in their possession represent a lifetime’s savings, a study of diamorphine transfer showed cotton paper to have
gift,aninheritance,theprofitsfromtradinginagreyeconomy, similarpropertiestobanknotes(averagecorrelationcoefficient
etc. They also maintain that the diamorphine contamination 0.84) in contrast to a wood pulp based copy paper (average
came to be on the banknotes by mechanisms unknown and correlation coefficient 0.79) (unpublished results).
unrelated to them. It is, therefore, useful to consider not only ContaminatedbanknotesweremodelledusingCrane’sCrest
whetheraseizureiscontaminatedtoanunusualdegree,butalso cotton paper (100% rag, 90GSM, Crane & Co. Inc. Dalton,
toassessthepatternofcontaminationontheseizedbanknotes MA). The paper was cut to the approximate size of a £10
and the likelihood that they became contaminated through sterlingbanknote.Thissubstitutionremovedtheneedtocorrect
means other than illegal drug activity. for the natural background levels of drugs on banknotes and
In this paper, we report on the contamination patterns of also eliminated differences between the quality of banknotes
banknotes subjected to three plausible scenarios associated (e.g. worn, new, dirty, clean). Indeed, it proved difficult to
with activities related to drug dealing. A further study obtain a sufficiently large number of banknotes in a similar
investigated the transfer of diamorphine from a highly physical condition.
contaminated note to adjacent notes over a period of 10 Threestageswereusedtosimulatedifferentstepsinadrug-
weeks. Lastly, a probabilistic model was fitted to the dealing scenario. Each stage was physically isolated on a
diamorphine contamination background database to permit a 30cm(cid:2)30cm sheet of laminated chipboard of a type
more reliable assessment of which contamination proportions commonly used in furniture manufacture. The intention was
may be classed as ‘‘unusual’’. that the amounts of diamorphine present on each of the three
surfaces were representative of those transferred by primary,
2. Materials, methods, techniques secondary and tertiary contacts with the drug.
In the first stage, approximately 10mg portions of street
All banknote analyses were performed using triple– heroin (Avon & Somerset Constabulary, Scientific Investiga-
quadrupole mass spectrometers (MDS Sciex, Concorde, ON., tions, approximately 10% diamorphine) were weighed into
Canada)withcustom-builtthermaldesorptioninletsconsisting cigarette papers to make 10 typical street-sized dealer’s
oftwometalplatesheatedto2858C.Insertionofabanknoteor ‘‘wraps’’. Samples were weighed using Diamond brand
paper swab, for approximately 1s, between the metal plates ‘‘professional mini’’ electronic scales, typical of the variety
causedvolatilisationofdebrisincludingdiamorphine,fromthe seized from suspected drugs dealers. Ten cotton paper notes
sample. Thevapours became entrained in ambient air flowing were counted onto this surface into a single pile, which was
into the APCI source of the instrument [12]. The mass theninvertedandthenotesrecounted.Subsequently,thenotes
spectrometer was programmed to monitor two product ions were analysed.
from the protonated molecule of diamorphine (m/z 370). The Inthesecondstagethewraps,preparedabove,wereplaced
productionsofinterestfordiamorphinewerem/z328andm/z on a surface previously shown to be clean. Ten cotton paper
268. Peak areas were obtained from the extracted ion notes were then counted,asbefore,onto this surface and then
chromatograms (Analyst 1.4, MDS Sciex, Concorde, ON., again onto a third, clean surface prior to analysis.
Canada) using softwarewritten in MATLAB (release 12, The Inthethirdstage, 10cotton papernoteswere countedonto
Mathworks, Natick, MA, USA) for the identification and the third surface and analysed.
integration of peaks [11]. Gloves were worn during the weighing and wrapping of
Instrument performance was monitored by injection of a heroin, but cotton paper notes were counted using bare hands
solution containing 2ng of diamorphine in methanol. Swabs becausethisappearedmorerealistic.Handswerewashedwith
from the analyst’s gloves and worktop were analysed, and copiousamountsofwarmwaterbetweencountingstages.The
96 K.A.Ebejeretal./ForensicScienceInternational167(2007)94–101
entire sequence was repeated using clean surfaces and cling [12].Thisnumberwas thennormalisedtothetotal number of
film to manufacture the wraps. banknotes in the sample (true positives per 100 banknotes) to
correct for sample size variation, and rounded to the nearest
2.2. Study 2—transfer between banknotes integer.
The arithmetic mean of the number of contaminated
To simulate the contaminants as realistically as possible, banknotes per 100 (m) was also determined from the same
particulate street heroin and household dust were used to data set. A Poisson distribution, using m as the predicted
contaminate clean paper. Household dust is likely to contain outcome (l) was then generated and compared to the
skin oils as found on banknotes [10] and may be expected to background database. Data were processed using SYSTAT
simulate the daily environment to which banknotes are 11 for Windows (SYSTAT software Inc., Richmond, CA).
exposed. A quantity of street heroin (as above) was mixed in
aplasticreceptaclewithadusttakenfromadomesticvacuum 3. Results
cleaner and passed though a 1.7mm sieve followed by a
0.5mm sieve. 3.1. Study 1—primary, secondary and tertiary transfer to
Anumberofcottonpapernoteswereagitatedmanuallyfora banknotes
periodofseveralminuteswiththespikeddust.Thenoteswere
withdrawn,shakentoremoveloosedepositsandtestedandthe Fig.1(aandb)showtheintensityofthediamorphineproduct
responsetodiamorphinewasmeasured.Theamountofheroin ions,as a function of time, detected on ten cotton paper notes
inthedustwasadjusteduntiltheresponsewasapproximatelyat fromeachofthethreestagessimulated.Thetrendsobservedfor
themid-pointofthedynamicrangeoftheMS–MSinstrument. cigarette paper wraps and cling film wraps can better be
Thismixturewasthenusedtospikefurthercottonpapernotes visualised from a histogram representing the relative, average
inthesamemanner.Afewofthespikednoteswereanalysedto diamorphinetransferforeachcategory(Fig.2).Thenumberof
check that sufficient diamorphine had been transferred to contaminatednotesdecreaseswiththelevelofcontactinboth
mimic the contamination levels typically detected on con- the cigarette paper and cling film wrap scenarios. Secondary
taminated banknotes observed in case work.
A sample consisting of 120 banknotes (£10 denomination)
wasdrawnoverthecounterfromabankinBristol,UKandone
end of each note analysed for the presence ofdiamorphine. A
recordwaskeptoftheorderinwhichthenoteswereanalysed,
andthisorderwasmaintainedwhiledividingthebanknotesinto
12groupsof10banknoteseach.Aspikedcottonnotewasthen
insertedexactlyinthemiddleofeachbundle(Fig.3),andthe
bundlesweresealedinsideclosefittingzip-lockbags.Allbags
werethensealedinsideatamper-evidentbagandstoredinthe
dark.
Over the course of 10 weeks, each bundle, except two
controlbundles,wascountedonaweeklybasis.Thezip-lock
bagsweredistributedtodifferentco-workerswhoopenedthe
bag and counted out the banknotes, including the spiked
note, at least once, in any manner they pleased, but without
altering the sequence of notes within the bundle. After a few
minutes, the banknotes were returned to their respective zip-
lock bags. In weeks 2, 4, 6, 8 and 10, two bundles were
analysed and removed from the pool. The analysis was
performedonbothendsofeachnote.Theremainingsamples
were resealed inside a tamper-evident bag and returned to
dark storage.
2.3. Study 3—comparing contamination with the
background database
One hundred and eighty-six bundles of banknotes, each
comprising 100 banknotes or more, were selected from the
background database. Diamorphine contamination on both
Fig. 1. Ion counts recorded for the m/z 268 product ion of diamorphine on
ends of the banknotes was measured using the procedure
cottonpapernotescountedon(tertiary,secondaryandprimary)surfacesusedto
described.Themeannumberofbanknotesineachsamplethat
simulateadrugdealusingwrapsof(a)cigarettepaper(b)clingfilm.Thesame
gaveatruepositiveresponsefordiamorphinewasdetermined trendwasobservedforthesecondionmonitored(m/z328).
K.A.Ebejeretal./ForensicScienceInternational167(2007)94–101 97
whetheronecandistinguishthevarious stagesofadrugsdeal
by the quantity of drug transferred.
In the two factor analysis the main factor was the type of
transferscenario(primary,secondaryortertiarytransfer).This
variable was responsible for approximately 65% of the total
variation in the sample set, as measured by the total sum of
squares. Its contribution was more than twice as large as that
from random fluctuations, showing that this factor is highly
significant (p<10(cid:3)14). The second factor analysed was the
typeofmaterialusedforthewraps.Thisfactorwasresponsible
for less than 5% to the variation but was still a statistically
significant source of difference (p<0.01). As might be
expected,theinteractionbetweenthetwofactorsissmallandat
anyreasonablychosenlevelofsignificancethetwofactorsare
not correlated.
The Tukey test identified which of the bundles were
significantly different with regards to the amount of drug
transferred.Thetestwasperformedindependentlyforeachion
onallsixbundles(threetransfertypesandtwowrapmaterials)
at a confidence level of 95%. The test reliably differentiated
betweenprimarytransferscenariosinonegroupandsecondary
andtertiarytransferscenariosinasecondgroup,i.e.thenotesin
Fig. 2. The mean responses for the product ions of diamorphine (summed) direct contact with spilled heroin on the cutting table were
detected on banknotes counted onto primary, secondary and tertiary stage significantly more contaminated than those from indirectly
surfaces. contaminatedbundles.Despitetheapparenttrendtowardsless
contamination for the tertiary stage it was not possible to
transfer from cigarette paper wraps produced about half the differentiate between secondary and tertiary transfer, based
number of contaminated notes compared to primary transfer. solely on the quantity of material transferred, at the 95%
However, for secondary transferfrom cling filmwraps almost confidence level. By reference to Fig. 2 it is, however,
all the notes in the bundle were contaminated, making it immediately clear that secondary and tertiary transfer can be
difficulttodistinguishprimaryfromsecondarytransferonthe distinguished on the basis of the proportion of contaminated
basisofcontaminationproportion.Tertiarytransferresultedin notes.
a single note becoming contaminated from both types of
wrapping material. 3.2. Study 2—transfer between banknotes
Therelativeintensityplot(Fig.2)showsthatthequantities
transferred in the various stages differ considerably and an Asignaltonoiseratioof4.5wasusedasathresholdforpeak
analysis of variance was performed to establish whether the detection. Peak areas for the twoions were summed. Thereal
apparent differences are statistically significant. For a given peakdataforeachendofagivenbanknoteorspikednotewere
experimentalsituation,i.e.foreachbundle,thedistributionof also summed to give a better representation of the total
contamination(peakareas)wasfoundtobeapproximatelylog- diamorphine contamination detected on the note. Every two
normal. Therefore a two factor analysis with replication weeksaverageofthedatafromtwobundles,wascalculatedto
(MicrosoftExcel97)andaTukeytest[14]wereperformedon arrive at a single value for the response of the MS/MS
thelogarithmsofthepeakareasforthem/z268andthem/z328 instrument to a banknote in a given environment.
productions.Thetestsestablishedwhichparametersinfluence Thediamorphinecontaminationdetectedonbanknotesata
themeanquantityofdrugdetectedoneachanalysedbundleand given sequential distance (Fig. 3) from the spiked note was
Fig.3. Arrangementofbanknotesandcottonpaperspikeinstudy2.
98 K.A.Ebejeretal./ForensicScienceInternational167(2007)94–101
calculated as the total contamination on the two contributing for diamorphine per 100 banknotes. The number of con-
notes. The degree to which diamorphine was transferred or taminatedbanknotesinabundlewasshowntofollowaPoisson
retained was expressed as a proportion of the total amount of distribution (x2=2.30, p=0.81) (Fig. 6). Using this Poisson
diamorphine detected (Fig. 4). distributionitispossibletoestimatethechancethatasampleof
In addition, the number of contaminated banknotes in a banknotes with a high degree of diamorphine contamination
bundle was plotted as a function of time (bar chart, Fig. 5). has been drawn from the background by chance. The highest
The amount of diamorphine detected on the spiked paper, proportions of contaminated banknotes observed in the 186
corrected for instrument variation, was also plotted (line backgrounds were two incidents of 6 contaminated banknotes
graph, Fig. 5). per100,andasingleincidentof8notesper100.Theseupper
values are consistent with those predicted by a Poisson
3.3. Study 3—comparing contamination with the distribution based on a predicted outcome (l) of 1.85
background database contaminated notes per 100 (Table 1).
Thechanceofaseizurehaving10contaminatedbanknotes
A total of 186 bundles of banknotes in the background per100,orhigher,isveryslightifthebanknoteshadasimilar
databasewere found to have an average of 1.85 true positives history to that experienced by banknotes in the background
Fig.4. Proportionofdiamorphinedetectedonthespikeandadjacentbanknotes.
K.A.Ebejeretal./ForensicScienceInternational167(2007)94–101 99
Table1
Thechancethatagivennumberofcontaminatedbanknotesaredrawnfrom
generalcirculation(givenaPoissondistributionwithl=1.85)
Numberofcontaminated Approximateoddsof
banknotesinasampleof100 drawingthesebychance
0 6:1
1 3:1
2 4:1
3 6:1
4 12:1
5 33:1
6 100:1
8 2000:1
10 50,000:1
representatruereflectionoftherangeofparticlesizesfoundin
the real world [15]. It is reasonable to speculate that larger
crystals may transfer more easily between banknotes as they
become trapped lessreadily between thefibres.Since mostof
the mass transferred will be in the form of larger crystals,
Fig.5. Totalnumberofcontaminatedbanknotes(excludingthespikednotes) applying diamorphine in solution is likely to represent a poor
detectedfromtwobundlesoftenbanknotesafterintroducingaspikednoteto model for real world contamination, even though the spiking
eachbundleandcountingonaweeklybasis.
would be more quantitative and less variable. Further,
applicationofasolutioncouldbeachievedbydipping(leading
to homogeneous distribution of the material) or via a syringe
(resulting in contamination concentrated in certain areas),
neither of which accurately reflects the likely distribution of
particulatematerial.Hence,althoughthemethodusedtospike
cottonnoteslacksbothrepeatabilityandreproducibilityitwas
considered the best model available.
In study 1 it was, perhaps, unsurprising that cotton notes
counted on a surface with visible deposits of heroin retained
significanttracesofdiamorphinewhichcouldsubsequentlybe
liberatedbythermaldesorption.Theintensityoftheresponses
observedfromthesenoteswassimilartothoseobservedwhen
analysing banknotes from suspected drugs dealers. Wraps
prepared on the same surface would be exposed to similar
Fig.6. Dataobserved(red)andpredictedbyaPoissondistribution(blue)for amounts of heroin and can reasonably be expected to retain
diamorphinecontaminationofbanknotestakenfromgeneralcirculation.(For
externaldeposits.Someproportionofthesedepositsmaythen
interpretation of the references to colour in this figure legend, the reader is
fall onto the stage 2 surface, becoming available for retention
referredtothewebversionofthearticle.)
by the cotton notes counted on this surface.
database.Thisimpliesthatanalternativeexplanationiscalled From Figs. 1 and 2, it is apparent that some secondary
for. transfer to the cotton notes has indeed occurred. Cotton notes
countedaftercigarettepaperwrapshadbeenincontactwiththe
4. Discussion surface,however,retainedmuchsmallertracesofdiamorphine
than those counted after cling film wraps had been in contact
Althoughquantitativeamountsofdiamorphinecouldeasily with the surface.
have been applied to the simulated notes as solutions, it was Different mechanisms are undoubtedly responsible for the
consideredthattheuseofimpure,solid‘streetheroin’samples retention of particles by paper and by cling film. In the latter
would provide a more realistic model for ‘real-life’ heroin case the mechanism is almost certainly electrostatic, possibly
contamination in several ways. First, solutions are likely to leadingtolargerinitialretentionandalargerresultingdeposit.
‘wick’intothefibresofthepaper,ratherthanremainingonthe In addition, the retention mechanism of the cling film may be
papersurface,orbeingtrappedintheintersticesbetweenfibres weakerthanthepaper,resultinginadepositthatismorereadily
aswouldbeexpectedofparticles.Inturn,thismayrenderthe lost.Incontrast,particlesretainedbypapersurfacesarelikely
diamorphinelessamenabletoliberationbythermaldesorption. to be retained as inclusions in surface features [10] or by
Secondly, the evaporation of the solvent would lead to the dissolution in surface dirt (finger grease, etc.) or water
formation of smaller crystal sizes than would typically be associated with the cellulose. These different mechanisms
present in ‘street heroin’, and most certainly would not may well exhibit different affinities for different crystal sizes
100 K.A.Ebejeretal./ForensicScienceInternational167(2007)94–101
resulting in preferential retention of different materials. It is firstcontactevent,experiencedbyallbundleswhenthespiked
possiblethatdepositsretainedbyboththecottonandcigarette paper was introduced. Subsequently, transfer is minimal, and
papersarestronglybound,notreadilylosttothestage2and3 each counting occasion results in losses from the bundles and
surfaces and, therefore, not available for subsequent retention more exposure to light and moist air that could increase the
bycleancottonnotes.Thisconjectureissupportedbythesmall naturaldegradationofthediamorphine[16–18].
depositsretainedbycottonnotescountedonthethirdsurface. In study 3, the database against which comparisons were
The statistical analysis shows that the quantities of drugs madecomprisesmainlysamplesdrawnfromoverthecounterat
transferred to the notes directly from the cutting table are banksthroughouttheUK.Asmallercomponentofthedatabase
significantly different from those transferred at the secondary consists of samples from small businesses such as pubs and
andtertiarystages.Thisinformationwouldbeoflimitedvalue shops.Theseareallconsideredtoberepresentativeofbanknotes
unless the study was performed at similar primary surface ingeneralcirculation.Therehasbeenextensivediscussiononthe
contaminationlevelsasobservedinreallife scenarios.Whilst degreetowhichbanknotesmixingeneralcirculation[12].Ifthe
this cannot be asserted, the aforementioned similarity in smallproportionofbanknotescontaminatedwithdiamorphinein
intensity observed in seized samples gives an indication that general circulation is indeed randomly distributed, then the
this may indeed be the case. probability of drawing a sample bearing a given number of
Instudy2thesourceofcontamination(thespike)waseasily contaminated banknotes can be described by a Poisson
identifiable at every stage of counting. The amount of distribution [19]. This is a good approximation for a binomial
diamorphine remaining on the spiked note through the study probabilityforlargenumberswheretheeventofinterestoccurs
varied from 94 to 99% of the total amount detected on all withalowprobability.Thereisnoneedtospecifysamplesizeas
banknotes in a bundle. Transfer occurred primarily from the longasanexpectedfrequencycanbecalculatedforthenumber
spiked note to one or both banknotes immediately adjacent ofeventsobserved[19].
(Adj1)toitinabundle.Wherethespikednotelostasmuchas6% Thestudyhasillustratedthatthechanceofdetectinghighly
ofitsdiamorphinetootherbanknotesinthebundle,theamount contaminated bundles of banknotes in samples drawn at
detectedonasingleadjacentbanknotewaslessthan6%.Thisis randomfromgeneralcirculation,isslight.Indeed,thedetection
in line with previous findings [10]. After 4 weeks and after 8 ofhighfrequenciesofdiamorphinecontaminationimpliesthat
weeks,diamorphinewasalsodetectedonabanknotetwosteps thebanknotesmayhaveoriginatedfromanisolatedpopulation,
(Adj2) and three steps (Adj3) away from the spiked note, such as drug dealers or drug users. It is possible that
respectively.Onbanknotesbetweentheseandthespikednote,no contaminated monies could, for example, be paid into a bank
contamination was detected, suggesting that transfer in these andsubsequentlywithdrawnbyaninnocentindividualpriorto
instancesmayhaveoccurredviatheinnersurfaceofthezip-lock mixing.Extensivemixingofsuchbanknoteswithbanknotesin
baginwhichabundlewasstored,orviathecountingsurface. general circulation would account for the low frequency
Since a single source of contamination was, in general, observedinthedatabase.Findingahighlycontaminatedbundle
foundtocontaminateuptotwobanknoteswithdiamorphine,it therefore cries out for an explanation.
mightbearguedthattwo-thirds ofthecontaminationdetected
in an exhibit could have arisen by this process. This would 5. Conclusions
assume minimal bunching of contaminated notes in a bundle.
To have an appreciable effect, banknotes with ‘‘primary’’ The limited studies discussed in this paper have produced
contaminationwouldneedtobedispersedthroughoutabundle some interesting findings, although further work will be
of otherwise uncontaminated banknotes. necessary to strengthen the conclusions that can be drawn.
Inthefirst8weeksofthestudy,thenumberofcontaminated Cotton notes in direct contact with surfaces bearing visible
banknotes in every bundle of 20 notes increased with the deposits of heroin were found to retain significant amounts
number of counting events. This was, however, concomitant diamorphine. Such deposits did not transfer readily between
withageneralincreaseinthediamorphinecontaminationofthe banknotes when mixed and counted in bundles (less than 6%
spikedpaperinthesebundles.Oneofthedrawbacksofusinga transfer). A single highly contaminated note is unlikely to
solidmaterialtocreatethespikednoteswasthattherewasno contaminatemorethanthetwoadjacentbanknotes.Sinceonlya
waytopreciselyregulatetheamountofdrugpresentonthem. smallproportionoftheinitialdepositistransferredtheprimary
The observed trend in the spike note intensity reflects the andsecondarydepositsarereadilydistinguishable.Theextentof
random variations in this process. transfer between banknotes appears to be dependent upon the
Acombinationoftheamountofdruginitiallypresentonthe amount of material on the initial spiked note but may also be
spiked note, and the number of mixing events, is probably dependent upon the physical handling of the notes, which can
closertothetruth.Inthe10thweek,asuddendropinthenumber increaseordecreasethesizeofsecondarydeposits.
ofcontaminatedbanknotesinthebundlesthathadbeensubjected Other materials, e.g. cling film and cigarette papers can
to weekly counting corresponded with a fall in the amount of transferdepositstoasecondarysurfacewheretheyareavailable
contaminationonthespikedpaper.However,thespikesizeswere for retention by cottonnotes.These secondary deposits do not
similar in the counted and uncounted control samples, but the then transfer readily to other surfaces to become available for
number of contaminated banknotes in that sample was higher. retentionbyothernotes.Hence,countingcontaminatednoteson
Oneexplanationmaybethatthemajorityoftransferoccursinthe a surface along with clean notes does not appear to spread