OPTIMIZATION OF A SEQUENTIAL
EXTRACTION SCHEME FOR SPECIATION
OF METALS IN FINE URBAN PARTICLES
A.J. FERNANDEZ ESPINOSA*, M. TERNERO RODRIGUEZ,
F. FERNANDEZ ALVAREZ, F.J. BARRAGAN DE LA ROSA and
J.C. JIMENEZ SANCHEZ
Depa men o Analy ical Chemis y, Facul y o Chemis y, Uni e si y o Se ille,
C. P o eso Ga cia Gonzalez s/n, 41012 Se ille, Spain
A sequen ial ex ac ion p ocedu e ac ioned in ou s eps was op imized wi h he pu pose
o imp o ing a p e ious scheme o de e mining he chemical o ms o me als in ine ai bo ne
pa icles. The op imiza ion was es ed on syn he ic samples and hen con i med on eal samples
o ine pa icles, op imizing he ype o eagen , i s concen a ion, pH, empe a u e and
ex ac ion ime. Syn he ic samples we e p epa ed in he labo a o y wi h compounds o eagen
g ade. The analy ical me hod applied consis ed o ex ac ing he me als om each sample in a
o a o wi h each ype o eagen unde di e en condi ions o he pa ame e s and hen
cen i uging be o e measu ing by ICP-AES. The imp o ed and p e ious scheme was checked
on eal samples and he di e ences we e signi ican in he pe cen ages o he soluble chemical
o ms in he imp o ed scheme (5.2% in on o 22.0% o Pb, 27.0% in on o 50.0%
o Cd and 32.4% in on o 9.0% o Ni) om 55.6, 0.23 and 1.48 ng m
—
3
o o al me al
concen a ions espec i ely.
Keywo ds:
Chemical specia ion; Me als; Ai bo ne; Op imiza ion; Fine pa icles; U ban
pollu ion
INTRODUCTION
U ban popula ions a e exposed o he me als in ai bo ne pa icles. When he
me al concen a ion is high, i can pose se ious isk o human heal h.
Howe e , de e mina ion o me al le els is usually limi ed o de e mining
*Co esponding au ho
he o al me al concen a ion. These measu es p o ide us in o ma ion
nei he on he chemical o ms. The e o e, pa icula e s anda ds based on
o al suspended pa icles (TSP) alone a e insu icien [1]. Knowledge o
he chemical specia ion o he me als is i al in unde s anding he e ec s
on he public heal h [2,3].
Cu en bibliog aphy on chemical specia ion o sedimen s, soils, sludges
and biological sys ems is wide han hose on a mosphe ic pa icles [4].
Many au ho s use ex ac ion schemes based on he Tessie ’s scheme [5,6].
A esea ch in which his scheme was applied on a mosphe ic pa icles
(wi h se e al modi ica ions) was pe o med in Ba celona (Spain) [7]. The
BCR
®
’s scheme (Bu eau o Ce i ied Re e ence) was published la e on [8]
and i was also applied o sedimen s [9] and sludges [10] and in oduced
impo an changes. In 1993–1995 we s udied he ac iona ion o 10
me als in he ae osols o Se ille using he scheme o Ba celona [11]
(see Table I).
A i s objec i e o his p esen wo k is o op imize a specia ion scheme
o ine pa icles in a ci y in luenced mainly by a ic, bu also by ea h
c us al pa icles mixed wi h pa icles om indus ial sou ces [12]. The e is
a g ea di e ence be ween a ma ix o sedimen s and ha o ai bo ne
pa icles. Sedimen s con ain high con en s o o ganic ma e , i on and man-
ganese oxides and silica es. A mosphe ic pa icles gene ally also ha e im-
po an le els o o ganic ma e . Pa icles con ain o he oxidised me als as
geological ma e ial. Also, pa icles con ain elemen al ca bon, sulpha e,
ni a e, chlo ide and ammonium [13]. Fo his eason, some s ages o he
Tessie ’s scheme is modi ied o pa icles [7,14,15]. Bo h Tessie ’s modi ied
TABLE I Specia ion scheme om he Tessie ’s modi ied scheme acco ding o Obiols
e al
.
(1986)
Me allic ac ion Expe imen al condi ions
1
Soluble and exchangeable
25 mL o 1% NaCl
mechanical agi a ion du ing 60 min a oom empe a u e
2
Ca bona es, oxides
and educible
3
Bound o o ganic ma e ,
oxidisable and sulphidic
25 mL o 0.04 M NH
2
OHHCl in 25% HAcO
1 h a 95
◦
C, agi a ing occasionally
25 mL o 0.02 M HNO
3
+
10 mL 30% H
2
O
2
90 minu es a 85
◦
C
+
3 mL 30% H
2
O
2
1 hou a 85
◦
C
+
5 mL o 3.2 M NH
4
AcO in 20% HNO
3
con inuous agi a ion 30 min a oom empe a u e
4
Residual
5 mL o conc. HNO
3
+
2 mL o conc. HCl
+
20 mL H
2
O
90 min a 95
◦
C, agi a ing occasionally
scheme and BCR
®
we e es ed by us in 1996–1998, in o de o ind ou
which expe imen al condi ions we e he mos app op ia e o pa icles [16].
A second objec i e is o de e mine he chemical o ms ha can be
a ailable o he humans. The me al bioa ailabili y was al eady s udied
by us in a p e ious wo k unde he aspec o he physical specia ion [17].
In his wo k he smalles pa icles we e he mo e bioa ailable and he e
we e oxic me als ha end o accumula e in hese pa icles (Ni, Pb, Cd).
The e o e, he op imiza ion o he chemical specia ion should be ca ied
ou in he ine pa icles, ha is o say, below 1
m
m (ISO 7708, 1995) [18].
The specia ion scheme we e op imized in o de o de e mine he chemical
o ms in which he me als p esen in he pa icula e ma e can become o
he espi a o y ac [19]. Wi h his aim, he expe imen al condi ions o
he scheme we e designed in sea ch o simila condi ions in o he lung
and i is necessa y o simula e he physical and chemical condi ions ound
in he espi a o y ac ( edox po en ial, empe a u e and pH mainly).
The op imiza ion uses syn he ic samples [20–22] applied o a mosphe ic
pa icles [23,24]. The samples a e p epa ed in he labo a o y mixing
chemicals o eagen g ade [25,26] and he me allic concen a ions a e he e-
o e, known. Op imiza ion is done using he syn he ic samples sepa a ely
and, in o de o con i m he op imized esul s, he p ocedu e is epea ed
on eal samples o pa icles.
EXPERIMENTAL
P ocedu e o he Op imiza ion o Scheme
The p ocedu e consis ed o applying di e en expe imen al condi ions o
hese solid mix u es, modi ying in hese di e en es s pa ame e s such as
he ype o eagen , i s concen a ion, pH, empe a u e and ex ac ion
ime. The op imal condi ion o he pa ame e was selec ed o achie e
hese known me al concen a ions. All he pa ame e s we e op imized by
plo ing he g aphs o he di e en es s o each me al ep esen ing he
eco e ies wi h espec o he heo e ical concen a ion ( he 100%) and
hen e alua ing he op imal alue o all he me als as a whole.
When his p ocedu e is applied o eal samples, whose me allic concen a-
ion is unknown, he eco e y o 100% o each es is de e mined aking he
op imal alue o he pa ame e as he g ea e me al concen a ion o he ou
de e mina ions co esponding o he maximum o he cu e o he whole o
all he me als, ha is he maximum concen a ion o me al ex ac ed o he
mos o he 11 elemen s (al hough se e al me als a e ex ac ed be e a
o he alue o he pa ame e ). Then he co esponding pe cen ages o he
o he alues o he pa ame e o each es a e calcula ed and hen a e aged
o he ou de e mina ions. I is because o his ha he a e aged alues a e
no necessa ily 100.0% accu a e.
To op imize he 5 syn he ic samples co esponding o he ou ac ions,
13 es s we e needed. The las pa ame e op imized, he ex ac ion ime,
was always done in eal samples o ob aining an op imal ime mo e in
ag eemen wi h he eali y, and all he pa ame e s op imized in he hi d
ac ion was also done in eal samples, a e he wo i s ac ions we e
op imized and applied on he same samples p e iously. Each es consis ed
o modi ying he alue o he co esponding pa ame e by p epa ing 8
di e en solu ions (excep only 3 ypes o he esidual ac ion).
The e o e, a o al o 99 di e en solu ions we e measu ed by quad uplica e
and a o al o 99
×
11
×
4
¼
4356 di e en concen a ions o he 11 elemen s
we e de e mined. When hese op imal condi ions we e ob ained, hey
we e applied on he eal samples (excep he ime o ex ac ion and all he
pa ame e s o he hi d ac ion al eady done) and, consequen ly, hey
cons i u ed a o al o 7 es s
×
8
×
11
×
4
¼
2464 di e en concen a ions.
I hese esul s ag ee wi h hose ob ained wi h he syn he ic samples,
his means ha he design o he syn he ic samples has been alid o he
expe imen a ion and his ac gi es alidi y o he esul ing expe imen al
condi ions.
Addi ionally his la ge amoun o da a (4356
+
2464
¼
6820) assu e he
quali y o he op imiza ion.
When he ou ac ions ha e been op imized sepa a ely, all he op imized
condi ions we e es ed oge he on he syn he ic sample 5 (‘ o al’), and in
his way o ob ain he inal imp o ed scheme. Also, he ou expe imen al
condi ions o he ou ac ions we e es ed on he syn he ic samples no
co esponding o each ac ion, o ins ance, applying he condi ions
co esponding o he i s ac ion o he syn he ic samples co esponding
o he second, hi d and ou h ac ion, checking ha he me als ex ac ed
ep esen only a negligible pe cen age. In his way we can e i y o example
ha he me als ha should appea in he esidual ac ion dono appea in
p e ious ac ions.
Finally, he scheme was applied on 10 eal samples collec ed in 10
sampling si es o Se ille o co obo a ing he esul s wi h he imp o ed
scheme and also i was applied on he same samples wi h he p e ious
scheme o ou p e ious wo k [11] o s udying he di e ences and hei
signi icances.
P epa a ion o Syn he ic Samples
Fi e mix u es ‘solubles’ (chlo ides, sulpha es and ni a es) o op imize he
i s ac ion, ‘oxides’ and ‘ca bona es’ (ca bona es and hyd oxica bo-
na es) o op imize he second ac ion, ‘ esiduals’ (pu e me als) o op imize
he ou h ac ion and one wi h all he p e ious compounds oge he
(‘ o al’) we e p epa ed. Hyg oscopic subs ances we e d ied in a dessica o
o 48 h o a oid he he mal decomposi ion o he mel . Besides he
dessica o , ca e was aken in handling se e al eagen s in o de o a oid
oxicological p oblems and igo ous eac ions wi h he wa e o he mois u e,
and he e o e, a lamina lux cabine was also used he e. Fo he hi d
ac ion, he op imiza ion was ca ied ou di ec ly on samples o ai bo ne
pa icles, as i is explained below.
Syn he ic samples we e p epa ed as ollows: 1 g o each compound was
ca e ully and accu a ely weighed and pu in o polyp opylene lask wi h a
h ead s oppe . A e ha ing weighed all he compounds, he esul ing
mix u e was mul i-colou ed, due o he di e en colou s o each eagen .
The lasks o each mix u es we e, hen, d ied again in he dessica o o
ano he 48 h, o elimina e he possible inco po a ion o mois u e du ing
he p ocess o weighing.
Finally, he s oppe was pu back in he ube and his was hen mechani-
cally shaken o 48 h in he e ical ib a o , a e which he ini ial colou s
u ned o monoch ome and he pa icle size became ine and homogenous.
This homogenei y was co obo a ed by op ical mic oscopy and i was also
e i ied mic oscopically ha his ime o 48 h was su icien o ob aining
a size equal o he ine pa icles.
In o de o ob ain me al concen a ions om each syn he ic sample, hese
we e ex ac ed wi h acid diges ion on 30 mg o sample accu a ely weighed in
polyp opylene ubes wi h a mix u e o ni ic and hyd ochlo ic acids (3 : 1)
using a wa e -ba h a 95
◦
C o 90 min. This de e mina ion was pe o med
in quad uplica e. Then he ou 100 mL lasks we e measu ed by ICP-
AES. The esul s we e compa ed wi h he heo e ical alues and hey
ag eed (see Table II).
Suspended Pa icle Sampling
Fine ai bo ne pa icles we e collec ed in qua z il e s (20.3
×
25.4 cm
2
) om
WHATMAN (QM/A) wi h a sample (MCV, Model CAV-A/HF) equipped
wi h a cascade impac o (MCV, Model IC/CAV). Pa icles en e he
impac o a a low a e o 68 m
3
h
—
1
. The impac o e ec i ely sepa a es

TABLE II Reco e ies in pe cen ages wi h espec o he heo e ical alues o he 11 elemen s in each syn he ic sample
Syn he ic sample
Pe cen age o heo e ical alue(%)
Me al
Ca
Fe
Mg
Pb
Cu
Mn
V
Ti
Ni
Co
Cd
Mean
1 Solubles
81.4
±
12.3 98.2
±
10.6 90.3
±
3.5 105.5
±
13.2 99.6
±
4.9 100.3
±
8.3
91.1
±
11.1 97.6
±
17.6 103.1
±
8.4 103.4
±
4.0
106.3
±
13.3 97.9
±
35.2
2 Ca bona es
104.4 ± 6.0 –
95.7
±
9.1 105.2
±
15.5 101.1
±
11.5
94.0
±
5.2
– – 100.6 ± 15.5
94.9
±
8.0
91.5
±
3.0
98.4
±
28.8
3 Oxides
92.0
±
6.3
11.4
±
5.7
94.4
±
7.8 103.0
±
7.2 101.1
±
4.8
82.1
±
6.3
112.3
±
12.5 17.1
±
3.1
95.0
±
7.3
99.5
±
7.0
25.8
±
5.2
75.8
±
23.3
4 Pu e me als
96.0
±
3.2
91.4
±
4.8
99.8
±
3.1 99.4
±
3.2
95.0
±
2.4
97.1
±
7.4
84.2
±
4.3
2.5
±
2.0
96.9
±
5.7
89.7
±
18.0
4.8
±
2.5
77.9
±
22.3
5 To al
82.3
±
14.1
68.9
±
11.9
95.4
±
9.4 96.7
±
10.6 94.0
±
12.2
90.6
±
4.7
94.2
±
19.5 43.6
±
15.3
95.1
±
18.3
97.6
±
8.9
70.6
±
17.1
84.5
±
42.4
64
A.J.F.
ESPINOSA
e
al.
he pa icles in six s ages and he las il e collec s pa icles smalle han
0.61 mm [17].
13 samples o pa icles need ul o he op imiza ion we e collec ed in
Reina Me cedes, a ep esen a i e a ea o Se ille, Spain [12,27]. This sam-
pling s a ion ep esen s a ic and indus ial emissions. These samples
we e collec ed un il we ob ained he necessa y weigh o pa icles, wi h sam-
pling pe iods o 96 h collec ing abou 6500 m
3
o ai and 600 mg o pa icles.
The 10 samples o es ing he imp o ed and p e ious schemes we e
collec ed om ou ne wo k o sampling s a ions [11].
Reagen s and Appa a us
Ca e was aken in handling he samples in o de o a oid con amina ion
p oblems wi hin a e ical lamina ai low cabine wi h a HEPA il e
om INDELAB (Model IDL-48V). Wa e ba h was om JULABO
(Model SW-20C). Cen i uge was om SIGMA (Model 3-15). Reagen s
and s anda d solu ions o me als we e om MERCK. Ve ical ib a o
o mixing he chemical eagen s was om SELECTA (Vib oma ic Model
384). Milli-Q-g ade wa e was om WATERS-MILLIPORE (Model Plus).
The sample ex ac ions we e analysed o 11 elemen s (Mg, Ca, Ti, V, Mn,
Fe, Co, Ni, Cu, Cd and Pb) by induc i ely coupled plasma a omic emission
spec ome y (ICP-AES) using a Fisons-ARL 3410 ins umen . One
mini o ch consumes a gon gas a a a io equency powe o 650 W,
consequen ly, i is capable o consuming a ew millili es o sample a a
low o 2.3 mL min
—
1
. This ac allows ha he me als usually measu ed
by FAAS and GFAAS in 50-mL lasks in ou p e ious wo ks can also
now be measu ed by ICP-AES wi h only 15 mL.
Me hodology o Chemical Analysis
Op imiza ion es s we e ca ied ou in polyp opylene cen i ugal ubes
by adding 30 mg o each mix u e accu a ely weighed and subsequen ly
applying he co esponding expe imen al condi ions. 15 mL o each eagen
was added. Ex ac ions we e ca ied ou in a o a o a 50 pm and a
ambien empe a u e. Cen i uga ion was pe o med a 5000 pm o
10 min. Finally, he ube was le o decan by pou ing he liquid in o
100 mL lasks o measu ing by ICP-AES.
The ma ix e ec due o he pa icle and il e ma ix was s udied
h ough he s anda d addi ion echnique in he ou specia ion ac ions.
The s anda d addi ion e i ied ha his ma ix e ec was null. Rega ding
he in e e ence ma ix due o he eagen s, he calib a ion cu es ha e
been ob ained wi h he same ma ix as each one o he ou specia ion
ac ions.
A e sampling, each il e was d ied and weighed [17] and hen cu in o 8
eigh hs, each o which was hen placed in a polyp opylene cen i uge ube.
The es s we e ca ied ou on each eigh h wi h he same p ocedu e as ha
o syn he ic samples. The only di e ence was he ollows: he di e en
es s o he second ac ion and subsequen s we e applied o he esidue
o he p e ious ac ion al eady op imized. Fo hese samples, a se o
unexposed il e s was analysed using he same p ocedu e in each ac ion
o he scheme. The mean unexposed il e alue was sub ac ed om each
eal sample. Fo es ing he scheme o e he ‘ o al’ syn he ic sample
and o e he eal samples he p ocedu e was o apply he ou ac ions in
a con inuous sequence.
RESULTS AND DISCUSSION
Op imiza ion o he Fi s F ac ion o he Scheme:
Soluble and Exchangeable
In o de o op imize he ype o eagen we know ha di e en schemes use
wa e [15], magnesium chlo ide o sodium ace a e [5], sodium chlo ide [7],
ace ic acid [8], ba ium chlo ide [28]. Tessie ’s modi ied scheme uses
sodium chlo ide and BCR
®
scheme [29] uses ace ic acid. The e o e, bo h
sodium chlo ide and ace ic acid we e es ed.
Fo op imizing he concen a ion o sodium chlo ide, 8 solu ions we e
p epa ed om 0.0 (wa e ) o 3.5%. Fo ace ic acid he 8 solu ions p epa ed
we e om 0.01 o 0.15 M.
15 mL o solu ion was added o each polyp opylene ube con aining he
syn he ic sample 1. The ubes we e closed and o a ed a 50 pm o 5 h.
The sodium chlo ide ex ac s he 100% and eco e ies a e independen
om he concen a ion and wa e (NaCl 0.0%)is he selec ed concen a ion.
Some schemes ha e also ound wa e as he bes ex ac an o soluble me als
[15,30]. The wo es s we e epea ed on wo eal samples wi h he same
esul s (see Fig. 1a).
Fo op imizing he pH, we keep in mind ha SO
2
le el in Se ille is
low [31] and ou measu es made in ainwa e h ough he ime ha e an
a e age alue o pH 7.3. Mo eo e , pH o no mal lung is 7.40, which is
FIGURE 1 (a) Ex ac ion o soluble and exchangeable me als om a eal sample a ying he
concen a ion o sodium chlo ide and ace ic acid. (b) Ex ac ion o oxides, ca bona es and
educible me als om a eal sample a ying he concen a ion o hyd oxylamine chlo ide. (c)
Ex ac ion o oxides, ca bona es and educible me als om a eal sample a ying he pH o
hyd oxylamine chlo ide. (d) Ex ac ion o bound o o ganic ma e , educible and sulphidic
me als om a eal sample a ying he concen a ion o ammonium ace a e. (e) Ex ac ion o
bound o o ganic ma e , educible and sulphidic me als om a eal sample a ying he pH o
ammonium ace a e. ( ) Ex ac ion o esidual me als om a eal sample wi h he h ee acid
mix u es.
he physiological pH o he human body. I he pH o he lung mucosa
changes, e en below one uni , acid-based diso de s can be caused, as a
espi a o y acidosis. Then, we will use a pH 7.4 o he wa e .
In o de o op imize he ime, 15 mL o wa e a pH 7.4 was added o 8
ubes con aining an eigh h o a new eal sample. The ubes we e, hen
quickly pu in o he o a o and one ube was aken ou e e y 30 min. Then
he esul s (see Table III) indica e ha he e y bes ime o ex ac ion is 3 h.
Op imiza ion o he Second F ac ion o he Scheme: Oxides,
Ca bona es and Reducible
Fo selec ing he ype o eagen i is necessa y o keep in mind ha many
specia ion schemes commonly use hyd oxylamine chlo ide in on o