Uploaded by A Ramesh Kumar

Mechanism of volatile hydride generation and atomization

advertisement
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
2005 © The Japan Society for Analytical Chemistry
1401
Reviews
Mechanism of Volatile Hydride Formation and Their Atomization
in Hydride Generation Atomic Absorption Spectrometry
A. RAMESH KUMAR* and P. RIYAZUDDIN**†
*Chemical Laboratory, Central Ground Water Board, South Eastern Coastal Region, E1, Rajaji Bhavan,
Besant Nager, Chennai 600090, India
**Department of Analytical Chemistry, University of Madras, Guindy Campus, Chennai 600025, India
The mechanism of volatile hydride generation (HG) and the formation of analyte atoms in the quartz cell atomizer used in
the determination of hydride-forming elements (As, Bi, Ge, Pb, Sb, Sn, Te etc.) by atomic absorption spectrometry
(AAS), have been critically reviewed. The nascent hydrogen mechanism failed to explain hydride generation under
different experimental conditions when tetrahydroborate (THB), amineboreanes (AB) and cyanotrihydroborate (CBH)
were used as reductants. Various experimental evidence suggested a non-nascent hydrogen mechanism, in which the
transfer of hydrogen directly bonded to boron to an analyte takes place. In electrochemical hydride generation (EcHG),
the reduction of the analyte species and subsequent hydrogenation was proposed. The mechanism of analyte atom
formation in a quartz tube atomizer has been explained by the following hypotheses: thermal decomposition, oxidation by
O2 and collisions by hydrogen free radicals. The free-radical mechanism satisfactorily explains most of the analytical
implications. The significant variation in the experimental conditions required to generate different analyte hydrides
makes it difficult to arrive at a generalized mechanism of hydride formation.
(Received December 15, 2004; Accepted May 16, 2005)
1
2
3
4
Introduction
1401
Historical Development of Hydride Generation 1401
Generation of Volatile Hydrides
1402
Hypotheses Concerning the Hydride Generation
Process
1402
4·1 The nascent hydrogen mechanism
4·2 Formation of hydroboran intermediates
4·3 Role of pH in hydride generation
4·4 Studies carried out with deuterated reagents
1 Introduction
Volatile hydride generation (HG) combined with atomic
absorption spectrometric (AAS) determination is a sensitive
analytical method for the elements As, Bi, Ge, Pb, Sb, Sn and
Te.1 The basic design of a hydride generation system with
subsequent detection by atomic absorption may be classified as
four steps. First, generation of the hydride; second, collection
of the hydride (if necessary); third, transfer of the hydride to the
atomizer; and fourth, decomposition of the hydride to the gasphase elemental form within the optical axis of the detection
system. Various designs and techniques were developed during
the last three decades to achieve these goals, which led to better
sensitivity and lower detection limits. Because of the less
interference, better selectivity and sensitivity, HG-coupled with
AAS detection is the better choice for quantification of the
† To whom correspondence should be addressed.
E-mail: riyazdr@yahoo.co.uk
4·5 Electrochemical hydride generation
5 Effect of the Reaction Matrix and Thiols
5·1 Effect of halide ions
6 Atomization of Volatile Hydrides in the
Quartz Cell
6·1 Atomization via thermal decomposition
6·2 Oxidation of the hydride
6·3 Free radical mechanism
7 References
1405
1406
1408
hydride-forming elements.
Though the principle of this
technique seems to be simple, the generation of volatile
hydrides and their subsequent reduction to the atomic state, is
still not completely understood.2 Many studies have been
reported on the determination of As, Bi, Hg, Sb, Se etc. in
natural water samples,3–12 geologic materials,13,14 flora and
fauna,15–18 but, relatively few studies have been reported on the
process of hydride generation and their detection in a quartz cell
atomizer.
2 Historical Development of Hydride Generation
The conventional flame atomization of hydride-forming
elements results in poor sensitivity, since the resonance lines of
these elements lie in the far-ultraviolet region of the spectrum.
With the application of an Ar–H2 flame to AAS in 1967, a much
improved sensitivity was obtained. This flame absorbs about
15% of the available light, compared with 62% for air–acetylene
flame at the 193.7 nm line of arsenic.19
1402
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
Holak20 first determined As by converting to AsH3, collected
it in liquid-nitrogen trap, which was then warmed, and the evolved
AsH3 was swept into the flame with a flow of N2. This greatly
minimized the matrix interference. Madsan21 collected AsH3 in
dilute AgNO3 and aspirated the resulting solution directly into
the flame as described by Khan and Schallis.19 Dalton and
Malanoski22 were the first to report on the direct aspiration of
AsH3 into a flame, and attained a detection limit of 0.1 µg.
A further improvement in the sensitivity was obtained by
Fernandez and Manning,23 by collecting the AsH3 in a balloon
reservoir and subsequent introduction into the flame. Manning24
was the first to introduce an accessory commercially with AAS
for determining As. Pollock and West25 extended it to Sb and
Bi. In 1972 the use of sodium tetrahydroborate (THB) was
reported by Braman et al.,26 who employed an emission in a
D.C. arc-discharge method. In 1973, Schmidt and Royar27
reported on the use of THB for the determination of As and Se;
it was extended to Ge,28 Sn29 and Te29 by others. All of these
methods continued to use an Ar–H2 flame. In 1974, Thomson
and Thomerson30 reported the use of THB with a flame-heated
quartz tube as the atomizer for As and Pb. The use of THB
dramatically increased the popularity of HG as an analytical
tool, exclusively for hydride-forming elements; recently, its use
has been extended to the generation of volatile species of
Cd,31,32 In,33 Tl,34 Cu,35 Ni,36 Zn,37–39 Au,40 Ag,37,38 and Pt.41
Various application methodologies did appear in the literature
describing improved analytical performances.42–46 However,
unfortunately the mechanism of hydride production and its
atomization in a quartz cell atomiser is not fully understood, and
the studies attempted so far to explain these phenomena form
the basis of this review article.
space of an electrolytic cell; and the process is considered to
take place in at least two sequential events:49–51
3 Generation of Volatile Hydrides
Various reducing agents and sources of nascent hydrogen have
been suggested in order to convert the element to its hydride.
These include Zn/HCl, SnCl2/HCl–KI, Mg/HCl–TiCl3 and
THB.47 The reactions leading to the formation of hydrides with
metal/acid reductant can be written as:
Zn + 2H3O+ → Zn2+ + 2H2O + 2H·,
Am+ + (m + n)H· → AHn + mH+,
where Am+ is an analyte.
THB is a particularly versatile reagent, which has become
widely used for its reducing and hydride transfer properties.48
Assuming the evolution of nascent hydrogen from the acid
hydrolysis of THB, the generation of hydrides is represented as
m+
→ H3BO3 + NaCl + 8H· A
→ AHn + H2 ↑.
NaBH4 + 3H2O + HCl 
However, studies supporting the non-nascent hydrogen
hypothesis propose that the hydride is formed by the action of
hydrogen bonded directly to boron through the formation of
some hydroboron intermediates.
According to this, the
derivatization process may be written as
→ Intermediates 
→ H3BO3 + H2,
THB + H3O+ + 2H2O 
THB/Intermediates + Analyte 
→ Hydride.
Hydrides have also been generated electrochemically
(EcHG).49–55 In EcHG the hydride is formed in the cathodic
Am+ + me– → A0.
The formation
metal/metalloid,
of
the
hydride
from
the
deposited
A0 + nH· → AHn,
where H represent the hydrogen atom, from the adsorbed H
atom, reduced H3O+ or reduced H2O.
Recently, D’Ulivo et al.56 generated hydrides using amine
boranes (AB) of the type L-BH3 (where L = NH3, tert-BuNH2,
Me2NH-, Me3N-) and sodium cyanotrihydroborate (CBH)
(NaBH3CN) as derivatizing agents. Though a variety of
reductants were suggested, THB is the most popular reductant
used in analytical applications.48
4 Hypotheses Concerning the Hydride Generation
Process
According to Dedina,57 the mechanism of the reaction leading to
the formation of hydrides is a matter of debate, and the different
hypotheses of mechanisms can be classified into two categories.
The first is based on the assumption that hydrides are formed by
the action of “atomic hydrogen” or “nascent hydrogen”. The
second includes all those mechanisms proposed by various
workers who disagreed with the nascent hydrogen mechanism;
it is called the “non-nascent hydrogen mechanism.” Based on
the published literature up to 1995, Dedina et al.57 concluded
that there was no convincing evidence for either of the two
classes of mechanisms.
4·1 The nascent hydrogen mechanism
From the time that Marsh58 conducted a test for As proposed
in 1836 and subsequent stibine generation by Bloxam59 in 1861,
untill recently this mechanism has been proposed by several
authors. Robbins and Caruso60 in 1979 suggested that the acid
decomposition of THB produces nascent hydrogen. Recently,
Carrero et al.61 reported this mechanism to explain aciddependant hydride generation. Indirect support for the nascent
hydrogen mechanism could be derived from studies of Wang
and Barnes.62 They described a mathematical model for the pH
dependency of hydride generation. Their theoretical predictions
agreed well with the experimental results, and it was concluded
that the efficiency of HG could be improved by the in situ
generation of nascent hydrogen.
Laborda et al.63 strongly objected to the nascent hydrogen
mechanism, mainly concerning the formation of nascent
hydrogen and the reduction of metal/metalloid, as indicated by a
change of the oxidation state. Considering that the estimated
standard potential for the H+/H couple is E0(H+/H) = –2.016 V,
neither THB [E0(H3BO3)/BH4–) = –0.482 V], nor Zinc
[E0(Zn2+/Zn) = –0.763 V] would be able to perform the
reduction of protons to atomic hydrogen. He further pointed out
that hydrides could also be generated in basic solutions64–67 and
in an electrochemical process.49–55 In the case of EcHG, the
hydrogen involved in the generation process does not have any
special reducing property, because the electrons provided at a
sufficient negative potential to the electrode are a reducing
agent. In addition, there is a lack of agreement in assigning the
oxidation number to the analyte/hydride, and hence knowing
what is actually reduced, the analyte or the hydrogen, becomes
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
doubtful. According to Laborda et al.,63 the mechanism of HG
could be written as
Am+ + (m + n)/8 BH4– + 3(m + n)/8H2O →
AHn + (m + n)/H3BO3 + 7(m – n)/8H+,
where m represents the oxidation state of the analyte and n the
coordination number of the hydride. The reaction would be
analytically expressed as a hydrogenation reaction of the type
A
m+
+ nH + (m + n)e → AHn.
+
–
1403
Table 1 Hydrolysis rate constants for hydroboran species
formed during acid hydrolysis (0.1 – 1.1 H+) of THB in an 88%
methanol–12% water solution
Hydroboran
species
Khyda = –(d ln[hydroboron])/
dt
Temperature/
˚C
H-BH3 (THB)
H2O-BH3
K1 = Too fast to be measured
K2 = 1.5 × 10–3 s–1 + 1.6 × 10–3
[H+]l mol–1 s–1
K3 ≈ 0
K3 = 1.7 × 10–2 s–1
K4 = 3.3 × 10–4 s–1
–78
–78
(H2O)2-BH2+
H2O-BH2-OH
H2O-BH-(OH)2
–36
–36
The fact that hydrides could also be generated under basic
conditions was sometimes put forth as evidence for a nonnascent hydrogen mechanism, since it is not compatible with the
metal/acid system. However, Dedina et al.2 stated that, after the
analyte is reduced in alkaline media, acidification generates the
analyte hydride, which is still compatible with the nascent
hydrogen mechanism. Qiu et al.67 generated stannane by
mixing an acidic analyte soluition with alkaline THB and
observed stannane up to pH 12. Their conclusion is that
stannane generation proceeds via a non-nascent hydrogen
mechanism, and stannane that generation and neutralization are
synchronous reactions. This statement, according to D’Ulivo et
al.,68 has led to the uncertainty of the reaction rate, and to the
objection that it could be possible that nascent hydrogen is
formed during the neutralization reaction, which thus poses
some doubts about the fact that the findings of Qiu et al.67 can
be considered as experimental evidence of the non-nascent
hydrogen mechanism.
D’Ulivo et al.68 generated stannane, stibine and bimuthine
using THB and AB complexes in the pH range of 4.3 to 12.7.
Based on kinetic calculations, they concluded that the nascent
hydrogen mechanism failed to explain the generation of
hydrides. According to them, under these pH values, the
undecomposed borane complexes of AB are the only available
species for hydride formation, and hence the transfer of
hydrogen bonded to boron most probably forms the hydride.
where L could be one or more groups among H2O, OH– and Cl–
and n is charge, which could be –1, 0 or +1, depending on L.
Under the unusual acidity employed in HG, the rate of acid
hydrolysis of the hydroboran species appears to follow
K1>K2>K3>K4. The rate constants of hydroboran species formed
during the acid hydrolysis (0.1 – 1.1 H+) of THB in an 88%
methanol–12% water solution, as reported by Wang and Jolly,74
is given in Table 1.
According to them, the analyte elements differ in their
selectivity to react with different hydroboran intermediates. In
aqueous HCl media, Hg2+ and Sb3+ are very reactive and
unselective towards all intermediates, implying that all three are
equally reactive. Te4+ reacts preferentially with the hydroboran
species formed in strong acid media, but Sn4+ is completely
unreactive towards the same species. The analyte substrate of
As3+, Se4+ and Bi3+ reacts selectively with some hydroboran
species. Thus, the effectiveness of the derivatization processes
depends on both the nature of the substrate containing the
analyte and the nature of the hydroboran intermediate.
4·2 Formation of hydroboran intermediates
Narsito et al.69 calculated the rate of formation (Kf) of
hydrides as a function of the THB concentration. The results
showed that, for Se, the values found for Kf were independent of
the THB concentration up to 0.1% of THB; for As3+ and As5+, a
slight dependency was noticed; and for Sb the dependency was
considerable. Based on these findings, Narsito et al.69 proposed
that hydride formation is not necessarily caused by
decomposition of the THB reagent, but by some intermediate
formed from its decomposition. The presence of these types of
intermediates has been suggested by various workers.70,71
Studies of Krevoy and Hutchins72 supported the existence of
an intermediate of the type H2-BH3 during the hydrolysis of
THB. According to D’Ulivo,73 this intermediate and the
precursor (BH4–) are the same independent of the pH, and loss
of the first hydrogen of BH4– is the rate-determining step.
Studies of Wang and Jolly74 on the decomposition of THB in a
water–methanol solution indicated the existence of mono-, diand tri-hydroboran intermediates [H2O-BH(OH)2, (H2O)2-BH2+,
Though the formation of such a type of
H2O-BH3].
intermediates was stated during the early sixties,71,75 their role in
the hydride derivatisation reaction has been suggested recently.
D’Ulivo et al.76 in a recent study attempted to generate
volatile hydrides from these hydroboran intermediates.
According to them, the decomposition of THB in aqueous
media follows a similar pattern as that of in a water–methanol
4·3 Role of pH in hydride generation
The role played by the hydrogen ion concentration in
analytical HG is two fold. First, it activates the analyte
substrate to suitable chemical form; second, the hydrolysis of
THB. Wang and Barnes62 described the theoretical pH
dependence of hydride generation. A theoretical modeling and
experimental study of As5+ and As3+ indicate that both reactions
are pH dependent. However, As5+ reduction occurs in a lower
pH region, which is restricted mainly by the dissociation of the
borohydride. The reaction condition of As3+ is relatively broad,
as has been verified in many experiments.77,78 Only a little
dissociation of BH4– is necessary for As3+ hydride generation.57
Several authors studied the formation of arsine from As5+ and
As3+ under different pH and THB concentrations.79–85 The
determination of arsenic species, exploiting pH-dependant
arsine generation is of interest to study by many authors.86–95 It
was stressed by many authors that the analyte species must be
fully protonated in order to generate arsine. Carrero et al.61
supposed that the reaction mechanism could be modified under
different acid concentrations.
They suggested that the
protonation of analyte-thiolate complexes is necessary to permit
its reaction with THB. The importance of the pH in activating
the analyte precursor can be easily understood from the work of
Bell and Kelly96 concerning the acid-catalysed reduction of
nitrite to N2O by a weak reducing agent, Me3N–BH3 (KH+ = 1.2
× 10–4 l mol–1 S–1). They found that the rate of formation of N2O
solution.74 Hence, by analogy, the stepwise decomposition of
THB in a strongly acidic solution at room temperature could be
K→
K→
K→
K→
1
2
3
4
[BH3] or L-BH3n 
L2BH2n 
L3BHn 
B(OH)3,
BH4– 
1404
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
increased by ~109 times by decreasing the pH from 5.85 to 0.3.
It was concluded that the rate-determining step is a hydride
attack to protonated nitrous acid, H2ONO+, or free NO+; this
produces the nitrosyl hydride, HNO, which is rapidly converted
into N2O by
reaction with THB. The other, leading to the formation of
SbD3, resulted from its reaction with B2D6, which was formed
by a reaction of THB and H2SO4, followed by deuterium
exchange.
Pergnatis et al.,103 in their studies on arsine generation using
duterated reagents and mass spectroscopy, observed that the
reaction of As3+ and As5+ with NaBD4 in HCl and H2O produced
AsD3 as the main product. When the reactions were performed
in NaBH4, DCl and D2O, the main product was AsH3. If
nascent hydrogen was involved in the above processes, all of
the possible arsines randomly deuterated of type AsHnD3–n
should have been formed. The findings of Pergnatis et al.103
strongly support the non-nascent hydrogen mechanism;
particularly, the direct transfer of hydrogen atoms from boron to
analyte species takes place.
2HNO → H2O + N2O.
HONO + H+ ↔ H2ONO+ ↔ NO+ + H2O
D’Ulivo et al.73 stated that considering HNO2 is a weak
inorganic acid, a similar mechanism would also take place for
the weak inorganic oxoacids of these metalloids, which are
precursors in the analytical HG technique. In the case of Sb3+,
by increasing the pH, the substrate would evolve into the
following species:
Sb3+ → SbO+ → SbO(OH) → SbO2–.
Because SbO(OH) is insoluble, the antimonite SbO2– that is
formed in alkaline solution is not reactive towards a THB
attack. This indicates that the role of the pH in analytical HG
could also be the formation of suitable analyte precursors.
D’Ulivo et al.56 reported that suitable analyte precursors are
obtained by increasing the acidity for Sb3+, As3+, Se4+ and Te4+,
while the opposite behavior is observed for Bi3+. They also
showed that the change in the efficiency of hydride generation
was correlated to the variation in the structure of molecular
species containing the analyte. Unsuccessful hydride generation
in alkaline media mostly results from the loss of reactive species
of the analyte, rather than from an inefficient degree of
decomposition of the borane complex.68
4·4 Studies carried out with deuterated reagents
The inconsistency of nascent hydrogen formation upon the
decomposition of THB under acid medium has been
demonstrated by various studies. Among these, studies carried
out with deuterated reagents are important. The existence of the
intermediate H2-BH3 indicates72 that molecular hydrogen, and
not atomic hydrogen, is evolved during THB decomposition.
Mesmer and Jolly97 carried out experiments on THB
decomposition with deuterated reagents.
The acid
decomposition of KBD4 in 3 mol l–1 H2SO4 yielded HD as the
main product (>90%), with 3 – 5% of D2 and H2. The
hydrolysis of NaBH4 in D2O, resulted HD as the main product
which represent a fraction >90%98 and 95%99 of the total
hydrogen evolved. LiAlH4, the homologous compound of THB,
has been used to prepare high-purity HD (99%) by the
controlled hydrolysis of LiAlH4 with D2O at low temperature,100
indicating a similar mechanism as that described above. These
studies indicate that the formation of nascent hydrogen is
incompatible with the hydrolysis of THB, since it would have
produced D2 and H2 as the main product in the above
experiments.
Craig et al.101 successfully derivatized methyl mercury
(CH3Hg+) to its hydride, CH3HgH, at pH 4 with NaBH4 and to
its deutrated form, CH3HgD, by a reaction with NaBD4.
Freund102 prepared stibine by slowly adding an alkaline solution
of Sb3+ (0.5 mol l–1) and KBH4 (0.4 mol l–1) to a solution of
D2SO4 (2 mol l–1) prepared in D2O. The products were
continuously stripped by nitrogen, trapped and purified. The
main products identified were SbH3 (>93%) and SbD3 (∼7%),
even though the total exchangeable hydrogen was 97%
deuterium. The author proposed that there should be two
pathways. One leading to the formation of SbH3, involves a
4·5 Electrochemical hydride generation
Electrochemical hydride generation (EcHG) is an alternative
to wet chemical methods, and has been studied in detail during
the last 10 years.104–110 EcHG has recently been reviewed by
Denkhaus et al.111 In EcHG the generation of hydrides is a
three-step reaction on the cathode surface; deposition of the
hydride forming elements onto the surface of the cathode and
their reduction; formation of the hydrides and subsequent
desorption of the hydrides. The deposition is connected with
the reduction of the analyte, which depends on the nature of the
cathode material.
A detailed mechanistic electrolytic hydride generation was
summarized by Denkhaus et al.112 The generation of hydrides
follows two different mechanisms; electrocatalytic and
electrochemical mechanisms, depending on the hydrogen
overvoltage of the cathode material.113–116 In both mechanisms,
the hydride-forming element must be reduced first and
deposited on the cathode surface.
Me + Xy+ + ye– → Me-X (Me, cathode surface; X, analyte)
The electrocatalytic reaction mechanism is involved in the
production of hydrogen atoms adsorbed on the surface of the
cathodes with a low hydrogen overvoltage, whereby the
discharge of hydronium ions to adsorbed H· atoms is facilitated
and is characterized by the Volmer equation,117–119
Me + H3O+ + e– → Me-H + H2O
(Volmer reaction; Me, cathode surface).
The recombination of the adsorbed hydrogen atoms to form H2
follows the Tafel reaction,120
Me-H + Me-H → 2Me + H2 (Tafel reaction).
To produce elemental hydrides, the adsorbed hydrogen atoms
must recombine with the hydride-forming element adsorbed.
This recombination reaction occurs stepwise, followed by
desorption of the element hydride from the cathode surface:
Me-H + Me-X → Me + Me-XH,
→ (n + 1) Me + XHn+1
Me-XH + nMe-H 
→ nMe+ + Me-XHn+1 
(Me, cathode surface; X, hydride forming element).
The recombination to H2 competes with hydride formation.
In electrochemical mechanism, the first step is the discharge
of H3O+ ions on the cathode surface (Volmer reaction). The
second step is the reaction of the deposited hydrogen atoms with
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
1405
the hydronium ions of the catholyte solution and subsequent
electron transfer within the cathode surface. This reaction is
called the Heyrowski reaction:121–123
probably due to electron release from the sulfur group, by
analogy with alkoxy-borane and hydroxy boranes (RO-BH3–,
HO-BH3–).
On the other hand, the formation of analyte-cysteine complex
was suggested by many authors48,132,134–136 in which the –OH
group of protonated oxyacids of the hydride-forming element is
replaced by the (–SR) group, which subsequently reacts with the
BH4– and generates the hydrides:
Me + H3O+ + e– → Me-H + H2O
(Volmer reaction; Me, cathode surface),
Me-H + H3O+ → Me-H2+ + H2O,
Me-H2+ + e– → Me + H2
(Heyrowski reaction; Me, cathode surface).
2R-SH + RnAsO(OH)3–n → RS-SR + RnAs(OH)3–n + H2O,
RnAs(OH)3–n + (3–n)R-SH → RnAs(SR)3–n + 2H2O,
The formation of hydrides ensues from the surface of the
cathode to the deposited hydride-forming element by charge
transfer and a subsequent reaction with the hydronium ions of
the electrolyte solution.
It is possible that electron-charge-transfer proceeds
simultaneously from the cathode surface to the adsorbed
analyte, whereas the reaction of the adsorbed analyte with the
hydronium ions runs stepwise.112 The overall reaction may be
written as
Me-X + me– + mH3O+ → XHm + mH2O + Me.
Salzberg et al.50 proposed a similar mechanism for AsH3,
where water molecules are reduced to H2, which is chemisorbed
on elemental arsenic, forming a covalent hydride.
5 Effect of the Reaction Matrix and Thiols
In HGAAS it is found that reaction matrix, type of acid, buffers
and the addition of certain reagents enhances the analyte signal
to a considerable extent. Such a signal enhancement in the
HGAAS was first reported by Boampong et al.124 Brindle et
al.125 reported the use of L-cysteine in the determination of
arsenic in HG-DCP-AES as a signal enhancer. A similar effect
was also reported for Sn126 and Ge.127,128 According to Chen et
al.129 L-cysteine can speed up the rate of reduction by THB,
thereby increasing the sensitivity. Anderson et al.130,131 suggested
that the sulfur containing ligand was important in the reaction,
and that the same effect was produced by thioglycerol, thiourea
and thioacetic acid. It has also been reported that L-cysteine
reduces the transition metal interference and acts as acid-base
buffer, enhancing better stability of the hydrides.132
Brindle et al.125 showed that there is a compound formed
between NaBH4 and the SH group of L-cysteine, which may be
the reagent, which is a more potent reducing agent than NaBH4
alone.
BH4– + RSH → RS-BH3– + H2
The formation of the above type of complex was assumed
because it may be generated by the reduction of the disulfide:124
R-S-S-R + BH4– 
→ RSH + R-S-BH3– (R-CH2CH(NH2)COOH)
In another study, Brindle and Le133 found that in an aqueous
solution of L-cysteine, pencicillamine and thioglycerol gave
thiolotrihydroborate(III) anions.
The formation of
thiolotrihydroborate(III) was confirmed by 11B NMR at
thiol/THB in the range of 0.66 to 0.83 M. The authors proposed
that the thiol ligand competes with the solvent for forming the
complex with BH3. D’Ulivo73 suggested that the enhanced
reactivity of thiol-borane complex with respect to THB is
RnAs(SR)3–n + BH4– → RnAsH3–n + (3–n)BH3 + (3–n)RS–,
BH3 + 3H2O → H3BO3 + 3H2 ↑.
These authors commented that the THB-thiol reaction did not
explain the fact that the arsine yield depends on the prereaction
between the analyte species and cysteine, prior to the addition
of THB. Lee et al.137 reported that, though intermediate As-S
compounds are formed, the final product, does not, incorporate
the thiol group.
According to Howard et al.,134 the thus-formed complexes are
less sterically hindered with respect to the THB attack than the
parent attack. Jolly138 in his work on the preparation of hydrides
of As, Ge, Sb and Sn discussed the importance of accessibility
of the hydride-forming atom to a THB attack. However, Carrero
et al.61 suggested that the protonation of thiolate complexes is
necessary to permit the reaction with BH4–. An important
conclusion, which could be drawn from the studies reported by
various authors, is that suitable substrate precursors are essential
to facilitate a BH4– attack. This seems to be reasonable.
The two reaction pathways supported by various studies seem
to be acceptable, at least with the current status of knowledge.
The specific role played by the thiol-analyte and thiol-THB
interaction in the derivatization process of As needs to be
studied further.
However, the thiol modified reaction does not produce any
appreciable signal enhancement in the generation of stibine139
and bismuthine140 in a continuous-flow system. D’Ulivo73
suggested that the lack of a detectable effect on stibine and
bismuthine generation should not necessarily be ascribed to the
lack of an interaction between the analyte and the thiols. On the
other hand, Sb3+, Bi3+ present a more pronounced cationic
character compared to As3+, and it is likely that their analyte
substrates (SbO+, Sb(OH)2+, BiO+) already present a suitable
reactivity towards a hydride attack of an unmodified hydroboran
species. For As3+, the species As(OH)3 is probably not reactive
enough for a hydride attack, and thiols modifies the analyte
substrates reactivity towards THB.
In the case of stannane generation, L-cysteine increases the
signal only at higher acidity values.141,142 However, L-cysteine
suppresses the signal during SeH2 and TeH2 generation, and
must be considered in some cases as interfering species.143 This
may probably be due to the formation of a much less reactive
analyte substrate of the type RS-Se-SR and RS-Te-SR.
5·1 Effect of halide ions
Agterdenbos et al.144 reported that halide ions (Cl–, Br– and I–)
have a beneficial effect on the generation of SeH2 by THB in a
continuous-flow reaction system. The authors studied the effect
of HCl, HNO3 and H2SO4 on SeH2 generation, and found that
with HNO3 and H2SO4 the addition of Cl– and Br– to the reaction
system produced a signal equivalent to that generated by HCl.
1406
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
The authors called it as a “catalytic effect”. The addition of I– to
the reaction mixture reduced the signal, which was ascribed to
the preferential reduction of Se4+ to Se. Interestingly, if I– was
added to BH4–, it enhanced the signal more than that produced
by other halide ions. D’Ulivo et al.73 stated that the effect of I–
could be due to the formation of the iodide–THB interaction,
which produced hydroboran species that were more resistant to
acid hydrolysis. Another possible explanation stated by
D’Ulivo et al.76 is that the additive–analyte interaction could
lead to the formation of a halogenated species, such as SeOX2,
SeX4 or SeX62–, in which the analyte substrate is more reactive
towards THB than the simple H2SeO3 species.
combination of the enthalpy values used for the calculation of
Kd is correct.
However, there are some inconsistencies that cause doubt
concerning this mechanism, one of which is the temperature,
itself. Whereas in heated quartz tubes, typically temperatures of
around 800 – 900˚C are found to be optimum for the
atomization of arsine,159,160 the same hydrides are atomized at
1700 – 1800˚C in a graphite-tube furnace.161,162 The second fact
was that the addition of oxygen or air to the purge gas increased
the sensitivity,163,164 and the highest sensitivities were found for
extremely fuel-rich hydrogen–oxygen flames burning inside
unheated quartz tube atomizers,165,166 with an optimum H:O ratio
of 5:1; the third fact was a pronounced influence of the quartz
cell surface on the analyte signal.167–169
6 Atomization of Volatile Hydrides in the Quartz
Cell
The first reference to using a quartz tube atomiser, in this
instance electrically heated (EHQT) as an alternative to an
Ar–H2 flame, appeared in a report of Chau et al.145 in 1972. In
1974, Thompson and Thomerson30 published a report on the use
of THB and a flame-heated silica tube as the atomizer. Similar
externally heated quartz tube atomizers were reported by
various authors.146–149 An internal flame, or flame-in tube (FIT),
which utilized the hydrogen-oxygen, was introduced by Simer
and Hageman.150 It does not need to be heated. Of the several
quartz tube atomizers reported, the most popular one is the
flame heated quartz tube (FHQT), because of its excellent
sensitivity, low cost and easy operation.2
The question of the atomization mechanism of gaseous,
covalent hydrides in heated quartz tubes, as frequently used in
the HGAAS technique, is not fully understood. It has been
found that various instrumental and experimental parameters
have a profound influence on the atomization mechanism. In
addition, there are some inconsistencies that arise in the
atomization among different hydrides, e.g. arsine and selenium
hydride require different temperatures and carrier gas
composition for obtaining the maximum signal.
6·1 Atomization via thermal decomposition
Soon after introducing quartz tube atomizers, the general
opinion, however appeared to be that the hydrides are atomized
by thermal decomposition. Thomson and Thorsby,151 for
example, referred to an “electrothermal atomization” of arsine
in a heated quartz tube. This opinion was probably due to the
similarity of electrically heated quartz tube atomizers and
flame-heated quartz tubes reported by various workers.152–156
The increased analytical sensitivity with increasing temperature
up to a maximum,156–158 as reported by various workers, might
led to this conclusion as being the atomization mechanism. The
thermal decomposition of hydrides are generally written as
AHn(g) → A(g) + n/2H2(g).
(1)
Agterdenbos et al.158 made a thermodynamic calculation of
the equilibrium constant values, Kd, for the decomposition of
SeH2 in the temperature range of 600˚C to 2000˚C.
SeH2 → Se + H2 ↑
(2)
The theoretical calculations predicted the decomposition to
occur at 800˚C and by experiment it was found that maximum
atomization occurred at 700˚C. This fairly good agreement
suggests that Eq. (2) governs the decomposition of SeH2 in the
heated quartz tube. This conclusion is correct only if the
6·2 Oxidation of the hydride
Bax et al.169 calculated the equilibrium constant of Eq. (1) for
the atomization of AsH3 from the free energies of the various
possible analyte species. Their results showed that: the
decomposition occurs at a temperature of 1200˚C if only As and
AsH3 are considered and at a temperature of 1500˚C if
polymeric species are included in the calculation. Practically,
however, appreciable atomization already occurs at 700˚C,
which suggests that arsine is not atomized according to Eq. (1).
Narsito et al.168 supposed that the decomposition reaction of
the type in Eq. (1) could be reversed by the addition of excess
hydrogen; instead atomization was efficient with an increase of
extra hydrogen for AsH3. However, increasing the amount of
hydrogen in selenium hydride atomization decreased the signal,
which the authors supposed to be due to a reversal of the
reaction in Eq. (2). For this reason, Bax et al.170 concluded that
arsenic behaves differently from selenium. Bax et al.170
experimentally calculated the Kd values for the decomposition
of selenium hydride (Kd = 0.26 atm), and found fairly good
agreement with the Kd value calculated from thermodynamic
data (Kd = 0.18 atm) by Agterdenbos et al.158 However, the
experimental value of Kd remains constant for higher
temperatures against the expectations of higher values. This
discrepancy did not support Eq. (2) as the atomization process.
The authors169 assumed that oxygen in the carrier gas oxidizes
the hydrides to atoms, as given by
4AsH3 + 3O2 → 4As + 6H2O.
Narsito et al.69 also suggested that the atomization of arsenic
and other elements, like Antimony hydrides, follows a similar
mechanism, as proposed by Bax et al.:169
AHn(g) + n/4O2(g) → A(g) + n/2H2O.
(3)
Narsito et al.69 made similar thermodynamic calculations for
the decomposition of SbH3, and found that SbH3 dissociated at
400˚C if only Sb and SbH3 were considered, and at about 900˚C
if the dimerization reaction was also included.
The
experimental atomization occurs at about 500 – 600˚C, which
did not agree with the theoretical calculations, like arsenic.
Several authors69,168–171 confirmed an increase of the analytical
signal by the addition of extra hydrogen and oxygen, which led
to an inference that these gases play a role during the
atomization process. According to Narsito et al.,69 the lowest
temperature at which the hydrides are decomposed or oxidized
to atoms is about the same for the three elements, i.e. As, Sb
and Se (about 700˚C). They assumed that some common factor
initiates the atom formation. According to them, the role of
oxygen is to oxidize the analyte hydride into the atom, as per
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
1407
Eq. (3), and the excess of oxygen decreases the analyte signal,
probably due to the reaction of the type
may be reatomized, but only upon another contact with H·
radicals.
Welz and Melcher181 carried out in-depth studies on the
atomization mechanism of arsine in an electrically heated quartz
tube atomizer. They found that purging the sample with
nitrogen, or boiling the solution for 15 min before the reaction
with THB, caused a decrease in the analytical signal. This led
to the conclusion that the oxygen in dissolved air favored
atomization. In another study,181 the authors reported that the
introduction of pure arsine into the quartz tube produced no
signal at all. The signal was resumed if a blank reaction
mixture having THB and HCl was injected simultaneously.
This indicated the same degree of atomization for both cases.
From these experiments, the authors concluded that the
hydrogen evolved by the acid decomposition of THB was
essential in the formation of analyte atoms. This was further
confirmed by signal enhancement, if extra hydrogen was added
with the stripping gas. The authors also reported that the quartz
cell surface affects the analytical sensitivity considerably if not
conditioned i.e. a treatment with HF.
From these studies, Welz and Melcher181 concluded that: 1)
hydrogen is essential and plays an active role in the atomization
of gaseous hydride forming elements in quartz cell atomizers
heated at 1000˚C; 2) oxygen supports atomization in the
presence of hydrogen, at least at lower quartz cell temperatures;
3) in the presence of oxygen, the effective temperature of the
vapor phase in the heated quartz cell is not higher than in pure
argon; 4) the maximum sensitivity can be obtained only in the
cleaner environment of a quartz cell conditioned with HF; and
5) in the absence of H2 arsine is decomposed, but not atomized
in a quartz cell. Based on these observations, the authors
proposed that the atomization of volatile hydrides in a heated
quartz cell must be due to collisions with free hydrogen
radicals, as follows:
2As + 3O2 → 2As2O3.
The role of hydrogen is to prevent the analyte atoms from
forming the oxide; moreover, the role of both gases may be to
form radicals, which probably take part in arsine oxidation.
6·3 Free radical mechanism
The effect of oxygen on the atomization of As and Sb
hydrides in the heated quartz tubes was first reported by
Gouldden et al.164 They reported that an increase in the
sensitivity could be obtained if oxygen was added to the purge
gas. Similarly, Siemer et al.172 based on their studies on the
optimization of arsine generation in flame-in quartz tube
atomizers, found that the sensitivity could be increased by using
hydrogen for stripping the arsine, rather than using argon or
nitrogen. According to Siemer et al.,172 carrier gases like Ar and
N2 dilute the analyte atoms within the atomizer, but no reason
was given for the observed sensitivity increase.
Dedina et al.173 studied the atomization of SeH2 in a cool
hydrogen-oxygen flame burning in an unheated quartz tube, and
proposed that atomization is brought out by collisions of the
analyte hydride with the hydrogen free radicals. The hydrogen
generated from the acid decomposition of THB, or added to the
reaction mixtures, is the source of hydrogen atoms, and is
produced in the primary reaction zone of the diffusion flame
according to the following equations:
→ OH· + O,
H· + O2 ←
(4a)
O· + H2 ←
→ OH· + H·,
(4b)
→ H2O + H·.
OH· + H2 ←
(4c)
MHx + H → MHx–1 + H2,
McCrum174 has stated that the equilibrium constant of Eq. (4c)
is very high and the hydrogen concentration is much higher than
the water concentration. Thus atomization is via a two step
mechanism with the predominating H· radicals:
SeH2 + H· → SeH + H2
∆H = –189 kJ/mol,
(5a)
SeH + H· → Se + H2
∆H = –131 kJ/mol.
(5b)
A further reaction that should be taken into account is
recombination,
Se + H· → SeH
∆H = ±305 kJ/mol.
(5c)
By taking into account Eqs. (5a) – (5c), equilibrium is attained
after a sufficiently large number of collisions with H-radicals.
With increasing atomic weight (e.g. As-Sb-Bi), fewer free
radicals are required for atomization, due to the higher
efficiency of the larger and less stable hydride molecules.
Dedina et al.173 made extensive studies on the atomization
mechanism of AsH3, SeH2 in various types of quartz tube
atomizers, i.e. EHQT,175 flame-in-gas-shield (FIGS),176,177
miniature diffusion flame (MDF),178 and a multiple microflame
quartz tube atomizer (MMQT).179 He speculated that the radical
mechanism of atomization appears to be identical to all types of
quartz-tube atomizers,180 and it proceeds only in a small part of
the atomizer volume. After that, the free atoms disappear from
the observation volume by forced convection and by chemical
reactions (free atom population decay). Once reacted, species
MH + H → M + H2.
The hydrogen radicals were probably formed by the same
reactions with oxygen as proposed by Dedina et al.173 for the
fuel rich oxygen-hydrogen flame.
The proposed mechanism satisfactorily explains the
observations made by several authors, that the need of hydrogen
for atomization and oxygen enhances the sensitivity of the
analyte signal. The authors suggested that in the absence of
hydrogen, AsH3 is decomposed, but not atomized until a
temperature of 1700˚C is reached, which is the typical
temperature used in graphite-furnace atomizers.
The
temperatures of quartz-tube atomizers favor the decomposition
of arsine with the formation of the tetramer or the dimmer, as
follows:
→ 2As2 ←→ 4As.
As4 ←
The authors did not rule out the possibility of oxide formation in
the absence of hydrogen.
The conclusions made by Welz and Melcher181 were
supported by Akman et al.182 They studied the atom formation
mechanisms in GFAAS with the direct introduction of arsenic
and arsine. In the direct injection method at a furnace
temperature of about 1700˚C, it was found that arsenic is
vaporized into AsO and atomized by dissociation of the oxide.
However, in the direct injection of arsine, AsH3 is decomposed
on the graphite surface before the atomization temperature is
1408
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
reached, and the metallic arsenic is vaporized as As4, which is
then decomposed to As2 dimers, and finally atomized by gasphase dissociation.
However, there were some objections to the free-radical
mechanism. Bax et al.169 raised the question that the reaction
requires the presence of H· radicals in at least twice as high as
the pressure of SeH2 molecules (approximately 10–7 atm).
However, at temperatures of around 650 – 700˚C,
decomposition of SeH2 already occurs, and at this temperature
the equilibrium concentration for H, calculated from
thermodynamic data for the H2 ↔ 2H· equilibrium, is about 100
or 1000-times lower.
The same authors criticized the
conclusions of Welz and Melcher181 that hydrogen radicals are
required for the decomposition, since no absorbance signal was
found if the absorption cell was not conditioned. Bax et al.170
reported that the amount of H· radicals present is probably by
far not sufficient for these reactions. Agterdenbos et al.183
reported that OH· radicals may play a role in the decomposition
reaction, since H2 and O2 gases are needed for atomization.
However, Welz and Melcher181 stated that such a role was
improbable, since the concentrations of OH· radicals are much
lower than H· radicals. However, the authors183 assumed that
the catalyst acts at even very low concentrations. In contrast,
Welz and Melcher181 speculated that in the heated quartz cell the
number of H radicals should increase with increasing
temperature as well as increasing oxygen concentration, up to a
certain optimum. Moreover, because radical recombination is a
considerably slower process than their formation, a concentration
well above the equilibrium values can be expected in a heated
quartz cell, in the absence of any radical scavengers.
The hydrogen free-radical mechanism is strongly supported
by various studies, mainly interference studies made by
Dedina184 on the atomization of AsH3 and SeH2. According to
him, inside the quartz tube the interferent may reduce the
population of H· radicals by accelerating their decay, thereby
reducing the fraction of atomized analyte hydrides. These types
of interferences mostly depend on the design of the quartz cell
and the gas flow. In addition, the interferent may accelerate the
decay of the free analyte atoms in the quartz tube, which is
called self-interference by D’Ulivo and Dedina.185 This type of
self-interference or analyte decay interference probably decides
the fate of free atoms.
The mechanism proposed by Dedina et al.173 and Welz and
Melcher,181 satisfactorily explains the need of hydrogen and
oxygen and their role in the atomization process. According to
them, oxygen enhances the formation of H free radicals. Welz
and Melcher181 explained the temperature-dependant sensitivity
by stating that quartz has a strong catalytic effect at
temperatures of around 1000˚C and that hydrogen radicals are
formed by the decomposition of H2 molecules at the quartz
surface. The concentration of H· free radicals will be
maintained above the equilibrium concentration, since the
competing process i.e. radical recombination is comparatively a
slow process.
4. M. Eva Moreno and C. Perez-Conde, Anal. Bioanal.
Chem., 2003, 375, 666.
5. H. W. Sun, J. Ha, J. M. Sun, D. Q. Zhang, and L. L. Yang,
Anal. Bioanal. Chem., 2002, 374, 526.
6. I. Ipolyi and P. Fodor, Anal. Chim. Acta, 2000, 413, 13.
7. M. O. Andreae, Anal. Chem., 1977, 49(6), 820.
8. J. Morda-Pineiro, M. Luisa Cervera, and Miguel de La
Guardia, J. Anal. At. Spectrom., 1997, 12, 1377.
9. M. L. Kokot, At. Absorption Newsl., 1976, 15, 105.
10. W. M. Mok and C. M. Wai, Water Res., 1989, 23, 7.
11. W. M. Mok, J. A. Riley, and C. M. Wai, Water Res., 1998,
22, 769.
12. J. Y. Cabon and N. Cabon, Anal. Chim. Acta, 2000, 418,
19.
13. J. Guimont, M. Pichelte, and N. Rheaume, At. Absorption
Newsl., 1977, 16, 53.
14. A. Kuldvere, Analyst, 1989, 114, 125.
15. A. Takatsa and A. Uchiumi, Analyst, 1998, 123, 73.
16. M. Burguera and J. L. Burgurea, Talanta, 1997, 44, 1581.
17. Z. Stejkovvec, J. T. Van Elteren, and U. D. Woroniccka,
Anal. Chim. Acta, 2001, 443, 277.
18. M. E. Moreno, C. Perez-Conde, and C. Camera, J. Anal. At.
Spectrom., 2000, 15, 680.
19. H. L. Kahn and J. E. Schallis, At. Absorption Newsl., 1968,
7, 5.
20. W. Holak, Anal. Chem., 1969, 41, 1712.
21. R. E. Madsen, At. Absorption Newsl., 1971, 10, 57.
22. E. F. Dalton and A. J. Malanoski, At. Absorption Newsl.,
1971, 10, 92.
23. F. J. Fernandez and D. C. Manning, At. Absorption Newsl.,
1971, 10, 86.
24. D. C. Manning, At. Absorption Newsl., 1971, 10, 123.
25. E. N. Pollock and S. J. West, At. Absorption Newsl., 1971,
11, 104.
26. R. S. Braman, L. L. Justen, and C. C. Foreback, Anal.
Chem., 1972, 44, 2195.
27. F. J. Schmidt and J. L. Royer, Anal. Lett., 1973, 6, 17.
28. E. N. Pollock and S. J. West, At. Absorption Newsl., 1973,
12, 6.
29. F. J. Fernandez, At. Absorption Newsl., 1973, 12, 93.
30. K. C. Thompson and D. R. Thomerson, Analyst, 1974, 99,
595.
31. A. Sanz-Medal, Y. Valdes-Hevia, M. C. Temprano, N. B.
Gracia, and M. R. F. De La Campa, Anal. Chem., 1995, 67,
2216.
32. X. W. Guo and X. M. Guo, J. Anal. At. Spectrom., 1995,
10, 987.
33. J. S. Busheina and J. B. Headridge, Talanta, 1982, 29, 519.
34. D. Yan, Z. Yan, G. S. Cheng, and A. M. Li, Talanta, 1984,
31, 133.
35. L. Ebdon, P. Goodal, S. J. Hill, P. Stockwell, and K. C.
Thompson, J. Anal. At. Spectrom., 1995, 10, 317.
36. R. E. Sturgeon, J. Liu, V. J. Boyko, and V. T. Luong, Anal.
Chem., 1996, 68, 1883.
37. A. S. Luna, R. E. Sturgeon, and R. C. Campos, Anal.
Chem., 2000, 72, 3523.
38. X. M. Guo, B. Huvang, Z. Sun, R. Ke, Q. Wang, and Z. B.
Gong, Spectrochim. Acta, Part B, 2000, 55, 943.
39. T. Matousek, J. Dedina, and M. Vobecky, J. Anal. At.
Spectrom., 2002, 17, 52.
40. H. Sun, R. Suo, and Y. Lu, Anal. Chim. Acta, 2002, 457,
305.
41. Y. L. Feng, J. W. Lam, and R. E. Sturgeon, Analyst, 2001,
126, 1833.
42. J. Y. Cabon and N. Cabon, Fresenius J. Anal. Chem., 2000,
7 References
1. S. J. Hill, J. B. Dawson, W. J. Price, I. L. Shuttler, C. M.
M. Smith, and J. F. Tyson, J. Anal. At. Spectrom., 1999, 14,
1245.
2. J. Dedina and D. L. Tsalev, “Hydride Generation Atomic
Absorption Spectroscopy”, 1995, Wiley, New York.
3. N. M. M. Coelho, A. Cosmen da Silva, and C. Moraes da
Silva, Anal. Chim. Acta, 2002, 460, 227.
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
1409
368, 484.
43. H. Wen, S. J. Ha, J. M. Sun, and D. Q. Zhang, and L. L.
Yang, Fresenius J. Anal. Chem., 2002, 374, 526.
44. J. F. Chapman and L. S. Dale, Anal. Chim. Acta, 1979, 111,
137.
45. H. Narasaki and M. Ikeda, Anal. Chem., 1984, 56, 2059.
46. S. A. Amankwah and J. L. Fasching, Talanta, 1985, 32(2),
111.
47. A. E. Smith, Analyst, 1975, 100, 300.
48. A. G. Howard, J. Anal. At. Spectrom., 1997, 12, 267.
49. L. Tolimson, J. Electrochem. Soc., 1964, 111, 592.
50. H. W. Salzberg and B. Goldschmidt, J. Electrochem. Soc.,
1960, 107, 348.
51. A. M. Espinosa, M. L. Tascon, M. D. Varyquez, and P. S.
Batanero, Electrochim. Acta, 1992, 37, 1165.
52. D. M. Heuber and J. D. Winefordner, Anal. Chim. Acta,
1995, 316, 129.
53. F. Laborda, E. Bolea, and J. R. Castillo, J. Anal. At.
Spectrom., 2000, 15, 103.
54. W. W. Ding and R. E. Sturgeon, Spectrochim. Acta, Part B,
1996, 51, 1325.
55. W. W. Ding and R. E. Sturgeon, J. Anal. At. Spectrom.,
1996, 11, 225.
56. A. D’Ulivo, V. Loreti, M. Onor, E. Pitzalis, and R.
Zamboni, Anal. Chem., 2003, 75, 2591.
57. J. Dedina, “Flow Methods in Gas-liquid Separations” in A.
Sanz-Medal (ed.), “Flow Analysis with Atomic
Spectrometric Detection”, 1989, Elsevier, Amsterdam, 237
– 273.
58. J. Marsh, New Edinburgh Philos. J., 1836, 229.
59. C. L. Bloxam, Q. J. Chem. Soc., 1861, 13, 12.
60. W. B. Robbins and J. A. Caruso, Anal. Chem., 1979, 51,
889A.
61. P. Carrero, A. Malave, J. Luis, M. Buguera, and C.
Rondon, Anal. Chim. Acta, 2001, 438, 195.
62. X. R. Wang and R. M. Barnes, Spectrochim. Acta, Part B,
1986, 41, 967.
63. F. Laborda, E. Bolea, M. T. Baranguan, and J. R. Castillo,
Spectrochim. Acta, Part B, 2002, 57, 797.
64. T. Wickstrom, W. Lund, and R. Bye, Analyst, 1995, 120,
2695.
65. T. Wickstrom, W. Lund, and R. Bye, J. Anal. At.
Spectrom., 1998, 10, 389.
66. R. Bye, J. Autom. Chem., 1989, 11, 156.
67. D. R. Qiu, C. Vandecasteelie, K. Vermeiren, and R. Dams,
Spectrochim. Acta, Part B, 1990, 57, 439.
68. A. D’Ulivo, C. Baiocci, E. Pitzalis, M. Onor, and R.
Zamboni, Spectrochim. Acta, Part B, 2004, 59, 471.
69. Narsito, J. Agterdenbos, and S. J. Santosa, Anal. Chim.
Acta, 1990, 237, 189.
70. R. E. Mesmer and W. L. Jolly, Inorg. Chem., 1962, 1, 608.
71. R. E. Davis, E. Bromels, and C. L. Kibby, J. Am. Chem.
Soc., 1962, 84, 885.
72. M. M. Kreevoy and J. E. C. Hutchins, J. Am. Chem. Soc.,
1972, 94, 6371.
73. A. D’Ulivo, Spectrochim. Acta, Part B, 2004, 59, 793.
74. F. T. Wang and W. L. Jolly, Inorg. Chem., 1972, 11, 1933.
75. J. A. Gardiner and J. W. Collat, J. Am. Chem. Soc., 1965,
87, 1692.
76. A. D’Ulivo, M. Onor, and E. Pitzalis, Anal. Chem., 2004,
76(21), 6342.
77. M. H. Arabab-Zaver, Analyst, 1980, 105, 744.
78. A. V. Shaikh and D. E. Taliman, Anal. Chim. Acta, 1978,
98, 251.
79. P. D. Wentzell, N. G. Sundin, and C. Hogeboon, Analyst,
1994, 119, 1403.
80. T. R. Rude and H. Puchelt, Fresenius Z. J. Anal. Chem.,
1994, 350, 44.
81. S. Nielsen and E. H. Hansen, Anal. Chim. Acta, 1997, 343,
5.
82. Y. Yano, T. Miyama, A. Ito, and T. Yasuda, Anal. Sci.,
2000, 16, 939.
83. J. Aggett and A. C. Aspell, Analyst, 1976, 101, 341.
84. R. S. Braman, D. L. Johnson, C. C. Foreback, J. M.
Ammins, and J. L. Bricher, Anal. Chem., 1977, 49, 621.
85. M. O. Andreae, Anal. Chem., 1977, 49, 820.
86. A. V. Shaikh and D. E. Tallman, Anal. Chim. Acta, 1978,
98, 251.
87. C. Felsman, Anal. Chem., 1979, 51, 664.
88. M. Verlinden, J. Barrt, and H. Deelstra, Talanta, 1980, 27,
633.
89. A. G. Howard and M. H. Arab-Zavar, Analyst, 1981, 106,
213.
90. T. A. Hinners, Analyst, 1980, 105, 751.
91. M. Yamamoto, K. Urata, K. Murashige, and Y. Yamamoto,
Spectrochim. Acta, Part B, 1981, 36, 671.
92. S. Nakashima, Analyst, 1979, 104, 172.
93. Y. Yamamoto and T. Kumamaru, Fresenius Z. J. Anal.
Chem., 1976, 282, 355.
94. P. W. Vijan and G. D. Wood, Analyst, 1976, 101, 966.
95. H. D. Fleming and R. G. Ide, Anal. Chim. Acta, 1976, 83,
67.
96. K. E. Bell and H. C. Kelly, Inorg. Chem., 1996, 35, 7225.
97. R. E. Mesmer and W. L. Jolly, J. Am. Chem. Soc., 1962,
84, 2039.
98. T. Freund, J. Inorg. Nucl. Chem., 1959, 9, 246.
99. W. L. Jolly and E. Mesmer, J. Am. Chem. Soc., 1961, 83,
4470.
100.I. Wender, R. A. Friedel, and M. Orchin, J. Am. Chem.
Soc., 1949, 71, 1140.
101.P. J. Craig, D. Mennie, N. Ostah, O. F. X. Donard, and F.
Martin, Analyst, 1992, 117, 823.
102.T. Freund, J. Am. Chem. Soc., 1991, 83, 2279.
103.S. A. Pergantis, W. Winnik, E. M. Heitmar, and W. R.
Cullen, Talanta, 1997, 44, 1941.
104.X. O. Wang, Z. X. Zhuang, P. Y. Yang, and B. L. Huang,
Microchem. J., 1995, 316, 88.
105.W. W. Ding and R. E. Sturgeon, Spectrochim. Acta, Part B,
1996, 51, 1325.
106.W. W. Ding and R. E. Sturgeon, J. Anal. At. Spectrom.,
1996, 11, 225.
107.W. W. Ding and R. E. Sturgeon, J. Anal. At. Spectrom.,
1996, 11, 421.
108.D. Schaumloffel and B. Neidhart, Fresenius Z. J. Anal.
Chem., 1996, 354, 866.
109.E. Bolea, F. Laborda, M. A. Bellara, and Y. R. Castillo,
Spectrochim. Acta, Part B, 2001, 56, 2347.
110.J. Sima, P. Rychlovsky, and J. Dedina, Spectrochim. Acta,
Part B, 2004, 59, 125.
111.E. Denkhaus, A. Golloch, X-M. Guo, and B. Huang, J.
Anal. At. Spectrom., 2001, 16, 870.
112.E. Denkhaus, F. Beck, P. Bueschler, R. Gerhard, and A.
Gollach, Fresenius J. Anal. Chem., 2001, 370, 735.
113.F. Beck, Chem. Ing. Tech., 1976, 48, 1096.
114.W. M. Lestie and J. A. V. Butler, Trans. Faraday. Soc.,
1936, 32, 989.
115.S. V. N. Szabo, Z. Phys. Chem., 1938, 181, 367.
116.K. F. Bonhoeffer, Z. Phys. Chem., 1924, 113, 199.
117.M. Volmer and H. Wick, Z. Phys. Chem., 1932, 172, 429.
118.J. Tafel, Z. Phys. Chem., 1905, 50, 641.
1410
ANALYTICAL SCIENCES DECEMBER 2005, VOL. 21
119.J. Tafel and K. Naumann, Z. Phys. Chem., 1905, 50, 713.
120.P. Rutschi and P. Delahay, J. Chem. Phys., 1955, 23, 195.
121.I. J. Heyrowski, Rec. Trav. Chim., 1927, 46, 582.
122.J. Horiuti and M. Ikusima, Proc. Imp. Acad., 1939, 15, 39.
123.G. Okamoto, J. Horiuti, and K. Hirota, Sci. Papers Inst.
Chem. Res., 1936, 29, 223.
124.C. Boampong, I. D. Brindle, X.-C. Le, L. Pidwerbesky, and
C. M. C. Ponzoni, Anal. Chem., 1988, 60, 1185.
125.I. D. Brindle and X.-C. Le, Anal. Chem., 1989, 61, 1175.
126.I. D. Brindle, X.-C. Le, and X.-F. Li, Analyst, 1988, 113,
1377.
127.I. D. Brindle and X.-C. Le, J. Anal. At. Spectrom., 1989, 4,
227.
128.I. D. Brindle and X.-C. Le, Anal. Chim. Acta, 1990, 229,
239.
129.H. Chen, I. D. Brindle, and X.-C. Le, Anal. Chem., 1992,
64, 667.
130.R. Anderson, M. Thompson, and E. Culbard, Analyst, 1986,
111, 1143.
131.R. Anderson, M. Thompson, and E. Culbard, Analyst, 1986,
111, 1153.
132.D. L. Tsalev, A. D’Ulivo, L. Lampugnai, M. Di Marco, and
R. Zamboni, J. Anal. At. Spectrom., 1996, 11, 989.
133.I. D. Brindle and X.-C. Le, Anal. Chim. Acta, 1990, 229,
239.
134.A. G. Howard and C. Salou, Anal. Chim. Acta, 1996, 333,
89.
135.B. Welz and M. Sucmanova, Analyst, 1993, 118, 1425.
136.A. Shraim, B. Chiswell, and H. Olszoay, Talanta, 1999, 50,
1109.
137.X. C. Le, W. R. Cullen, and K. J. Reimer, Anal. Chim.
Acta, 1994, 285, 277.
138.W. L. Jolly, J. Am. Chem. Soc., 1961, 83, 335.
139.A. D’Ulivo, L. Lampugnai, G. Pelligrini, and R. Zamboni,
J. Anal. At. Spectrom., 1995, 10, 969.
140.A. D’Ulivo, E. Bramanti, L. Lampugnai, and R. Zamboni,
Spectrochim. Acta, Part B, 2001, 56, 1893.
141.H. W. Chen, W. Yao, D. X. Wu, and I. D. Brindle,
Spectrochim. Acta, Part B, 1996, 51, 1829.
142.A. D’Ulivo, E. Bramanti, L. Lampugnai, and R. Zamboni,
Spectrochim. Acta, Part B, 2001, 56, 1893.
143.K. Marcucci, R. Zamboni, and A. D’Ulivo, Spectrochim.
Acta, Part B, 2001, 56, 393.
144.J. Agterdenbos, J. T. M. Van Elteren, D. Bax, and J. P. Ter
Heege, Spectrochim. Acta, Part B, 1986, 41, 303.
145.R. C. Chau, G. P. Barron, and P. A. W. Baumgarner, Anal.
Chem., 1972, 44, 1476.
146.D. L. Collett, D. E. Fleming, and G. A. Taylor, Analyst,
1978, 103, 1074.
147.D. E. Fleming and G. A. Taylor, Analyst, 1978, 103, 101.
148.J. W. Robinson, R. Gracia, G. Hindman, and P. Selvin,
Anal. Chim. Acta, 1974, 69, 203.
149.W. H. Evans, F. J. Jackson, and D. Dellar, Analyst, 1979,
104, 16.
150.D. D. Siemer and L. Hagemann, Anal. Lett., 1975, 8, 323.
151.A. J. Thomson and P. A. Thoresby, Analyst, 1977, 102, 9.
152.M. Verlinden, H. Deelstra, and E. Adriaenssens, Talanta,
1981, 28, 637.
153.D. Booke Hatfield, Anal. Chem., 1987, 59, 1887.
154.F. D. Pierce, T. C. Lamoreaux, H. R. Brown, and R. S.
Frazer, Appl. Spectrosc., 1976, 30, 38.
155.F. D. Pierce and H. R. Brown, Anal. Chem., 1977, 49,
1417.
156.R. D. Wauchope, At. Absorption Newsl., 1976, 15, 64.
157.B. Welz and M. Melcher, Anal. Chim. Acta, 1981, 46, 17.
158.J. Agterdenbos, J. P. M. Van Noort, F. F. Peters, D. Bax,
and J. P. Ter Heege, Spectrochim. Acta, Part B, 1985,
40(3), 501.
159.M. Fishman and R. Spencer, Anal. Chem., 1977, 49, 1599.
160.B. Welz and M. Melcher, Spectrochim. Acta, Part B, 1981,
36, 439.
161.W. A. Maher, Anal. Chim. Acta, 1981, 126, 157.
162.M. M. McDaniel, A. D. Shendrikar, K. D. Reiszner, and P.
W. West, Anal. Chem., 1976, 48, 2240.
163.P. N. Vijan and G. R. Wood, Talanta, 1976, 23, 89.
164.P. D. Goulden and P. Brooksbank, Anal. Chem., 1974, 46,
1431.
165.J. Dedina and I. Rubeska, Spectrochim. Acta, Part B, 1980,
35, 119.
166.A. Mayer, C. Hofer, G. Tolg, S. Raptis, and G. Knapp,
Fresenius Z. J. Anal. Chem., 1979, 296, 337.
167.D. C. Reamer, C. Veillon, and P. T. Tokousbalides, Anal.
Chem., 1981, 53, 245.
168.Narsito and J. Agterdenbos, Anal. Chim. Acta, 1987, 197,
315.
169.D. Bax, J. T. VanElteren, and J. Agterdenbos, Spectrochim.
Acta, Part B, 1986, 41, 1007.
170.D. Bax, F. F. Peters, J. P. M. Van Noort, and J.
Agterdenbos, Spectrochim. Acta, Part B, 1986, 41, 275.
171.B. Welz and M. Melcher, Anal. Chim. Acta, 1981, 46, 17.
172.D. D. Siemer, P. Koteel, and V. Jariwala, Anal. Chem.,
1976, 48, 836.
173.J. Dedina and I. Rubeska, Spectrochim. Acta, Part B, 1985,
35, 119.
174.M. McCrum, “Principles of Atomic Absorption
Spectrometry”, 1999, GBC Scientific Equipment, Australia,
44 – 45.
175.J. Dedina and B. Welz, J. Anal. At. Spectrom., 1992, 7, 307.
176.J. Dedina and A. D’Ulivo, Spectrochim. Acta, Part B, 1997,
52, 1737.
177.A. D’Ulivo and J. Dedina, Spectrochim. Acta, Part B, 1996,
51, 481.
178.J. Dedina, A. D’Ulivo, L. Lampugnani, T. Matousek, and
R. Zamboni, Spectrochim. Acta, Part B, 1998, 53, 1777.
179.T. Matousek, J. Dedina, and A. Selecka, Spectrochim. Acta,
Part B, 2002, 57, 451.
180.T. Matousek, J. Dedina, M. Johansson, and W. Frech,
Spectrochim. Acta, Part B, 2000, 55, 151.
181.B. Welz and M. Melcher, Analyst, 1983, 108, 213.
182.S. Akman, O. Genc, and T. Balkis, Spectrochim. Acta, Part
B, 1982, 37(10), 903.
183.J. Agterdenbos, J. P. Van Noort, F. F. Peters, and D. Bax,
Spectrochim. Acta, Part B, 1986, 41, 283.
184.J. Dedina, Anal. Chem., 1982, 54, 2097.
185.A. D’Ulivo and J. Dedina, Spectrochim. Acta, Part B, 2002,
57, 2069.
Download